Cambridge History Byzantine Empire.pdf - PDFCOFFEE.COM (2024)

THE CAMBRIDGE HISTORY OF THE B Y Z A N T I N E E M P I R E c. 5 0 0 – 1 4 9 2

Byzantium lasted a thousand years, ruled to the end by self-styled ‘emperors of the Romans’. It underwent kaleidoscopic territorial and structural changes, yet recovered repeatedly from disaster: even after the near-impregnable Constantinople fell in 1204, variant forms of the empire reconstituted themselves. The Cambridge History of the Byzantine Empire tells the story, tracing political and military events, religious controversies and economic change. It offers clear, authoritative chapters on the main events and periods, with more detailed chapters on outlying regions and neighbouring societies and powers of Byzantium. With aids such as maps, a glossary, an alternative place-name table and references to English translations of sources, it will be valuable as an introduction. However, it also offers stimulating new approaches and important findings, making it essential reading for postgraduates and for specialists. jo nat h an sh e pa rd was for many years a Lecturer in History at the University of Cambridge, and was a Fellow of Selwyn College and of Peterhouse. He is the coeditor (with Simon Franklin) of Byzantine Diplomacy (1992), co-author (also with Simon Franklin) of The Emergence of Rus, 750–1200 (1996), author of Nespokoini s’sedi: b’lgaro-vizantiiska konfrontatsiia, obmen i s’zhitelstvo prez srednite vekove [Uneasy Neighbours: Bulgaro-Byzantine Confrontation, Exchange and Co-existence in the Middle Ages] (2007) and editor of The Expansion of Orthodox Europe: Byzantium, the Balkans and Russia (2007). Shepard is Doctor Honoris Causa of St Kliment Ohrid University in Sofia.

THE CAMBRIDGE H I S TO RY O F T H E B Y Z A NT I N E E M P I R E c. 5 0 0 – 1 4 9 2 Edited by

JO NAT H A N S H E PA R D

c a m b r i d g e u n i ve r s i t y p re s s Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, S˜ao Paulo, Delhi Cambridge University Press The Edinburgh Building, Cambridge cb2 8ru, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521832311 C Cambridge University Press 2008

This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2008 Printed in the United Kingdom at the University Press, Cambridge A catalogue record for this publication is available from the British Library Library of Congress Cataloguing in Publication data The Cambridge history of the Byzantine Empire c. 500–1492 / edited by Jonathan Shepard. p. cm. Includes bibliographical references and index. isbn 978-0-521-83231-1 1. Byzantine Empire – History – 527–1081. 2. Byzantine Empire – History – 1081–1453. I. Shepard, Jonathan. II. Title. df571.c34 2008 949.5 02 – dc22 2008038886 isbn 978-0-521-83231-1 hardback

Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

CONTENTS

List of maps List of illustrations List of tables Preface

page ix xi xiv xvii

G E N E R A L I NT RO D U C T IO N jonathan shepard Formerly Lecturer in History, University of Cambridge i Approaching Byzantium

2

ii Periodisation and the contents of this book

21

iii Other routes to Byzantium

53

iv Smoothing the way and short-cuts to Byzantium: texts in translation

76

P A R T I : T H E E A R L I E R E M P I R E c. 5 0 0 – c. 7 0 0 1 Justinian and his legacy (500–600) an drew louth, Professor of Patristic and Byzantine Studies, University of Durham

99

2 Eastern neighbours 2a Persia and the Sasanian monarchy (224–651) zeev rubin, Professor of Ancient History, Tel-Aviv University

130

2b Armenia (400–600) r. w. thomson, Formerly Calouste Gulbenkian Professor of Armenian Studies, University of Oxford

156

v

vi

contents

2c The Arabs to the time of the Prophet l awrence i. conrad, Professor of the History and Culture of the Middle East, Asia–Africa Institute, University of Hamburg

173

3 Western approaches (500–600) john m oorhead, McCaughey Professor of History, University of Queensland

196

4 Byzantium transforming (600–700) an drew louth

221

P A R T I I : T H E M I D D L E E M P I R E c. 7 0 0 – 1 2 0 4 5 State of emergency (700–850) m arie-france auz e´ py, Emeritus Professor of History, University of Paris VIII

251

6 After iconoclasm (850–886) shaun tougher, Senior Lecturer in Ancient History, Cardiff School of History and Archaeology, Cardiff University

292

7 Religious missions sergey a. ivanov, Professor of Byzantine Literature, Moscow State University

305

8 Armenian neighbours (600–1045) t. w. g reenwood , Lecturer in Mediaeval History, University of St Andrews

333

9 Confronting Islam: emperors versus caliphs (641–c. 850) walter e. kaegi, Professor of History, University of Chicago

365

10 Western approaches (700–900) m ichael mc cormick, Francis Goelet Professor of Medieval History, Harvard University

395

11 Byzantine Italy (680–876) thom as s. brown, Reader in History, University of Edinburgh

433

12 The middle Byzantine economy (600–1204) m ark whit tow, Lecturer in Modern History, University of Oxford

465

13 Equilibrium to expansion (886–1025) jon athan shepard

493

contents

vii

14 Western approaches (900–1025) jon athan shepard

537

15 Byzantium and southern Italy (876–1000) g . a. loud, Professor of Medieval Italian History, University of Leeds

560

16 Belle e´poque or crisis? (1025–1118) m ichael angold , Professor Emeritus of Byzantine History, University of Edinburgh

583

17 The empire of the Komnenoi (1118–1204) paul m agd alino, Bishop Wardlaw Professor of Byzantine History, University of St Andrews

627

18 Balkan borderlands (1018–1204) paul stephenson, Reader in Medieval History, University of Durham

664

19 Raiders and neighbours: the Turks (1040–1304) d. a. korobeinikov, Wolfson College, University of Oxford

692

P A R T I I I : T H E B Y Z A NT I N E L A N D S I N T H E L AT E R MIDDLE AGES 1204–1492 20 After the Fourth Crusade 20a The Greek rump states and the recovery of Byzantium m ichael angold 20b The Latin empire of Constantinople and the Frankish states dav id jacoby, Emeritus Professor of History, The Hebrew University of Jerusalem

731

759

21 Balkan powers: Albania, Serbia and Bulgaria (1200–1300) al ain ducellier, Emeritus Professor of History, University of Toulouse – Le Mirail

779

22 The Palaiologoi and the world around them (1261–1400) an geliki e. l aiou, Dumbarton Oaks Professor of Byzantine History, Harvard University

803

23 Latins in the Aegean and the Balkans (1300–1400) m ichel bal ard, Emeritus Professor of History, University of Paris I – Panth´eon-Sorbonne

834

viii

contents

24 The Roman orthodox world (1393–1492) an thony bryer, Emeritus Professor of History, University of Birmingham Glossary (including some proper names) Genealogical tables and lists of rulers List of alternative place names Bibliography Notes on use Abbreviations Primary sources Manuscripts Secondary works General and frequently cited works Part I (c. 500–c. 700) Part II (c. 700–1204) Part III (1204–1492) Unpublished theses Picture acknowledgements Index

852

881 906 930 936 936 938 946 983 984 990 1017 1084 1117 1119 1124

MA P S

1 2 3a, b 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Physical geography of the Byzantine world page 12 Middle Byzantine ‘hot-spots’ 35 Archaic, and other less familiar, names 91 An empire of cities: Byzantium c. 500 100 East Rome into west: expansion under Justinian 110 Constantinople in the earlier Byzantine period 113 Sasanian Persia 131 The Armenian lands in the earlier Byzantine period 158 Pre-Islamic Arabia and its northern neighbours 181 Lands of the empire in the west in the sixth century 197 Byzantium transforming: the empire towards the end of the seventh century 222 The empire in the eighth and ninth centuries 252 The empire under militarised rule: army units and embryonic themes, earlier eighth century 262 Administrative organisation: themes in the later ninth century 263 Byzantine religious missions 306 Armenia 591–850 334 Armenia and imperial expansion 850–1045 350 The expansion of Islam 632–850 366 Byzantium versus Islam: the zone of direct confrontation 371 Western neighbours c. 700–c. 1025 396 Italy 700–900 434 The Byzantine economic world 466 The empire in the tenth and eleventh centuries 494 Administrative organisation of the empire at the end of Basil II’s reign c. 1025 534 Southern Italy in the tenth century 561 Cross-currents: Byzantine building- and decorative works in the later tenth and eleventh centuries, and the course of the First Crusade 594 ix

x

list of maps 27 Constantinople in the middle Byzantine period: new building-work, major repairs and embellishments 28 The empire and its neighbours in the twelfth century 29 The Balkans: physical geography and regions 30 The imperial mesh in the Balkans, eleventh and twelfth centuries 31 Central Asia, the Abbasid caliphate and the emergence of the Seljuqs 32 The coming of the Turks: Asia Minor c. 1040–c. 1100 33 Byzantine Asia Minor, the Seljuqs of Rum and other Turks 34 The coming of the Mongols 35 Asia Minor c. 1265 36 The Greek-speaking rump states 1204–1261 37 The Latin empire of Constantinople and the Frankish states 38 Albania 39 Theodore Angelos’ seizure of Thrace 1225 40 Spanning the Egnatian Way 41 Klokotnitsa and Ivan II Asen’s counter-offensive 1230 42 Manfred of Hohenstaufen’s acquisitions c. 1257, and subsequent dowry settlement by Michael II Angelos of Epiros 43 Serb expansionism in the late thirteenth century 44 The empire reconstituted and lost: Byzantium (a) in the late thirteenth century, and (b) in the 1330s 45 The fall of western Asia Minor to the Turks in the first half of the fourteenth century 46 Towns and trade in the fourteenth century 47 Under pressure from the Serbs, then Turks: Byzantium in the mid- to late fourteenth century 48 Greeks and Latins in the Aegean and Balkans 49 Ties material and spiritual: the Roman orthodox and the Romance-speaking worlds 50 Mount Athos and Mara 51 The Pontos in the fifteenth century 52 Names of peoples, archaic or less familiar

596 635 666 667 695 700 712 720 725 732 760 780 786 789 790

795 802 807 808 819 828 836 848 873 877 882

ILLUSTRATIONS

1 2 3 4 5 6 7a 7b 8a 8b 9a 9b 10 11a, b 12 13 14 15 16 17 18a, b 19 20 21 22 23 24 25a, b 26 27 28

The Holy Mountain, Athos page 46 The walls of Constantinople 57 The interior of St Sophia, Constantinople 112 Pedestal from the Hippodrome of Constantinople 121 Shah Peroz being invested with two diadems 138 Inscription from the church of St Gregory at Aruch 162 Plan of fort at Timgad (Thamugadi) in North Africa 204 Justinianic fort at Timgad 205 Mosaic of Justinian from San Vitale, Ravenna 210 Mosaic of Theodora from San Vitale, Ravenna 211 The imperial fleet burns the ships of Thomas the Slav 234 Twenty-first-century Greek fire 234 Nomisma of Justinian II 236 Gold coins of Constantine VI and Irene 275 Zodiac from Ptolemy’s ‘Handy Tables’ 280 Manuscript of the Uspensky Gospels 281 Theodore the Stoudite, Leo V, Patriarch Nikephoros and iconoclasts 290 Michael III racing near the church of St Mamas 295 Basil I being crowned by the Angel Gabriel 303 Fresco from Tokale Kilise church in Cappadocia 317 Reliquary pectoral crosses from Hungary 323 Early example of Glagolitic script 330 Early example of Cyrillic script 331 View from within the walls of Ani 335 The palatine church of Aruch 343 The palatine church of Aght‘amar 356 The cathedral church of Ani 363 Dinars of ‘Abd al-Malik 383 The Dome of the Rock, Jerusalem 384 Mosaic of Theodore the Stoudite from the monastery of Nea Moni, Chios 403 Silver coin of Michael I 418 xi

xii

list of illustrations

29 Greek translation of Gregory the Great’s Dialogues 30 Fresco of a Seraph’s head from Santa Maria Antiqua, Rome 31a The Byzantine hilltop fortress of Sardis 31b Marble column drums reused for the walls at Sardis 32 Nomismata of Romanos I Lekapenos 33 Ivory of Constantine VII being crowned and blessed by Christ 34 King Abgar of Edessa receiving the mandylion from Christ 35 Byzantine cavalry pursuing fleeing Muslims 36 The head of Nikephoros II Phokas being shown outside the palace 37 Basil II in parade armour 38 Negotiations between Bardas Skleros, the caliphate and Basil II 39 Ivory of Romanos II and Bertha-Eudocia being crowned by Christ 40 Ivory of Otto II of Germany and Theophano being crowned by Christ 41 View of Amalfi’s hinterland 42 Miniature of Asaph teaching from the book of the law 43a, b The monastery of Nea Moni, Chios: general view and mosaic 44 Nomisma (histamenon) of Isaac I Komnenos 45 A Byzantine view of the world (miniature) 46 Alexios I Komnenos receiving the Panoplia dogmatike 47a, b Alexios I Komnenos’ reformed coinage: a nomisma and a low-value tetarteron noummion 48 Monastery of Christ Pantokrator, Constantinople 49 Manuel I Komnenos and his second wife Maria of Antioch 50 Seal of Leo, imperial spatharokandidatos 51 Peter Deljan being proclaimed ruler by his fellow Slavs 52a, b The Armenian Gospel Book of 1066, Sebasteia: title page and miniature of St Mark writing his gospel 53 Fifteenth-century picture-map of Constantinople and the Bosporus 54 Illustration from the Tomiˇc Psalter 55 Illustration from the Radomir Psalter 56 Thirteenth-century ‘teratological’ letter design 57 Wall-painting of Michael VIII and his family, church of Mother of God, Apollonia

428 439 471 472 507 514 515 518 521 523 524 543 551 578 584 593 604 606 614 620 634 643 669 671 693 753 797 798 799 800

list of illustrations 58 The Deesis mosaic in St Sophia, Constantinople 59 Mosaic of the Grand Logothete Theodore Metochites, Chora monastery, Constantinople 60 The shape of the future: a square-rigged cog 61 The basileus under western eyes: portrait medal of John VIII Palaiologos by Pisanello 62 Bull of union, Florence, 1439 63 Engraving of Mistra in the seventeenth century 64 The fall of Constantinople, 1453, by Panayotis Zographos 65 Grant by Mehmed II to the Genoese of Galata, 1453 66 Portrait of Sultan Mehmed II 67 Europe under Ottoman eyes: copy of Ptolemy’s map of Europe

xiii 826 826 847 854 855 861 866 867 870 876

TA B L E S

g enealogical tables and lists of rulers Many of the dates and some of the names included in these tables are debatable, and these lists are therefore somewhat provisional. 1 Byzantine emperors in Constantinople (c. 500–1204 and 1261–1453) page 906 2 Rulers of the Greek ‘rump states’ (1204–1461) i Nicaea 908 ii Greek rulers in the western provinces (Epiros and Thessaloniki) 908 iii Emperors of Trebizond and Grand Komnenoi 908 3 Patriarchs of Constantinople (381–1502) 909 4 Popes of Rome (c. 450–c. 1500) 912 5 Eastern rulers i Sasanian Persia 916 ii Umayyad caliphate; Abbasid caliphate 917 iii Armenian princes: the principal Bagratuni and Artsruni lines 918 iv Turks: the Seljuq dynasty 919 v Turks: the Seljuq sultanate of Rum 920 vi Turks: the Ottoman beylik and sultanate (c. 1282–1566) 920 vii Mongols: Genghis Khan and his descendants 921 6 Western rulers i Frankish emperors/senior co-emperors 921 ii Western emperors of Saxon origin, and their successors 922 iii Lombard princes of Capua-Benevento and Salerno 923 iv Norman rulers of southern Italy 925 v Latin emperors in Constantinople 926 vi Rulers of the Frankish principality of Achaia 926

xiv

list of tables 7 Balkan rulers i Bulgarian rulers ii Serbian rulers (over the core area of Raˇska) iii Hungarian rulers list of alternative pl ace names

xv 927 928 929 930

To Nicola

PREFACE

This is a short preface for quite a lengthy book, but it is a means of paying tribute to those principally involved in the development, shaping and production of The Cambridge history of the Byzantine empire (or CHBE). Like the empire itself, the process of formation has been protracted, without a clear-cut starting-point, and such sense of direction as has been attained owes more to collaborative effort than it does to untrammelled autocracy. Given the sizable number of persons contributing in one way or another, the preface’s brevity entails a mere sketch of those without whose help and advice CHBE would have been a far more onerous and lengthy task. It was Bill Davies who originally encouraged me to take on remodelling materials already available, and several anonymous readers helped structure the volume. Michael Sharp took over from Bill at Cambridge University Press and he has been an extremely patient and supportive editor, ably assisted at various times by Liz Davey, Sinead Moloney, Liz Noden and Annette Youngman. Particular thanks should go to the following key players: Bernard Dod, our indefatigable and eagle-eyed copy-editor, whose attention to detail and wise counsel averted many a mishap; to Barbara Hird, our expert indexer, whose care and clarity have created a valuable additional pathway to Byzantium; to Patricia Jeskins, our assiduous proofreader; and to David Cox, our cartographer, whose splendid maps are closely integrated with the text of our chapters. For bibliographic help I have to thank the following colleagues, who have supplied references and answered tiresome queries with speed and good grace: Jean-Claude Cheynet, Florin Curta, Peter Frankopan, Judith Gilliland, Michael Gr¨unbart, Paul Herrup, James Howard-Johnston, Elizabeth Jeffreys, Lester Little, Margaret Mullett, Angel Nikolov, Paolo Odorico, Maureen Perrie, G¨unter Prinzing, Charlotte Rouech´e, Maciej Salamon, Alexios Savvides, Teresa Shawcross, John Smedley, Tsvetelin Stepanov, Alice-Mary Talbot, George Tcheishvili, Ida Toth, Vladimir Vavˇr´ınek and Mark Whittow. I should also like to thank the staff at the Bodleian, Taylorian Slavonic, Sackler, Oriental Institute and the other Oxford libraries, as well as the staff of the University Library in Cambridge. xvii

xviii

preface

Colleagues who clarified various points along the thousand-year trek, or who freely provided access to unpublished materials of value for this work include Jane Baun, Jeffrey Featherstone, Paul Fouracre, John Haldon, Rosemary Morris, Pananos Sophoulis and Monica White. Particular thanks are due to Catherine Holmes, Mike Maas and Andrew Roach, who read the introduction and some of the chapters that follow, and who warned of culs-de-sac and quicksands to be charted or – hopefully – avoided. On the technical side, help with translation and transliteration was given by Lawrence Conrad, Jeffrey Featherstone, Tim Greenwood, Mona Hamami and Marina Kuji´c. Jenny Perry saved me on several occasions when Macs failed to talk to PCs, and vice versa, while Nigel James of the Bodleian initiated me into the mysteries of digital map-making. Locating and sourcing illustrations was made easier through the assistance of Nancy Alderson, Michel Balard, Theodore van Lint, Cyril Mango, Nicholas Mayhew, Dorothy McCarthy, Denys Pringle, Michael Stone and Robert Thomson. Particular thanks go to our neighbours, Vanessa and Peter Winchester, to whom I am indebted for several pictures of Constantinople. These thanks should be accompanied by apologies for a certain lack of sociability in recent years – and extended to all remaining friends. It is a commonplace to thank one’s immediate family for their help and endurance in these endeavours. However, I must single out my wife, Nicola, who took on the role of editorial assistant on the project without, I think, appreciating the sheer scale of activity involved. As I have often pointed out to her, this could be seen as due penance for failing to attend my lectures on Byzantium and its neighbours all those years ago in Cambridge! Without Nicola, the volume would probably not have been published this decade, and I am profoundly grateful for her patience, counsel and support. However, those most indispensable are the volume’s contributors. The chapters whose first incarnation was in The Cambridge ancient history or The new Cambridge medieval history have been joined by important new contributions expanding and elaborating on relevant themes. But it goes without saying that, notwithstanding all the help and advice received along the way, I take responsibility for such mistakes or errors as may have crept into the finished work. notes on using this volume Our approach to transliteration may induce unease among some colleagues – and invite charges of inconsistency – but we have tried to make proper names and technical terms accessible to the English-speaking world wherever possible. Greek has been transliterated and bars have been

preface

xix

used to distinguish ¯eta from epsilon and o¯mega from omicron in the case of individual words and technical terms, but abandoned for proper names. Greek forms of proper names have generally been adopted in Parts II and III – Komnenos instead of the Latinised Comnenus, for example – in contrast to Part I, set in late antiquity, when Latinised names seem appropriate. In general, we have adopted a ‘b’ and not ‘v’ when transliterating the Greek letter b¯eta. However, where a name is more or less domiciled in English usage, we have let it be, e.g. Monemvasia and not Monembasia. Where the names of places are probably so familiar to most readers in their Latinised forms that the use of a Greek form might distract, the Latinised form has been retained in Parts II and III – Nicaea instead of Nikaia, for example. Familiar English forms have been preferred out of the same consideration – Athens not Athenai, for example – and in Part III, when the empire’s possessions were being taken over by speakers of other tongues, the place names now prevalent have generally been preferred – Ankara instead of Ankyra, for example. Arabic diacritics have been discarded in proper names, with only the ayn (’) and hamza (‘) retained in the form shown, on the assumption that the diacritics will not help non-Arabic readers and may actually distract from name recognition and recall; however, full diacritics have been retained for individual words and technical terms. We have tried to be consistent yet accessible in transliterating other key scripts, such as Armenian and Cyrillic, using for the latter a modified version of the Library of Congress system. Detailed notes on how to use the bibliography can be found below at pp. 936–8. Chronological sectioning for the secondary bibliography is – like the periodisation of history itself into mutually exclusive compartments – rather arbitrary. The bibliography of secondary works should therefore be treated as a whole and the reader failing to find a work in one section should try the others. The Glossary and Tables are not intended to be comprehensive guides. The Glossary offers a selection of the technical terms, foreign words and names of peoples and institutions appearing in CHBE. But wherever possible, these are explained in the context of a chapter and only the more problematic proper names have a Glossary entry (see also Maps 3 and 52). Likewise, the lists of rulers and genealogies have been kept to a minimum, since they are available in more specialised works. The list of alternative place names is intended to help the reader locate some towns and regions which were known under radically different names by diverse occupants or neighbours, and to offer modern equivalents where known. The maps are designed to reconcile accessibility for anglophone readers with a sense of the form prevalent during the chronological

xx

preface

section of CHBE in question, not wholly compatible goals. The maps are intended to be viewed as an ensemble, and readers unable to spot a place in a map positioned in one chapter should look to adjoining chapters, or (aided by the list of alternative place names and the index) shop around.

G E N E R A L I NT RO D U C T IO N

I N T R O D U C T I O N – PA RT i

APPROACHING BYZANTIUM jon at han s he pard

Many roads lead to Byzantium, ‘the New Rome’, and guidance comes from dozens of disciplines, including art history and archaeology, theology and expertise in stone inscriptions, coins or handwriting. Indeed, those general historians who act as guides have themselves often majored in other fields, such as ancient Greece and Rome, the medieval west, the Slav or Mediterranean worlds, and even the Italian renaissance. The surest fact about the elusive ‘New Rome’ is that it lasted over a thousand years, albeit with a fifty-seven-year dislocation from 1204. Across this millennium, the questions of how, why and where the empire survived, receded and (most importantly) revived as a more or less functioning organism – and as an idea – underlie this book. We take a narrower road than the one chosen by this volume’s predecessor, The Cambridge medieval history IV,1 whose first part recounted political, military and ecclesiastical history in detail from 717 until the end of the empire, and devoted several authoritative chapters to neighbouring peoples and powers; its second part contained thematic chapters, on for example law, government, the church, music, the visual arts and literature. No such comprehensive treatment of Byzantium’s culture will be attempted here. Our chapters follow the fortunes of the empire, as shifting politicomilitary organisation and as abiding ideal and state of mind, but do not attempt portrayal of Byzantium and its civilisation from every angle; however, some important alternative approaches to its history are sketched in the third section of this introduction (see below, pp. 53–75). Our narrative picks out those occurrences salient to the political organism, with an eye for the many problems, external and internal, facing the upholders of imperial order from their capital in the New Rome. Unfashionable weight is given to individual emperors’ characters, and to the statecraft of such giants as Justinian (527–65), Leo III (717–41), Basil I (867–86) and Basil II (976–1025), Alexios I Komnenos (1081–1118) and Manuel I Komnenos (1143–80). Their diverse, often successful, solutions to problems 1

Hussey (ed.) (1966–7).

2

i. approaching byz ant ium

3

of governance are outlined, and a recurring theme is the pragmatism of Byzantium’s rulers in coping with plague, financial straits and the inroads of ‘barbarians’, and also with unexpected problems of success. The dynamics of these improvisations, abrupt overhauls and longer-term shifts are traced through the course of events rather than through detailed analysis of institutions as such, a justifiable approach given that the precise workings of so many of Byzantium’s institutions – from the army to provincial administration – are so hard to determine and highly controversial. Topics of relevance to Byzantine political culture are brought into the narrative, from religious devotions to patronage of the visual arts, and the broader, provincial society revolving around that of the metropolis is outlined. Thematic chapters look at the economy and Christian missions, and there is treatment of several societies, elites and powers that had long-term dealings with Byzantium. Here, too, coverage is less than comprehensive: for example, no chapter is dedicated to ties between the empire and the lands of the Rus. But enough is provided to demonstrate the impact of Byzantium on various cultures of world significance: the world of Islam, the Eurasian and the Slav worlds, and the Christian west. The aim is to outline and analyse interaction rather than to recount every known detail of relations with a particular state. The importance of Byzantium to neighbouring or newly forming societies and powers emerges more clearly when their individual situations and needs are taken into account. This is particularly true of the tortuous interrelationship with the Christian west across the centuries, and the vitality of the exchanges, cultural as well as ecclesiastical and political, between ‘Latins’ and ‘Greeks’ is brought out in full here. The chronological range of our chapters spans from just after the formal termination of the western half of the Roman empire (476) to the fifteenth century, when the Christian west was viewed by some Byzantines as a potential saviour from the Turks. This broad yet careful sweep takes in the numerous communities and towns of Greek-speakers who came under new rulers after the empire’s collapse in 1204, sometimes Venetians or Frenchspeakers, sometimes Bulgarian or Serbs. The ebb and flow of the imperial dominions in the thirteenth and fourteenth centuries is presented in more detail than is usual with this kind of survey, and it shows up qualities of the Byzantine body politic too easily overlooked: its ‘variable geometry’, a capacity to function quite effectively even without the use of apparently vital members; and resilience, its constituent parts realigning themselves with imperial dominion more or less of their own accord, without much prompting from the top. The conspectus offered here, at once authoritative and unusually wideranging, should yield some fresh insights to specialists in, and postgraduate

4

general introd uction

students of, the Byzantine world. But it also has something to offer newcomers to the enigma variations of Byzantium. No prior knowledge of the subject, or indeed of pre-modern history, is presupposed, and every effort has been made to provide guidelines for readers whose mother tongue or first foreign language is English. Translations of primary texts are cited in the footnotes where available, and a guide to sources in English translation is offered in the fourth section of this introduction (see below, pp. 76–90). Our introduction is divided into four sections, The first – this one – looks at Byzantine notions of empire, their tenacity in the face of adversity and the significance of religious rites for believers at grass-roots, constituting Byzantium’s special blend of faith and power. It concludes with a discussion of the nature of the interrelationships between outsiders and insiders, and of their bearing on the broader question of the Byzantine identity. The second section addresses the book’s time-frame and considers possible alternatives. It is followed by a survey of the book’s three main parts, which run from c. 500 to c. 700, c. 700 to 1204 and 1204 to 1492. Themes running through chapters that may, at first sight, seem rather disparate are picked out, part by part. The chapters are not surveyed in strict order of their sequence in the book: thus the topic- or region-specific chapters of Part II are considered en bloc, after the chapters forming the main narrative spine. Part III’s contents, lacking a single fixed point, and encompassing a wide variety of populations and polities, receive fairly lengthy treatment without close adherence to the order of the chapters. The third section outlines other possible approaches to those taken in this book, which mostly follow the course of recorded events of political, ecclesiastical or military significance for the empire. The outline draws attention to some more or less recent introductions to art, institutions and the human condition among the Byzantines. It is nonetheless slanted towards topics germane to the idea or substance of empire, whether political imagery, size of armies, or castration. The fourth and final section of the introduction addresses some of the problems of approaching Byzantium without benefit of Greek and offers short-cuts that may help towards the study – and teaching – of the empire’s story: historical atlases covering Byzantium and neighbouring peoples, chronologies, art-historical lexicons and whole dictionaries devoted to the subject. Far more works penned by the Byzantines or about the Byzantines by contemporary outsiders are available in English translation than is generally realised and further translations are underway. These make aspects of Byzantium readily accessible to newcomers from the English-speaking world, and this section of the introduction points to some of the online guides to English-language translations now available.

i. approaching byz ant ium

5

n otions of empire, resilience and religion The phenomenon of Byzantium has multiple connotations and even the name which its rulers used of their polity, ‘Roman’, was controversial.2 ‘Greeks’ was the name by which they and their subjects were known to many of their neighbours. This was a reflection of the language in everyday use in Constantinople and provincial towns and in which most imperial business was done from the sixth century onwards. To Goths fanning Italians’ prejudices, ‘Greeks’ carried intimations of frippery and rapaciousness (see below, pp. 214–15). Yet a certain readiness to accept the empire’s claim to be ‘Roman’ surfaces spasmodically among Frankish courtiers, for all their fulminations to the contrary (see below, p. 397). And while some Arabic writers in the Abbasid era stressed the Byzantines’ cultural inferiority to the ancient Greeks or Romans,3 Rum (‘Romans’) was the name by which Muslims called the Byzantines, and the Turkish potentates who made themselves masters of south-central Anatolia from the late eleventh century became known as sultans of Rum.4 The very terms Rome and Roman had overtones of unimpeachably legitimate sovereign authority, evoking the greatest empire the world had yet seen. Fantastic as popular notions might be concerning the imagery of classical monuments in Constantinople,5 Byzantine rulers still acted out triumphal parades through its streets and enlisted the citizens’ support in staging them, manifesting the classical Roman concept of ‘eternal victory’.6 Less flamboyantly, the City’s water-supply kept flowing through an intricate network of pipes and cisterns established in the sixth century, to standards set by Roman engineers. The workings of this system, ensuring the pure water vital to Constantinople’s survival, were seldom if ever set down in writing,7 and in fact the importance of this state secret features in a late thirteenth-century treatise on Byzantine political thought.8 In contrast to mundane matters of pipelines, the supernatural protection enjoyed by the ‘God-protected City’ of Constantinople was a leitmotif of imperial pronouncements from the seventh century onwards,9 2 The term ‘Byzantium’ only came into use in the sixteenth century, when it was introduced to distinguish the medieval eastern Mediterranean state from the ‘Roman’ empire of antiquity. Byzantium is a Latinised form of the name of the city chosen by Constantine the Great (306–37) to be his residence, Byzantion, renamed Constantinople after him. 3 El-Cheikh (2004a), pp. 103–11. 4 See below, p. 708; EI, VIII, s.v. Saldjuks, pp. 948–50 (C. E. Bosworth). 5 Parastaseis syntomoi chronikai, ed. Preger; tr. Cameron and Herrin; Dagron (1984b). 6 McCormick (1990), pp. 3, 21–31, 205–8; Morris (2003). 7 Greek fire’s workings are likewise ill-documented: see below, pp. 233–4; Haldon and Byrne (1977); Haldon (2006a); Pryor and Jeffreys (2006), pp. 607–31. 8 Crow et al. (2001); Crow and Bayliss (2005); Angelov, D. G. (2004), p. 520; below, pp. 114, 471, 485. 9 Fenster (1968), pp. 97–8, 104 and n. 2.

6

general introd uction

becoming engrained in the consciousness of Christians in the eastern Mediterranean world. The dedication of the new City by Constantine the Great in ad 330 symbolised his conversion to Christianity and was commemorated each year on 11 May.10 Constantine’s espousal of Christianity marked a new beginning not just for the emperor but for all mankind, whose spiritual salvation now became his avowed concern. Bishop Eusebius of Caesarea, Constantine’s counsellor and biographer, interpreted the turningpoint thus, laying the foundations for an ideology that would treat the history of the church as being coterminous with the bounds of the Roman empire.11 The emperor thus became a pivotal figure in God’s grand design for believers and unbelievers alike, and the conception gained monumental expression in stone from Justinian’s building of St Sophia in Constantinople (see below, pp. 111–12, 114). Justinian’s building-works were undertaken when, for all the pressures from external enemies on several fronts, military feats could still bring confirmation that the Christian God conferred victory, and churchmen ranged far and wide on missions to bring remaining groups of pagans within the emperor’s fold (see below, pp. 307–12). The association of the empire of the Christians with the future of mankind remained vital even when the tide abruptly turned and, following a Persian occupation, the empire’s eastern provinces were overrun by bands of Arab warriors in the mid-seventh century. Formerly deemed poor, divided and readily manipulable by the Romans, these Arabs now acted in concert, united in responding to their own revealed truth, as conveyed by God to the prophet Muhammad (see below, pp. 173–95, 365–9). Little more than a generation later, Pseudo-Methodius12 explained ‘the Ishmaelites’’ extraordinary victories as God’s punishment on the Christians for their sins. He prophesied that ‘the Ishmaelites’ would carry all before them until the emperor awoke ‘like a man from sleep after drinking much wine’, arose and put them to flight; the emperor would subsequently make for Jerusalem, and his arrival there would lead to the appearance of the anti-Christ and Christ’s second coming.13 The text was soon translated from Syriac into Greek and the surviving version contains an interpolation alluding to actual Arab expeditions against Constantinople of the late seventh or early eighth century.

10 DC, I.79 (70), ed. Reiske, I, pp. 340–9; ed. and French tr. Vogt, II, pp. 143–50; Dagron (2000), pp. 60–71. 11 Eusebius of Caesarea, Life of Constantine, II.3–5, II.44–61, IV.74–5, ed. Winkelmann, pp. 48–50, 66–72, 150–1; tr. Cameron and Hall, pp. 95–7, 110–15, 182; Dvornik (1966), II, pp. 614–22; Brock (1994), p. 70. 12 A seventh-century Syriac author, who wrote in the name of the fourth-century bishop of Patara. 13 Pseudo-Methodius, Apocalypse, [13], 11, ed. Aerts and Kortekaas, I, p. 174; below, p. 247. See also Psalms 78: 65.

i. approaching byz ant ium

7

It also represents the Ishmaelites as momentarily entering the City before the emperor’s resurgence.14 The Arabs never did penetrate the walls of Constantinople and so these events were not, strictly speaking, relevant to Pseudo-Methodius’ prophecy. But the interpolation reflects widely held Byzantine beliefs: that they were acting out events foretold in sacred writings, and empire and capital were closely bound up with the fate of mankind.15 Sudden strikes against the City by barbarians such as the Rus in 860 were interpreted as divine punishment for its sins,16 and after Constantinople’s fall to the Crusaders in 1204, many believed this was God’s warning that the Byzantines should mend their ways before He showed His displeasure terminally (see below, p. 735). Faith and empire could no longer be held to be indissoluble to the same extent after 1204, yet eastern orthodox emperors remained at large and upon seizing control of Constantinople in 1261, Michael VIII Palaiologos (1258–82) presented himself as a new Constantine: his success in occupying the City was in itself a mark of God’s favour towards him and of God’s mercy for His people. Apocalyptic writings and sayings, some deriving from Pseudo-Methodius, circulated widely among orthodox Greek- and Slavonic-speakers alike. The Byzantine emperors’ predicament in the face of Ottoman Turk advances from the mid-fourteenth century onwards, the collapse of other orthodox polities and then, in 1453, the City’s fall to these Ishmaelites, appeared to bear out the prophecies. These developments could be aligned with other computations that earthly time would cease upon expiry of the seventh millennium from the creation, a date corresponding with the year 1492.17 Such computations were commonplace in the higher echelons of the church, and Patriarch Gennadios II Scholarios (1454–6, 1463, 1464–5) foretold doomsday on 1 September 1492. He thus assumed the City’s occupation by infidels could only be provisional, now that the empire was no more. Meanwhile, at grass-roots, orthodox Christian faith was integral to Roman identity; even today, a villager in north-eastern Turkey can explain that ‘this was Roman country; they spoke Christian here’ (see below, pp. 852, 853). Thus Byzantium is best viewed as an amalgam of communities of religious ritual and faith in the power of God, and of administrative institutions and defence works, some kept to a high degree of efficiency.18 True 14 Pseudo-Methodius, Apocalypse, [13], 9–10, ed. Aerts and Kortekaas, I, p. 172 (text); II, p. 49 (commentary). Already in 654, a large Arab fleet may have advanced far towards Constantinople before being destroyed by a storm: O’Sullivan (2004). See also below, p. 372, n. 17; Magdalino (2005), p. 42. 15 Alexander (1962), pp. 341–55. 16 Photios, Homilies, ed. Laourdas, pp. 29–52; tr. Mango, pp. 82–110. 17 Nicol (1979); Magdalino (1993b), pp. 27–8; Polyviannyi (2000), pp. 207–8, 218–23. 18 On ‘political orthodoxy’ among the Byzantines as well as orthodoxy’s doctrinal and ritual mould, see Beck (1978), pp. 87–108.

8

general introd uction

believers, however far removed from the material protection of the imperial authorities, could hope for spiritual salvation and perhaps physical protection through prayer, regular celebration of the eucharist and access to the holy. As with the bread and wine bringing the body and blood of Christ to mankind, other rites of worship and also the decor and layout of the structure within which they were celebrated symbolised higher things, the medley standing for an infinitely superior, harmonious whole. Willingness to see providential design in the domed interior of a Byzantine church was articulated by Maximus the Confessor, and it was further elaborated upon by Patriarch Germanos I (715–30) in his influential treatise on the liturgy. Theological meaning was assigned to even the humblest example of ecclesiastical architecture and its interior furnishings: proceedings inside the church building mirrored those in heaven.19 The ‘corporate consciousness’ generated by rites revolving round the liturgy could hold communities of Christians together, so long as priests could be mustered to perform the church services. In a sense, therefore, imperial governmental apparatus was superfluous, and orthodox communities could carry on even under barbarian occupation. This was the case in the thirteenth and fourteenth centuries, when the populations under Frankish or Italian rule were still, in their hearts, ‘turned towards Greek matters’. Such ‘Greek matters’, which did not distinguish very sharply between this world and the next, gave Marino Sanudo, a fourteenth-century Venetian observer, grounds for unease (see below, p. 778). In similar spirit the eminent holy man, Neophytos, ignored the Latins’ occupation of his island of Cyprus, and as Catia Galatariotou has remarked, judging by his writings alone, one ‘would be forgiven for believing that Cyprus never ceased to be a province of Byzantium’.20 Byzantine writings about the apocalypse offer little coverage of rebounds of imperial power before the final awakening from drunken sleep, but individual emperors showed resilience, sometimes recovering territories after generations of barbarian occupation. An emperor’s expectations of acceptance and collaboration from the orthodox under outsiders’ rule could be misplaced, as in the case of Manuel I Komnenos (see below, pp. 716–17). But after the Latin occupation of Constantinople and the emergence of rival orthodox emperors, widely scattered populations still proved receptive to the idea of belonging to the original Christian Roman empire. Not even the well-organised, culturally accommodating regime of the Villehardouin lords of the Peloponnese could counteract this magnetism, and Marino 19 Germanos I, Historia ecclesiastica, ed. and tr. Meyendorff; Taft (1984), p. 111–26; see below, pp. 111– 12, 244. On the (sometimes varying) interpretations of Maximus and Germanos, see Mathews (1971), pp. 113–15, 121–2, 140–4, 150, 159–60. See also D´eroche (2002), pp. 177–80 and, for later developments, ˇ cenko, N. P. (2006). Sevˇ 20 Galatariotou (1991), p. 218.

i. approaching byz ant ium

9

Sanudo’s apprehensions were voiced at a time when the Palaiologoi were gaining ground on the peninsula (see below, pp. 803–33, 860). Only outsiders with overwhelming military might, bonded together by distinctive religious beliefs and able to count on numerous like-minded enthusiasts, had fair prospects of implanting themselves lastingly in the ‘God-protected City’. This conjuncture did not come about swiftly or inevitably: the subtle, tentative quality of Mehmet II’s (1444–6, 1451–81) measures even after his capture of Constantinople in 1453, suggests as much (see below, pp. 858, 865–72). This is not to claim that the amalgam of faith-zone, imperial idea and state apparatus which the Byzantine empire represented was an unqualified asset, or that it was sustainable indefinitely. The bonds were coming apart as Athonite monks and some senior churchmen and officeholders denounced the overtures to the Roman papacy which beleaguered emperors, pressured by raisons d’´etat, were constrained to make. The implacable opponents of ecclesiastical subordination to the Latins accused John VIII Palaiologos (1425–48) of betraying orthodoxy when he accepted a form of union with Rome at the Council of Florence in 1439 (see below, pp. 862–3). Perhaps other, un-imperial socio-political structures could better have served the earthly needs of Greek-speaking orthodox in the thirteenth and fourteenth centuries, allowing for the development of their burgeoning urban centres, trading enterprises and litt´erateurs.21 But the plasticity, even virulence, of the orthodox Roman order during its protracted decomposition goes some way to answering the question of why the empire lasted so long. inside out: emperors, outsiders and roman orthod ox identit y The relations of Byzantium with the Christian west loom large through the chapters that follow, tracing political, military and ecclesiastical encounters and exchanges. This does not necessarily mark over-simplification of the issues for the sake of narrative formatting. To recount Byzantium’s relationship with all the peoples and areas around it in equal measure would not be feasible, given the kaleidoscopic movement of the peoples and, in many cases, the dearth of source-materials for their relations with the empire. The only institution whose dealings with Byzantium can be tracked continuously across a thousand years is the papacy, offering an alternative universalist scheme of things. The minutiae of this relationship are not analysed or recounted here, but Byzantino-papal relations form a baseline 21 See below, pp. 43–4, 45, 49–50, 824–5, 830–3. The flurry of late Byzantine writings on political economy signals interest in alternative constitutions, as well as reinforcement of the imperial order: Angelov, D. G. (2007); Gaul (forthcoming); see below, p. 862.

10

general introd uction

for Byzantium’s relations with the Christian west, a story offering extensive windows on, if not a key to, the empire’s longevity. Time and again, they also show how ‘Old Rome’ and its adherents impinged on the empire’s domestic affairs. There was an epic turning of the tables in the balance of power and wealth between Byzantium and the west from the sixth century, when Justinian’s armies restored most of Italy to his dominion, through to the eleventh century, when emperors could still harness western martial and commercial resources on their own terms, and up to the thirteenth and fourteenth century, when westerners often, but not invariably, had the upper hand. By the late Byzantine era, the empire was in many ways an economic colony of the west, the Genoese and Venetians controlling the islands and other strategically important vantage-points in the Aegean, backed up by formidable naval resources and exchanging manufactured goods for primary produce. The renown of western arms was such that Manuel II Palaiologos (1391–1425) spent years touring the west in hopes of military aid.22 Yet by this time much of the Peloponnese had been restored to imperial dominion after decades of Frankish rule in the thirteenth century, and – against the Turkish odds – ‘hot-spots’ such as Thessaloniki still aligned themselves with the emperor in Constantinople under the encouragement of their church leaders (see below, pp. 857–9). In tracing these shifts in power one glimpses the silhouette, if little more, of that ‘silent majority’ of orthodox Greek-speaking country-dwellers whose customs and beliefs stood in the way of occupiers’ maximal exaction of resources and consolidation of their regimes. In its way, the imperviousness of ‘Greek matters’ to land-based Latin warlords and churchmen offers as strong a clue as any to the reasons for the resilience of the Byzantine empire (see above and below, pp. 777–8). Yet it also stood in the way of Palaiologan emperors seeking some form of union with Rome (see below, pp. 829, 863–4). This work pays pronounced attention to emperors’ dealings with nonmembers of their empire, those considered not quite ‘Romans’ for one reason or another, laying it open to the charge of undue attention to ‘Byzantium’s foreign relations with little regard for its internal history’.23 This plaint cited the then-published volumes of the New Cambridge medieval history and is pertinent, seeing that over half our chapters derive from contributions made to that series; the series’ framing of the middle ages is maintained in this work.24 Moreover our chapters, in line with the New Cambridge medieval history, aim to present the interplay between 22 23 24

See below, p. 829; Barker (1969), pp. 171–99; Nicol (1974); Mergiali-Sahas (2001b), pp. 56–7. Treadgold (2003), pp. 1002–3. Chapters 1, 3 and 4 were published in substantially similar form in NCMH, I; chapters 10 and 11 in NCMH, II; chapters 13, 14 and 15 in NCMH, III; chapters 16 and 17 in NCMH, IV; chapters 20a,

i. approaching byz ant ium

11

socio-economic developments, the turn of events and vicissitudes of successive political regimes – the stuff of narrative. There are, as emphasised above, many roads to Byzantium, but the trails left by contacts with outsiders are numerous and quite well documented. They bear closely on Byzantium’s one undeniable characteristic, its durability, and on our opening questions: how on earth did the empire last so long, as political entity and as idea? The empire was continuously confronting armed outsiders, and constructing a balanced account of this requires frequent recourse to non-Byzantine sources. So attention to alternative polities seems not merely excusable, but advisable, particularly since those which veered between merging with and separating from Byzantium often provide invaluable information about the empire’s internal affairs. Four considerations may support this proposition. Firstly, a geopolitical fact no less important for being obvious: Constantinople lay at the hub of many routes by land and sea. Constantine the Great chose Byzantium because major military highways converged there and because its accessibility by sea would facilitate provisioning of the increased population he envisaged for his new residence. For almost 300 years corn supplies were regularly shipped from Egypt, free of charge. But the assumption that overwhelming advantage would lie with the emperor against all comers already needed qualification in Justinian’s era. Once Byzantium became a kind of empire sans fronti`eres, the very accessibility of Constantinople and its environs exposed citizens to abrupt arrivals of aliens. Even lulls were apt to be rudely interrupted by the onset of ‘barbarians’, as for example the appearance off the City walls of 800 Rus or Scandinavians who refused to disarm and whose ships had to be dealt with around 1025.25 And the speed with which Suleiman ibn Qutlumush (1081– 6) and his Turkomans advanced north-westwards along the military road towards the Bosporus in 1075 shows the mixed blessings of the highways inherited from ancient Rome.26 The state of emergency generated by the Arabs’ onset eased after the seventh and eighth centuries, but the challenge posed by potentially formidable foes on two or more fronts at a time never wholly lifted.27 Goings-on among outsiders were therefore of keenest concern to imperial statesmen. Through maintaining a stance of eternal vigilance against barbarians, they could hope for loyalty and order among the City’s inhabitants. The capital was, in effect, permanently in a frontline position and this raises a second aspect of the empire’s involvement with outsiders: every generation or so Constantinople’s citizens faced a major ‘barbarian’ incursion 20b and 21 in NCHM, V; chapters 22 and 23 in NCMH, VI; chapter 24 in NCMH, VII; and chapters 2a, 2b and 2c in CAH, XIV. 25 Skyl., ed. Thurn, pp. 367–8; French tr. Flusin and Cheynet, p. 305. 26 See below, p. 707. 27 Haldon (1999a), pp. 37–8.

12

general introduction

Map 1 Physical geography of the Byzantine world

i. a pproaching byz a nt iu m

13

14

general introd uction

or at least an alert.28 The more fertile tracts of territory in the provinces were mostly either at risk of raids from Muslims or juxtaposed to Slavonicspeaking populations. Those few which were not, such as the inner sanctum around the Sea of Marmara or the north-eastern Peloponnese (see below, p. 501), were of considerable economic and fiscal value to the empire, enabling it to carry on. In fact the very fragmentation of Byzantium’s territories from the seventh century onwards made it the harder for marauders to hit all the prize areas simultaneously. With a modicum of naval capability, the imperial government could tap these fertile areas’ resources and maintain an administrative infrastructure and armed forces of a sort. Revenues reliant on agrarian produce, porous borders and painstaking (and therefore slow) methods of assessing and collecting taxes in consultation with locals were not wholly incompatible with one another (see below, p. 63). But in such circumstances the government could seldom afford very large, full-time armed forces, and the more convincing estimates favour a generally modest scale.29 This brings us to a third aspect of the emperors’ ready recourse to external regimes and keen interest in direct dealings with them: the value of military manpower from other societies, whether as individuals in the imperial forces, companies serving alongside them, or self-sustaining hosts attacking Byzantium’s enemies on home ground. Sizable field armies recruited from ‘Romans’ and geared to combat were not only costly to equip and maintain. They also posed a standing temptation for ambitious generals. Military coups, apprehended and actual, formed part of the empire’s heritage from ancient Rome and the double-edged qualities of glorious victories won by generals, however trustworthy, underlie Justinian’s differing treatment of Narses, who as a eunuch was debarred from the throne, and Belisarius (see below, pp. 206, 208). During the Byzantine emperors’ centuries-long confrontation with their Muslim counterparts they were ever watchful of their strat¯egoi (see below, pp. 259, 266, 380–1, 394). These provincial governors had sweeping powers, but neither the revenues nor high-calibre manpower sufficient to make a bid for the throne easy. Themselves disposing of finite military resources, emperors had good reason to concern themselves with elites and power structures other than their own. It was not merely a matter of cost-effectiveness, substituting battle-hardened ‘barbarian’ brawn for that of Christian Romans, nor even that outsiders were generally less likely to show enthusiasm for attempts on the throne. Diplomacy amounted to negotiating arrangements with external or subordinate powers and with other elements not quite – or not 28 29

Beck (1965), p. 13; see below, pp. 260, 299. Haldon (1999a), pp. 99–106; see below, pp. 60–1.

i. approaching byz ant ium

15

at all – Roman. This was an activity that an emperor could direct from his palace, relying on court counsellors and hand-picked agents, notably the basilikoi who often acted as his emissaries to another potentate or notable. In this way the emperor could swiftly mobilise armed units, even whole armies. They served his ends but with minimal employment of his administrative apparatus, and payment was at least partly conditional upon results. Thus the ‘flat-management’ style discernible in central governmental bureaus of middle Byzantium suited the emperor’s dealings with outsiders particularly well. And in this special relationship of the emperor with barbarians lies a fourth reason for our paying particular attention to un-Roman peoples beyond the City walls. It is in the field of diplomacy that Byzantine statecraft can claim responsibility for a text without any known precursor from the ancient Roman epoch. The title of De administrando imperio (‘Concerning the governance of the empire’) given by a seventeenth-century scholar to Constantine VII’s handbook addressed to his son Romanos II (959–63) has been criticised as a misnomer, since internal affairs feature only briefly, far more coverage being devoted to ‘the nations’ (ethn¯e), outsiders beyond his direct dominion. But the highly personal nature of the text does not make it unrepresentative: Constantine’s order of priorities registers where palace-bound emperors saw their strengths lying. Constantine’s rhetoric in his preface demonstrates the way in which workaday considerations of costeffectiveness could be dignified into positive affirmations of the emperor’s ascendancy, couched in biblical tones: God has raised up Romanos ‘as a golden statue on a high place’, ‘that the nations may bring to thee their gifts’ and bow down before him (Psalms 17.34, 71.10, 32.14).30 Through the incessant reception of embassies from other potentates, the emperor could demonstrate his authority in majestic form and signal his hegemony to subjects as well as to outsiders. In addition, and with less ceremony, he dealt directly with individual foreign notables. The logothete of the Drome was the first official to have an audience with the emperor in the Chrysotriklinos each morning, and he had a further session every evening. External affairs and matters arising from them were the logothete’s principal brief, and one reason for his close attendance on the emperor was probably the steady flow of outsiders through this hall. The Book of ceremonies treats the reception there of ‘several foreigners’ as routine.31 These were not necessarily ambassadors, representing another potentate, but individuals. Such face-to-face encounters enabled the emperor to forge personal ties with a wide range of notables, encounters 30 31

DAI, preface; Shepard (1997), pp. 90–4. DC, II.1, ed. Reiske, pp. 520–2. A high-ranking official (ho epi t¯on barbar¯on) was responsible for barbarians in the empire, especially Constantinople itself: Seibt and Wassiliou, Byzantinischen Bleisiegel, II, pp. 50–1.

16

general introd uction

which might involve bestowal of a court-title but had no necessary institutional framework. Through his ‘diplomacy of hospitality’ the emperor could make the acquaintance of individuals who might return to a position of prominence in their home society – or might return to acquire as much. Besides, there was always the possibility that a visiting ethnikos would opt to remain at Constantinople, becoming the emperor’s doulos, even ultimately a Roman. Young barbarians from across the steppes or from the other end of Europe were apt to spend stints at court.32 The princely and noble families among the Armenians offered particularly rich pickings for talent-spotters at Constantinople, and lower-born individuals could rise through merit, usually initially military, in the emperor’s service. The families of the Kourkouases and the Lekapenoi are examples of such recruitment. Instances of Armenian princes and, still more strikingly, of middle-ranking notables holding court-titles while resident in their homeland will feature in chapters below.33 The Armenian lands and their multifarious links with Byzantium were to an extent a special case, but similar processes were underway on most approaches to Byzantium other than central and south-eastern Anatolia in the era of the jihad. They underline the way in which governance shaded into dealings with separate societies and cultures. During the earlier middle ages military governors supplemented central officials in treating with Slavonic-speaking and other non-Roman notables on the outskirts of Thessaloniki, Dyrrachium and other fortresses and strongholds on the Balkan and Peloponnesian coast, while headmen of Slav groupings such as the Belegezitai were termed archontes and given responsibilities as well as titles. In this way, and complemented by ecclesiastical organisation, imperial enclaves very gradually extended their reach to the point where taxes were imposed or services exacted.34 In the western portions of the Byzantine ‘archipelago’ what might be termed ‘internal diplomacy’ was continually in play, operating by devices not dissimilar to the higher-profile encounters of the emperor with potentates and notables in the Chrysotriklinos or Magnaura at Constantinople. Thus encounters and negotiations of many kinds between the emperor and his senior officials and outsiders – whether informal meetings, ties solemnised by a court-title, or actual administrative posts – were the sinews of Byzantine governance. This networking process was necessarily unending, occurring at many different points and social levels across the imperial dominions, not merely the capital. This is one reason why the question of Roman identities is so complex. A senior army commander, Philaretos 32 33

Shepard (2006d). See below, pp. 340–2, 345, 347, 357–8, 509, 709; ODB, II, pp. 1203–4 (A. Kazhdan); Savvides (1990); Brousselle (1996). 34 Turlej (2001), pp. 105–24; Seibt (2003a), p. 462; below, p. 258.

i. approaching byz ant ium

17

Brachamios, could carve out a power structure having markedly Armenian characteristics to the point where he was dubbed first of the Armenian rulers of Cilicia by a later Armenian chronicler.35 And a century earlier the sons of an Armenian kom¯es in the imperial armed forces had transmuted into leaders of a Bulgarian insurrection against Byzantine occupation, the Kometopouloi (see below, p. 522). Collation of Byzantine with western sources shows several persons prominent in the imperial service, intellectual life and even the Byzantine patriarchate in the eleventh and twelfth centuries to have had close Italian connections, if not actually to have been of Italian origin.36 It is by considering some of the other elites with which the imperial court had so much to do that one may hope to understand the workings of the Byzantine empire. If this attention to ‘foreign relations’ appears excessive, such is the price of prying into the human, and not very institutionalised, organs of that empire. Byzantium’s workings involved compromise and accommodation on the part of both outsiders and imperial authorities. The latter were in practice willing to make concessions. For example, the Rus trading in Constantinople in the tenth century were allowed to have their disputes with Byzantines resolved partly in accordance with Rus custom,37 while the Armenian princes allocated territories in eastern Anatolia had commands over sizable communities of fellow Armenians, maintaining their own culture and church organisation.38 At any stage in the course of these encounters, individual outsiders could opt for Roman ways and religious orthodoxy in their entirety. Hence the need to keep orthodoxy clear and pure, and to be on guard against deviance. It is no accident that lists of ‘the errors of the Latins’ (i.e. western Christians) began to be circulated at the very time when westerners were becoming a familiar sight in the larger Byzantine towns and on highways, and when social intercourse with them was on the rise.39 It was, in fact, their ongoing accommodation of exogenous groups and individuals within the empire in varying degrees of assimilation and their flexibility in dealings with them and with externally based traders, elites and potentates that made Byzantine rulers so adamant concerning certain prerogatives. So long as key marks of uniquely legitimate hegemony were reserved, all manner of concessions – jurisdictional, territorial, honorific – might be vouchsafed according to circumstances. Foremost among these ‘brandmarks’ was the name of ‘Roman,’ with all its connotations of cultural and moral superiority, antiquity, rightful sovereignty and, from Constantine the Great’s time on, manifest Christian destiny (see above, p. 6). It is 35 36 37 39

D´ed´eyan (2003), I, pp. 6–7, 69. See also Yarnley (1972); below, pp. 707, 709. Magdalino (2003); Magdalino (2007a). 38 D´ Stein-Wilkeshuis (1991). ed´eyan (1975); see below, pp. 360, 692, 701. Kolbaba (2000); Kolbaba (2001); Kolbaba (2006); see below, pp. 73–4.

18

general introd uction

no accident that the Byzantines reacted promptly to those external rulers and their emissaries (usually western) who impugned their monopoly of Romanness, whether by terming the basileus ‘Greek’ or by purporting to brand their own regime Roman (see below, pp. 417, 432, 540, 545). From the same considerations, efforts were made to maintain consistent protocols, terminology and, even (for centuries at a time), media in formal communications of the basileus with other rulers. As Anthony Bryer observes, John VIII was still styling himself ‘emperor and autocrat of the Romans’ and signing in purple ink at the council of Florence in 1439.40 Court ceremonial and indeed the whole ambiance of the emperor’s ‘sacred palace’ in Constantinople, its orders of precedence, titles, vestments and other trappings, were likewise presented as quintessentially ‘Roman’. As the chapters below suggest, the style of the court could alter as new emperors sought to distance themselves from immediate predecessors, and certain authority symbols changed appearance over time. Yet even emperors invoking ‘renewal’ to legitimise their regime tended to present themselves as ‘new Constantines’, harking back to the very first Constantine.41 Conscious efforts were made to use de luxe baths, antique dining styles, buildings and other monuments, together with chariot-racing and spectacles patently associated with ancient traditions for the grander state occasions.42 Such observances seem mostly to have continued until the twelfth century. Some involved sizable numbers of Constantinople’s citizens as well as the elite,43 and the games and races occasionally yet regularly held in the Hippodrome symbolised the emperor’s ‘marriage’ to his City as well as his other attributes, such as eternal victory (see below, p. 521). Even banquets in the palace drew hundreds of invited guests, and the purpose of official orders of precedence was to maintain ‘good form’ and order (taxis) against the ever-present risk of confusion and loss of imperial composure.44 But there was also a sense that the imperial court was the repository, breeding-ground and citadel of true Romanness, the place where those ‘born in the purple’ would first see light of day.45 The conviction that being raised in the palace conferred moral qualities as well as legitimacy was volubly expressed by a prime (and far from disinterested) beneficiary, Constantine VII Porphyrogenitus. Decrying his former 40 See below, p. 853 and fig. 62. For continuities and changes in diplomatic forms, see D¨ olger (1938– 9); D¨olger and Karayannopulos (1968), pp. 89–107; Kresten (1992–3); Kresten (1998); Kresten and M¨uller (eds.), ‘Die Auslandsschreiben der byzantinischen Kaiser’; Schreiner (2005). 41 Magdalino (ed.) (1994). 42 Mango (1981a), p. 352; Dagron (2000); Featherstone (2006). 43 Morris (2003), pp. 241–2, 253. 44 Philotheos, Kletorologion, ed. and French tr. Oikonomides, pp. 82–3; McCormick (1985), p. 5; Kazhdan and McCormick (1997), pp. 175–6; Oikonomides (1997a). 45 Attempts to concretise ‘the purple’ as birth in the Porphyra chamber in the palace were, however, subsequent to the notion that special qualities were inherent in those born to reigning emperors: Dagron (1994).

i. approaching byz ant ium

19

co-emperor, Romanos I Lekapenos (920–44) as ‘common and illiterate’, he opined that only ‘those raised up in the palace’ were imbued with ‘Roman customs from the very beginning’, as if the court were a kind of crucible of Romanness.46 Classical, Attic Greek was also prized by Constantine, aware as he was of his own deficiencies in writing it.47 Attic was the dialect in which orations and other formal statements were composed for delivery at court occasions, and in which official accounts of emperors’ deeds were composed. Thus the Byzantine court, with its regard for ‘good form’ and preoccupation with continuity, religious orthodoxy and linguistic correctness might seem to epitomise a ‘mandarin’ political culture. Literary works from this quarter are among the readiest sources for the general history of the empire (see below, p. 58). Such priorities and shibboleths are, however, best viewed against a background of barbarians frequenting the imperial court, ad hoc arrangements continually being made with useful potentates, and titles bestowed on outsiders with barely a smattering of spoken Greek. The proportion of families in the ruling elite comprising first-, second- or third-generation immigrants probably made up around a quarter of the total.48 The number of persons of external stock who made it, or almost made it, to the imperial throne is striking. Romanos Lekapenos’ uncouthness made an easy target for Constantine VII’s jibes since he was of quite recent Armenian origins. But the Porphyrogenitus was himself the grandson of a low-born opportunist, conceivably of Slavonic stock; the tendentious ancestry claimed for Basil ‘the Macedonian’ in the Life of Basil composed under Constantine’s auspices even represents him as of Armenian kingly descent.49 Once sole occupant of the throne, Basil I had displayed his orthodox piety and staged triumphs to parade his supposed qualities of victorious generalship.50 He also undertook spectacular works to restore churches in and around Constantinople and to refurbish the Great Palace, the setting for imperial ceremonies.51 Basil’s measures were designed to legitimise a palace coup, but they demonstrate how certain ‘core values’ such as doctrinal orthodoxy and regard for palace ceremonial lent themselves to assimilation by highly ambitious, capable outsiders. Basil’s adaptation and manipulation of establishment forms and conventions was extraordinarily skilful, enabling him to work, charm and perhaps sleep his way to the very top. But his career pattern was played out less spectacularly – and through more straightforward merits such as military talent – by many individuals intent on merely attaining the higher reaches of the imperial establishment, or gaining a footing there for their offspring. Many were members, if not from the dominant family, of elites 46 48 49 51

47 Sevˇ ˇ cenko, I. (1992a), pp. 178–80. DAI, ch. 13, pp. 72–3; see below, pp. 508–9. Kazhdan and Ronchey (1999), pp. 346–7. 50 McCormick (1990), pp. 154–7, 212–23. Life of Basil, pp. 212–13, 215–16; see below, p. 294. Ousterhout (1998), pp. 115–19, 129.

20

general introd uction

beyond Byzantium’s borders, external or internal.52 Thus one of Basil’s early patrons, the widow Danelis, appears to have belonged to the ruling family of a Sklavinia in the Peloponnese. Basil’s way of thanking her upon seizing power was to confer a court-title on her son and to stage a reception in the Magnaura, befitting ‘someone of substance and distinction who is at the head of an ethnos’.53 The concern with ‘form’ and general inclination to stand on ceremony of imperial Byzantines were, unquestionably, obstacles to casual infiltration by outsiders belonging to different cultures. Their presence in sizable numbers in the imperial milieu was predicated by the ‘diplomacy of hospitality’. An abiding apprehension was that this might lead to dilution of the ‘Roman customs’ which were integral to Byzantium’s credentials for hegemony. Such apprehensions are seldom vented in as many words in extant written sources. But they go far to explaining the limitations of the historical sources emanating from the Byzantine establishment, their preference for a classicising prose style and tendency to present events in terms of antique or scriptural precedents. The insistence on taxis in the more functional works composed in palace circles is, in fact, an index of the pressures making for the reverse. Prominent among those pressures’ drivers was the steady stream into Constantinople – and, often, out again – of outsiders, whether from the ‘outer territories’ beyond the City walls54 or out-and-out ethnikoi. The maelstrom of constant interaction between the imperial leadership and significant outsiders and alternative power structures underlies the glassy surface that establishment-derived literature tends to present to us. This interaction, the opportunities as well as the problems it posed for Byzantium’s rulers, is a theme running through the chapters of this book and it has a bearing on the empire’s longevity. 52 53 54

Brousselle (1996), pp. 47–50, 53–4; Garso¨ıan (1998), pp. 59–61, 66, 88. ˇ cenko, I. (1992a), p. 192 and n. 68. Life of Basil, pp. 217–18; Sevˇ Magdalino (2000b); pp. 149–52. For Skylitzes’ sense of virtual ‘home counties’ of the ‘Romans’ in the vicinity of Constantinople itself, see Bonarek (2003).

I N T R O D U C T I O N – PA RT i i

PERIODISATION AND THE CONTENTS OF THIS BOOK jon at han s he pard

when did byz antium end – or begin? Byzantium is an elusive phenomenon because so many of its constituent parts altered in place and over time. The overarching fac¸ade of the imperial order remained, with certain fixed points: religious doctrine, use of the Greek language, and the City of Constantinople itself. But many other elements were mutable – from court fashions, administrative methods and commercial undertakings, to forms of warfare or territorial possessions. Byzantium’s distinctive qualities lie in this interplay between the fixed and the changeable, the expendable and the non-negotiable, ensuring its endurance across a millennium or so, longevity which only the Chinese and Japanese empires can unequivocally be said to have surpassed. However, even the chronological limits of the Byzantine empire are contentious. In a material sense, the Constantinopolitan-based emperor could be regarded as powerless, politically dead by the time Sultan Mehmed II’s technicians closed the Bosporus and trained their guns on the City in 1453. Yet alternative or affiliated imperial regimes were still functioning, and to all appearances the empire of Trebizond and the despotate of the Morea could have carried on indefinitely, even flourished, had the Ottomans not determined to put paid to them, too, while reducing other robust polities in the Balkans to tributary status (see below, pp. 831–2, 860–1). And the idea of the central place of the empire and the City in God’s scheme of things persisted among the orthodox well after 1453. From that point of view, 1492 – when the world had been predicted to end following upon the empire’s fall (see above, p. 7) – seems as good a date as any to conclude. And it is not wholly coincidental that 1492 saw the discovery of the New World: Christopher Columbus, himself of Genoese stock, was sailing a refined version of the type of cog which plied directly between Genoese Chios, England and Flanders until the Turks began putting pressure on their trading activities in the Levant (see below, pp. 847–8).

21

22

general introd uction

Our story might accordingly begin with the new covenant between God and mankind which Constantine the Great (306–37) made upon accepting the Christian religion and basing himself in the city of Byzantion. That is when the emperor became a figure of universal value to influential Christian churchmen such as Eusebius (see above, p. 6). Triumphalist notions about the Christian empire’s destiny and hopes of individual spiritual rebirth started to filter through the lettered and propertied classes of the Roman Mediterranean and other strata of society, providing a sense of purpose and consolation through military setbacks and periodic devastation. In other words, something of the amalgam of Christian faith and eschatological hopes that characterised medieval Byzantium was already being mixed in the fourth century, when the Roman empire encompassed much of continental Europe, was a formidable presence in Africa and western Asia and still harboured notions of conquering Persia. To begin the story with Constantine among his bishops has all the more to recommend it, in that the Christian empire’s longevity and perseverance through a variety of changes of fortune and circumstances is the connecting theme of this book. Besides, Constantine’s conversion is roughly the point where several other authoritative surveys of Byzantium begin, whether focused on the ups and downs of the Byzantine state and its ruling classes;1 on the thought-world of the faithful and the dissenters of Byzantium;2 or dealing with culture and society as well as matters of state.3 However, both practical and theoretical considerations have discouraged us from beginning with the fourth century. Constantine accepted Christianity in 312 but the processes by which Christian observance became irreversible, an indispensable attribute of Romanness, were intricate and protracted. At the time of Constantine’s death in 337 and for many decades to come, the majority of the population were non-Christian. The diffusion of Christianity can partly, but only partly, be charted through the injunctions of senior churchmen, the edicts of emperors and the feats of holy men. The decisions of individuals, families or communities to adopt the Christian faith and forms of worship could be made for many different reasons, not least peer-group pressure. These processes are seldom set out in reliable detail in our surviving sources, and such records as there are come from highly partisan writers.4 The fifth century saw the construction of important platforms and spectacular pinnacles of Christian empire that would be admired and utilised by 1 3

2 Mango (1980). Ostrogorsky (1968). Cavallo (ed.) (1997); Treadgold (1997); Angold (2001); Treadgold (2001); Mango (ed.) (2002); Harris (ed.) (2005); Gregory (2005); Haldon (2005b); Cameron, Averil (2006b). 4 See Brown, P. (1998); contributions to Lenski (ed.) (2006); Casiday and Norris (eds.) (2007).

ii. peri odisation and contents of t his book

23

much later regimes in the Christian west as well as the east. The ‘rhetoric of empire’, already well worked upon by Eusebius, Themistius, John Chrysostom and others in the fourth century, was further elaborated.5 A vibrant court culture and ceremonial accrued around the figure of the emperor ensconced in his ‘sacred palace’, the majesty and dignitaries of his court evoking the heavenly court above.6 The monuments of this architecture of empire took both material and institutional form, from the walls of Constantinople, built for Theodosius II (402–50) (fig. 2), to the almost as massive law-code, the Codex Theodosianus, that he promulgated. This law-code marks a milestone in emperors’ attempts to codify law and governance across the spectrum of society, providing for church property and the jurisdiction of bishops and the religious observances and way of life of ordinary subjects. An entire book of the Codex is devoted to religious issues, heretics, Jews and pagans among them.7 These new materials of empire-building did not, however, make unreservedly for the consolidation of imperial power. The leadership of the church was prone to bitter disagreements over elements of doctrine such as the interrelationship of the divine and human qualities of Christ. These controversies periodically reached boiling-point and assemblies of patriarchs and bishops were convened under the supervision of emperors to try and reach an agreement. Of these ‘universal’ – ecumenical – councils, the council of Chalcedon (451) stands out as of particular importance. Its outcome was a formula concerning Christ: that He was ‘recognised in two natures’ while also ‘in one person and hypostasis’. This was acceptable to the papacy, being very close to the terms which it had formulated, and Emperor Marcian’s commissioners pressed the council to accept it. Serious fault-lines, however, remained both among eastern churchmen and between easterners and the papacy.8 The divisions would reopen and become still more acrimonious in the following century. A case could be made for bringing these achievements and controversies within the compass of this book, treating ‘the Byzantine empire’ as already in place in the fifth century. However, such identification of the empire’s development and well-being with the formal elaboration of Christian doctrine by councils and the spread of Christian observance in everyday life raises three major difficulties. Firstly, as already stated, Christianity spread along multifarious channels and its effects – or otherwise – on social attitudes and behaviour patterns in town and country varied greatly between 5 7

6 Cameron, Averil (2002b); McCormick (2000), pp. 156–63. Cameron, Averil (1991b). Theodosius, Codex Theodosianus, XVI, tr. Pharr et al., pp. 440–76. On the making of the Codex, see contributions in Harries and Wood (eds.) (1993). For the impact of Christianisation on the family, see Giardina (2000). 8 Gray (1979); Allen (2000), pp. 814–16; Gray (2005), pp. 221–5.

24

general introd uction

communities and regions. The onset of the new religion in its various guises has been much discussed in recent English-speaking scholarship and might seem to provide grounds for studying the Christianising empire of the fifth and sixth centuries en bloc. But scholarly voices have also sounded in favour of closer attention to the nuts and bolts of empire, institutions of governance such as the law and its enforcement, the state apparatus for revenue raising and expenditure, and coinage.9 These institutions remained in working order across much of the eastern empire throughout the fifth century, and the continuing pax romana rested on impressive reserves of military manpower, coordinated to awesome effect. So long as the empire presented obvious and overwhelming advantages of martial strength, prosperity and public welfare, these material benefits spoke for themselves. Christian preachers and holy men might inveigh against alternative cults, indifference, materialism and – in matters of discipline and doctrine – against one another, and their written outpourings have survived in bulk, as has the Christian framing which orators and senior churchmen now provided for imperial power. But while that power still appeared to underwrite general well-being out of its own vast resources, in the heterogeneous and multi-cult towns and settlements of the eastern Mediterranean region,10 Christian worship and observance had a wide range of alternative connotations for their inhabitants – whether as an optional extra supplementing other devotions; an imposition; a familial or communal tradition of cult practices and obligations; or an avenue for individual spiritual development. Christian court culture and splendiferous trappings supplemented, embellished and enhanced imperial power, rather than virtually substituting for it. Faith and worship were a valued asset in bringing the emperor victories and the empire dominance, but they were not yet generally seen as vital to the empire’s survival: the empire did not yet, in the fifth century, amount to a faithzone.11 Secondly, many shades of Christian belief, practices and organisation were developing under their own momentum, on a geographical scale extending far beyond the empire’s frontiers. The ferment of Christianity in the fertile crescent and other parts of the orient posed obstacles for the Roman emperor as well as openings. When Armenia’s King Tiridates IV adopted Christianity early in the fourth century, the Armenian church organisation and distinctive Armenian script provided buildingblocks for the development of a separate political identity. Yet occasionally 9 Millar (2000), pp. 754, 757–9; Giardina (1999); Marcone (2004), pp. 30–6. For a response (focusing on problems of periodisation) see Bowersock (2004), with a rejoinder by Giardina (2004). See also Harries (1999); Cameron, Averil (2002b), pp. 180, 190; Fowden (2002), pp. 683–4. 10 On widespread well-being in the eastern provinces, see Whittow (1990); below, pp. 467–9. 11 See Brown, P. (1998), p. 653.

ii. peri odisation and contents of t his book

25

prospects opened up of bringing Armenia – ever a region of keen strategic interest – under Roman hegemony, if only Armenian churchmen would subscribe to imperially approved church doctrine (see below, pp. 169–70, 337–8). Persia is another example of how Christianity was something of a double-edged sword for the Roman empire. The Sasanians offered safe haven for dissidents, vociferously at odds with the established church and (often) with the imperial authorities; by the sixth century the Nestorians made up a substantial portion of the Persian population and Persianoccupied Nisibis was a school for dissenters from the imperial line. Yet there flickered the prospect of further Christian converts in Sasanian ruling circles and it was not inconceivable that key individuals might opt for Chalcedonian orthodoxy (see below, pp. 136, 142–4, 311). Meanwhile, and less spectacularly, ruling families and local communities adopted Christianity in the Arabian peninsula, Abyssinia and the Sudan for a variety of reasons, sometimes thanks to proselytisation by sects which operated in rivalry with missionaries sponsored by the emperor (see below, pp. 180, 188–9, 308–11). These movements and cross-currents among other societies and powers posed anomalies and challenges to an empire purporting to embody Christianity on earth. It has therefore seemed appropriate to include chapters which look back in detail to the more important developments on the empire’s eastern approaches around the time of Constantine’s conversion.12 They put in perspective the church councils of Ephesus (431) and Chalcedon and those of the sixth century,13 and also the tug of culturoreligious forces working on imperial decision-making from east and west. However, a balanced presentation of the fifth century for its own sake would require full coverage of the western half of the empire, too, and this constitutes a third reason why overall treatment of the fifth-century empire is not attempted here. Law and order ceased to be the sole preserve of the imperial authorities in the west long before the abdication of the last legitimate emperor, Romulus Augustulus, in 476. In the west, the adoption of Christianity as the official religion of the empire appeared to usher in political turbulence and disorder, rather than consolidating military effectiveness, state-maintained infrastructure and prosperity, as it did in the east. Furthermore issues such as the diffusion of power; the levels of law and order sustained and of everyday violence; and the calibre of urban living and economic activity in the Mediterranean world and the Roman provinces further north are highly contentious.14 The contrasts and cross-currents between the eastern Mediterranean world and the Christian 12 14

13 See below, ch. 1. See below, chs. 2a, 2b, 2c, 7. See, for example, Liebeschuetz (2001); Cameron, Averil (2002b); Brown, P. (2003); Ward-Perkins (2005); Heather (2005); Wickham (2005). The question of whether conditions in the Germanic kingdoms were more violent than in the empire preceding them, together with comparison with the eastern empire, feature in contributions to Drake et al. (eds.) (2006).

26

general introd uction

west are a key theme of this work, but the dissolution of empire in the west has distinctive, often quite local, explanations. The broader implications for the eastern empire of the formation of more or less ‘barbarian’ regimes in the central and western Mediterranean regions will be discussed below (see ch. 3). That their existence was unprecedented, posing new problems yet also diplomatic and strategic openings for the rulers of Constantinople, is hard to deny, and this goes some way towards justifying the starting-point of this book around ad 500.15 We have therefore begun our story around the time when Byzantium first stood alone as a working Christian empire, surrounded by potentially formidable predators. Those seeking balanced treatment of the economic, social and politico-administrative history of the earliest centuries of the Christianising Roman empire have only to turn to the three final volumes of the Cambridge ancient history, which have advanced the bounds of classical antiquity up to around ad 600.16 They will also find the progress of the Christian faith and its practices traced from its beginnings, across the length and breadth of the Roman empire and beyond, in the Cambridge history of Christianity. The first volume includes accounts of Constantine’s reign and the first council of Nicaea.17 Also of use are discussions by individual scholars or teams of conference speakers on the problems of the sense in which late antiquity may be said to have ended and the Byzantine empire begun, of how far the sixth century marks an end or a beginning.18 part i: the earlier empire c. 500–c. 700 The age of Justinian: flexibility and fixed points in time of uncertainty (Chapters 1–4) In the sixth century, imperial armies were still large, the infantry tactics and military units of Rome’s heyday were still in use, and they functioned on the strength of an urban economy whose structure was older still (see below, pp. 99–100). Expeditionary forces reconquered the coastline of north Africa and southern Spain and took back Sicily and Italy; their spoils bolstered Justinian’s (527–65) triumphalist claim to have restored the Roman empire to former worldwide glories (see below, pp. 201–3, 207, 208–10). Yet these were protestations in the face of uncertainties arising from plague, natural disasters, incursions of armed outsiders and internal religious dissent. 15 16 17

On periodisation, see Cameron, Averil (2002b); Fowden (2002). CAH, XII; CAH, XIII; CAH, XIV. Cameron, Averil (2006a); Edwards (2006). See also contributions in Casiday and Norris (eds.) (2007). 18 Allen and Jeffreys (eds.) (1996); Cameron, Averil (2002b), pp. 165, 190–1; Mango (2002b), pp. 2–5. The effective starting-point of Whittow (1996) is c. 600.

ii. peri odisation and contents of t his book

27

Characteristics of Byzantium following the seventh-century ‘transformation of a culture’19 can already be discerned in the era of Justinian – notably the fusion of faith and imperium; penny-pinching and a cast of defensiveness behind imperial bluster; and the assumption that a correct approach to the divine held the key to earthly imperial as well as spiritual salvation. The uncertainties of the sixth century made divine sense, if one accepted the numerous predictions of the end of the world then in circulation.20 While individual responses ranged from the traditional to the Christian, involving amulets, relics and incantations,21 church-going congregations and monasteries looked to the scriptures, priest-directed worship and holy men. In enumerating the fortified towns and refuges furnished by Justinian for rural populations in the Balkans, Procopius acknowledged the inevitability of barbarian incursions: yet he also stressed that the emperor was manifestly doing everything within his powers to protect, offering his subjects both a literal and spiritual safe haven (see below, p. 111).22 Thus the imperial order joined forces with faith and public acts of worship to offer a modicum of security: it is likely that by the later sixth century, images of the Mother of God and of the saints were being venerated with mounting intensity and orchestration.23 The emperor also offered underpinnings for social peace and order in the form of clear, accessible codification and distillation of Roman law (see below, pp. 107–9). A peculiar blend of military triumphalism, strenuous intercession for divine support and careful husbanding of assets helped the Byzantines survive as a collective the drastic turn of events in the seventh century and beyond. The medieval empire’s components were scattered and disparate, from the basileus in his God-protected City down to the inhabitants of fortified towns and self-sufficient, semi-pastoral hill-country kingroups in Anatolia or the Balkans. Their material circumstances and degrees of security varied considerably. But a substantial proportion even of the country-dwellers were within reach of refuges of some kind, and also of churches. Since the blend began to be brewed in Justinian’s era – when elaborate earthly measures of protection for the civilian population were instituted, first put constantly to the test and found only partly wanting – so do our opening chapters. They also take full account of the empire’s eastern neighbours and rivals, current and to come. Persia’s rulers, the Sasanians, made much of their victories over the Romans, defining their 19 21 22

20 See below, pp. 121, 122; Meier (2003), pp. 73–100, 373–87, 405–26, 459–70. Haldon (1997a). Krueger (2005), pp. 302–10. Pr B, IV, tr. Dewing and Downey, pp. 218–315; Gregory (1992 [1993]), pp. 246–50; Sarantis, ‘The Balkans during the reign of Justinian’ (DPhil thesis, 2005). On Anastasius’ building programme and appreciation of the need for a network of fortifications, see Haarer (2006), pp. 65–70, 109–14, 230–45. See also Dunn (1994). 23 Cameron, Averil (1978); Cameron, Averil (1992c), pp. 18–20; Cameron, Averil (2000), pp. 12–13; Belting (1994), pp. 109–14, 134.

28

general introd uction

own power in terms of these. Yet their institutional base may not have been quite as firm as this implies, while substantial minority groups within their realm worshipped the Christian God (see below, pp. 144, 153–5). The coexistence and cultural interaction of these two great powers prefigures that of Byzantium and the Abbasid caliphate, whose court in Baghdad drew on Persian customs, political thought and high culture.24 The Arabs in the age of the Prophet Muhammad lacked the Persians’ sophistication, yet their capacity for literacy, diplomacy and organised warfare was more advanced than hostile Romans, or their own later writers, allowed. To that extent their adroitness in exploiting the aftermath of ‘the last great war of antiquity’ between Byzantium and Persia is perhaps unsurprising (see below, pp. 174, 193–5). By the seventh century, the Armenians had long been Christian. The inventor of their distinctive script, Mashtots‘, based it on the Greek alphabetical model. He had received a Greek education, and Christian Armenia’s literary culture drew heavily on the fourth-century Greek fathers as well as Syriac writings (see below, p. 161). But the Armenians had their own church hierarchy, headed by a catholicos, and the princely and noble families in mountain strongholds debarred Romans and Sasanians alike from outright control over their respective sectors in Caucasia. For Justinian and his successors, the Armenian church posed a conundrum as intractable as was the papacy to their west: Christian, notionally beneath their umbrella, and yet highly articulate and prepared to defy the emperor and his senior churchmen on matters of doctrine (see below, pp. 171–2). The Armenians stood in the way of the idea of a Christian church coterminous with the empire even as, individually and collectively, they made an extraordinary contribution to its workings.25 Justinian’s legacy was, then, a singular concoction in unpredictable circumstances. Its supreme and understated asset was flexibility, the capability to withstand military setbacks through a blend of material safeguards, ad hoc diplomacy, spiritual purity, ideological vision – and bluff. The ‘beacon’ was not only St Sophia but Constantinople itself, where law and order were upheld and where the unceasing rites of empire and worship were performed, shielded by imperial orthodoxy (see below, pp. 111–12, 114). The emperor as beacon-keeper could still convincingly take charge of these essentials, although in reality he was unable to direct the course of events in all his provinces. Justinian’s reign can therefore be seen as prologue and scene-setter for all that was to come, until the City of Constantinople actually did fall to barbarians, albeit fellow Christians, in 1204. In many ways the sixth century was the starting-point of the cycles of rebuffs and recoveries that characterised the middle Byzantine period. 24

Kennedy (2004a).

25

Charanis (1961); Garso¨ıan (1998).

ii. peri odisation and contents of t his book

29

An alternative starting-point for our story might indeed have been the sensational events of the mid-seventh century. The chapters below subscribe to the widely held view that the eastern empire underwent massive shocks in the seventh century: thereafter things were never quite the same again, for all the restoration of order in many provinces and the semblance of Roman continuity maintained in the capital. The Arabs’ overrunning of the Levant and Egypt halted inflows to Constantinople of taxes and resources from what had been by far the richest provinces of the empire, dislocated distribution networks and military funding, and in the words of a mid-seventh-century text left the empire ‘humiliated’.26 Few, if any, men of letters could see the point of celebrating imperial deeds in the guise of classical heroics. Grand historical narratives in the mould of Thucydides, such as Procopius’ or Theophylact Simocatta’s, and rhetorical poems such as George of Pisidia’s in praise of Heraclius’ campaigns against the Persians in the 620s, could scarcely be cast from collapsing frontiers and incessant improvisation. As Averil Cameron has pointed out, much was still written, but with regard to the world of the spirit and the transcendent meaning of things, sermons, theological tracts and disputations.27 The lights go out, so far as straightforward narrative is concerned, and our main surviving Byzantine accounts of events from around 640 onwards were not composed before the early ninth century. Yet this change in source-materials does not necessarily imply a corresponding rupture in every single aspect of governance or of spiritual priorities for all the inhabitants of the empire at that time. The differences in civil administration and military organisation which are clear from our sources for the ninth century cannot be dated precisely, and few scholars now subscribe to George Ostrogorsky’s thesis that systematic military reforms and creation of a theme system were carried out by Heraclius in immediate response to the Arab invasions (see below, pp. 239–40, 266). The shifts of overall responsibilities to military commanders (strat¯egoi) and their staffs in the provinces may well have been provisional and fluctuating, with independent civilian authorities still functioning through the eighth century. The sixth and seventh centuries show sufficiently similar administrative arrangements still in place and important processes of change continuously underway to be viewed together in one part. Moreover, as Andrew Louth shows in Chapter 4, disputes about doctrine went on being fought out by churchmen under the emperor’s eye in the mid-seventh century and an ecumenical council was convened in his City 26 Doctrine of Jacob the Newly Baptised, ed. and French tr. Dagron and D´ eroche, p. 168. On the drastic measures apparently needed in the 630s to pay troops in Palestine and Syria with revalued copper coins, see Schulze et al. (2006), pp. 17–24. 27 Cameron, Averil (1992b), pp. 85–6, 104.

30

general introd uction

near that century’s end (see below, pp. 234–5); difficult to cross in its lower reaches, the Danube continued to act as barrier, if not formal border, until the Bulgars installed themselves south of the river in the early 680s; and Carthage, an imperial administrative centre and strategic key to the central Mediterranean, only fell to the Arabs in 698. Until around that time, imperial statesmen may well have reckoned that the Arabs’ extraordinary advances would eventually be repulsed, or would ebb away. It therefore seems defensible to bracket the seventh century together with the sixth as the time when the Christian empire first demonstrated its capacity to go through massive earthly vicissitudes, military triumphs and sudden reversals. For all the sense of imperial Roman continuity that Justinian’s propaganda conjured up, his genius lay in providing for conditions of incessant change. part ii: the midd le empire c. 700–1204 The course of events: Byzantium between shocks and rebounds (Chapters 5–6, 13, 16–17) Any boundary drawn across conditions of flux is arbitrary, and several chapters in Part II delve back into seventh-century events, as background to the problems facing emperors once warfare on their eastern approaches became unremitting. Armies had to be stationed across the Anatolian plateau, combat-ready yet potentially self-sufficient, and emperors needed to forestall defections to the Arabs by those forces’ commanders. The balance between maintaining military effectiveness and ensuring trustworthiness already coloured Byzantine political thinking and strategy in Justinian’s era. But the problem gained a new edge from the Arabs’ ongoing challenge and, as Walter Kaegi shows, emperors were very fortunate that comparable tensions dogged the Muslim leadership and stymied its capacity for major invasions (see below, pp. 365, 373, 375, 392). By around 700 the Muslims were tightening their hegemony over Armenia after a brief revival of imperial influence there (see below, pp. 345–6). And in 705, Justinian II (685–95, 705–11) forcibly regained his inherited throne in Constantinople, aided not by a ‘Roman’ army, but by the Bulgars, now installed in the former province of Moesia. The emperor’s special relationship with barbarians as an alternative to his own forces would become a hallmark of the medieval empire. The deep-seated state of emergency is set out in detail by Marie-France Auz´epy, who shows how Leo III (717–41) and Constantine V (741–75) recast the formula for state survival set out by the first Justinian. Through reforming the army and identifying it very closely with their own regime, the Isaurian emperors allayed risks of a coup d’´etat and provided a strong right arm for state power, even while recognising the limits of the material

ii. peri odisation and contents of t his book

31

defences affordable for Romans living in the provinces. They also provided the wherewithal, in the form of lower-denomination silver coins, for greater recourse to taxes raised in money (see below, p. 270). The sweeping powers of the autokrat¯or and his agents were tempered by concern for justice, providing a vent for the aggrieved through the channel of the emperor’s courts, but also ruthless punishment for proven malefaction (see below, pp. 275–7). The spiritual welfare of the emperors’ subjects was also catered for systematically, with numerous new sees founded. Furthermore, the ‘idols’ deemed to have incurred God’s wrath – and consequent disasters for the empire – were denounced and, eventually, destroyed. Thus iconoclasm is fitted by Auz´epy within a broader context of crisis, and her chapter as a whole illustrates the imperial order’s capacity for renewal. The fruits of this renewal ripened in the decades following the rulers’ final abandonment of iconoclasm in 843, while the Abbasid caliphs no longer led or funded massive incursions into Asia Minor. The need to purge contaminating idols had lost its urgency, while devotion to images for accessing the divine was fervent in some quarters of the church. Shaun Tougher’s chapter demonstrates the standing of churchmen after the restoration of icon-veneration. Patriarchs could still be unseated from their thrones, like Photios (858–67, 877–86) in 867 (see below, p. 301). But churchmen and monks had stood up for icons, some earning the status of ‘confessors’, persons who had suffered persecution for true belief, albeit not death. One such churchman was Theophanes Confessor, the author of a chronicle that is one of our main sources for eighth- and early ninth-century Byzantine history. Commemoration of the restoration of icons to favour was celebrated annually at the Feast of orthodoxy (see below, p. 290). The gradual expansion in the material and demographic resources available to the emperors from the mid-ninth century onwards was therefore tempered by the esprit de corps and general repute of churchmen as orthodoxy’s guardians. The limits of the emperor’s ‘space’ were symbolised in the routes he did, and did not, take on his way to the liturgy in St Sophia.28 It may be no accident that one of the earlier – and victorious – eastern expeditions launched by Basil I (867–86) was directed against dualists, the Paulicians, as if to demonstrate his orthodox credentials in the drive against heretics. Basil’s expeditions against the Muslims of Melitene and Tarsus were, however, less successful, and his parading of his piety and generalship was at least partly designed to camouflage humble origins and a blood-soaked throne (see below, pp. 294–6). Equally, Byzantine defence installations could do little to curb the depredations of Muslim raiders who had the nearby island of Crete as a safe haven and potential emporium for slave trading from the 820s on (see below, pp. 499–500). Yet their ability to 28

Dagron (2003), pp. 95–114.

32

general introd uction

sustain themselves through raiding implies fairly rich pickings to be had. This accords with other hints of economic vitality, for example the code for officials supervising trading and craft activities in the capital – the Book of the eparch, issued or reissued under Leo VI (886–912).29 Nonetheless, Byzantium’s armed forces were fully stretched in containing Muslim land raids. And the Christianisation of the Bulgars in the Balkans from c. 865 onwards rendered their polity more cohesive and militarily formidable than ever, even if their receipt of baptism from Byzantine priests made them nominally ‘spiritual sons’ of the Byzantines, and notionally deferential.30 With valuables and manpower leeching away to Muslim land raiders and pirates, Byzantium was hard put as ever to conduct large-scale campaigns on two fronts at once (see below, pp. 498– 500). Even after the death of Symeon of Bulgaria in 927 eased Byzantine concerns about its western neighbour, offensives to the east were limited in scale and largely confined to removing thorns from the flesh. A kind of equilibrium prevailed, compounded by the emperors’ reluctance to entrust their generals with armies of full-time soldiers schooled in aggressive warfare. Such an army could easily be turned against an emperor and this was, in effect, what happened after the rampages of the amir of Aleppo, Saif alDawla, became insufferable. Within a few years of the codification of the status of theme-soldiers’ military holdings,31 the raising of more full-time soldiers and switching of tactics to full-scale offensives, Crete was regained – and its conqueror, Nikephoros II Phokas (963–9), was sitting on the imperial throne. There is little doubt that the army’s size increased markedly in the later tenth century.32 This reinforced the challenge which ambitious army commanders posed to the young emperors claiming the right to rule through birth in the purple, Basil II (976–1025) and Constantine VIII (1025–8). Basil eventually quelled the revolts of his generals and associated his regime with the army to an extent unparalleled since the iconoclast soldieremperors. The protracted resistance of the Bulgarians to his attempts to impose hegemony provided opportunities for the exercise of war leadership in person. While the epithet of ‘Bulgar-slayer’ was only applied to Basil much later,33 his Bulgarian wars enabled him to square the circle and maintain larger armed forces, spectacularly intimidating neighbours on all sides, without falling prey to rebellion (see below, fig. 37 on p. 523). And the continually mounting agrarian and commercial prosperity and 29 31 33

30 D¨ See below, p. 497. olger (1940); see below, pp. 299, 318–20. 32 Haldon (1999a), pp. 84, 103–5. Haldon (1979), pp. 45–65. Stephenson (2003a), pp. 81–96.

ii. peri odisation and contents of t his book

33

population size of the enlarged empire was most probably sufficient to sustain this army. What is less clear is whether the empire’s customary methods of painstaking tax-collecting and transmuting of revenues into soldiers’ pay were well geared for the armies that Basil II amassed. Such negotiable fiscal transactions required very many officials, and a significant increase in their numbers is suggested by the profusion of their seals in this period. Moreover, Basil set a precedent as ‘happy warrior’ and expansionist, without providing a male heir: his successors had to cope with a certain legitimacy deficit as well as with broader issues of strategy, the role of the armed forces and finding means of paying for them. The vitality, wealth, yet vulnerability of eleventh-century Byzantium is brought out in Michael Angold’s chapter. Culturally the empire was a hive of creativity, from the visual arts to literature. The volume of law-cases concerning money, property and inheritance is registered in a textbook assembled from a senior judge’s rulings and opinions, the Peira (literally, ‘trial, experience’).34 And Constantine IX Monomachos’ (1042–55) institution of a law school at Constantinople represented an attempt to ensure welltrained jurists and administrators for state service in an era of widespread litigation (see below, pp. 598–9). Byzantium had not seen such a pitch of general material well-being and diversity of faiths and cultures beneath the imperial aegis since the seventh century. The analogy holds good in strategic terms, too. In the mid-eleventh century, as in the 630s, the emperor could justifiably believe that his foes were subjugated or reduced to virtual impotence (see below, pp. 227–8). Yet then, without much warning, emperors found themselves combating raiders on three fronts: although the Pechenegs were more or less absorbed into the Balkans, the Normans in the west and above all the Turks in the east were not so amenable. In default of an incontestably legitimate dynasty ruling in Constantinople, several generals fancied for themselves the role of imperial saviour, for which there was pressing need. Disagreements over strategy and uncertainty as to the nature or intentions of the enemy were compounded by rivalries between generals and within the now labyrinthine Constantinopolitan court establishment. That Byzantium lacked flexibility in its response to external challenges at a time of internal tensions and inflated bureaucracy is not so surprising. More striking is the alacrity with which Alexios I Komnenos (1081–1118) recovered from the strategic mistakes of his early years as emperor and learned from them. He proceeded to reorganise his army, abolish many court titles and effectively debase the coinage. The empire had, after all, 34

Peira, ed. Zacharia von Lingenthal, pp. 11–260.

34

general introd uction

lost control of much of Anatolia to the Turks and was correspondingly impoverished: in cutting his imperial coat to fit diminished cloth, Alexios was pragmatically responding to severely reduced circumstances. There were precedents from earlier reigns for such economies and recourse to ‘flat-management’ style, as there were for the simultaneous emphasis on piety and plain living that Alexios made a hallmark of his regime (see below, p. 618). The empire’s material losses made correct worship all the more important, and although the church was now vocally resistant to emperors’ tampering with doctrine, Alexios and his descendants still saw themselves as guardians of doctrine, shepherds of their subjects’ souls (see below, pp. 617– 18 and fig. 46). This was also the case with Manuel I Komnenos (1143–80). Manuel displayed prowess in astrology, jousting and war in equal measure.35 His virtual ‘cult of personality’ included placing Christ Emmanuel on his earliest coins, a visual pun on Manuel’s name, while the list of subjugated peoples associated with Manuel on an inscription in St Sophia evoked the titulature of Justinian’s era.36 In emphatically aligning his regime with doctrinal purity and regularity of worship, Manuel resembled Justinian. The blend of expansionist bravado and inspired opportunism with tacitly defensive measures and ad hoc fortification-work belonged to a great tradition (see below, pp. 637–9, 642–4, 684, 685). And the Komnenian empire’s reversion to a pattern of far-flung strongholds and outer and inner zones of imperial orthodox order in some respects evokes the state of emergency of the late seventh and eighth centuries (see below, pp. 261, 264, 653–4). In the twelfth century, too, the imperial presence could be concentrated in ‘hot-spots’, the more fertile lands and strategically important points, where protection and exactions were more intensive, in contrast to those districts, maritime or inland, that were left exposed to barbarian incursions or occupied by outsiders. Manuel Komnenos still had formidable armed forces37 and a navy at his disposal, and these could well have helped him and also his successors gain new vantage-points, tap the burgeoning commerce of the eastern Mediterranean and forge alliances (see below, pp. 638–9, 645). Two twelfth-century developments complicated matters. Firstly, the political stability and administrative workings of Byzantium were now entwined with the extended family of the Komnenoi, together with a number of related families (see below, pp. 657–8). Lands, fiscal privileges and senior military posts were gathered in their hands, and for all the resultant advantages of cost-cutting and political cohesiveness, the expectations of individuals and branches of the family were high, mutually competitive, and 35 36

See below, pp. 637, 644, 646; Jones and Maguire (2002), pp. 113–18, 136–9. 37 Birkenmeier (2002). DOC, IV.1, pp. 281, 296; Mango (1963a), pp. 324 (text), 330.

Sea of Azov

Venice

Cherson

B L A C K Ragusa

Bari

Ohrid

on ym Str

Amalfi

Dyrrachium

Maritsa

Adrianople

Gulf of Taranto

THESSALY

Nikopolis

Patras

SICILY

(lost by c.900)

Anchialos Sinope Constantinople Sea of Marmara

Thessaloniki Avlona

Amastris

Nicaea

Amorion Smyrna

d Maean

Ephesos

Sebasteia Akroinon

Caesarea

er

Chonai

Melitene Ikonion

Plain of Argos

Monemvasia

Trebizond

Ankyra

Dorylaion

Lemnos

Lesbos Euboea Chios Thebes

S E A

CILICIA

Attaleia (Antalya)

Seleukia

Antioch

RHODES Stronghold Economically significant areas Land over 1000 metres 0 0

100

200 100

300 200

400

CRETE CYPRUS

500 km 300 miles

Map 2 Middle Byzantine ‘hot-spots’: sketch map of key strategic strongholds and fertile, well-populated lands of particular economic interest to the imperial authorities, yielding revenues or other material benefits; few were under continuous imperial control throughout the middle Byzantine period, and some outliers were then ruled by local elites

36

general introd uction

proliferating. This lessened the flexibility that the imperial administration had traditionally shown in attuning tax assessments to a property’s current capability to pay them.38 The effect of extensive tax exemptions, piling tax burdens on those left unprotected by privileges of one sort or another, was neither healthy for state finances nor conducive to longer-term political stability. Secondly, the twelfth-century imperial authorities had to contend with western Europeans of a different stamp from those of the earlier middle ages. The westerners were themselves fragmented and many individuals were primarily concerned with trading opportunities or a career rising high in the basileus’ service. Yet the intimacy of some western venturers with the Komnenoi and their successors paved the way for displaced members of the imperial family or pretenders to seek aid from western potentates and from causes with agendas of their own. Alexios Angelos’ fateful bid in 1201–2 for help from western leaders, one of whom was his brother-inlaw, was from this perspective nothing out of the ordinary, but it triggered the capture of Constantinople by the Fourth Crusaders. Already in 1185 a kinsman of Manuel I Komnenos, together with a pretender claiming (falsely) to be Alexios II, Manuel’s son, had given King William II of Sicily (1166–89) a pretext for sending an expedition that easily took Thessaloniki and only failed to reach Constantinople through overconfidence.39 Around 1184 another authentic Komnenos, Isaac, had taken control of Cyprus and started issuing coins in his own name, and it was a western crusader, King Richard I of England (1189–99), not the Constantinopolitan emperor, who eventually dislodged him. Thus some of the empire’s choicest lands and fortified towns were proving to be highly vulnerable, or self-sufficient imperial entities, a foretaste of conditions after 1204, and indeed after the restoration of imperial status to Constantinople at the hands of Michael VIII Palaiologos (1258–82). Taking stock: the economy, religious missions, border regions and significant others (Chapters 7–12, 14–15, 18–19) The question of economic conditions on the eve of the Latins’ seizure of Constantinople is discussed by Mark Whittow in one of the ten topicor region-specific chapters in Part II. Byzantine economic history has undergone intensive enquiry, and The economic history of Byzantium: from the seventh through the fifteenth century published in 2002 provides an authoritative summing up.40 The work’s three volumes contain (besides 38 That there should be regular, even equitable procedures at a time of maximal exactions was still a concern of the Komnenoi: Magdalino (1994), pp. 107–14; see below, p. 63. 39 Brand (1968), pp. 161–71; Angold (2003a), pp. 40–1, 84; Phillips (2004), pp. 90–4. See also below, p. 687. 40 EHB. For a concise yet informative and wide-ranging overview, see now Laiou and Morrisson (2007).

ii. peri odisation and contents of t his book

37

much else) syntheses on economic and non-economic exchange, the role of the state in the economy, and the periodisation of Byzantine economic history, as well as studies on the urban economy, both in Constantinople and in the provinces, and also surveys of economic life in the countryside, and of prices and salaries.41 Taking account of all this, Whittow shows that there remains room for discussion over the main lines of Byzantium’s economic development. In particular, our ever-expanding archaeological database suggests that the material impoverishment and demonetarisation of the provinces in the seventh and eighth centuries may not have been quite as drastic as often supposed, and thus that the undeniable economic recovery of the ninth and tenth centuries may have started from a higher base-line (see below, pp. 478, 483–4). Whittow reopens the question of the relationship between this recovery and the condition of peasant-proprietors. Such proprietors could be of substance, and imperial novellae referring to them as ‘poor’ (pen¯etai) denote their vulnerability to encroachments by the well-connected rather than material penury (see below, p. 489). Imperial pronouncements concerning their vital benefit to the state had their rationale, whereas the eventual amassing of prime properties by a few well-connected and privileged families was of questionable compatibility with the state apparatus’ longer-term workings (see below, pp. 490–1). Unlike economic affairs, Byzantine missions received limited scholarly attention in the twentieth century. Sergei Ivanov’s chapter is the first survey in English of the full sweep of missionary activity from Justinian’s time to the Palaiologan period.42 Ivanov questions the strength of the Byzantines’ impulse to spread the word to peoples beyond their borders, and shows that the initiative for missions often came from external potentates. The Byzantine state seems to have been better geared to the Christianisation of individuals or groupings of non-Romans now seeking careers in its service, or who had settled en masse within its environs. By contrast, Byzantineborn churchmen such as Theophylact of Ohrid assigned to far-flung sees were at their most eloquent in expressing discomfort with their barbarous surroundings.43 The emperor’s role of indomitable defender of ‘the Christians’ was projected in court ceremonial as vividly as his image of being the equal of the apostles, and here at least, as Walter Kaegi shows, rhetoric bore some resemblance to reality. The forces of Islam were arrayed against the once mighty Christian empire, which they claimed to have superseded. Devising administrative means of coping with Muslim incursions was of paramount 41 Laiou (2002c); Oikonomides (2002); Laiou (2002d); Dagron (2002); Matschke (2002a); Lefort (2002); Morrisson and Cheynet (2002). 42 See also Ivanov (2003). 43 John Mauropous, Letters, ed. and tr. Karpozilos; Theophylact of Ohrid, Letters, ed. and French tr. Gautier; Mullett (1997); see also below, pp. 321, 672–3.

38

general introd uction

concern for Constans II (641–68) and subsequent emperors. Warfare with the Muslims was unremitting for centuries, the orthodox Christian convictions of the majority population in Asia Minor supplementing the Taurus mountain range and cold winters in discouraging permanent Arab occupation of Anatolia. Iconoclast emperors repeatedly led expeditions against the Muslims in person; and the early Abbasid caliphs, in contrast to their immediate Umayyad predecessors, were also intent on leading expeditions against the Byzantines themselves (see below, p. 388). The raiding and counter-raiding between the arch adversaries came to form a rhythm, even if the caliphs could still deal knock-out blows to imperial prestige as late as the mid-ninth century (see below, pp. 391–2). The Byzantines’ caution in exploiting the caliphate’s internal difficulties with large-scale military initiatives was matched by the Armenian princes, generally wary of bringing down the wrath of their Muslim overlords. Yet, as Timothy Greenwood shows, the boundaries between Byzantine and Armenian faith and church organisation were more fluid than Armenian narrative historians lead one to suppose. While Photios’ project for formal union between the churches in the ninth century came to nothing, the Constantinopolitan patriarchate extended its organisational reach into what had been the preserve of Armenian churchmen during the tenth century, and writers on behalf of princes not subscribing to the Chalcedonian line on Christ’s nature could still show fulsome admiration for the basileus (see below, p. 357). Such intermingling was not to the emperor’s unmitigated advantage: the ties between leading Byzantine generals and Armenian princes brought them additional military manpower, and Basil II’s involvement with Caucasian affairs was impelled partly by considerations of self-defence (see below, pp. 358–9). The emperor’s interest in the Latin Christians of the central and western Mediterranean regions was likewise stimulated partly by their capacity to intervene in his own affairs, especially as the pope’s spiritual standing entitled him to pronounce on even fairly minor disputes concerning elections within the Constantinopolitan patriarchate. Beneath the formal ecclesiastical boundaries, exchanges between Greek-speaking eastern orthodox populations and communities in Sicily, southern Italy and the Byzantine lands remained active even after the Muslim conquest of Sicily. The prospect of southern Italy succumbing to Sicily’s fate in the later ninth century and becoming a springboard for Arab incursions into Dalmatia and the Aegean prompted Basil I’s decision to restore the southern Adriatic ports and strategically significant inland power-nodes to imperial dominion.44 For almost 200 years, strongholds and eventually extensive tracts of territory on the peninsula came under Byzantine administration. The population of regions 44

Shepard (1988b), pp. 70–2.

ii. peri odisation and contents of t his book

39

such as Apulia was mostly Latin-speaking, its ultimate spiritual head being the pope, while Lombard customs prevailed in the courts.45 This hardly disqualifies southern Italy from attention and yet, as has justly been remarked, the source-material for this part of the empire has still to be fully exploited in many works on Byzantium.46 The seepage of imperial elements and eastern Christian culture into many strata and spheres of Italian life, from the papacy downwards, is demonstrated in Thomas Brown’s chapter. The trajectory of imperial power can only be described as ‘recessional’, and local elites and the papacy had to fend for themselves against Lombards and later Muslim maurauders. But, as Brown shows, ‘le snobisme byzantinisant’ was current among some leading families irrespective of their ethnic origins; commercial ties linked other points with the eastern empire; and even as the papacy aligned its own ideology with Frankish imperium, ‘Rome remained within the Byzantine cultural orbit’ (see below, p. 448). All this had to be taken into account by the Carolingians when trying to bring northern and central Italy within their dominions, as rightfully part of their empire. Many elements in Byzantine religious culture were of interest to churchmen hailing from north of the Alps, not least the utility of Greek for clarifying phrases in the Bible or of the church fathers. As Michael McCormick shows, the militarily robust iconoclast emperors provided a foil for Carolingians and their counsellors, intent on framing an empire to their own specifications yet impeccably Christian (see below, pp. 417–18, 424–5, 431). The working model of such an empire to the east could hardly fail to excite in them emulation, and occasional adaptations. The phenomenon of Frankish arms, letters and church organisation stimulated the papacy to take a firmer, more confident, line in its own dealings with the Constantinopolitan patriarchate and emperors. Things came to a head when in 863 Pope Nicholas I (858–67) took against Photios; the ensuing rift was both symptom of, and further stimulus to, the Byzantine church’s sense of its own exalted status.47 The Frankish behemoth that loomed behind the papacy’s fulminations was, however, disintegrating by the 880s, whereas Byzantium’s naval vessels could still sail to relieve Rome from Muslim raiders. Byzantine dominion began to coagulate and then extend northwards from the heel of Italy. As is pointed out in Chapter 14, the Byzantine expedition to oust Muslim pirates from the Garigliano valley south of Rome in 915 was mounted in tandem with warriors supplied by local magnates and with the papacy’s cooperation. A century later, the katepan¯o Basil Boioannes managed to intervene in the Garigliano valley and destroy the fortress of a papally backed magnate off his 45 47

46 Morris (1995), pp. 5–6. Martin (1993), pp. 48–53, 531–2, 709–11. See below, pp. 420–1; Dagron (1993); Chadwick (2003).

40

general introd uction

own bat (see below, pp. 538, 558). Emperor Henry II (1002–24) retaliated in 1022 but his attempt to cut the Greeks down to size was no more lastingly effective than his recent predecessors’. The resuscitation of the western empire in 962 by the Ottonian dynasty from Saxony had unleashed challenges, explicit and implicit, to Byzantium, but Liudprand of Cremona’s pronouncements on the subject strike a note of defiance rather than fullthroated confidence. In fact the Ottonian emperors found many uses for Byzantine luxury goods and authority symbols in devising a political culture for their newly amassed dominions (see below, pp. 546, 549–50, 554–5). The Ottonians provided the princes of Capua-Benevento and other potentates in south-central Italy with a powerful, yet fitful, counterforce to the Byzantine presence in the peninsula. The principalities of CapuaBenevento and Salerno, and the duchies of Naples, Gaeta and Amalfi seem to have been quite stable through the first two-thirds of the tenth century. They were, however, vulnerable to wrangles over the succession and other disputes within the respective ruling families, and power and resources were becoming diffused among the families of counts and other masters of castelli (see below, pp. 571–2, 579–80). In the case of these principalities and duchies, as with so many other elites and political structures bordering on Byzantium, their amoeba-like characteristics and the highly personal nature of leadership placed them at a disadvantage compared with the continuity of a unitary state. The basileus’ strongholds ensured his potential military presence, while through diverse diplomatic devices, operated by his indigenous officials and local sympathisers and also at his own court, he kept tabs on established leading families and forged ties with significant newcomers. The power-play of Byzantine Italy is fairly well documented and bears comparison with that in the middle Byzantine Balkans, for which archival evidence is poor. There, too, the imperial government maintained its interests with the help of centrally appointed agents, local elites, potentates ensconced in discrete political structures and mobile groupings whose military capability could be temporarily harnessed. Paul Stephenson’s chapter illustrates the traditional workings of steppe-diplomacy and shows how imperial strategy after Basil II’s conquest of Bulgaria envisaged hegemony over the Balkans: a network of routes and a series of zones, with the innermost receiving fairly intensive administration, fiscal exactions and protection, while the outer ones were left more to their own devices, under local notables (see below, pp. 664–9, 670, 673–5). Imperial attention and resources could be devoted to those zones where external threats or internal rebellions arose, and in many ways this flexible arrangement worked. Defensive measures and diplomacy succeeded in repulsing or deterring Norman incursions into Dalmatia and beyond for some time after their seizure of southern Italy. Byzantine emperors also exploited divisions within the

ii. peri odisation and contents of t his book

41

Hungarian royal family to curb rising Hungarian power. Manuel I Komnenos even appropriated a strategically significant portion of the Hungarian lands for a while (see below, pp. 642, 684–5). Yet as Stephenson shows, the emperors’ hold over much of the Balkan interior was loose-meshed, and Manuel’s preoccupation with the intentions of well-resourced Latin potentates and crusading ventures reflects awareness of this. But diplomatic d´emarches cost gold, and westerners were no longer bought cheaply or lastingly. The Byzantines generally tried to reconcile non-Greek-speaking populations to their rule by keeping taxes low. But in 1185–6, resentment over higher taxes fuelled an uprising of ethnic notables and provincial Greek-speakers, which took on separatist tendencies and transmuted into the resurrection of an independent Bulgarian power (see below, pp. 656, 687–8). The outlook for Byzantium’s eastern provinces was transformed abruptly by the coming of the Turks. By the mid-eleventh century, there was quite heavy reliance on local elites in the borderlands and a not unreasonable assumption that military threats from Islamic regions could be contained.48 The vigorous opportunism of Turkish chieftains and individual war-band leaders offset their lack of military cohesiveness and of regularly raised revenues. The drastic reform of military organisation needed to cope with the Turks was beyond the capacity of mid-eleventh-century Byzantine regimes (see below, pp. 600–1, 603, 607). Not that the empire was lacking in a series of outer zones on its eastern approaches any more than it was in the Balkans, as Dimitri Korobeinikov shows: Armenian local notables and the king of Georgia could still be enlisted to the imperial cause, George II (1072–89) being swayed by a sizable concession of strongholds and territories (see below, p. 705). Manuel I Komnenos was also adept at local-level diplomacy in Asia Minor and his personal ties with Turkish dynasts furthered stabilisation of the borders. Stability, however, made established rulers such as Kilij Arslan II (1156–92) even more militarily formidable, and Manuel’s attempt to overturn the Seljuq Turkish powerbase at Ikonion (Konya) led to crushing defeat at the battle of Myriokephalon in 1176 (see below, p. 716). Fortunately for the empire, the Seljuqs and other more established Turkish leaders showed little inclination to descend from their abodes 1,000 or so metres above sea-level in the Anatolian plateau. Not even the dissipation of imperial power after 1204 changed this state of affairs. The imperial Byzantine ‘rump state’ that formed around Nicaea co-existed fairly easily with the Seljuqs of Rum. It was the Mongols’ arrival and pressure in eastern Asia Minor that precipitated a chain reaction of migration among the Turcoman nomads and, in the early 1300s, the breakdown of residual Byzantine 48

See below, pp. 600, 607–8, 674; Haldon (2003b), pp. 60–74.

42

general introd uction

defences in the western coastal plains (see below, pp. 723–4, 726). This is yet another example of how far-away events could have drastic repercussions, upsetting the best efforts of the empire’s guardians. part iii: the byz antine l and s in t he l ater middle ages 12 04– 1492 Embers of empire (Chapters 20–24) By the time Byzantium’s defences in Bithynia in north-west Asia Minor succumbed to the Ottoman Turks, an ‘emperor of the Romans’ had once again been resident in Constantinople for some fifty years. To perceptive contemporaries, Michael VIII Palaiologos’ seizure of Constantinople from the Latins appeared ill-starred (see below, pp. 753, 804), and they would seem to have had a point. Recovery of the traditional seat of empire may have brought Michael personal prestige, but organising its defence and everyday maintenance proved to be heavy burdens on state finances and diplomacy. His son and heir, Andronikos II (1282–1328) had neither strategic flair nor trustworthy generals to cope with affairs in Asia Minor or the repercussions of the Mongols’ inroads there (see below, p. 726), and he anyway lacked resources to fund a navy. The vicissitudes of Constantinople-based regimes, whether of Latin or Byzantine emperors, reflected the demise of the command economy which had made the City such an omnivorous centre of consumption up to 1204. Thereafter its rulers were unable to collect taxes from numerous far-flung provinces, or to orchestrate a wide range of manufacturing and trading activities to their regimes’ advantage. The City’s inherent geographical advantages now provided it with economic buoyancy, in default of overriding state power. Constantinople became a meeting-point of externally based trading enterprises, mainly Italians’. Their self-interest drove the exchanges and determined the alignment of trade routes, and they pocketed the profits.49 The reinstallation of a ‘Roman’ basileus in Constantinople in 1261 might change the dominant outsiders from Venetians to Genoese, but not the dynamics of a now almost ‘globalised’ economy: the leading Italian commercial families and enterprises were not amenable to control by any one territorial state, and the Constantinopolitan emperor’s ability to rake off proceeds through taxing goods or transactions was gravely impaired. Across the Golden Horn from the City, in Pera (also known as Galata), the Genoese ran their own, fortified, trading centre. The Genoese and Venetians alike were prominent in fourteenth-century Constantinopolitan court ceremonial, a mark of their involvement in the latest permutations of empire. 49

Jacoby (2005a); see below, pp. 776–7.

ii. peri odisation and contents of t his book

43

Byzantine emperors, wherever they were now installed, would never again be able to amass resources or exercise purchasing power on a scale that made state service the main route to status and wealth. Their ability to reward and to coerce was correspondingly diminished and empire became more ‘virtual’, a matter of voluntary adherence and belief, than had been the case before 1204. But this did not put paid to the idea of empire, and in fact demographic trends, agricultural production and commercial activity seemingly continued on an upward course throughout the lands of the former empire until the mid-fourteenth century (see below, pp. 818, 820– 2). These provided material supports for a variety of political structures, under the leadership of scions of what had been imperial, or imperiallyconnected, families in Constantinople before 1204. So long as they resisted the temptation to make a name for themselves by recovering the City, they could enlist local elites, sport imperial trappings and count on a measure of popular acceptance and even armed backing in what had been outlying provinces, for example, Epiros. In addition, ambitious leaders of non-Greek-speaking Christian polities in Caucasia and the Balkans sought to legitimise and enhance the standing of their regimes, emphasising the sanctity of members of their dynasty or of other local saints and shrines close-linked with their rule. They, too, tried to make their respective realms coterminous with a church province or patriarchate. And a Greek-speaking basileus established himself in distant Trebizond and managed to stay aloof from bids for the throne in Constantinople.50 The kaleidoscopic swirl of Byzantine-born claimants to empire, splinter groups of Greek-speaking communities, orthodox Slav nation-builders and Frankish warlords does not lend itself to neat narrative rendition. Full treatment of all the different local situations would require a volume to itself. This is one reason why the conditions of flux following Constantinople’s fall in 1204 tend to be set apart from the general history of the Byzantine empire. The chapters in Part III offer an outline of political events in the main Byzantine dominions (except Trebizond) up to the fifteenth century, but no attempt is made to replace or duplicate detailed narratives already available in English.51 Instead, chapters are devoted to some of the principal beneficiaries from the events of 1202–4: the western European conquerors and colonisers; Italian and other merchants in the Aegean; Serbian and Bulgarian rulers contending for control of the Egnatian Way and outlets to the sea; and Albanian chieftains. Quite extensive coverage is given to matters of trade, emporia and trade routes. These illustrate the volatile nature of the commerce that yielded the most spectacular wealth. Several elites, 50 51

See below, pp. 731, 779; Bojovic (2001); Eastmond (2003a); Eastmond (2004). See, in particular, Nicol (1993).

44

general introd uction

would-be imperial Greek dynasts in the Balkans, Serbian and Bulgarian potentates and Latin men-at-arms, did business with, as well as competing against, one another, making military and marriage alliances in numerous permutations. At a time when few boundaries were really closed, or power centres firmly rooted, the prospects for a regime or simply a local family could be transformed by appropriation of a prosperous port, or a new deal with the Venetians or the Genoese. If this seldom made for stable political structures, it tended to stimulate rather than stifle new trading nexuses. New axes also formed directly between the former provinces of Byzantium, and Greekspeaking traders and sailors played a significant part in developing and operating these networks, albeit on a secondary plane to westerners.52 The new conditions prompted local, lower-value commerce and offered opportunities for other forms of intercourse between Greek-speaking imperial subjects or their descendants and Latins in Aegean coastal towns. This spectacle, together with fear for their orthodox souls, may well have stimulated the movement towards extreme asceticism and a hardening of the line against the Latins discernible in some Latin-frequented commercial centres, for example fourteenth-century Thessaloniki, a city in socio-cultural ferment (see below, pp. 47, 820, 823–4, 857–8). One region that temporarily insulated itself against such cultural contamination was that of Nicaea under the Lascarid emperors, in the first half century or so following Constantinople’s fall. They eschewed lavish consumption and ruled in a style somewhat reminiscent of the soldier-emperors of the eighth century.53 Restrictions against trading with the Latins were enforced and the state’s objective was self-sufficiency. Nicaea’s mostly agrarian economy and its character of a frontier society facing the Turks made for an effective fighting force, while also sustaining a robust and variegated court culture (see below, pp. 739, 751). Under the capable generalship of Michael Palaiologos, warriors from Nicaea defeated what was, in a sense, their opposite number among the Latins, the Franks of Achaia, at the battle of Pelagonia in 1259 (see below, p. 749). The Nicaeans’ victory is the more striking for the fact that their adversaries, under the leadership of the Villehardouin family, included the best-organised among the Frankish occupiers of the Byzantine lands. The qualities of the Villehardouin regime are brought out in David Jacoby’s chapter. The Villehardouin princes’ dealings with the Italian entrepreneurs were sometimes fraught and at first they had their differences with other Frankish lordships. But they came to arrangements of mutual advantage with, for example, the Venetians, while also courting the cooperation of Greek-speaking landowning elites and leaving 52 53

Morrisson (2005); see below, pp. 818, 820–1, 842–3. Angold (1975a); see also below, pp. 739–40.

ii. peri odisation and contents of t his book

45

orthodox churches and churchmen mostly unmolested, a prudent stance given their own limited numbers. However, the demands placed on peasants were less constrained by law: most tenants on estates became legally unfree, and their disputes could now usually be heard only in seigneurial courts. They lacked access to public courts proceeding by Romano-Byzantine law, which seem to have functioned right up to 1204 (see below, pp. 772–3). The Villehardouin leadership deliberately fostered a sense of regional identity, accommodating indigenous archontes within their political culture, and members of these families fought on their side at Pelagonia. Nonetheless, the Villehardouins continued to lose ground to the Palaiologoi in the later thirteenth and earlier fourteenth centuries.54 The importance of the Peloponnese to the Palaiologoi is shown by the fact that their territories there were usually allocated to the sons of the emperor. The ‘despotate of the Morea’ is not recorded as being of much fiscal value to the Constantinopolitan government, but its long-term economic viability and the cultural vitality maintained at the despots’ court in Mistra made this more a beacon than an outpost of the orthodox Roman empire. The despotate has been described as a ‘success story’ of late Byzantium (see below, p. 860), and part of its buoyancy came from the agreements that were made with Latin powers and trading interests. Essentially, the Byzantines marketed their wheat, honey and other primary produce to Italian traders ensconced on the coast, and catered efficiently for newly established trading posts such as Clarence (Glarentza) (see below, pp. 835, 841, 845). In doing business with the westerners without losing political autonomy or doctrinal orthodoxy, the despotate improved upon the example of Nicaea, demonstrating the resilience of ‘virtual empire’. Another success-story, likewise rather undersung in relation to the empire because unchronicled by Byzantine narrative historians, is that of the heterogeneous monks of Mount Athos in this period. Copious writings flowed from the pens of ascetics who resided for a while or were trained there, for example Gregory of Sinai, Gregory Palamas, Evtimii (a future Bulgarian patriarch) and Kallistos, a future patriarch of Constantinople. They recounted the lives and miracles of one another, composed texts for use in worship, denounced the Latins or polemicised with fellow orthodox over other theological matters such as the possibility of experiencing the Divine Light, a basic tenet of the hesychasts.55 The heavenly kingdom and the means by which individuals could train themselves for exposure to the divine – through prayer, contemplation and abstinence – were of paramount concern to these monks, transcending earthly dangers and 54 Shawcross, ‘The Chronicle of Morea’ (DPhil thesis, 2005); Shawcross (forthcoming b); see below, p. 772. 55 See below, pp. 823, 857; Meyendorff (1964); Meyendorff (1974c); Krausm¨ uller (2006).

46

general introd uction

Figure 1 The Holy Mountain, Athos

powers. Yet the Holy Mountain also acted as a focal point for orthodox potentates: the orthodox emperors in Constantinople and Trebizond, as well as Georgian, Bulgarian and Serb rulers, believed that veneration of Athonite monks in general, and patronage of individual houses in particular, offered them a means of gaining both God’s favour and their own subjects’ respect (see map 50 below, p. 873).56 If this polycentric orthodox world was riven by fierce political, territorial and ethnic rivalries, common religious beliefs, saints’ cults and axioms of church discipline maintained strands of unity. Athos and affiliated monasteries served as a ‘workshop of virtue’.57 The frequency of contacts between far-flung monasteries58 was facilitated by the proliferation of routes and affordability of travel that followed on from the Latins’ dominance of the Aegean and the Black Seas. The capacious 56 See the emperor of Trebizond’s explanation for founding a house on Athos in 1374: Actes de Dionysiou, ed. Oikonomides et al., p. 60. See below, p. 791. 57 Register des Patriarchats, ed. Hunger et al., II, no. 56, pp. 428–9; R´egestes des actes du patriarcat, ed. Grumel et al., no. 2309; Nicol (1979), p. 19; Meyendorff (1981), pp. 115, 128–30; Krausm¨uller (2006); below, pp. 827, 831. 58 Evans (ed.) (2004), pp. 11–12 (introduction); Goth´ oni (2004), pp. 60–4; Shepard (2006c), pp. 17– 18, 36–40.

ii. peri odisation and contents of t his book

47

ships of Italian merchants could ferry sizable parties of Rus churchmen from the lower Don to Constantinople, or on to Thessaloniki and thus the neighbourhood of Athos. And the wanderings of holy men such as Gregory of Sinai from Thessaloniki to Chios and beyond may well have been made in Italian vessels.59 Thus, paradoxically, a community of faith and spiritual role models gained in intensity and range both from the weakness of polities unable effectively to regulate sea traffic, and from the ubiquitousness and drive of Latin merchants and their trading partners in quest of profits. Not that many senior churchmen in the Constantinopolitan patriarchate or Athonite leaders saw merit in the fragmentation of earthly powers. Besides seeking individual emperors’ support in wrangles over hesychasm, church appointments and property-ownership, these churchmen upheld the idea of empire as an article of faith. It was more than a matter of finding a compliant figurehead at a time when the patriarchate’s own stock and organisation were riding high. Allegiance to a Christian Roman emperor on earth, and specifically in the ‘holy city’60 of Constantinople, was a characteristic that distinguished true Romans from mere Latins, whose brand of Christian observance seemed to bring them so many material advantages and sharp debating points. The empire that Constantine the Great had instituted was, after all, part of God’s design for the redemption of mankind, and those who stayed loyal to the idea were at the same time ‘true believers’, orthodoxoi. More positively, and less time-specific, senior orthodox churchmen could hold up the imperial order projected in Constantinople through ceremonial and liturgical worship as a kind of ‘icon’, prefiguring the divine order in heaven. Even if the late Palaiologan empire appeared to be confined within the City’s walls, the capital’s endurance of siege conditions had venerable precedents.61 The empire had repeatedly survived almost total submersion beneath alien occupiers and invaders, its enclaves standing out above the flood as a kind of archipelago (see below, pp. 226–7, 255–7, 259–60, 610– 12). The successive phases of fragmentation and territorial reconfiguration gained meaning and purpose from a standpoint attuned to liturgy, the constant re-enactment of sacred time by means of key texts and symbols in a church building, a miniaturised heaven (see above, p. 8). This perspective enabled churchmen and laity alike to see beyond current setbacks and material want to the ultimate victory of the emperor and all he stood for. Patriarch Antony IV (1389–90, 1391–7) voiced it in his letter to Grand Prince Vasilii I of Moscow (1389–1425) in 1393, when he insisted on the ‘commonality’ of the church and ‘the natural emperor, whose legislations 59 Kallistos, Life of Gregory of Sinai, ch. 15, ed. Pomialovsky, p. 33; Balfour (1982), pp. 44–7, 52–3; Majeska (1991), pp. 36–7. 60 MM, II, p. 361; R´egestes des actes du patriarcat, ed. Grumel et al., no. 3112. 61 Cameron, Averil (1979b).

48

general introd uction

and regulations and ordinances are held in regard across all the inhabited world’, in contrast to ‘particular local rulers’, like Vasilii, ‘besieged by the unbelievers and himself taken captive’, a dig at Vasilii’s recent spell as a hostage of the Golden Horde.62 By this line of thinking, which many on Athos shared, church and empire stood for ethical and political principles of universal validity, and Constantinople was still their exemplary centre. Whether Rus and other Slavic-speaking potentates fully subscribed to this line is questionable, but one should not underestimate the readiness of some of their clergy, at least, to put an exceptional valuation on the liturgical rites in St Sophia or to associate them closely with the emperor. Thus Ignatius of Smolensk wrote a detailed eyewitness account of the coronation there of Manuel II Palaiologos (1391–1425) in 1392. He interrupts his description of the liturgy to ask ‘who can express the beauty of this?’ in terms akin to those of Rus emissaries who had reported back to Vladimir of Kiev after witnessing a service in St Sophia some 400 years before: ‘We knew not whether we were in heaven or on earth . . . We only know that God dwells there among men.’63 For even the most educated Rus churchmen, Byzantine political culture was only a remote aspiration, but south Slav potentates were eager to appropriate details of Byzantine inauguration ritual to sacralise their own regimes. Translated texts containing the basic prayers and procedures are known from fourteenth- and fifteenth-century manuscripts: the translations into Slavonic were probably carried out in the fourteenth century, if not earlier.64 While these appropriations signal Bulgarian and Serb leaders’ ambitions to gain divine sanction for their own authority and for the right solemnly to delegate to subordinates in the manner of the basileus, they also imply a kind of gold-standard status for the rites of rulership celebrated at his court. This did not stop them from doing battle with the basileus’ armies or occupying his former territories and provincial towns, as the Bulgarian Ivan II Asen (1218–41), and the Serb rulers Stefan Uroˇs II Milutin (1282–1321) and Stefan Duˇsan (1331–55) did with panache.65 Divinely sanctioned authority was not, however, gained quite so straightforwardly. Overweening as individual potentates’ personal pretensions might be, many of their churchmen and subject populations still saw in the tsar’s court in Constantinople a model of legitimate monarchical rulership, even a reflection of the celestial order. As Alain Ducellier notes, the victorious Milutin effectively remodelled his court ceremonial and panoply of authority symbols on Byzantine lines at the time of marrying 62 63 64 65

MM, II, pp. 191, 192; tr. in Barker (1969), pp. 108, 109; Crummey (1987), p. 58; see below, p. 852. Ignatius of Smolensk, tr. in Majeska, Russian travelers, pp. 104–5, 110–11; PVL, p. 49; RPC, p. 111. Biliarsky (1993), pp. 125–7, 133, 139; Biliarsky (2001), pp. 72–4, 85–8. See below, pp. 788, 790–2, 801–2; Soulis (1984), pp. 6–11, 25–7, 35–47.

ii. peri odisation and contents of t his book

49

Simonis, daughter of Andronikos II Palaiologos (1282–1328), in 1299.66 Already wealthy, Milutin could now legitimately bedeck his wife and selected nobles in ‘imperial garments and gold belts’ and in imperial purple, sending them on progresses through his lands.67 The belief that preeminence rightfully belonged to the ‘emperor of the Romans’ resonated among the monks of Athos and even the most ambitious of Serbian predators, Stefan Duˇsan, had to take heed while appropriating Byzantine-ruled towns in Macedonia and claiming to be chief protector of the Holy Mountain.68 Stefan Duˇsan’s conquests and prestige owed much to the military failings and penury of the Byzantine empire. As Angeliki Laiou shows, these weaknesses were partly self-inflicted, a consequence of bitter divisions within the Palaiologan dynasty and civil war between the regents of a minor, John V Palaiologos (1341–91), and a formidable figure who for a while took the helm, John VI Kantakouzenos (1347–54) (see below, pp. 809, 810–11, 822– 4). The mid-fourteenth century saw an unmistakable turn for the worse in the empire’s fortunes, as pressures from Turks, Serbs and other external powers mounted, while revenues fell far short of the emperor’s outgoings. One symbolic indignity was the cessation of issues of gold coins from some point between 1354 and 1366 onwards: striking gold coins bearing his image had been a prerogative of the emperor in the New Rome’s heyday (see below, pp. 809–10). A mid-fourteenth-century observer bemoaned the loss of territories: ‘Now it is we who are enslaved by all those people who were . . . [formerly] . . . under our sway.’69 The Ottomans, in contrast, conducted a war-machine formidably well calibrated for continual operations. The Byzantine emperor became a tribute-payer and thus vassal of Sultan Murad I (1362–89), but this bought only temporary respite, and for nearly ten years from 1394 Constantinople was under siege (see below, pp. 827–8, 832). Deprived of a forceful legitimate monarch by Duˇsan’s untimely death in 1355, the Serbs’ new polity itself fell prey to internal dynastic rivalries and regional secessions, while the Serbs’ defeat at the battle of Kosovo in 1389 might suggest that the Ottomans were all but unstoppable (see below, p. 852); likewise with the Turks’ annihilation of a large crusading army at Nikopolis in 1396 under the leadership of Murad’s son and heir Bayazid I (1389–1402). The survival of the Byzantine empire into the fifteenth century could plausibly be put down to luck and its very harmlessness in Ottoman eyes. Yet the loose-knit, almost federal, empire of the Palaiologoi was not necessarily worst-adapted for obstructing the Ottomans. A case may even 66 67 68 69

´ See below, pp. 801–2; see also Malamut (2000), pp. 500–5; Cirkovi´ c, (2004), pp. 49–52. ˇ Danilo II et al., Zivoti kraljeva, ed. Daniˇci´c, pp. 96–7; Malamut (2000), p. 503. Obolensky (1971), pp. 255–6. ˇ cenko, ‘Alexios Makrembolites’, p. 213 (text), p. 225 (tr.). Sevˇ

50

general introd uction

be made for its resilience. From behind his Roman walls, the emperor could still seek out his enemy’s enemy in the diplomatic tradition of Justinian. The arrival of the Central Asian conqueror, Timur, in Anatolia in 1401 probably owes something to Manuel II’s d´emarches towards him in conjunction with western emissaries.70 Timur’s crushing of Bayazid’s army at the battle of Ankara in 1402 and the subsequent squabbles between Bayazid’s sons eased the pressure on Byzantium, and some Byzantines invoked another equally venerable tradition, the intervention of the Mother of God. The despatch of icons and relics – ‘reliquary diplomacy’ – was pursued with as much vigour by emperors and senior churchmen after Nikopolis as before 1396; their efforts were directed at both western and eastern sympathisers, potential providers of military manpower or treasure.71 Institutionalised links were forged in the late fourteenth century with the church organisations of nascent Wallachian and Moldavian principalities beyond the lower Danube, and as late as the 1430s the Serbian despot George Brankovi´c (1427–56) constructed a fortified residence at Smederevo on the lines of one recently built in the City walls at Constantinople by, most probably, his fatherin-law, Theodore Palaiologos Kantakuzenos.72 Other marriage-ties bound the Serbian political elite with that of the empire of Trebizond, and this network was, towards the mid-fifteenth century, extended to the Ottoman ruling family, too (see below, pp. 872, 874). Given the Ottomans’ problems with finding military manpower for the Balkans, and the limited number of Muslims residing west of the Aegean and the Bosporus in the later fourteenth and earlier fifteenth centuries, it was conceivable that the strands and strongholds of orthodox dynasts and supporting populations might be tweaked together in such a way as to thwart the ‘Ishmaelites’, denying them sufficient captives, plunder or revenues to maintain their war machine. If the Turks proved ultimately unstoppable, this owed much to the Ottomans’ methods of ‘harvesting’ Christian children and firing the ‘new army’ of janissaries with zeal for further conquests (see below, p. 858). The underlying ties of faith and allegiance between emperor, Greek-speaking Romans and even sometimes the Slavonic-speaking orthodox had survived earlier inundations and, when occupying elites and armies faltered, resurfaced with a vengeance (see below, pp. 785, 798–9). The empire without frontiers lost vital nutrients at grassroots with each successive ‘child levy’ and s¨urg¨un, haemorrhaging as debilitating as the holes blasted in Constantinople’s walls by Turkish guns in 1453.73 70 71

Obolensky (1963), pp. 47–52; Barker (1969), pp. 183, 504–9; Jackson (2005), pp. 238–40. Barker (1969), p. 408; Mergiali-Sahas, (2001a), pp. 56–9; Baronas (2004), pp. 85–7; Vassilaki (2005); Baronas (2007). 72 Shepard (2006c), pp. 26–8; Peschlow (2001), p. 401 and illust. 19. 73 Runciman (1965), pp. 97, 99, 104, 116–17; see below, pp. 858–9.

ii. peri odisation and contents of t his book

51

The care the Ottomans showed in drawing on the human resources of the empire’s former provinces is as revealing about Byzantium as it is about their own organisational talents. The Ottoman war machine was vastly more formidable than the one that had enabled the Fourth Crusaders to seize the City and, unlike the Crusaders, the Turks had long dominated its hinterland. But they needed to draw heavily and confidently on their Balkan possessions for revenues and manpower before taking on the task of capturing Constantinople and administering in and from it. They were not going to repeat the experience of the Crusaders, who had had to contend with Greek and Slav populations of, at best, uncertain loyalty, to the west and south-west of Constantinople. The forbearance of the sultans and their counsellors from attempting a direct assault in the first half of the fifteenth century was partly due to internal political tensions. But it also suggests that the Byzantine empire had other strengths besides the near-impregnability of its ‘reigning city’. Embers could still flare up in unlikely places and outliers metamorphose into new centres, as Mistra showed signs of doing with the help of its commercial and cultural ties with the Italian world.74 The sultans were assiduous in courting acceptance from Athonite monasteries by confirming their landed possessions and right to go their own spiritual way and, as Bryer shows, once Mehmed II (1444–6, 1451–81) had captured the City, he showed ambivalence in his quest for cooperation from senior churchmen, from the patriarchate downwards (see below, pp. 869, 871–2). At a material level, he confirmed the Genoese trading privileges within days of Constantinople’s fall (see below, fig. 65 on p. 867). The Genoese deal can, like Mehmed’s compact with the orthodox church, be viewed as a measure of the old empire’s decomposition, its unravelling into discrete ecclesiastical, monastic, regional and commercial sectors. Yet to dwell only on these negatives would be to overlook the variable geometry that had long been characteristic of the empire sans fronti`eres (see above, p. 3). Middle Byzantine emperors had mostly managed the balancing act between Greek-speaking religious orthodox insiders and other princes, populations and powers until the preponderance of western resources and organisational skills made the balance virtually unsustainable. The loose-knit, dynastic mini-empires emerging after the catastrophe of 1204 were structured differently from their illustrious predecessor, and the ‘emperor of the Romans’ reinstated in Constantinople in 1261 could not call up the administrative or military apparatus of the past. In fact the malfunctioning of late imperial governance was the despair of some of those who sought to operate it or who had written on its behalf,75 while from 74 On the ‘half-way house between a Greek polis and an Italian renaissance seigniory’ envisaged by its leading thinkers, some of whom hoped for military aid from the west, see Ronchey (2006), pp. 321–2. 75 See, e.g. Sˇ evˇcenko, I. (1961); Nicol (1979), pp. 75–83; Kolbaba (1995).

52

general introd uction

the end of the fourteenth century, numerous craftsmen – goldsmiths, gold wire-drawers, shipwrights and also medical doctors – saw better prospects in the west and set up successful enterprises as far afield as London.76 The imperial order did not, however, survive by institutions alone. In its capacity to engage the sympathies, belief or commercial concerns of quite disparate, scattered groupings the Palaiologan empire showed a certain continuity with its earlier incarnation. Not for the first time the patchwork qualities of the Byzantine empire made outright conquest and long-term occupation by even the most resolute outsiders an expensive, potentially self-defeating business. The Ottomans’ step-by-step approach to the conquest of the Byzantine empire and its affiliates bears witness to this. So do the studied ambiguities and concern for legitimacy in the eyes of their new subjects of Mehmed the Conqueror and his immediate successors. 76

Harris (1995a), pp. 156, 164–88.

I N T R O D U C T I O N – PA RT i i i

OT H E R R O U T E S T O B Y Z A N T I U M jon at han s he pard

Our chapters touch on many matters and subject-areas handled at greater length elsewhere and, without aiming to be comprehensive, some of the more important alternative approaches to Byzantium are outlined below. For the most part, only fairly recent publications will be mentioned, as their bibliographies usually cite earlier studies. church history The Byzantine church’s history has been expounded by scholars in connection with ideology, political affairs and relationships with other churches, the church being considered as administrative institution and more generally, as element in urban and rural society.1 Likewise monasteries great and small, together with monks as individuals and as groups, have been studied from numerous angles: as property-owners, spiritual oases in provincial and Constantinopolitan society, sober counsellors or individual troublemakers. Besides the useful general introductions to the editions of documents from the archives of Mount Athos, collections of studies on Athos, saints and individual monasteries have been published; and monographs have been dedicated to holy fools and to the relationship between monks and laymen.2 The broader spectrum of eastern Christian belief, worship, everyday experience and expectations is also receiving scholarly attention, and contributions relating to the Byzantine world feature in volumes dedicated to medieval Christianity in general.3 An entire volume of the Cambridge history of Christianity is dedicated to eastern Christianity after c. 1050.4 1 Dvornik (1966); Runciman (1977); Nicol (1979); Hussey (1986); Herrin (1987); Angold (1995); Morris (ed.) (1990); Dagron (2003). 2 See the series of publications of the archives of the individual monasteries on Athos, e.g. Actes de Lavra, ed. Lemerle et al., I, pp. 13–48 (introduction); Actes du Protaton, ed. Papachryssanthou, esp. pp. 17–109 (introduction); Hackel (ed.) (1981); Mullett and Kirby (eds.) (1994); Mullett and Kirby (eds.) (1997); Morris (1995); Bryer and Cunningham (eds.) (1996); Ivanov (2006). 3 See Dagron et al. (eds.) (1993); Vauchez et al. (eds.) (1993); Mollat du Jourdin et al. (eds.) (1990); Krueger (ed.) (2006); Noble and Smith (eds.) (2008). 4 Angold (ed.) (2006).

53

54

general introd uction

Congregational worship, priest-led ways of communicating with God, entering into His presence and gaining the intercession of the saints, were of vital concern to the Byzantines, from emperors to provincial peasants, and there are authoritative guides to the liturgy and church services.5 But individuals – whether monks or laypersons – also sought immediacy with the holy for themselves, and relics and icons offered access: pilgrimages to shrines were a feature of Byzantine life, and relics and relic-containers of one sort or another were prized in the imperial palace and at grass-roots, whether to bring spiritual fulfilment, physical salvation or simply material wellbeing.6 visual media Icons – more or less formulaic likenesses of otherworldly beings, sacred events and scenes – offered the Byzantines access to the holy par excellence, and although reviled as idols by some emperors (see below, pp. 278–84), they became engrained in private piety and collective imprecation. After the Mother of God’s protection of her City of Constantinople in the seventh century, icons representing her were revered and, eventually, panel icons were processed regularly through Constantinople’s public spaces, helping to render them and the City yet more sacred.7 Icons were deemed truer than words in conveying the divine. The sense that their contrasting brightness and shade, yet stable basic forms, could relay sacred happenings and communicate spiritual essentials was strong; it is notable in, for example, late Byzantine art, when directly experiencing the energies and uncreated light of God was the ambition of prominent ascetics.8 Integral to private devotions, ritual routines and theological truths, icons were painted on wood or walls, or portrayed in mosaics, ivory or metalwork, and from the ninth century onwards the beings on them were generally identified by inscriptions.9 Significantly, they were not sharply distinguishable in style from images of emperors, past and present, and an emperor 5 6

Taft (1978); Taft (1992); Taft (1984). Horn´ıcˇkov´a (1999); Durand and Lafitte (eds.) (2001); Durand and Flusin (eds.) (2004); Wolf et al. (eds.) (2004); proceedings of a symposium on ‘Pilgrimage in the Byzantine empire, 7th–15th centuries’, published with an introduction by A.-M. Talbot, DOP 56 (2002), pp. 59–241; Lidov (ed.) (2003); Klein (2004); Gr¨unbart et al. (eds.) (2007). 7 Sevˇ ˇ cenko, N. P. (1991); Sevˇ ˇ cenko, N. P. (1995); Angelidi (1994); Angelidi and Papamastorakis (2000), Weyl Carr (2000) and other contributions in Vassilaki (ed.) (2000); Papaioannou (2001); see below, n. 46, p. 129); Vassilaki (ed.) (2005); Gerstel and Talbot (2006), p. 87; Lidov (2006). 8 See below, p. 823; James (1996), pp. 80–5, 96–101, 117–23, 139–40; Cormack (2000); Franses (2003), p. 823; contributions in Evans (ed.) (2004). 9 On icons and the Byzantines’ ways of painting and viewing them and relating them to texts, see Mango (1963b); Mango, Art of the Byzantine empire; Talbot Rice (1968); Maguire (1981); Maguire (1996); Belting (1994); Cormack (1985); Cormack (1997a); Cormack (2000); Rodley (1994); Cutler and Spieser (1996); Lowden (1997); Mathews (1998); Brubaker (1999a); Barber, C. (2002).

iii. other routes to byz antium

55

could be shown in the company of Christ or a saint (see below, fig. 33, p. 154). A particularly fine mosaic of Christ graced St Sophia from soon after Michael VIII Palaiologos (1258–82) restored empire to Constantinople (see below, fig. 58, p. 826), while Michael demonstrated the imperial presence at newly regained points through wall-paintings, as at Apollonia, south of the strategic base of Dyrrachium (Durazzo) on the Adriatic coast (see below, fig. 57, p. 800). Michael VIII’s projection of his authority far and wide through visual media belongs to a great tradition, involving coins, seals and the minor arts, reaching back beyond Justinian to the heyday of imperial Rome. The ways in which the emperor and his order were portrayed and idealised are discussed and illustrated in specialised but accessible studies as well as in more general works.10 That beauty and superlative technical expertise should be attributes of imperial power was a tenet of Byzantine thinking until virtually the end. Constantine VII Porphyrogenitus (945–59) could claim that ‘all beauty and adornment had been lost to the empire’ for want of due attention to ceremonial. He was in fact taking a sideswipe against his detested former co-emperor, who had manipulated political imagery against him.11 The grand halls for the reception of visitors, the gardens, feasts, exotica and religious rites experienced, and the ‘diplomatic gifts’ presented at court or sent to notables and potentates further afield have enjoyed considerable scholarly attention.12 The Constantinopolitans’ penchant for dignifying workaday or dilapidated buildings with silks and other splendid hangings has also been noted. Wealth in this flexible – and portable – form became the hallmark of the elite. The minor arts and ceremonial could cover for the limitations and condition of structures of brick and stone. This held true not only of the capital but also of citadels in ancient cities and strongholds in outlying regions, which could be reoccupied and refurbished when threats loomed.13 The authorities’ alertness to the impact of sights on outsiders is registered in a text for receiving envoys in the capital: if they came from greater powers, they were to be shown the ‘masses of our men, good order 10 See below, pp. 111, 207, 273–4, 501–3. See also, beside the classic work of Grabar (1936), Spatharakis (1976); Walter (1978); Magdalino and Nelson (1982); Cormack (1985), pp. 179–214; Brubaker (1985); contributions in Evans and Wixom (eds.) (1997); Ousterhout (2001); Grierson (1982); contributions to the series SBS; Cheynet (2005). 11 DC, preface, ed. Reiske, p. 4; ed. and French tr. Vogt, I, p. 1; see below p. 509 and fig. 32. 12 Cormack (1992); Lowden (1992); Maguire (ed.) (1997); Cutler (2001), pp. 261–4; Schreiner (2004); Littlewood et al. (eds.) (2002); Anca (2005); Prinzing (2005); Reinsch (2005); Tinnefeld (2005a); Bardill (2006); Bauer (2006); Featherstone (2006); Klein (2006); Luchterhandt (2006); Schreiner (2006); Maguire and Maguire (2007), pp. 29–57. 13 Holmes (2002a), pp. 97–9, 103–4; Morris (2003), pp. 244–9; Haldon (2005c), pp. 77–8; Featherstone (2006); see also below, p. 486.

56

general introd uction

of our weaponry and the height of our walls’.14 In the empire’s later years, mosaicists could still portray in St Sophia the emperor wearing a crown and vestments replete with gemstones. Yet, as Nikephoros Gregoras deplored, his actual crown and vestments were ‘make-believe (phantasia)’, ‘made of gilded leather . . . and decorated with pieces of glass of all colours’. Here again, one art or craft could substitute for another in the imperial kaleidoscope, to keep up appearances. A peculiarly Byzantine blend of faith, self-belief and expectations of ultimate vindication underlay such improvisations.15 The choicest of the visual arts, crafts and architecture were reserved to display imperial majesty, superlative craftsmanship and beautiful artefacts denoting possession of supernatural powers and legitimate authority. Some of the highest-quality imperial silks named their place of manufacture near the Great Palace or the emperor reigning when they were made.16 Such association of extraordinary skills, technical and aesthetic, with hegemony is characteristic of numerous pre-industrial societies,17 and to many Byzantines reverence for the emperor appeared interwoven with service of God, however firmly churchmen drew the line. By and large the imperial authorities and the leading monks and churchmen were, from the mid-ninth century onwards, in alignment as to what was acceptable ‘official’ and religious art. Their command of skills and resources meant that they could set the tone and contents of the more elaborate, public examples of the visual arts. The forms, decorative programmes and ritual significance of ecclesiastical and monastic buildings have received scholarly attention, and the prominence of churches in studies on Byzantine art and architecture is not wholly an accident of survival: the empire was well- (if not over-)stocked with churches and monasteries from at least the time that Justinian was building more churches in Constantinople than strictly pastoral needs warranted.18 But not all buildings were commissioned by churchmen or the imperial authorities. Private secular architecture after the seventh century is known to us only from occasional mentions in literary sources and from archaeology. Further excavations should shed light on the material facts of life in Byzantine towns and even, eventually, in rural settlements, which have mostly as yet only been identified from field

14 Peri strategias (‘Strategy’), ch. 43, ed. and tr. Dennis, Three Byzantine military treatises, pp. 124–5; tr. in Lee and Shepard (1991), p. 30. 15 NG, XV.11.4, ed. Schopen and Bekker, II, pp. 788–9; German tr. van Dieten, III, pp. 170–1; Hetherington (2003), pp. 164–5. 16 Lopez (1945), p. 7; Muthesius (1995), pp. 56–64; Muthesius (1997), pp. 34–43. 17 Helms (1993); Trilling (1997). 18 Mango (1990), p. 52. See also Talbot Rice (1968); Beckwith (1979); Krautheimer and Curˇ ´ ciˇc (1986); Mango (1979); Lowden (1997); Rodley (1985); Rodley (1994); Freely and C ¸ akmak (2004); Cutler and Spieser (1996); Mathews (1998); Ousterhout (1999); Cormack (2000).

iii. other routes to byz antium

Figure 2 The walls of Constantinople, often repaired but basically late Roman in design and technique

57

58

general introd uction

surveys.19 Likewise collation of excavated artefacts with long-studied objets d’art, wall-paintings or even manuscript illuminations is beginning to highlight other kinds of subject-matter in the representational arts, unofficial visual statements which could veer far from the ‘party-line’ of court orations, sermons and other literary set pieces. Ceramics can be particularly eloquent in revealing the fancies, fantasies and humour of Byzantines having little or no connection with the imperial-ecclesiastical establishment.20 literature We do not glean very much about society or life in general in towns and settlements outside the capital from surviving literature, that is, from writings in Greek composed for more than ephemeral purposes. No term in use among the Byzantines corresponds precisely with our ‘literature’ and what they wrote has been termed a ‘distorting mirror’, designed to reflect other than reality.21 Works recounting the deeds and reigns of emperors could amount to extended narratives, purporting to be ‘Histories’ while retaining strong rhetorical traits, for example the Life of Basil (see below, pp. 292, 294). Such works tended to emanate from court circles, whereas chronicles, less polished presentations of events, often from a religious angle, were less committed to an establishment viewpoint, and were much read (see below, pp. 82, 103). The Byzantines’ writings vary greatly in intricacy of style and in the kind of Greek they use, and fashions and preoccupations changed over time. Rhetorical and grand historical works were written in classical – ‘Attic’ – Greek, for reading or declaiming primarily among members of the metropolitan elite. Thanks to private secondary schooling, the handful of senior officeholders, churchmen and scholars were at home with an all but dead language far removed from the everyday Greek spoken in the countryside or even in the capital’s streets.22 Authors writing in these circles presupposed familiarity with the antique world23 but could cross-cut to figures or themes from the Scriptures or to sayings from the church fathers. The collections made of these sayings, like the full-length chronicles, some sermons and many saints’ Lives, tended to be written in plainer Greek,24 more akin to the spoken word. 19 Whittow (1995); Whittow (1996b); Bouras (2002); Sanders (2003), pp. 396–7; Bakirtzis (2003), pp. 54–6, 64; contributions to Dark (ed.) (2004); Dark (2004); Dark (2005); see also below, pp. 477–8. 20 Maguire and Maguire (2007). Marginal drawings and paintings in manuscripts could also convey orthodox messages vividly, even grotesquely: Corrigan (1992). 21 Mango (1975b). 22 See below, pp. 86, 212, 238, 511–12. See the letter-collection of a tenth-century Constantinopolitan teacher: Anonymi professoris epistulae, ed. Markopoulos (contents summarised by Browning (1954), pp. 402–25). See also Lemerle (1986); Constantinides (2003). 23 Hunger (1969–70). 24 For saints’ Lives, see Pratsch (2005a), pp. 405–7.

iii. other routes to byz antium

59

This sprawling, still partly unpublished, body of literary materials is not easy to categorise, and perhaps the most authoritative general history of Byzantine literature remains that of Karl Krumbacher.25 Nonetheless, several histories of branches of Byzantine literature are available, as are histories of particular periods,26 and the later twentieth and early twenty-first centuries saw studies on the subject burgeoning. Some are wide-ranging survey projects, or introductions,27 while others examine Byzantine rhetoric, poetry and letter-writing,28 besides more technical issues such as palaeography, epigraphy and the nature and uses of Byzantine books (codices) and libraries.29 Byzantine literature and texts written in Byzantine Greek are more approachable by students, now that the classical Greek–English Lexicon of Liddell and Scott is reinforced by such works as the Lexikon zur byzantinischen Gr¨azit¨at.30 It is becoming clear that poems such as the tale of the border-lord Digenis Akritis (in its surviving versions) are the product of complex interplay between litt´erateurs in the capital and the composers of stories and ballads and reciters of songs at popular level.31 Some acquaintance with letters might be expected at village level, and while the priest was likeliest to be capable of functional literacy, laypersons could have reading skills, or access to social superiors possessing them, for example through confraternities.32 It was perhaps partly via confraternities or comparable groups that texts in everyday Greek recounting visits to the next world and visions of the wicked receiving punishment circulated. It is quite possible that they were countenanced by churchmen, venting grievances about the workings of church and secular administration, yet counteracting dissidents overtly opposed to the imperial order.33 Such a cellular structure of orthodoxy has to be deduced, and is not directly attested in our sources, yet it probably constitutes an important strand in the fabric of Byzantine society. Such

25 26

Krumbacher (1897). Beck (1959); Beck (1971); Politis (1973), pp. 1–43; Hunger (1978); Kazhdan and Franklin (1984); Kazhdan (1999). 27 Agapitos (1991); Littlewood (ed.) (1995); Beaton (1996); Odorico and Agapitos (eds.) (2002); Odorico and Agapitos (eds.) (2004); Rosenqvist (2007). 28 Hatlie (1996); H¨ orandner and Gr¨unbart (eds.) (2003); Alexiou (1982–3); Alexiou (1986); Lauxtermann (2003–7); Maguire (1981); Mullett (1997); Mullett (2003); Dennis (1997); Littlewood (1999); Jeffreys, E. (ed.) (2003); Jeffreys, E. (2007). 29 Wilson (1996); Cavallo et al. (eds.) (1991); Cavallo and Mango (eds.) (1995), De Gregorio and ˇ cenko, I. (2002). Kresten (eds.) (1998); Waring (2002); Sevˇ 30 LBG. 31 Politis (1973), pp. 23–5; Beaton and Ricks (eds.) (1993); Digenis Akritis, ed. and tr. E. Jeffreys, pp. xiv–xviii, xli–xlix, liv–lvii (introduction). 32 Browning (1978); Holmes (2002b); Holmes and Waring (eds.) (2002); Jeffreys, E. (2007), pp. 169– ˇ cenko, N. P. 70; Confraternity of Thebes, ed. and tr. Nesbitt and Wiita, pp. 373–9; Horden (1986); Sevˇ (1995). 33 Baun (2000); Baun (2007); Baun (2008).

60

general introd uction

hidden strengths of the empire are what Byzantine literature in its broadest sense can intimate. army and ad ministration The institutions comprising the army, tax-collection and other administrative apparatus and the law are more familiar. Some deliberately evoked ancient Rome, and inscriptions on coins – themselves a clear symbol of continuity – styled rulers ‘emperors of the Romans’ from around 812 onwards (see below, fig. 28 on p. 418). The organisation and role of the Byzantine navy have been set out in authoritative works.34 But the army has received the lion’s share of scholarly attention, in part reflecting the coverage of military matters in Byzantine literary sources. Military history features in many of our chapters, and the tactical manuals available in translation are noted below (see below, pp. 87–9). The formal units, prescribed methods of fighting and even some pay rates are known from snapshots in particular sources, and certain developments, the metamorphoses of the seventh and eighth centuries and the revival of large-scale offensive warfare in the tenth, are beyond reasonable doubt.35 Likewise with the retrenchment carried out by Alexios I (1081–1118); the capability of Manuel I Komnenos’ (1143–80) forces; and the robustness of the armed forces in Lascarid Nicaea and during Michael VIII’s Constantinopolitan regime.36 Nonetheless, major questions about the army remain unresolved and sometimes contentious. Aspects of the arrangements for maintaining a pool of operational and potential military manpower in the provinces are opaque, probably because of their flexibility and the late date when they were formally codified. But it is clear that for a full-time core force, iconoclast emperors and their successors relied on ‘Byzantine praetorians’, elite units generally stationed in or near the capital; and to be enrolled in the military registers in the provinces brought remuneration and status as well as potentially heavy obligations.37 More controversial is the question of the armed forces’ size in the medieval period. The figures provided by contemporary Arabic writers and occasional Byzantine references would suggest operational field armies of 80,000 or more. But such figures jar with Byzantine chronicles’ assumptions about the difficulty of campaigning on more than one front at 34 35

Ahrweiler (1966); Pryor (1988); Pryor (2002); Pryor (2003); Pryor and Jeffreys (2006). See below, pp. 236–7, 239–41, 266–9, 517–18. See also McGeer (1988); McGeer (1991); McGeer (1995); K¨uhn (1991). See also, more generally on the earlier and middle Byzantine army, Treadgold (1992); Treadgold (1995); Scharf, (2001); Haldon (1999a); Haldon (2001a); Haldon (ed.) (2007). 36 Bartusis (1992); Birkenmeier (2002). See also below, pp. 612, 619–21, 716–17, 747, 749. 37 Haldon (1979); Haldon (1984); Haldon (1993); see also below, pp. 268–9.

iii. other routes to byz antium

61

a time, and an abiding imperial concern was to impress upon outsiders that Byzantine armies were larger than in fact they were.38 The discrepancies in figures probably reflect not only imperial disinformation, but also actual fluctuations of various kinds – in the empire’s population size; in the number of units of outsiders employed for short-term campaigning; and in the authorities’ resort to ad hoc call-ups of all remotely serviceable males. Such call-ups might be made in dire emergencies, or even for occasional offensives.39 Arms-bearers originating from societies attuned to violence played an important part in maintaining the empire’s security from Justinian’s era onwards, the Armenians being pre-eminent.40 They seldom receive extensive attention in Byzantine narratives; even the 6,000 or so Rus warriors sent to the aid of Basil II c. 988 are known to us mainly from non-Byzantine sources (see below, p. 525). This was an era of imperial expansionism, but in earlier periods, too, externally based warriors were employed for specific operations, temporarily swelling the ranks of imperial forces. The question of the figures for the Byzantine armed forces bears heavily on the history of the empire’s administration. The forces were the largest item of expenditure, providing much of the raison d’ˆetre for the apparatus for raising revenue and spending it. If, as seems likely, the empire could get by with modest-sized, highly disciplined armed units for much of the time, counting on a modicum of cooperativeness from eligible military manpower, suppliers and carriers in those places under threat, financial outlay was correspondingly limited. This combination of cost-effectiveness and reliance on cooperative locals lessened the need for a sizable administrative apparatus. Direct supervision from the capital could be focused on the districts that were more fiscally lucrative or the most strategically important, a form of ‘hot-spots’ and ‘cold-spots’ or inner and outer zones of governance discernible in varying permutations and regions throughout Byzantium’s history (see below, pp. 498–501, 653–4, 664–5, 668, 827–8). The outlines of central administration from the late seventh and eighth centuries on are only dimly discernible. They seem to comprise departments of senior officials dedicated to particular tasks such as revenue-raising or expenditure, but with overlapping functions and without a firmly cast hierarchy of great offices of state.41 Their activities could be readily scrutinised by the emperor and his closest associates and counsellors, a costeffective form of flexible ‘flat-management’ provided that the volume of business was fairly limited, the emperor or his closest associates reasonably 38 Compare Treadgold (1995), pp. 64–78 with Haldon (1995b); Haldon (1999a), pp. 101–6; see above, pp. 55–6 and n. 14. 39 See below, pp. 265–9, 502; Haldon (1993); Haldon (1999a), pp. 105–6, 234–7. 40 Charanis (1961); see below, pp. 124, 168, 337, 357–8, 364, 665. 41 See below, pp. 238–9, 273; Brandes (2002a); Haldon (2003a).

62

general introd uction

assiduous. The names of the higher or more durable offices are known to us. But details come mainly from the orders of precedence of title- and office-holders at palace receptions, and we lack texts clearly setting out functions and lines of accountability in full.42 This deficiency is partly made up for by the survival of many lead seals belonging to senior officeholders. A major step towards matching such seals with what else is known of the central administration was made by Vitalien Laurent, followed up by other sigillographers, and the series Studies in Byzantine sigillography, notably volumes 7 and 8, offers useful additions and updates. Work on the prosopography of the middle Byzantine period is collating seals with what the written sources relate about individuals’ career patterns. This can yield statistical data as well as case studies of individuals working in the administration, and the online database is designed to offer means of access to non-specialists.43 The forementioned orders of precedence also list the strat¯egoi and other senior officials serving in the provinces but expected to attend court functions quite regularly. Collation of these with Byzantine narratives and Arabic sources yields a rough picture by the ninth century. The strat¯egoi were military commanders at the head of armed units. Their judicial, levying and requisitioning powers were sweeping but did not permanently supplant other, more painstaking forms of tax-collection: this was primarily the task of officials answerable to the administration in Constantinople.44 The scope of the strat¯egoi within their respective themes is not wholly clear, and the territorial extent of the themes is seldom delineated precisely, perhaps because they were slow to assume fixed, territorial form. One clear development is the creation of smaller command units, known as kleisourai (literally, ‘passes’), to firm up defences in the Taurus mountain regions.45 Towns and other fortified population centres were fixed points in later seventh- and eighth-century administration, being also the likely sites of apoth¯ekai, state depots for storing revenue proceeds such as grain, and for issuing supplies and probably also equipment to soldiers.46 But the dealings, formal and informal, of state agencies with outlying countrydwellers emerge from our sources only fitfully. The authorities could seldom 42 43

LPB (containing an extensive commentary); useful tables of functions in Haldon (2005c). Laurent (ed.), Corpus des sceaux, II (= L’Administration centrale) is a collection of seals of central officeholders; DOS; Seibt and Wassiliou, Byzantinischen Bleisiegel, II; SBS 7–8; PMBZ, I (for prosopography to 867) and II (to 1025, forthcoming); Prosopography of the Byzantine world (http://www.pbw.kcl.ac.uk). 44 See below, pp. 269–71; JG, tr. McGeer (introduction); Oikonomides (1996a); Oikonomides (2002), pp. 995–1004; Brandes (2002a), pp. 505–10; Haldon (2003a). 45 ODB, II, p. 1132 (A. Kazhdan); Skirmishing, ed. and French tr. Dagron and Mih˘ aescu, pp. 219, 240–3 (commentary); Haldon (1999a), pp. 79, 114. 46 On the role of the apoth¯ekai between c. 650 and c. 730, see below, pp. 271–2; Brandes (2002a), pp. 300–5, 418–26, 505.

iii. other routes to byz antium

63

guarantee full protection to those far removed from strongholds or fortified refuges.47 l aw and justice Something of the way in which peasant-proprietors were expected to resolve issues of property-ownership, animal husbandry, theft, injury or damage emerges from the Farmer’s law (or Nomos georgikos), a text whose date of composition and status remain open to discussion. At any rate, the Farmer’s law seems to have long been a working document, laying down norms for dispute settlement within local communities. The prescriptions are detailed and presuppose regular taxation, implying governance that was loose-meshed but under the authorities’ ultimate oversight. The text was later translated into Slavonic.48 It is in key with procedures set out in two tax-collectors’ handbooks, and is not inconsistent with the texts concerning methods of measuring land for purposes of tax assessment. The latter seem to have been at their most accurate in measuring smaller plots.49 The handbooks imply that the individual contributions towards the tax burden imposed on a fiscal unit were ultimately for its members to determine among themselves. The government’s concern was that the tax be paid, due allowances being made for lands devastated by enemy raids and abandoned by their owners: these were eventually – usually after some thirty years of non-payment of taxes – declared klasmata and they could be reallocated by the state, through sale, renting-out or gift. The productive value of these lands was reviewed from time to time, keeping the central administration abreast of changes – and potential gains for its coffers: klasma-land could be sold by the state to new, tax-paying proprietors.50 The texts relating to taxation offer the viewpoint of officialdom, but the dynamics of the middle Byzantine economy and society glimmer through their assertions and prescriptions. The quality of justice and the workings of the law in Byzantium are no less murky, but modern studies shed some light.51 Here, too, affairs in the capital are far better illuminated than elsewhere, and while the Basilika 47 48

See below, pp. 265–6, 498–9, 502. Farmer’s law, ed. and Russian tr. Medvedev et al.; ed. and tr. Ashburner; see below, pp. 264, 488–9; ODB, II, p. 778 (A. Kazhdan); Lefort (2002), pp. 279–81; G´orecki (2004). 49 D¨ olger, Beitr¨age, pp. 114–23; tr. in Brand (1969), pp. 48–57; Karayannopulos, ‘Fragmente’, pp. 321– 4; tr. in Brand (1969), pp. 57–60; G´eom´etries du fisc byzantin, ed. and French tr. Lefort et al., pp. 223–4, 235, 252–5, 263–5 (commentary); Lefort (2002), p. 272; Oikonomides (2002). 50 On klasmata and the government’s concern with restoring cultivation and revenue yield from unproductive lands, see G´orecki (1998), pp. 244–54; JG, tr. McGeer, p. 14 (introduction); Lefort (2002), pp. 281–3; Oikonomides (2002), pp. 995–6; Morris (2006b), pp. 25–30. 51 Laiou and Simon (eds.) (1994); Karlin-Hayter (1990); Stolte (1998); Macrides (1999); Stolte (2003– 4 [2005]).

64

general introd uction

project of revising Justinian’s corpus of laws begun in Basil I’s reign laid down markers for the entire empire, some novellae issued by Leo VI were primarily concerned with Constantinople, as was the Book of the eparch (see below, pp. 301–2, 497–8). That written rulings were being issued by senior officials according to principles of Romano-Byzantine law in distant borderlands is indicated from southern Italian materials, and Athonite beneficiaries of tax exemptions and other imperial privileges were, in the eleventh century, taking care to have them confirmed by successive new regimes.52 How disputes were settled among peasants at grass-roots, and the redress available to them in the event of unlawful actions by the well-connected ‘powerful’, are harder to track down. This bears on the general question of the mesh of imperial administration at grass-roots.53 There is reason to think that in some borderlands and newly acquired regions in the tenth and eleventh centuries power structures were left largely intact, with local elites or administrators raising exactions and resolving disputes with few departures from past practice.54 A degree of devolution was customary in the Greek-speaking zones of the empire, too, and diverse rivalries were played out among members of local elites. The already well-connected could pull strings at provincial level or in the imperial court in Constantinople; the newly well-to-do could purchase them, with an eye to further enhancing their local position. Or the rights and possessions of lesser folk could be overridden roughshod, without judges or other officials lifting a finger.55 Loose-meshed as local self-governance may have been, courts of justice and other embodiments of imperial solicitousness were not invariably beyond the reach of provincial smallholders with a grievance or under unlawful pressure. The fertile lands of the western Asia Minor theme of the Thrakesioi long remained largely the preserve of smaller proprietors, and the prosperous region of Thebes, while partly in the hands of substantial landowners, still accommodated proprietors of more modest means in the eleventh century (see below, p. 489). We lack direct evidence that this was due to legal process and regard for the spirit of the laws. But where archival evidence survives, in the form of the deeds involving Mount Athos’ monasteries, there are indications that communities of peasants did not always complain in vain.56 The Athonite monks became major landowners in eastern Macedonia from the tenth century onwards, enjoying direct access to imperial circles. Yet the tax exemptions and other privileges for their properties which emperors issued did not spare them judicial investigations and 52 53 54

Morris (1986), pp. 135–7, 143–6; Morris (1995), pp. 140–2, 296–7. The breadth of the mesh is emphasised in Neville (2004). See, however, Morris (2006a). Stephenson (2000); Holmes (2001); Holmes (2005), pp. 368–91, 440–7; see below, pp. 570, 668– 70, 706. 55 Neville (2004), pp. 105–18, 136–56. 56 Morris (1986), pp. 131–5, 141–6; Morris (2006a). See also Magdalino (1994).

iii. other routes to byz antium

65

hearings, conducted by local judges or at Constantinople. Findings did not always favour the well-connected.57 Comparable patterns may emerge from further investigation of the eleventh-century Peira. Peasants and other unprivileged persons are depicted as vulnerable to encroachments and unlawful seizure of their lands, animals or chattels from the ‘powerful’. But the judge Eustathios Rhomaios’ rulings document how he sought to adjudicate disputes over crops and boundaries between what were sometimes quite small-scale proprietors, applying principles of Romano-Byzantine law to current circumstances. The well-connected or well-to-do had the advantage, but the courts could redress the balance.58 The Peira’s rulings were expected by its compiler to apply throughout the empire and the text seems to have been much consulted. This may corroborate the impression that reports of the demise of peasant-proprietors are much exaggerated; novellae declaring the worth of a prosperous peasantry to fisc and army were not necessarily dead letters.59 These issues are material not just to legal or economic history, but also to the empire’s capacity to continue raising taxes on lands and possessions through thick and thin. The unchronicled majority of the population, peasant-proprietors raising livestock and growing wheat and other crops, were the source – by way of land taxes – of the greater part of the state’s regular incomes. Equally, chronicles and saints’ Lives depict fertile regions like, for example, Asia Minor’s Aegean coastal areas as vulnerable to external raiders in the earlier tenth century, while Thessaly’s plains and other ‘hotspots’ underwent Hungarian and Bulgarian incursions through the later years of that century. And yet, as Rosemary Morris observes, ‘Byzantine bureaucrats in Constantinople and their provincial representatives soldiered on’,60 tax assessments were negotiated, and revenue streams trickled in. The unarticulated nexuses of local pride, peer-group pressure and religious belief accompanying this anomaly underpinned the empire and those who – voluntarily, habitually or perforce – maintained them made up Byzantium’s ‘silent majority’. They are not treated in detail here, but their existence underlies the chapters that follow. societ y: gend er and eunuchs The social fabric to which Byzantium owed its resilience drew on diverse human resources and the nature of that diversity is worth considering. An 57 58

Morris (1986), pp. 146–7; Morris (2006b), pp. 34–7. Peira, XL.12, XLII.18–19, XL.1–4, ed. Zacharia von Lingenthal, pp. 167, 177–8, 165–6. For a less sanguine interpretation of the Peira, see Litavrin (1977), pp. 179, 187–90, 193. See also Oikonomides (1986b); ODB, III, pp. 1617, 1793 (D. Simon); Magdalino (1994), pp. 102–5. 59 See below, pp. 489, 492; JG, ed. Zepos and Zepos, I, p. 209; ed. Svoronos, no. 3, p. 85; tr. McGeer, p. 55. 60 Morris (2006b), p. 23; see below, pp. 500, 525–7.

66

general introd uction

elemental difference is that between men and women. The Byzantines’ assumptions and demarcations on matters of gender are now receiving attention, as are the specific experiences and activities of women.61 In some respects, such as life expectancy at birth, their condition seems to have resembled men’s – living to between their late twenties and early thirties – with expectancy rising markedly (to perhaps their late forties) for those surviving their first five years on earth.62 These estimates apply, however, to the early fourteenth century. Demographic and other social and economic data for the middle empire, from the eighth to the thirteenth centuries, is scanty for men and women, while the data available for the early period is not really comparable.63 Exegesis and comparison of the roles of men and women in Byzantine society and culture with due allowance for all the variations in class, place and time is correspondingly difficult. In the fifth and sixth centuries, women of wealth, status and also position in public life are quite well attested,64 and it is no accident that Theodora is portrayed with her female retinue on one side of the sanctuary of the church of San Vitale in Ravenna, in equal majesty to her husband (see below, fig. 8b, p. 211). But for subsequent centuries, the picture darkens in nearly every sense. Already under Justinian, the church’s influence on imperial laws was becoming marked, with a ban on the performance of judicial duties by women and abandonment of divorce by mutual consent.65 Thereafter, little evidence survives for verification of the restrictions on women imposed by canon law, or the idealised portrayals of holy women in their Lives.66 Not that the picture is wholly dark. Women retained the right to own extensive landed properties as well as chattels during the middle Byzantine period, and strong-minded individuals of substance occasionally surface in narrative sources, for example the widow Danelis.67 Lower down the social scale, scraps of archival information such as tax registers take for granted the role of women – often but not invariably widows – owning land in peasant communities, heading households and paying taxes.68 Women in the capital had important economic roles in crafts and trades, including weaving and silk-working, could walk freely in the streets and occasionally joined with menfolk in rioting against unpopular regimes.69 In the betterattested sphere of religious life, there is evidence of women as writers of hymns, and founders of nunneries in their private houses, and they probably 61 See, for example, the AHRB Centre for Byzantine Cultural History’s Gender Project (http://www.byzantine-ahrb-centre.ac.uk/Projects/Gender.htm); Talbot (1997); Smythe (2005). 62 Dennis (2001a). 63 Laiou (2002b), pp. 51–2. 64 Brubaker (2005). On the legal background to women’s status, property and the family, see Arjava (1996); Giardina (2000). 65 Stolte (1999); Humfress (2005), p. 181. 66 Holy women of Byzantium, ed. Talbot. 67 Laiou (1992b); Cheynet (2000); see above, pp. 19–20; below, pp. 294–5. 68 Neville (2006), pp. 77–83. 69 Garland (2006).

iii. other routes to byz antium

67

played a distinctive part in maintaining the veneration of icons.70 Most prominent of all were the women occupying or close to the throne who saw to the restoration of icon-veneration.71 In addition to the augusta, women had a formal part to play in court ceremonial. Nonetheless, the bias of middle Byzantine normative texts, political narrative and even the tax registers is against the independent status of women as individuals, acting in their own right. Empresses who lost their husbands usually remarried, were sidelined (as in the case of Zoe (see below, pp. 504, 505–6)), or eventually were themselves dethroned (as with Irene (see below, pp. 277–8)). The sense that a woman’s place is in the home is taken for granted in the Strategikon of the eleventh-century general Kekaumenos. Wife and daughters should be kept in their chambers if a man’s friend comes to stay because he will seduce them, given the chance.72 Perhaps significantly Kekaumenos, so free with his advice for men, does not seem to have followed up his stated intention of writing a text on how women should conduct themselves. His world is state-centred: serious money, top jobs and social status come from the public sector. Those prepared to apply themselves to military matters or judicial duties will go far – posted to successive places dotted across the empire. A wife is, at best, an adjunct in one’s career. So long as career structures and spectacular riches revolved around state service and access to the court, the role of ‘high-fliers’ was reserved for men. But one might expect a change when the state’s role as employer and determinant of rank began to fray. One straw in the wind may be observed at the very top, from the late eleventh century onwards. The Komnenoi ceased to rely on an elaborate hierarchy of court-titles, and family ties became more important as bonds of governance. It is no accident that women become more prominent in the new regime of households and affinities, starting with Anna Dalassena, who ‘drove the imperial chariot’ while her son Alexios I was absent on campaign.73 During the twelfth century several women of the Komnenian clan and its affiliates exercised extensive powers of literary patronage; some were themselves accomplished litt´erateurs. While this was partly a measure of the wealth and opportunities ever available in Constantinople, it also reflected the enhanced role of the family in high politics, and the multifarious influences that a woman could exercise on behalf of her children or other relatives. The daughter of Manuel I, Maria Komnena, ‘reckless and masculine in her resolution’, took the initiative 70 Silvas (2006); Herrin (2006); Herrin (1982); Herrin (1994); Cormack (1997b); Herrin (2000a), pp. 4–5. 71 See below, pp. 287–91; Herrin (2001). 72 Kek., ed. and Russian tr. Litavrin, pp. 218–21. 73 Al., VII.1, ed. Reinsch and Kambylis, I, p. 103; ed. and French tr. Leib, I, p. 123; tr. Sewter, p. 118; see also below, p. 612.

68

general introd uction

in an attempted coup d’´etat against her stepmother, also named Maria.74 Anna Komnena, herself blessed with outstanding literary talents, also had a taste for power. She was allegedly ‘chief instigator’ of a plot to dethrone her younger brother John II (1118–43) in favour of her own husband. Anna’s Alexiad was itself one more round in the power-play, looking back in anger long after her plot’s failure.75 The dynamics of power shifted again in the thirteenth and fourteenth centuries, and economic activity proliferated across the imperial or formerly imperial lands. Resources were now diffused in a medley of political centres, aristocratic households and commercial concerns, while laws upheld the legal rights of widows, and dowries were increasingly convertible into liquid assets. It is in this period that Byzantine women’s initiatives and activities are documented most fully, whether as founders of long-lasting convents, managers of commercial enterprises, money-lenders, midwives or medical practitioners.76 It may well be that the shrinking of resources and career opportunities in the emperor’s palm had favourable repercussions for certain classes of women. Widows of substance or good family enjoyed considerable independence while the lives of elite women in general were now less geared to spouses’ careers and itinerancy in the emperor’s service. Familial ties no longer had to compete so hard with the alternative prospects of drastic enrichment or social advance through office. Matriarchs such as Theodora Synadene could, in the thirteenth and fourteenth centuries, make elaborate provision for the women of their family by way of typika for the convents they founded.77 Another category of difference was imposed by human hands rather than nature, the act of castration that created eunuchs. Eunuchs were not unreservedly admired in Byzantium, and could be denounced as ‘ignoble’, unfit to govern (see below, p. 519). But individual eunuchs feature in many episodes narrated below, from Justinian’s general Narses onwards. The office of ‘chamberlain’ (parakoim¯omenos) was usually pivotal in the government and, like other senior posts involving the emperor’s bedchamber, it was reserved for eunuchs. Eunuchs were employed by noble families in their households and some eunuchs were, in the twelfth century, themselves of good family. But they were associated most prominently with the emperor’s cause, dedicated to state service rather as monks were to the service of God. In fact some monks and churchmen were eunuchs, gaining renown for their piety, 74 75

NC, ed. van Dieten, I, pp. 230–2; tr. Magoulias, pp. 130–1. NC, ed. van Dieten, I, pp. 10–11; tr. Magoulias, p. 8. On Anna, see contributions in GoumaPeterson (ed.) (2000). On female patrons and connoisseurs of the arts in Komnenian circles, see Laiou (1981), pp. 253–4; Jeffreys, E. (1982); Jeffreys and Jeffreys (1994); Jeffreys, E. (1998); Garland (ed.) (2006). 76 Laiou (1981), pp. 234–47; below, pp. 814–15, 830; Connor (2004), pp. 263–77. 77 Connor (2004), pp. 266, 277–308.

iii. other routes to byz antium

69

especially in the tenth to twelfth centuries.78 Intimate with the imperial house and presumed (quite often wrongly) to have forsaken all familial and carnal ties, eunuchs symbolised a hierarchy revolving around the emperor. As counsellors-cum-agents of policy, they also suited the kind of ‘flat-management’ that became characteristic from the seventh century on. Power nodes might form around them, but these did not harden into hereditary coteries even in the case of Basil Lekapenos, the veteran parakoim¯omenos (see also below, pp. 238–9, 277, 295–6, 505, 519–20, 524, 531–2). It is probably no coincidence that eunuchs lost their prominence as trusty servants after Alexios I Komnenos called upon his extended family to fulfil the most pressing imperial needs. Members of the Komnenos clan, starting with Alexios’ own mother, combined domestic ties with governance and military and civil commands, maintaining the semblance of a unitary state. Manuel I Komnenos did, it should be noted, reverse the trend and brought eunuchs back into governance. But in the late empire eunuchs mostly had lowlier ranks in the imperial and patriarchal households.79 societ y: dissid ence and outsiders Men, women and eunuchs answered the description of ‘Romans’ comfortably enough provided that their religious faith and ritual were orthodox, they acknowledged themselves to be the emperor’s douloi (a somewhat ambiguous term),81 and they could manage spoken Greek. A ranking order of precious vestments distinguished the upper echelons of members of the Byzantine empire, while certain conventions of clothing were observed by non-elite men and women for most of its history.82 There were, however, other types of person who, whether tacitly, through open dissent, or through living in discrete groupings, diverged from religious, ethical or social norms.83 Some had valuable contributions to make to the empire 80

78 Ringrose (1999); Sid´ eris (2002); Tougher (1997a); Tougher (2002); Ringrose (2003), pp. 117–27; Smythe (2005), p. 164; Tougher (2006). 79 Gaul (2002), pp. 200–1, 208–9; see below, pp. 612, 657. 80 On notions of masculinity and what was expected of men in Byzantine society, see Barber, C. (1997). 81 In the sixth century, an imperial official working at Corinth could style himself on inscriptions the emperor’s ‘faithful servant’ (pistos doulos; gn¯esi¯os douleu¯on) as a measure of his own status, and doulos retained connotations of access to the emperor throughout the medieval period: Feissel and PhilippidisBraat (1985), 279–81; Gregory (1993), pp. 12–14; Pazdernik, ‘A dangerous liberty’ (PhD thesis, 1997); Pazdernik (2005), pp. 203–5. However, doulos had other connotations, from ‘slave’ to non-Roman princes beyond the borders who ‘in servitude’ (doulik¯os) acknowledged the emperor’s hegemony, for example the Serbs and the Croats: DAI, chs. 31, 32, pp. 150–1, 160–1. See also Treitinger (1956), p. 227 and n. 84; ODB, I, p. 659 (A. Kazhdan); LBG, p. 407 (douleia, douleusis); glossary below, p. 888. 82 Lopez (1945); Maguire (1997); Ball (2005), pp. 37–56, 79–89, 102–4. 83 For surveys of some alternatives to orthodox society and thought, see contributions to Garland (ed.) (1997); Smythe (ed.) (2000).

70

general introd uction

in the economic sphere, while the presence, real or supposed, of the unRoman in the Byzantines’ midst had its ideological uses, providing the emperor with vivid foils. Not all these categories of nonconformists – usually minorities within the empire – were self-declared or acting in open concert. Homosexuality fell foul of Roman and church law and its practice is unlikely to have found very much sympathy in rural communities. Emperors were occasionally accused of homosexual tendencies by contemporaries or by later historians: Michael III (842–67) was one such (see below, p. 295 and n. 23). The monastic vocation and its extensive network of remote, male-dominated communities beckoned to those seeking to sidestep their family’s expectations of marriage and to escape from the things of this world; for very many, they offered access to the divine. Nonetheless, some rule-books of monasteries forbade beardless youths and even eunuchs from approaching their houses, for fear of the temptations they might pose.84 One form of unacceptable difference virtually endemic in Byzantium’s political and religious culture was heresy. Generally this charge of dissidence or error (from haeresis, ‘sect’) was levelled by monks or members of the imperial-ecclesiastical establishment against those held to be breaching orthodox doctrine or ritual; the charge could serve as the small change of political discourse. Several chapters of this book recount how successive earlier emperors sought to reconcile churchmen who disagreed profoundly over the finer points of defining the nature of Christ, only themselves to be accused of heresy. Then, in the eighth and earlier ninth century, the emperors’ efforts to purge the empire of ‘idols’ – icons – aroused opposition and they themselves were styled arch-heretics after icons were reclassified as orthodox in 843 (see below, pp. 117–19, 122–3, 228–9, 231–2, 287–91). Communities of heretics could, however, profess an alternative creed in certain contexts, especially where the Roman orthodox were thin on the ground. For example Paulician dualists were transplanted from eastern Anatolia to the Thracian borderlands and, in the later tenth and eleventh centuries, Syriac and Armenian monophysites were encouraged to settle in newly won Byzantine territories (see below, pp. 288–9, 297, 532–3, 677, 783 and n. 25). These monophysites formed their own church organisation, the catholicos of the Syriac Jacobites being encouraged to base himself in imperial territory.85 The sovereign confidence of Basil II (976–1025) and his immediate successors that these heterodox could be brought beneath their imperial umbrella says something for Byzantium’s vibrancy at that time. But it is consistent with a tradition whereby the emperor had discretion 84 Galatariotou (1987), pp. 121–2; ODB, II, pp. 945–6 (J. Herrin); Smythe (1999); Smythe (2005), pp. 164–5; Tougher (1999); Jordan (2000), pp. 67–71; Ringrose (2003), pp. 112, 126. 85 Dagron (1976), pp. 187–93.

iii. other routes to byz antium

71

to license certain forms of diversity: he thereby demonstrated the universal reach of his rule, while himself remaining a paragon of orthodoxy. Incoming aliens who accepted orthodox Christianity could be assigned fertile lands to work, pay taxes or perform military service from, as with the Pechenegs in the 1040s.86 Longer-term organised communities of heretics, non-believers or other aliens were left to areas of little economic consequence to the government, for example the warlike Melingoi in the Taygetos mountains of the southern Peloponnese, who still spoke Slavonic and maintained a distinct identity in the thirteenth century, or the Vlachs, Romance-speaking pastoralists of the uplands.87 Not all of them were confined to the empire’s ‘cold-spots’, however. The Jews occupied a district across the Golden Horn from Constantinople itself and some resided in provincial towns and Cyprus.88 The Jews were a special case, anomalous remnants of a faith that Christians thought their religion had superseded; learned proponents of an earlier version of monotheism and priesthood; and a convenient scapegoat for the empire’s woes in times of adversity as, for example, in the seventh century when Heraclius launched a drive against them.89 Unlike some unorthodox, the Jews were not predisposed to proselytise and they lacked powerful coreligionists beyond Byzantium’s borders. So while subjected to occasional drives for purification, they were seldom suspected of being actively hostile towards the empire. The Jews are, then, an example of how minorities of the unorthodox and alien could define the essence of empire through exemplifying error and its price. But the history of the Jews in Byzantium is far from static. Jewish goldsmiths, silk-dyers and other craftsmen were an asset, not least because of their ties with co-religionists across the Muslim world, commercial nexuses at once detectable and taxable.90 In fact the Jews’ fortunes amount to a barometer of Byzantium’s general well-being. Jewish immigrants offer examples of a different breed of outsider that rising prosperity in the medieval era attracted, firstly to Constantinople and later to provincial towns. It is no accident that, despite individual Jews’ initial dismay at the Byzantines’ conquest of Crete in 961, subsequent decades saw many Jews drawn to the empire by the prospects of security and favourable trading conditions it held out.91 From around the tenth century onwards, various other groups of outsiders were frequenting the capital, travelling mostly by 86 87 88

Skyl., ed. Thurn, p. 459; French tr. Flusin and Cheynet, p. 380; see also below, pp. 328, 674. See below, pp. 258, 664, 687; Ahrweiler (1962b), pp. 3–4, 7–10; Winnifrith (1987). Starr (1939); Sharf (1971); Jacoby (1995); de Lange (1992); Greek Jewish texts from the Cairo Genizah, ed. and tr. de Lange. 89 See below, pp. 116, 241, 247; Sharf (1971), pp. 107–12, 116–21; Bowman (1986); Maas (1990); Cameron, Averil (2002a); de Lange (2000); de Lange (2005a); de Lange (2006), pp. 172–7. 90 Goitein (1967–93), I, pp. 42–63, 211–14, 266–72; Muthesius (1995), pp. 245–53; below, p. 474. 91 Holo (2000); Jacoby (2000a); Ankori (1959), pp. 163–4; Sharf (1971), pp. 107–27.

72

general introd uction

sea and staying more or less in touch with home ports. ‘Syrian’ and other Muslim traders, Bulgarians and Rus from the north, and merchants from Italian towns such as Venice and Amalfi frequented the capital.92 In presiding over this process, emperors showed characteristic flexibility, alert to the benefits which the outsiders’ activities could reap for their own treasury coffers and also to the leverage that could be exerted on outsiders once they had a stake in the empire’s economy. These externally based traders were, almost literally, paying tribute to the resources and purchasing power concentrated at the imperial capital from the tenth century on. Their presence was yet another token of the basileus’ worldwide sway. His toleration of them in the capital was akin to his role of lord and ringmaster of exotic creatures, symbolised by the mechanical birds and lions at receptions for outsiders in the Great Palace.93 This, however, presupposed a fixed ring, whose creatures would neither evolve nor multiply beyond measure, a presupposition undermined by events unfolding in the wider world. The mounting engagement of external traders with Constantinople’s markets and the rising volume and value of transactions there were not necessarily harmful to the empire’s interests. Through the eleventh and twelfth centuries emperors showed astuteness and ingenuity in harnessing outsiders’ specialist talents and economic dynamism to their own advantage. But the emperors’ balancing act between, on the one hand, guarding doctrinal and ritual purity, security and well-being for the ‘silent majority’ and, on the other, licensing the presence and idiosyncrasies of aliens living within or frequenting the capital was a delicate one. The balancing act presupposed pliability on the outsiders’ part, and that the emperor was master in his own house. Such balancing also called for outstanding qualities of statecraft from each successive emperor in turn. outsid ers withi n From the mid-eleventh century on, the foresaid preconditions began to change as the wealth and numbers of outsiders frequenting Constantinople rose, while some orthodox churchmen and, especially, monks took exception to the rites and ways of western Christians. First hints of what was to come include the outbreak around 1042 of violence between Constantinople’s citizens and Arab, Jewish and other non-Roman traders, followed by the emperor’s ban on their residence inside the City; and the popular support Patriarch Michael I Keroularios (1043–58) mustered in taking his stand against the papal legates in 1054. Whether or not Keroularios physically 92 93

Magdalino (2000a), pp. 219–21; Balard (1976); Reinert (1998); Shepard (2006b). Liudprand, Antapodosis, VI.5, ed. Chiesa, p. 147; tr. Wright, pp. 206–7; Brett (1954); Trilling (1997), pp. 222–30; Maguire and Maguire (2007), pp. 41–5, 54–5.

iii. other routes to byz antium

73

closed the Latin churches in Constantinople, it is likely that an increase in their numbers, itself a register of Latins’ commerce there, made their distinctive rites more of an issue than had previously been the case.94 Ample reserves of authority – material and moral – remained within a manipulative emperor’s grasp, and the Latin west’s multifarious facets could be kept in play yet apart from one another, as Manuel I Komnenos showed.95 Nonetheless, western naval capability, martial adventurism and papal aspirations to Christian leadership coalesced in the events culminating in the capture of Constantinople by crusaders in 1204. This was, in part, a matter of long-privileged outsiders who could be deemed ‘insiders’ – the Venetians – vindicating their rights within the empire.96 Intensive intermingling of outsiders’ affairs with Byzantium’s, and emperors’ familiarity with western churchmen would still further characterise the empire Michael VIII Palaiologos restored to Constantinople in 1261. His pressing forward with the Union of Lyons is understandable in light of the threat that Charles of Anjou appeared to pose to his regime, but it earned him execration from orthodox monks and many churchmen.97 In the aftermath of 1204, Byzantine clerical writers were voluble in denouncing their western counterparts and warning orthodox lay folk of the impious conduct and unhallowed rituals of Latin Christians in general. Lists describing ‘the errors of the Latins’ had begun to circulate in the era of Michael Keroularios, and became fuller in the later twelfth century, and more numerous. But it was the thirteenth century that saw the lists lengthen and proliferate.98 This bespeaks a hardening of the line against outsiders. The church filled the vacuum once the emperor proved wanting in the role of upholder of religious orthodoxy. One may therefore view the orthodox church’s anti-Latin stance as a reaction to the experience of, in effect, being colonised by western Christians. This was, after all, the period when Marino Sanudo expressed concern that populations under Latin rule were still, at heart, given up to ‘Greek matters’ and hostile to their new masters (see above, p. 8). Yet the very proliferation of the ‘lists of the errors of the Latins’ suggests that orthodox writers may then have been engaging in a competition for souls whose outcome was not utterly assured. The faithful might yet succumb to Latin ways out of ignorance or lack of clarity as to the points 94 Bar Hebraeus, Chronography, tr. Wallis Budge, I, p. 203; Runciman (1955), p. 52–67; Shepard (1978–9), p. 174; Kaplan (1997), pp. 170–1; Kolbaba (2005), pp. 40–1; see below, pp. 601–2. The events of 1054 did not gain a place in Byzantine historiography and, under the Komnenoi, the church leadership was on too tight an imperial rein for Keroularios’ stand to be held up as exemplar: Kolbaba (2003). This, however, suggests how potentially provocative the presence of westerners was becoming at street level. 95 See below, pp. 644–5; Magdalino (1993a). 96 Angold (2003a), pp. 50–8, 75–101; Magdalino (2007a). 97 See below, pp. 752–3, 755–8, 803–4. 98 Kolbaba (2000), pp. 15–16, 25–9, 170, 173–88; Kolbaba (1997).

74

general introd uction

of difference, or they might be tempted deliberately to opt for a western affiliation, on material or intellectual grounds. The very stridency of the condemnations of the association or marriage of orthodox with Latins in the ‘lists’ suggests that day-to-day contacts between orthodox lay persons and Latins were not uncommon, at least in the towns.99 In other words, dividing lines may not have been so clear-cut or so uncrossable as one might at first sight suppose. One can reasonably treat the ‘lists’ as a sign of new uncertainties and opportunities available following the dissolution of the imperial envelope that had contained the orthodox for so long. Political boundaries were now fluid in the thirteenth century and the empire had anyway long ceased to be more or less coterminous with the faith-zone it had effectively been in the early middle ages. From this point of view the ‘lists’ represent the justified apprehensions of rigorist orthodox churchmen and their elaboration of culturo-religious identity, in default of the taxis provided by the imperially guided state.100 Yet the ‘lists’ also suggest how loose-knit the identity of the medieval Byzantines had actually been hitherto or rather how little was spelled out in writing or tabulated, and how much was a matter of liturgical rituals and ceremonies revolving round a few core values, beliefs and traditions. In other words, even the more or less unthinking ‘conformists’, faithful subjects of the emperor, were perhaps a more variegated bunch than they themselves were fully aware. Beneath the imperial umbrella and the outward and visible symbols of religious orthodoxy, a medley of assumptions, local customs and religious devotions could comfortably co-exist.101 und ercurrents of byz antium This matches the impression given by other scraps of evidence concerning the subjects of the middle Byzantine basileus: of undercurrents at various depths of society uncharted even by surviving tax registers and treatises.102 These points of view, assumptions and practices were not necessarily consciously contradictory to the tenets of the ruling establishment, while even outright dissenters might have no conception of a viable alternative to the apparently irreversible scheme of things. But this very lack of elaborate 99 See also Kolbaba (2000), pp. 17, 28, 38–9, 139–40, 152; Kolbaba (2006), pp. 209–12. See above, pp. 43–4, 45, 46–7. 100 For the apparent mutual compatibility of Greek and Latin liturgical music and the use, in the fourteenth century, of the scholastic method of argument by orthodox writers, see respectively Lingas (2005); Russell (2006). 101 Beck (1978), pp. 103–6; Kolbaba (2000), pp. 46, 69–72, 95, 104–17; Cameron, Averil (2006b), pp. 96–8, 112–15, 121–5, 129–32. 102 Maguire and Maguire (2007); Baun (2007a); Shepard (2007). See also Beck (1978); Mango (1980), pp. 88–104.

iii. other routes to byz antium

75

definition of ‘orthodoxy’ was what made it so necessary for writers to spell out the rites, the do’s and don’ts of orthodoxy and ‘the errors of the Latins’ once the imperial order slackened and variants were to hand.103 All this suggests the multiplicity of approaches that the modern enquirer may take towards the empire, and how much of importance, at once mutable and elusive, remains to be uncovered behind the Roman fac¸ade. 103

On differing conceptions of religious orthodoxy among intellectuals, see Magdalino (2006).

I N T R O D U C T I O N – PA RT i v

SMOOTHING THE WAY AND SHORT-CUTS TO BYZANTIUM: TEXTS IN TRANSLATION jon at han s he pa rd

Byzantium at first sight looks inaccessible to those approaching for the first time, especially without Greek or Latin, or one of the modern languages spoken in regions closely associated with the empire. Native English-speakers may feel like ‘barbarians’ before the walls of Constantinople, excluded and daunted. Yet as with the great City, so with the subject, portals and gateways are available and the newcomer can reach some of the landmarks surprisingly fast, arriving at positions not all that much inferior to those of life-long devotees. The reasons are at once straightforward and specific to some of the main types of the surviving literary and other source-materials. Nothing like a full guided tour of sources available in English translation is attempted here, but the curious should be able to follow the directions towards more detail about them. Some of the more general introductions to the subject are noted below (pp. 90, 94). sourcebooks Straightforward considerations first: there are several collections of excerpts from sources, providing historical introductions as well as translations. They make a good first port of call for students, or for teachers who are themselves non-specialists but are thinking of offering a class or two on Byzantium. The earlier period, roughly corresponding to our Part I, is well served by sourcebooks. Michael Maas covers most aspects of life in the Byzantine sphere from the era of Constantine the Great’s conversion until the Arab invasions of the seventh century, general remarks being interwoven with extracts from relevant texts.1 Maas gives details of websites dedicated to more specialised source-guides and collections of texts. A wide-ranging assortment of texts bearing on religion, whether Christian or non-Christian, is provided by Douglas Lee with substantive introductory paragraphs,2 and collections of texts relating to doctrine and the disputes and councils arising 1 Readings in late antiquity, ed. Maas. (This work, like most others cited in this section, features in our bibliography of primary sources; the remainder are in the bibliography of secondary works.) 2 Pagans and Christians, ed. Lee. See also the texts with multi-authored introductions in Religions of late antiquity, ed. Valantasis.

76

iv. tex ts in transl ation

77

therefrom are available.3 The empire’s eastern frontier is the subject of a very full narrative sourcebook.4 The middle and later Byzantine periods – effectively our Parts II and III – are covered in their entirety by very few sourcebooks. The contrasting civilisations of Byzantium and Islam are presented by Charles Brand, while Deno Geanakoplos supplies a broad overview of the Byzantine world from Eusebius’ time until the Italian Renaissance.5 Sourcebooks focusing on particular themes are more plentiful, for example the well-chosen collections of saints’ lives in Byzantine defenders of images, and in Holy women of Byzantium.6 The former is devoted to the iconoclast controversy, for which other translations and authoritative guidebooks exist.7 Fields in which the Byzantines had close dealings with other peoples have generated sourcecollections, for example, on medieval trade,8 the Christianisation of the Slavs,9 the world of Islam,10 the Normans or crusading.11 These can be illuminating, even while offering different perspectives, often hostile towards the Byzantines. The loss of so many written source-materials from Byzantium is one reason why we depend heavily on outsiders for knowledge of, for example, the layout of Constantinople itself, fortunately a subject of keen interest to pious Rus travellers.12 But there is something about Byzantium, whether as political structure or cultural atmosphere, that resists categorisation or orderly review in the manner of, say, imperial Rome. And now both sourcebooks and general guides to sources in translation have rivals on the internet. A reliable general guide to printed translations was provided by Emily Hanawalt,13 but future guides and source-collections will probably appear mainly in cyberspace. Online guides offer accessibility together with high-quality scholarship, as witness the collections of Paul Halsall and Paul Stephenson.14 An authoritative online survey of translations of saints’ Lives 3 For the era from the death of Constantine the Great to the council of Chalcedon: Creeds, councils and controversies, ed. Stevenson and Frend; coverage up to the eighth century in Nicene, ed. Wace and Schaff. 4 The Roman eastern frontier, ed. Lieu et al. 5 Icon and minaret, ed. Brand; Byzantium, ed. Geanakoplos. 6 Byzantine defenders of images, ed. Talbot; Holy women of Byzantium, ed. Talbot. 7 Icon and logos, ed. Sahas; clear, detailed guidance to all forms of source-material in Brubaker and Haldon (2001). 8 Medieval trade, ed. Lopez and Raymond. 9 Medieval Slavic lives, ed. Kantor; Kiril and Methodius, ed. Dujˇ cev; Monumenta Bulgarica, ed. Butler. 10 Islam, ed. Lewis. 11 Normans in Europe, ed. van Houts; First Crusade, ed. Krey; P. Halsall’s Crusade sources in translation (www.fordham.edu/halsall/source/cdesource.html). See also contributions to Whitby, Mary (ed.) (2007). 12 Russian travelers, ed. and tr. Majeska. 13 Hanawalt, Annotated bibliography of Byzantine sources. 14 For the collection of translations (into western languages) made available by Halsall in the Internet Medieval Sourcebook, see: www.fordham.edu/halsall/byzantium. For the collection made by

78

general introd uction

in print is also provided by a bastion of Byzantine studies in the Anglophone world, the Dumbarton Oaks Research Center in Washington, DC.15 Internet guides are open to constant updating, an asset that may have its disadvantages. But they are well suited to Byzantium, in their ability to bring together sources and resources widely scattered across disciplines and geographical space, ready for use by newcomers or by long-time scholars. And, as a medium, the internet offers direct and flexible access to important source-materials, since the visual arts and archaeological data can be presented in various degrees of detail, in high definition but at minimal cost. art and visual media The electronic medium is all the more important for introducing students to sources because Byzantium was such a self-consciously ‘visual’ culture. For the ruling elite, display and portrayal were invaluable in projecting imperial ideology. And in the religious sphere, accurate representation of Christ, the saints and sacred scenes conveyed doctrine, provided instruction and edification, but also transmitted the divine in most truthful form.16 Certain icons were, from the middle Byzantine period onwards, venerated for themselves working miracles, and ordinary icons were often supposed to possess special powers. Partly for this reason, the veneration of icons became the subject of controversy (see below, p. 282). The polemics generated reveal the many shades of Byzantine thinking on the question. Besides the works already noted, excerpts from texts concerning the iconoclast controversy are provided by Cyril Mango’s The art of the Byzantine empire. This magisterial collection covers most aspects of the visual arts, including buildings and building-works, and the pithy commentary offers a guide to the Byzantines’ writings about imagery.17 The writings, like the images themselves, usually tell us more about the Byzantines’ beliefs and ideals, their notions of what religious doctrine should be, or the awe that buildings or mosaics ought to inspire, than they disclose of ordinary people’s reaction to them or of how things actually were. The writings were mostly penned by the more learned members of society. Likewise the political imagery and the court ceremonial represent the order of things as projected by the ruling elite, its agents and aficionados, rather than political realities, everyday affairs or Stephenson, with links to other sites, see: http://homepage.mac.com/paulstephenson/trans.html. For the earlier period, the Society of Late Antiquity’s site makes a good starting-point: www.sc.edu/ltantsoc. 15 The regularly updated survey of Byzantine saints’ Lives available in translations into English and other western European languages may be found at http://www.doaks.org/research/byzantine/ translations byzantine saints lives.html. 16 See above, p. 54. For example, Photios, Homilies, ed. Laourdas, pp. 167, 170; tr. Mango, pp. 290, 293–4; Barber, C. (2002), pp. 135–7; James (1996); see now James (ed.) (2007). 17 Mango, Art of the Byzantine empire.

iv. tex ts in transl ation

79

the living conditions of the unprivileged. This is the case even if details of, for example, ordinary people’s clothing can be gleaned from study of the paintings in churches.18 With visual imagery, then, as with a great deal of surviving Byzantine literature, one often encounters an ideal scheme of things, what leading lights in the Byzantine church and empire wanted to be seen, rather than a wide range of witnesses as to what actually happened.19 But these representations are at least approachable by newcomers, whether through looking at religious and political imagery or through reading in translation the prescriptive works, idealised portrayals of saintly lives, and orations and histories emanating from the imperial-ecclesiastical circles. Various possible readings and interpretations are possible, with nuances and allusions being more apparent to the learned, or to those steeped in eastern orthodox religious traditions and practices. But first impressions of these portrayals are not necessarily far removed from those which their creators sought to evoke, while the message of the directly prescriptive texts is often plain enough. l aws, t y p i k a and saints’ l i v e s Imperial laws were systematised by Justinian. His Institutes and Digest are available in translation, as are several important later legal texts or decrees, including the Book of the eparch issued under Leo VI’s auspices and the novellae of tenth-century emperors on peasant landholdings.20 The concept of legislation informed works of administrative regulation such as the Book of the eparch, and these in turn shade into detailed administrative prescriptions or treatises, such as two texts for tax-collectors.21 Regulations governing church life were issued by church councils and patriarchs as well as by individual emperors, and the acts of the ecumenical councils are available in translation.22 Collections of the rules and regulations issued in the medieval period by Byzantine churchmen and specialists in church law – canonists – have 18 19

Ball (2005); see also Parani (2003). Unauthorised tastes and subject-matter are, however, discernible in some visual forms: Maguire and Maguire (2007). 20 For fuller details of the works cited in this and subsequent footnotes, see Abbreviations or Bibliography of Primary Sources. Justinian, Corpus iuris civilis, tr. Birks and McLeod, Institutes; tr. Watson, Digest; Ecloga, tr. Freshfield; Ecloga privata aucta, tr. Freshfield; Farmer’s law, ed. and tr. Ashburner; Eparch, tr. Freshfield; JG, partial tr. McGeer; Rhodian sea-law, tr. Freshfield. 21 The two handbooks are tr. in Brand (1969), pp. 48–57, 57–60. See further, on the concept of regulations having the force of law: Magdalino (1997). 22 Nicene, ed. Wace and Schaff, XIV. The highly problematic text of the acts of the seventh council (Nicaea 787) is partially available in a more recent translation: Icon and logos, ed. Sahas. See also Brubaker and Haldon (2001), pp. 233–7.

80

general introd uction

not received English translations.23 However, the regulations for monasteries’ administration and liturgical observances, typika, are well served by translators. Together with the surviving order for the liturgy prescribed for the monastery of Theotokos Evergetis in Constantinople, they set out in varying degrees of detail what founders envisaged for their monasteries.24 Considering the broad cross-section of laypersons concerned with monks and monasteries during the middle Byzantine empire and beyond,25 the typika are of great historical importance. They present a spectrum of spiritual aspirations that were widely respected, if seldom fully attained, among the Byzantines. So, in their way, do the Lives of saints, with due allowance made for their authorial agendas, literary genres, frequent aversion to specifics of place and time, and conceptions of truth other than the literal or earthly. The Lives were widely appreciated for their transcendent spiritual examples and instruction, and were intended to convey a higher reality than life as actually lived. But one should note that some give details of persons and events verifiable from other sources, and the very desire of the hagiographer to make the case for his (or her) subject could entail reference to their actual situations and the problems they faced. For example, in his Life of Lazaros, a stylite for forty years on Mount Galesion, the contemporary author Gregory the Cellarer gives evidence of hostility towards his hero among members of the church hierarchy, ‘and even within his own monastic community’. The Life also makes ‘important allusions to historical events and personages in the world outside the monastery’, and offers an at least plausible portrayal of men, women and everyday country matters in eleventh-century Asia Minor.26 sermons and orations Besides the idealised lifestyles of hagiography,27 sermons provided the Byzantines with guidelines for praying, living in this world, and enduring. Some, composed by church fathers such as Gregory Nazianzen, became elements in the liturgy, read out during services, while others from 23 This is scarcely surprising, seeing that the Nomokanones, compilations of secular laws (nomoi ) and ecclesiastical regulations (kanones), did not amount to a system of canon law. Important collections were, however, made, notably the Nomokanon of fourteen titles, and magisterial commentaries were written, for example by the nomophylax of St Sophia, Theodore Balsamon in the twelfth century: see ODB, I, pp. 372–4 (A. Schminck), pp. 248–9 (A. Kazhdan); ODB, II, pp. 1490–1 (A. Schminck); Macrides (1990); see below, pp. 616–17. See also pp. 241, 245. 24 Complete translation of extant texts in Byzantine monastic foundation documents, ed. Thomas and Hero. See also Synaxarion of the monastery of the Theotokos Evergetis, ed. and tr. Jordan (2 vols. to date). 25 Morris (1995). 26 Gregory the Cellarer, Life of Lazaros, tr. Greenfield, pp. 4, 69 (introduction). 27 See above, n. 15; see also Pratsch (2005a).

iv. tex ts in transl ation

81

antiquity remained familiar to the Byzantines.28 New sermons continued to be composed in Attic style and kept, and a few of those designed for special occasions or recounting specific events have been translated.29 A translation has yet to be made of the sermons written and delivered by Leo VI, who approached his pastoral duties as ruler with high-minded diligence. But a full exegesis of their form and contents is available.30 Leo’s sermons were delivered before his court, the setting for the delivery or performance of many of the Byzantine elite’s literary creations. The sermons, orations and verse-poems furnished a steady, solemn, usually upbeat note to proceedings. Some were written for recurrent religious festivals. Others marked state occasions or recent events, and orations could be more or less unsolicited, currying favour or – more especially during the later empire – advocating a policy, seeking to persuade. Only a tiny proportion of these presentations survives – and not necessarily in the form in which they were first delivered. The little that has been translated into English tends to celebrate specific recent events, for example, the rededication of St Sophia in 562; the building of a palace bathhouse for Leo VI; the treaty with Bulgaria in 927; or Manuel II Palaiologos’ funeral oration on his brother, Theodore.31 Nine orations of Arethas, some after-dinner speeches, others solemnly welcoming the arrival of relics in Constantinople, have been edited with English summaries.32 And a career-making speech in praise of Nicaea delivered before Andronikos II by the young Theodore Metochites in 1290 has been translated, together with one composed by a future Nicaean emperor, Theodore II Laskaris (1254–58).33 One of court oratory’s functions was to review current affairs and the recent past, accentuating the positive and setting ups-and-downs within the empire’s long history and manifest destiny. It is no accident that some men of letters prominent as speech-writers and -givers at court also composed for the historical record, notably Michael Psellos, Eustathios 28 See, e.g., John Chrysostom, Homilies, tr. Hill; Gregory Nazianzen, Select orations, tr. in Nicene, ed. Wace and Schaff, VII; tr. Vinson. 29 Photios, Homilies, tr. Mango; Nicholas I Mystikos’ sermon lamenting the sack of Thessaloniki (904), in his Miscellaneous writings, ed. and tr. Westerink, pp. 9–17. See Sironis (1998) and other contributions in Cunningham and Allen (eds.) (1998); Cunningham (2003). 30 Antonopoulou (1997). 31 Paul the Silentiary, St Sophia, partial tr. Lethaby and Swainson; Leo Choirosphaktes, ‘On the bath’, ed. and tr. Magdalino; Dujˇcev, ‘Treaty of 927 with the Bulgarians’; Manuel II Palaiologos, Oration, tr. Chrysostomides. See also Magdalino (1993a), pp. 413–70; Dennis (1997); Webb (1999); Jeffreys, E. (2007), pp. 172–4. See also on public uses of rhetoric, H¨orandner (2003); Jeffreys, M. J. (2003) and other contributions in Jeffreys, E. (ed.) (2003). 32 Arethas, Orations, ed. (with English summaries) Jenkins et al. 33 Theodore Metochites, ‘Nicene oration’; Theodore Laskaris, ‘In praise of the great city of Nicaea’. Excerpts from orations and other works relating to political economy are translated in Social and political thought in Byzantium, ed. Barker; see below, n. 49.

82

general introd uction

of Thessaloniki and Niketas Choniates. Unfortunately their orations lack English translations, unlike their histories of reigns or events of their own times.34 historical writing in and out of court A few other grand presentations of imperial deeds from insiders at court are – or soon will be – available to Anglophone readers, for example the Life of Basil, Anna Komnena’s highly personal portrayal of her father Alexios I (the Alexiad), and the grand logothete George Akropolites’ account of events between the fall of Alexios III in 1203 and Michael VIII’s restoration of Constantinople to Roman imperial status in 1261.35 Anna, however, carried out her work after leaving court life, Niketas Choniates revised his history shortly after Constantinople’s fall, and in fact major historical compositions often come from the fringes of the court, from writers formerly at the centre, or ensconced in administrative, legal or ecclesiastical niches rather than at the dizziest heights. A slight distancing from the very top facilitated composition of well-informed, more or less ostensibly favourable presentations of current emperors’ deeds or the reigns of an ongoing dynasty. This generally holds true of historians of the era of Justinian and his successors – Procopius, Agathias and Theophylact Simocatta – and also holds for the period when the empire’s fortunes were once again related in formats reminiscent of classical historians, as by Leo the Deacon and John Kinnamos. All these have English translations.36 Adding the translations of church histories and works generally labelled chronicles, with their diverse priorities and perspectives – for example, the works of John Malalas, Evagrius Scholasticus, the Paschal chronicle, Theophanes the Confessor, Patriarch Nikephoros I (806–15) and John Skylitzes – one obtains a continuous account of the earlier and middle empire’s history available in English.37 34 Psell., tr. Sewter; Eustathios, Capture of Thessaloniki, tr. Melville Jones; NC, tr. Magoulias. Substantive extracts from the orations of Eustathios are translated, with exegeses, in Stone (2001); Stone (2003a); Stone (2004). The original version of Niketas’ History was probably written ‘upon the request of the court circle of Alexios III’: Simpson (2006), p. 203. 35 Life of Basil, ed. and tr. Sevˇ ˇ cenko; Al., tr. Sewter; GA, tr. Macrides. 36 Pr W, ed. and tr. Dewing; Agathias, Histories, tr. Frendo; TS, tr. Whitby and Whitby; Leo the Deacon, History, tr. Talbot and Sullivan; John Kinnamos, History, tr. Brand. Certain histories now available in English did enjoy direct imperial patronage, e.g. Marcellinus, Chronicle, tr. Croke; Menander the Guardsman, History, ed. and tr. Blockley; see Rapp, C. (2005). See also Treadgold (2007), pp. 227–30, 293–5. 37 John Malalas, Chronicle, tr. Jeffreys et al.; Evagrius Scholasticus, Ecclesiastical history, tr. Whitby; Paschal chronicle, tr. Whitby and Whitby; Theoph., tr. Mango and Scott; Nikeph., ed. and tr. Mango; ˇ cenko. Skyl., tr. Wortley (in preparation). See also Life of Basil, ed. and tr. Sevˇ

iv. tex ts in transl ation

83

let ters, poems and l ampoons To some extent, then, the non-specialist is quite well served, and translated letter-collections or poems of major figures or government employees occasionally supplement the forementioned narratives. Letters collected for publication were partly intended to show their authors’ membership of the politico-religious elite and their familiarity with both the scriptures and classical lore. But their stylised qualities and contrived archaisms do not necessarily void them of straightforward historical content. This is especially so with the collected letters of the patriarchs Nicholas I Mystikos (901–7, 912–25) and Athanasios I (1289–93, 1303–9) and of Emperor Manuel II Palaiologos, while other writers such as Leo of Synada, John Mauropous and Gregory Akindynos disclose something of the goings-on in the imperial-ecclesiastical complex.38 One ‘statesman by day, scholar by night’, the grand logothete Theodore Metochites, sought consolation for loss of effective power through his Poems ‘to himself ’.39 There are no worthy successors to Procopius’ Secret history,40 but the pomp and pieties of court provided an arena for political differences, personal rivalries were keen, and undercurrents of criticism and satire flowed on. The currents occasionally surface, as in Psellos’ penportraits in Fourteen Byzantine rulers, where Psellos states that all emperors’ actions are ‘a patchwork of bad and good’, and proceeds to lampoon emperors such as Constantine IX whom his orations had praised to the skies.41 Former emperors’ foibles and misdeeds were fair game after a change of dynasty, as Michael III’s (842–67) posthumous reputation attests (see below, pp. 292, 295–6). And whole dynasties of emperors are castigated by iconodule writers such as Patriarch Nikephoros and Theophanes the Confessor. accoun ts of the christian empire and its precursors; other ‘chronicles’ Theophanes’ work exemplifies the rather different perspective of those writers whose main concern was God’s plan for mankind since the creation, and matters bearing directly on faith and the church. The realms of the ancient Persians, Assyrians and Macedonians played a part in their story, as did all the vicissitudes of Israel, and the deeds of early Christian Roman emperors were of interest. But these authors’ approach to the latter was coloured by 38 NM, ed. and tr. Jenkins and Westerink; Athanasios, Correspondence, ed. and tr. Talbot; Manuel II Palaiologos, Letters, ed. and tr. Dennis; Leo of Synada, Correspondence, ed. and tr. Vinson; John Mauropous, Letters, ed. and tr. Karpozilos; Gregory Akindynos, Letters, ed. and tr. Hero. For the letters of Theophylact of Ochrid, as yet untranslated into English, see above, n. 43 on p. 37. 39 Theodore Metochites, Poems, ed. and tr. Featherstone. 40 Procopius, Secret history, ed. and tr. Dewing; tr. Williamson. 41 Psell., tr. Sewter, p. 167; Dennis (1997), pp. 134, 138.

84

general introd uction

their maintenance of what the authors deemed to be orthodoxy: heretical or simply immoral emperors were condemned and their deviation might be expected to incur God’s wrath; the author’s task was, in that case, to chronicle their misdeeds, and the consequent disasters. Theophanes’ composition covers the period from the reign of Diocletian (284–305) until 813, but he was finishing off the opus of George Synkellos, whose universal chronicle had aspired to an accurate continuous chronology of events since Genesis.42 Eusebius and John Malalas had also traced events from the beginning, with somewhat different agendas, and this form of broad-sweep writing was popular with the Byzantines. Later writers such as the twelfth-century John Zonaras continued the tradition, tacking on synopses of events closer to their own time (albeit without yet gaining English translators).43 Thus the recording of political and ecclesiastical occurrences was partly done by churchmen or pious laymen who accepted the empire as divinely instituted but whose standpoint towards individual emperors was semi-detached, when not rather hostile. A different perspective comes from works focused on the empire’s earthly fortunes, but likewise drawing on heterogeneous texts over a lengthy period. One such is that of John Skylitzes, covering the period from Michael I’s accession in 811 until the later eleventh century. Skylitzes recast a medley of earlier works, including part of the text known as Theophanes Continuatus and, probably, the memoirs of a general, written in his disgruntled retreat and recounting his exploits from Armenia to Sicily.44 Thus events of the earlier and middle empire are relayed by a variety of voices now available in English. Their coverage is, admittedly, mainly of politico-military and ecclesiastical affairs, and the variations all pay at least lip service to the idea that the ideals of orthodoxy and empire were interdependent, although not all writers considered them identical. non-‘roman’ accounts and didact ic texts The non-specialist, forewarned, will probably find the differences in nuance between historical accounts written from court and from severely Christian perspectives more illuminating than confusing. Two other types of evidence supply contrasts or supplements to what is on offer from Byzantine narratives. Firstly, accounts penned by persons with no ambition whatsoever to be considered ‘Roman’ reveal much about Byzantine history, as the frequent citations from them in chapters below will attest. Here one may note that several conflicts or confrontations between the empire and external 42 43

George Synkellos, Chronography, tr. Adler and Tuffin, pp. xxxv–xlii, xlvi–xlviii. John Zonaras, Annales [Epitomae historiarum], ed. Pinder and Buttner-W¨obst. See ODB, III, p. 2229 (A. Kazhdan); Mango (1988/89). 44 Skyl., tr. Wortley (in preparation); Holmes (2005), pp. 125–52; Shepard (1975–6).

iv. tex ts in transl ation

85

forces are related from the opposing sides’ viewpoints, or in contemporary sources that were composed by third parties. A prime example is the Arab conquests in the seventh century, recounted by an eyewitness Egyptian bishop, John of Nikiu, and by the contemporary Armenian author now known as Sebeos, as well as by later Byzantine and Muslim writers.45 Other examples might include the encounters, diplomatic and military, between Byzantine emperors and the emergent German and Rus leaderships of the second half of the tenth century,46 while the First and Fourth Crusades each inspired a classic Byzantine set piece as well as vivid eyewitness accounts from westerners.47 Adversarial situations and battles are the stuff of narrative, and outside observers or travellers – other than ninthand tenth-century Muslim writers48 – have less to say about peaceful forms of exchange between Byzantium and its neighbours, or about the internal structure of the empire. This brings us to a second type of evidence that may draw the newcomer closer to the inner workings of the empire. It is neither narrative nor descriptive of Byzantium, but consists of didactic texts ranging from general theoretical considerations, maxims and counsel to precise technical instructions. In some ways these texts resemble the Byzantine source-material discussed above, seeing that they could be termed idealising or aspirational. They prescribe how things ought to be done, rather than describing things as they were. They do not amount to archival data, functioning organs of the empire in use. But the durability of some of the texts suggests that they appeared relevant, of potential invocation or practical application. The format could also allow a writer to voice opinions on contemporary issues of politics and society as well as on the abstract or the technical. This in itself gives them historical source value. Furthermore, some touch on issues of life and statecraft that seldom ranked as suitable subject-matter for formal historical compositions. Only a few examples will be cited here, not least because the Byzantines closely followed – and copied – the instructions of the ancients on so many subjects, grammar, mathematics, medicine and warfare among them. Attempts were, however, made to update received 45 John of Nikiu, Chronicle, tr. Charles; Seb., tr. Thomson and Howard-Johnston, I; Theoph., tr. Mango and Scott; Nikeph., ed. and tr. Mango; al-Tabari, Ta’rikh, tr. Yarshater et al. 46 Leo the Deacon, History, tr. Talbot and Sullivan; Skyl., tr. Wortley (in preparation); Leg., tr. Scott; RPC, tr. Cross and Sherbowitz-Wetzor. 47 The set piece by Anna Komnena (Al., tr. Sewter) may be compared with the Gesta Francorum, ed. and tr. Hill, composed by a First Crusader. For the Fourth Crusade, compare NC, tr. Magoulias with western participants: Villehard., tr. Shaw; Robert de Clari, Conquest of Constantinople, tr. McNeal; Gunther of Pairis, Capture of Constantinople, tr. Andrea. The Second and Third Crusades also receive lively treatment from (Second Crusade): Odo of Deuil, Expedition, ed. and tr. Berry; Kinn., tr. Brand; (Second and Third Crusades): NC, tr. Magoulias. A relatively temperate account of Crusaders’ dealings with the Byzantine government is supplied by the Latin archbishop of Tyre: William of Tyre, Chronicon, tr. Babcock and Krey. See also Whitby, Mary (ed.) (2007). 48 El-Cheikh (2004a).

86

general introd uction

wisdom in light of changing circumstances; occasionally a wholly new text was composed. Fortunately, the Byzantines’ more original texts and major revamps tend to attract English translators. A notable example of political thought couched as recommendations to an emperor dates from Justinian’s era, Agapetus’ Mirror of princes. Ernest Barker translated extensive sections, together with excerpts from orations and other texts bearing on political thought for eras up to the last decades of the Byzantine empire. Among the works translated and commented on by Barker are Gemistos Plethon’s ‘Address to Manuel Palaiologos on affairs in the Peloponnese’, and his ‘Treatise on laws’.49 One duty of the emperor himself was to set a moral lead, and his injunctions could have the status of solemn precepts or law. Leo VI expanded on this notion not only with his sermons, but also with other writings, including two treatises on military tactics, drawn largely from earlier texts. One of his main sources, the Strategikon of Emperor Maurice, is available in English as is the ‘constitution’ on naval warfare taken from Leo’s Tactica, but the rest of Leo’s oeuvre awaits its translator.50 Leo’s son Constantine went further still, commissioning a lengthy series of excerpts from classical and early Byzantine historical texts, each collection devoted to one topic, for example ‘plots against emperors’, ‘virtue and vice’ and ‘instructive sayings’. Constantine thereby displayed his unique access to book-learning, but his preface is addressed to ‘the public’. The texts are mostly in fairly straightforward Greek, and the lists of excerpted authors provided at the start of, probably, each set will have facilitated quick consultation. The ‘public’ probably consisted in practice mainly of persons in state service, who might benefit from picking up guidebooks, user-friendly both for practical expertise and for the broader ethical and cultural hinterland of empire. One of the few extant sets of excerpts is devoted to ‘embassies’, presumably being designed for persons involved with diplomacy in one way or another. The lengthy excerpts from a sixth-century historian of diplomatic exchanges are coherent enough for them to have been published in translation, partially reconstructing the now-lost original.51 Constantine VII’s regard for the written word as a means of enhancing good form and order is shown by the compilation on court ceremonies he commissioned. His sideswipe at Romanos I Lekapenos (920–44) for shortcomings in ceremonial (see above, p. 19) hits on a fundamental question: how to maintain stately continuity while accommodating the dynamics 49 Agapetus, Mirror of princes, partial tr. Barker; Gemistos Plethon, ‘Address to Manuel Palaiologos’ and ‘Treatise on laws’, partial tr. Barker. 50 Maurice, Strategikon, tr. Dennis; see below, pp. 498–9. For this ‘constitution’ (or chapter), see Leo VI, Naumachica, ed. and tr. in Pryor and Jeffreys (2006), pp. 483–519. 51 Men., ed. and tr. Blockley. No set of excerpts has been translated as such, but see on them Hunger (1978), I, pp. 244, 310–12, 326, 361–2; Lemerle (1986), pp. 323–32; below, pp. 511–12.

iv. tex ts in transl ation

87

of power shifts, finding room for one-off events and exceptional circumstances. The Book of ceremonies, while invoking the harmony of movement that God gives to all creation, draws partly on memoranda arising from particular occasions. Scholars have detected the layers of adaptation and improvisation underlying this, and the English translation now in preparation should make this plain.52 Constantine’s concern to uphold imperial decorum and exclusivity at all events emerges equally from the treatises on imperial expeditionary forces compiled for him.53 Most striking of all, however, are the prescriptions for divide-and-rule and other techniques of statecraft in his De administrando imperio.54 Romilly Jenkins’ translation conveys the generally plain style both of the source-materials assembled by Constantine and his aides, and of the emperor’s own written ‘doctrine’. We glimpse the Realpolitik behind the scenes, the presupposition that the barbarians for whom the grand receptions were staged were driven mainly by greed, fear and mutual rivalries. A ruler’s personal assumptions and calculations about his polity are captured, albeit in snapshot form, to a degree virtually unparalleled among pre-modern states. And all in the name of taxis. encyclopaedias and lexicons Constantine VII’s significance in commissioning digests of useful knowledge from ancient texts is generally acknowledged, but he was tapping into and trying to direct an intellectual trend under way long before his time, which Paul Lemerle has termed ‘encyclopedism’.55 A collection of late antique texts on the care and medical treatment of horses that Constantine had revised has been expounded in detail, although not yet translated into English.56 A major compilation made shortly after Constantine’s time was a lexicon, the Suidae lexicon, containing entries on a variety of words, names and subjects, mostly classical and scriptural, in alphabetical order. This is being translated online.57 m ilitary and other instruct ive manuals Several texts dealing with military matters have been translated into English, and their varying degrees of indebtedness to ancient tactical manuals assessed. Essentially, the Byzantines borrowed extensively but made 52 DC, tr. Moffatt and Tall (in preparation); for the translation of an important chapter, see Featherstone, ‘Di’ endeixin’, pp. 81–112. See McCormick (1985); McCormick (1990); Dagron (2000); Morris (2003). 53 Constantine Porphyrogenitus, Three treatises, ed. and tr. Haldon. 54 DAI, tr. Jenkins. 55 Lemerle (1986), pp. 309–46; ODB, I, pp. 696–7 (A. Kazhdan). 56 McCabe (2007). 57 www.stoa.org/sol.

88

general introd uction

adjustments to fit current circumstances, with Emperor Nikephoros II Phokas issuing fresh prescriptions on, for example, equipping, training and deploying heavy cavalry and other, lighter categories of cavalrymen.58 Another work, on siege warfare, makes classical techniques of building and operating siege-engines readily visualisable and applicable for present-day operations against the Arabs.59 These texts mostly date from the decades following Constantine VII’s reign, registering the switch towards more sustained offensive warfare on enemy territory. The change is signalled in the preface to the manual on Skirmishing commissioned by Nikephoros II: the tactics prescribed here ‘might not find much application in the eastern regions at the present time’, now that the Muslims are being driven back; but a written record is desirable, in case ‘Christians need this knowledge’ again, and have to contend with raiders ranging freely across Asia Minor.60 Another shift in priorities features in Campaign organization, a work envisaging warfare in Bulgarian territory and assuming that the emperor will be in command. It probably dates from the earlier part of Basil II’s reign.61 Similar provision for new circumstances is made in Nikephoros Ouranos’ Taktika, written while he was military governor of Antioch in the early eleventh century. Ancient military texts together with Leo VI’s and Nikephoros Phokas’ treatises are supplemented by chapters on, for example, cavalry warfare and sieges. These chapters, which have been expertly translated, cover ‘the full range of contemporary Byzantine military operations’ in the region of occupied Syria.62 Prescriptive handbooks could be more discursive. One such, conventionally termed the Strategikon of Kekaumenos, we have noted above (p. 67). This contains edifying maxims, tips on household management and social relationships, and counsel about serving as a judge in the provinces. Kekaumenos’ bias is, however, towards officers’ training: he had himself been a senior commander in the mid-eleventh-century army. An English translation is in preparation, supplementing the Russian translation.63 No precise analogy to Kekaumenos’ work is known. But it survives in just one manuscript. Similar sets of instructions could well have been composed by commanders or civilian officials, without the good fortune of manuscript 58 Nikephoros II Phokas, Praecepta militaria, ed. and tr. in McGeer (1995), pp. 12–59 (text); pp. 181–8, 211–17, 226–9 (commentary). 59 ‘Heron of Byzantium’, Parangelmata poliorcetica – Geodesia, tr. in Sullivan, Siegecraft, pp. 26– 151 (text), pp. 1–24 (introduction); see also De obsidione toleranda, ed. van den Berg, tr. Sullivan, pp. 150–263. 60 Skirmishing, ed. and tr. Dennis, p. 147. 61 Campaign organization, ed. and tr. Dennis, pp. 289, 291, 305 (text), pp. 242–3 (introduction). 62 Nikephoros Ouranos, Taktika, chs. 56 through 65, ed. and tr. McGeer (1995), pp. 89–163 (text); p. 81 (introduction). 63 Kek., ed. and Russian tr. Litavrin; tr. Rouech´ e (in preparation).

iv. tex ts in transl ation

89

survival. In fact, comparable stylistic traits, rhetorical devices and didactic tone characterise some of the military treatises discussed above; they are also discernible in Skylitzes’ chronicle. Kekaumenos dismisses unnamed rivals in stressing that his work stems from ‘authentic experience’, presenting ‘things not in any other Strategikon or any other book’.64 At the same time he presupposes readers’ familiarity with heroes such as Scipio Africanus and Belisarius. His work opens a window on under-chronicled Byzantine officialdom, on men educated in grammar and rhetoric, but not to the highest level. Opinionated and idiosyncratic Kekaumenos may have been, but his value-system was probably common to many of the empire’s servants. They were interested in relating recent developments to the classical past, preoccupied with issues of technique and policy, yet also disposed to pass useful knowledge, topped with pieties and worldly wisdom, on to their juniors in age or status.65 This political culture could act as a bonding mechanism, providing middling officials scattered across outposts of empire with a common stock of know-how, anecdotes and semi-learned allusions. A certain esprit de corps was thereby fostered. But this was no closed body. The military manuals and other practical works imply concern to introduce newcomers or successors to the systems they will have to operate. Most also place presentday norms and practices within the framework of the ancients, still deemed past-masters. The very fact that the counsel was set down in writing suggests that processes of training and dissemination went on beyond the confines of formal education. The attempts at spelling out military techniques in plain words, simplifying classical terminology, also bespeak ambitions for learning, for self-improvement, on the part of individuals coming from outside the gilded circles. In other words, the instruction manuals themselves constitute evidence of the means whereby the upwardly mobile could hone their military and other skills, gain a certain polish, and ultimately rise higher in the empire’s service, especially during its era of expansion, the tenth and eleventh centuries. They would need Greek to understand the manuals and most would be Byzantine-born. But individuals among neighbouring elites, or visitors to the empire, could manage some Greek, written as well as spoken. Didactic texts would have been of use to, for example, the young Norman noble who learnt not only Greek at court but also veterinary medicine for horses and birds in the mid-eleventh century.66 64 65

Kek., ed. and Russian tr. Litavrin, pp. 172–3, 164–5. Rouech´e (2003), pp. 27–8, 33–7; Rouech´e (forthcoming); seminar paper by Catherine Holmes, ‘Literacy and written culture in Byzantine political culture’ (17 June 2005, Oxford). 66 Chronique de Sainte-Barbe-en-Auge, ed. Sauvage, pp. 56–7. The information about this visitor to Constantinople is contained in a commemorative note added at the end of the (uncompleted) chronicle. See also Ciggaar (1996), p. 180; Amsellem (1999).

90

general introd uction

Paradoxical as it might seem, texts covering military matters could become available to outsiders. In fact a section in Kekaumenos’ Strategikon directly addresses a toparch, a local potentate in the borderlands. He is advised to be wary of the emperor’s blandishments, paying just one visit to Constantinople if he values his independence. The fate of an incautious toparch is recounted, by way of warning, and another section features the wiles of one of Kekaumenos’ own ancestors, a toparch in the Armenian borderlands who outwitted imperial commanders.67 Thus a senior military officer could proudly recall Armenian family roots and envisage sympathetically a contemporary toparch’s viewpoint. There is no reason to doubt Kekaumenos’ overriding loyalty towards the emperor, or that his prime self-identification was Roman. But Kekaumenos had not wholly relinquished ties with another culture, an alternative identity, and in that sense he exemplifies the multiple or mutable personae of many serving in the empire’s higher echelons, especially the armed forces. His ‘life and opinions’, while personal to the point of idiosyncracy, do much to explain Byzantium’s sinews of governance (see above, pp. 15–16). Kekaumenos’ injunctions, with other more technical treatises, are now becoming available to Anglophones; in reading these works, the newcomer to Byzantium can gain a direct impression of what it was to make oneself a Roman. short-cuts to byz antium There are many forms of short-cut to the study of Byzantium, literally so by way of atlases. Most aspects of its historical geography are authoritatively covered by John Haldon in The Palgrave atlas of Byzantine history,68 and the early phases of the empire and of the Christian church are charted in detail in The Barrington atlas and The atlas of the early Christian world.69 Detailed historical atlases of neighbouring peoples and regions are also available in English;70 likewise with the religious and other movements from outside that had some bearing on the empire’s fate.71 Online guides are likely to extend horizons further, in terms not only of geography but also of art and visual culture.72 The Oxford dictionary of Byzantium covers virtually every aspect of Byzantium across the ages, from the spiritual to the archaeological, while a broad canvas is presented in The Oxford handbook of Byzantine studies.73 Several 67 69 70 71 72

68 Haldon (2005c). Kek., ed. and Russian tr. Litavrin, pp. 316–19, 186–7. Talbert et al. (eds.) (2000); van der Meer and Mohrmann (1966). Hewsen (2001); Rapp and Awde (eds.) (forthcoming). Kennedy (ed.) (2002); Riley-Smith (ed.) (1991). See, for example, Byzantine Links listed on the website of the Society for the Promotion of Byzantine Studies (www.byzantium.ac.uk). 73 ODB, 3 vols.; Jeffreys et al. (eds.) (2008, forthcoming).

RUS LITHUANIAN-DOMINATED RUS (C14)

BE S

FRANCIA

BIA RA t SA ru

MORAVIA NORICUM

P

GAUL

PANNONIA

MOLDAVIA

CASTILE

BOSNIA RASKA

ILL YR

IA

DOBRUDJA

MOESIA

ˆ

LA

WALLACHIA

C DO UE NG

RUMELIA (C16 on)

ZETA

EP IR

O

S

THRACE

THESSALY

HE

LL AS

BOEOTIA

ACHAIA

[THE MOREA]

FR

KOSOVO EDONIA MAC

IU

LI

IQ

YA

ISTRIA

RIA GU LI

TUSCANY

ET RU

FRA

NG

THE PENTAPOLIS

OC

CRO AT IA

DA

ROMAGNA

A RI

IF

R

ETIA VEN

HO IL

LM

AT

RIO

LY R

IC

IA

MARCHE ABRUZZI MOLISE TERRA DI CAPITANATA LAVORO CA APU MP LIA AN IA D OBAR IA L A NG CILENTO LUCANIA CALABRIA/ BRUTTIUM

0 0

250

500 250

750

1000 km

500 miles

L I B YA

Map 3a Archaic, and other less familiar, names: the central Mediterranean world

N

UM

92

iv. t ex ts in t ra ns l at ion

general introduction

93

KHAZARIA

A

CRIMEA GO TH I

S E A

ARMENIA

(see inset 1)

R

Fergana

IZ

M xu

O

A S E

(ASIA MINOR)

N

(see inset 2)

THE PONTOS

A

I A

CAUCASIA ANATOLIA

W

S P

KH

C A

B L A C K

TRANSOXIANA

s

HYRCANIA

KURDISTAN

AL-JAZIRA M

SYRIA

ES

O

PO

Nishapur

GURGAN

KHORASAN PARTHIA

M

HAWRAN

MEDIA TA

Ghazna

IA

LEVANT

KH UZ N TA IS

ASHURESTAN

M

T

H

I nd u s

OUTREMER

ES

HA

N

BLACK SEA

F A R S

E

BITHYN

MYSIA

IA

GALATIA

N

PHRYGIA

A

T E

JD

H

LYDIA IONIA

C

H

CARIA

AZ IJ

CAU

HA ZIA

ZIC A LA

BLACK SEA

KL

IBERIA I JET AR

CHALDIA TAO

200

400 200

600 400

800

H I M YA R

1000 km 600 miles

(Theodosioupolis, Erzurum)

Map 3b Archaic, and other less familiar, names: Asia Minor and the Middle East

S

ax

es

SHI RA K

APAHUNIK / LOWER MEDIA TARON N KA Lake Van RA PU ARZANENE AS V

Karin

Kars VANAND

AZIA NT K ART LI A I N

Ar

Trebizond

INSET 2

C A ALANIA SU S M OU

SUANIA

ETHIOPIA

CIA

A

BK

DO

PAM ISAURIA PH YL I C IA I CI L

Rhodes A

A

A PP

PISIDIA LYCIA

ABYSSINIA/

INSET 1

A GONI HLA PAP

SI

AL

W

NI

BA

NI

A

K

THE MUGHAN

AZERBAIJAN

94

general introd uction

other introductory multi-authored works or broad synopses appeared early in the twenty-first century.74 The main papers and abstracts of the Twentyfirst International Congress of Byzantine Studies (2006), together with other proceedings published shortly afterwards, summed up the scholarly state of play across the field, the greater part of these papers having been presented in English.75 The economic history of Byzantium, covering the Byzantine world from the seventh to the fifteenth centuries, has already been mentioned (see above, pp. 36–7). So have the many accessible introductions to the art and archaeology of Byzantium, the introduction to alternative forms of imagery by Maguire and Maguire (2007) among them (see above, n. 19 on p. 79). Entries on all forms of Byzantine art history (in German) are supplied by the Reallexikon zur byzantinischen Kunst,76 while studies in English as well as other western languages on virtually every aspect of late antique history and culture are published in the compendious Aufstieg und Niedergang der r¨omischen Welt.77 A pithier synopsis emerges from the thematic essays and articles in Bowersock et al. (eds.) (1999), covering the antique Christian and Islamic worlds from the mid-third century until the end of the eighth. A chronology of salient political, military and ecclesiastical events year by year from 330 until 1461 is provided by A chronology of the Byzantine empire.78 These works can be supplemented in highly flexible ways by the online Prosopography of the Byzantine world.79 This offers a full, reliable chronology for most of the eleventh and twelfth centuries and gives details about individuals eminent or obscure. Its gateways also open up to the enquirer a range of thematic topics: for example, ‘murder’ will bring up a list of all those persons said to have been murdered during that period. Traditional reference works for the cultures and religions most closely linked with Byzantium remain of value as introductions and suppliers of background information, notably The Oxford classical dictionary, The Oxford dictionary of the Christian church and The early Christian world.80 The revised edition of The encyclopedia of Islam contains many entries on places within the Byzantine empire, encompassing the period before they came permanently under Muslim rule.81 The Dictionnaire d’histoire et de g´eographie eccl´esiastiques contains entries on places, rites and persons of significance

74 Angold (2001); Mango (ed.) (2002); Morrisson et al. (eds.) (2004); Harris (ed.) (2005); Gregory (2005); Cameron, Averil (2006b); James (ed.) (forthcoming); Cheynet (ed.) (in preparation); Laiou (ed.) (in preparation); Stephenson (ed.) (in preparation). See also above, nn. 1–3 on p. 22. 75 ACIEB 21 (3 vols.). Other publications include Litavrin (ed.) (2006); Franklin and Mavroudi (eds.) (2007); SBS 10 (forthcoming); Whittow (ed.) (forthcoming). 76 RbK. 77 ANRW. 78 Venning (ed.) (2005). 79 PBW. For the earlier period of 641–867, see Martindale (2001); PMBZ, I. 80 Hornblower and Spawforth (eds.) (2003); Cross and Livingstone (eds.) (2005); Esler (2000). 81 EI.

iv. tex ts in transl ation

95

in the eastern church, while the Lexikon des Mittelalters gives balanced treatment in depth of Byzantium as well as the west.82 Those wishing to follow the short-cuts through to the point of learning something of the language as written and spoken by the Byzantines have a number of choices. They may start with the classical Attic Greek to which members of the elite aspired, or with New Testament Greek, which is not so far removed from the everyday language of the earlier medieval Byzantines. Standard grammars and self-help courses offer instruction in these forms of Greek. Good introductions are also available for persons wanting to trace the historical connections between Byzantine Greek and the Greek in use today, or to learn something of the grammar of the modern language.83 The Greek script is explained in detail for newcomers as well as specialists by contributions to Greek scripts, while the new lexicon of Byzantine Greek, supplementing the classical dictionary of Liddell and Scott, is nearing completion.84 Finally, those who embark on systematic self-tuition or who contemplate offering a lecture or two or even a course on Byzantine history may turn to offerings in the online ‘overnight expert’ series. One of these is dedicated to the teaching of Byzantium by non-specialists. It provides some suggestions for essay questions or coursework, together with reading lists, and it points out where a Byzantine dimension can usefully be added to standard western medieval teaching topics. The closely related history of the Armenians is also covered in this series.85 The short-cuts mentioned in this section should help make basic facts and historical issues reasonably clear and communicable to and by nonspecialist teachers. They and their students have online access to sources in English translation and to guides to those sources (see above, pp. 77–8 and nn. 14, 15), while Byzantine landscapes, buildings and imagery can be accessed cheaply and accurately. In that sense, the many roads to Byzantium are wide open to travellers as never before. 82 83

DHGE; LexMA. Browning (1983); Farmakides (1983); Horrocks (1997); Holton et al. (2004) (with references to other coursebooks). 84 Easterling and Handley (eds.) (2001); LBG; see also above, p. 59. 85 http://www.heacademy.ac.uk/hca/resources/detail/teaching byzantium; http://www.heacademy. ac.uk/hca/resources/detail/teaching medieval paper armenia.php

PA RT I

T H E E A R L I E R E M P I R E c. 5 0 0 – c. 7 0 0

CHAPTER 1

JUSTINIAN AND HIS LEGACY (500–600) an drew lou t h

an empire of cities The beginning of the sixth century saw Anastasius (491–518) on the imperial throne, ruling an empire that was still thought of as essentially the Roman empire, coextensive with the world of the Mediterranean. Although Anastasius ruled from Constantinople over what we call the eastern empire, the western empire having been carved up into the ‘barbarian kingdoms’, this perspective is ours, not theirs. Through the conferring of titles in the gift of the emperor, and the purchasing of alliances with the wealth of the empire – wealth that was to dwarf the monetary resources of the west for centuries to come – the barbarian kings could be regarded as client kings, acknowledging the suzerainty of the emperor in New Rome, and indeed the barbarian kings were frequently happy to regard themselves in this light (see below, p. 198). The discontinuation of the series of emperors in the west, with the deposition of Romulus Augustulus in 476, was regarded by very few contemporaries as a significant event; the notion that east and west should each have their own emperor was barely of a century’s standing, and the reality of barbarian military power in the west, manipulated from Constantinople, continued, unaffected by the loss of an ‘emperor’ based in the west. The empire that Anastasius ruled was still the Mediterranean world as it had been since classical times in more than just a political sense: it consisted of a world whose basic unit was the city which, with its hinterland, formed a self-sufficient economic and even cultural unit. Although shorn of the political powers of the old city-state, the notables of the city still exercised considerable political influence and the provincial governors, appointed from the same social class as these notables, frequently found it more effective to recognise local influence than to challenge it. The cities – with fora, theatres, courts and opportunities for education – formed the seedbed for the educated elite who held posts in the imperial administration, often returning to the cities to enjoy the essentially rural wealth generated by their country estates. All this was to change from the sixth century onwards, though there is a good deal of debate about the rate at which this change took place. 99

100

the earlier empire

1. just inia n a nd his legacy ( 50 0 –60 0 )

Belgrade

Sirmium

Tomi

Adrianople Constantinople

Thessalonica

Nikomedia Nicaea

Pergamon

Corinth

Athens

Sardis Philadelphia Smyrna Ephesos Miletos

Aphrodisias Halicarnassus

Ptolemais

Alexandria

0 0

100

200 100

300 200

400

500 km

Nil

e

Memphis

Oxyrhinchus

300 miles

Map 4 An empire of cities: Byzantium c. 500 Cambridge Histories Online © Cambridge University Press, 2008

101

102

the earlier empire

The city was also the basic unit of the Christian church. From the end of the second century Christianity, which from the start had been a predominantly urban phenomenon, had developed an organisation based on the city and its hinterland; it was led by a single officer, called a bishop, who was appointed for life.1 With the gradual Christianisation of the Roman empire from the fourth century onwards, the bishop became a considerable figure among the notables of the city. He was sometimes appointed defensor civitatis, that is the leader or ‘judge’ of the city, and he regularly exercised the functions of this post, even when not officially appointed to it. Despite the decline of the city as an economic and cultural entity,2 the link between bishop and city was to continue. Christianity had never been a particularly peaceful religion, and the importance it attached to correctly formulated beliefs, combined with its increasing social influence as fewer and fewer inhabitants of the empire resisted the pressure to embrace Christianity, meant that well before the sixth century Christian belief had become both a cause of social, political and cultural divisions, and a means of articulating them. Modern historians are shy of regarding religious belief and practice as the reason for social and political divisions, and in general they may well be right, but it is undeniable that in this period division was often expressed and understood in religious terms. As we shall see, issues of religious difference are woven into the narrative of sixth-century history. It is important to understand the basis for these differences before going on to consider other explanations for social, political and cultural divisions that were expressed in these terms. Religious conflict is a theme to which we shall often return. religious divisions and o ur sources f or the six th century Anastasius inherited, and promoted, religious divisions that were to cast a long shadow over the Christian Roman empire. These religious divisions derived in the first instance from the council of Chalcedon (451), which attempted to settle long-standing differences about how godhead and manhood were united in Christ. The fathers of the council were almost entirely Greek, while the pope of the day, Leo I (440–61), played an important role through his legates. A formula acceptable to the papal legates was eventually agreed, which they regarded as endorsing the teaching of Cyril, the great 1 Translation from one city to another was forbidden by canon 15 of the council of Nicaea, although there were rare exceptions. 2 The question of the decline of the late antique city, and how such decline is to be interpreted, really becomes critical in the seventh century: see below, pp. 221, 224. For two general accounts, see Mango (1980), pp. 60–87; Liebeschuetz (1992).

1. j ustinian and his legacy (500–600)

103

patriarch of Alexandria (412–44), who was held in the highest regard by all but a small minority of the eastern bishops. But as a hard-won concession to the papal legates, Chalcedon recognised the unity of Christ’s person ‘in two natures’. This is not a phrase found in Cyril, but was taken from a papal letter – the so-called ‘Tome of Leo’ – which was received by the council. This concession spoilt the achievement of Chalcedon; many Christians, especially in Syria and Egypt, felt that the council had betrayed, rather than endorsed, Cyril. Rejection of the decision of Chalcedon often took violent forms: Juvenal, bishop of Jerusalem, needed imperial troops to make a safe entry into his episcopal city; and Proterios, appointed to replace Cyril’s successor who had been deposed by the council, was murdered by the mob. The violence that often accompanied these religious differences was regularly fostered by monks, who were increasingly becoming a force to be reckoned with in the Christian empire. After unsuccessful attempts to enforce Chalcedon, in 482 Emperor Zeno (474–91) issued a statement of belief with the intention of securing unity (called the Henotikon), which disowned Chalcedon, though it fell short of condemning the council. The Henotikon was the work of Acacius, patriarch of Constantinople (471–89), and Peter Mongos ‘the Hoarse’, patriarch of Alexandria (477, 482–89). However Rome, and the Latin west in general, was not willing to disown what it regarded as the council of Pope Leo; the promulgation of the Henotikon thus provoked the Acacian schism with Constantinople, named after its patriarch, which lasted until the death of Anastasius. For the Henotikon remained imperial policy during the reign of Anastasius who, if anything regarded the edict as too moderate, since he promoted those who rejected the Henotikon for not explicitly condemning Chalcedon. Our sources for the sixth century, although on the face of it plentiful, leave much to be desired. Histories on the classical model survive intact, in contrast to the fragmentary fifth-century histories, and these include Procopius’ Wars, the Histories of Agathias and Theophylact Simocatta, and substantial extracts from the History of Menander the Guardsman. These are complemented by chronicles – a new form of history writing of Christian inspiration – such as those by John Malalas (which only survives in an epitomised form) and Marcellinus, as well as the later Paschal chronicle (630) and the Chronicle of Theophanes (dating from the early ninth century, but incorporating earlier material). Church histories evolved from the form of the chronicle, and the main sixth-century example is that of the Antiochene lawyer, Evagrius Scholasticus. Such Christian history writing regarded the traditions of saints’ Lives as important, and there is a good deal of hagiographical material relating to the sixth century. Much of this is valuable for the social, as well as the religious, history of the period, notably the collections by Cyril of Scythopolis and John Moschus, together with

104

the earlier empire

the lives of individual saints (for example of the stylites or Theodore of Sykeon). To these can be added texts that are written, or survive, in Syriac, representing the views of those non-Chalcedonian Christians (monophysites) excluded from the imperial church by the drive towards a form of Chalcedonian orthodoxy promoted by Justinian and his successors. These include saints’ Lives by Zacharias of Mytilene, which were originally written in Greek, although his Church history does not advance into the sixth century; and a collection of saints’ Lives and a Church history by John of Ephesus, who wrote in Syriac.3 There is also an anonymous eighth-century chronicle, attributed to Pseudo-Dionysios of Tell-Mahre, and the twelfth-century chronicle of Michael the Syrian. Traditionally, the tendency has been to take the classicising histories at face value as a basic record, to be supplemented, with varying degrees of caution, from the chronicles and ecclesiastical sources.4 The trend of recent scholarship, however, has been to pay much more attention to the intentions and bias of the classicising historians, with the result that we now see in these sources a variety of sharply defined perspectives on the sixth century, rather than a straightforward narrative record that can be used as a basic framework.5 Archaeology is an important resource, not least over major imponderables, such as the decline (or survival) of the city, economic prosperity and climatic change. In addition we can also draw information from epigraphy, coins and seals, and make use of the evidence (still little used) that remains embedded in the conservative, yet developing, liturgy of the churches. Accounts of the second half of Anastasius’ reign indicate mounting popular unrest, ostensibly because of the emperor’s religious policy. Behind this may lie growing economic difficulties and an increasing sense of insecurity within the empire. At the beginning of the sixth century the long peace with Persia, the traditional enemy of the Roman empire, and indeed of its predecessors, came to an end. The Persians’ failure to restore Nisibis to the Roman empire, in accordance with a treaty made with Emperor Jovian in the fourth century, led the East Romans to withhold tribute payments; this, in turn, prompted the Persians to invade the Roman empire in 502 and they quickly took a number of frontier towns, including the city of Amida (see below, p. 135). To begin with, Roman resistance was weakened by a divided command, and it was not until 505 that the Romans recovered Amida. The weakness of the Mesopotamian frontier revealed by this war was remedied by the building of the fortress at Dara, close to the frontier and a few 3 The third part of John of Ephesus’ Historia ecclesiastica survives in a single manuscript, while the first two parts survive in fragmentary form, incorporated into later Syriac chronicles. 4 This is Gibbon’s method (1776–88), still used by Bury (1923) and even Jones (1964). 5 See, notably, Cameron, Averil (1985).

1. j ustinian and his legacy (500–600)

105

miles from Nisibis; it was called Anastasiopolis, after the emperor. In the north, too, there were threats from invaders in the early sixth century, and archaeological evidence suggests that the fortresses which Procopius says were built along the right bank of the Danube in the reign of Justinian (527–65) were at least begun by Anastasius.6 The riots venting opposition to Anastasius’ religious policy were triggered by a matter of liturgy. From the middle of the fifth century, the chant called the Trisagion (‘holy God, holy strong, holy immortal, have mercy on us’) had become a popular part of the liturgy in the east. In Syria this chant was understood to be addressed to God the Son; in order to underline the belief of those rejecting Chalcedon’s distinction between the two natures in the Incarnate Son, the phrase ‘who was crucified for us’ was added to the chant, affirming their conviction that in Christ, God himself had embraced human suffering (a doctrine called theopaschism). In Constantinople, however, the chant, with its triadic form, was understood to be addressed to the Trinity, so such an addition seemed to imply that the divine nature itself was subject to suffering. Behind the differing texts of the chant, there lay genuine mutual misunderstanding, but that only made each side’s sense of the other’s error more acute. When Anastasius directed that the theopaschite addition should be included in the Trisagion, it provoked a riot between non-Chalcedonian monks chanting the amplified form and the clergy and people of Constantinople. This led to popular demands for the deposition of the emperor, demands only quelled by the emperor himself facing the mob in 512, without his diadem, and inspiring an acclamation of loyalty. In the following year the emperor faced a further challenge to his authority from Vitalian, a military comes, who claimed to represent the reaction of the orthodox to the policies of the emperor. Although unsuccessful in his challenge to the throne, he outlived the emperor. the rise of justinian and the quest ion of his ‘grand d esi gn’ Anastasius died in 518, leaving the question of his succession undecided. He was succeeded by Justin I (518–27), a peasant from Illyria, who had risen through the ranks to become count of the excubitors. He was uneducated, perhaps even illiterate, and Procopius would have us believe that the real power behind the throne was Justin’s nephew, Peter Sabbatius, who took the name of Justinian; Justin had earlier brought him to the capital and lavished an expensive education on him. It is hard to say how true this is, for there is no independent evidence to support the claim.7 6

Poulter (1983), p. 97, cited by Cameron, Averil (1985), p. 220, n. 90.

7

Honor´e (1978), p. 7.

106

the earlier empire

Justin’s first act was to repudiate his predecessors’ attempts to achieve unity among the Christians by ignoring, or even implicitly condemning, the council of Chalcedon: the Henotikon was revoked and Chalcedonian orthodoxy became imperial policy. Justin announced his election and religious policy to Pope Hormisdas (514–23), who sent legates to Constantinople; a council was held there to confirm the ending of the Acacian schism and to condemn those who had promoted it. These included not only Acacius and those successors who had agreed with him, but also – and in this exceeding papal demands – the emperors Zeno and Anastasius. Prominent non-Chalcedonian monophysites, including Severus of Antioch and Philoxenos of Mabbug, were deposed and exiled. Reconciliation with Rome only reopened the wounds that the Henotikon had tried to heal, but very soon a refinement of Chalcedonian orthodoxy was put forward that was to become the focus of Justinian’s endeavours to achieve religious unity. A group of monks from Scythia, led by John Maxentius, brought their proposal to Constantinople: it involved supplementing the Chalcedonian definition with the affirmation that ‘one of the Trinity suffered in the flesh’. This affirmation would appeal to the monophysites’ conviction of the indivisible unity of Christ, which had found expression in the theopaschite addition to the Trisagion. Justinian was attracted by this proposal and sent the monks off to Rome, where they failed to convince Pope Hormisdas, though others found it acceptable, notably Dionysius Exiguus and Boethius. The proposal remained dormant until the 530s, when Justinian’s religious endeavours began in earnest. In spring 527 Justin fell ill, and Justinian was proclaimed augustus in April; four months later Justin died, and Justinian succeeded him. His reign lasted until 565, thirty-eight years in all – or forty-seven, if one includes his stint as the power behind Justin’s throne. This was an exceptionally long reign and its duration would have been an achievement in itself. But there was much else besides: reform of the legal code; reconquest of Roman territories in North Africa, Italy and Spain; grandiose rebuilding projects, notably the rebuilding of the centre of Constantinople, including the Great Church of the Holy Wisdom, St Sophia; the closure of the Platonic Academy in Athens; and a religious policy culminating in the fifth ecumenical council, held at Constantinople in 553 (or, to adopt a different perspective, in his lapse into heresy in his final months). The temptation to see all these as parts of a jigsaw which, when correctly fitted together, yield some grand design is hard to resist. And then there is glamour, in the person of Theodora, the woman he married. In doing this, Justinian circumvented the law forbidding marriage between senators and actresses; even Procopius acknowledges her beauty, while regarding her as a devil incarnate. He wrote a malicious account of Theodora’s meddling in the affairs of state in his Secret history. Procopius also relates how

1. j ustinian and his legacy (500–600)

107

during the so-called Nika riot in 532, when Justinian was terrified by the rioting against his rule and was contemplating flight, Theodora persuaded him to stay and face either death or victory with the dramatic words, ‘the empire is a fair winding-sheet.’ All this prepares the way for assessments of Theodora that rank her with Byzantine empresses like Irene or Zoe, both of whom (unlike Theodora) assumed imperial power in their own right, albeit briefly.8 The ‘grand design’ view of Justinian’s reign sees all his actions as the deliberate restoration of the ancient Roman empire, though a Roman empire raised to new heights of glory as a Christian empire confessing the orthodox faith. According to this view, reconquest restored something like the traditional geographical area of the empire; law reform encapsulated the vision of a Christian Roman empire, governed by God’s vicegerent, the emperor; the capital’s splendid buildings, not least the churches, celebrated the Christian court of New Rome, with the defensive buildings described by Procopius in the later books of his Buildings serving to preserve in perpetuity the newly reconquered Roman world. The defining of Christian orthodoxy, together with the suppression of heterodoxy, whether Christian heresy or pagan philosophy, completes the picture. In discussing Justinian’s reign it is therefore difficult to avoid the notion of a grand design. Virtually all our literary sources reflect something of this idea. It is there in Procopius (even the Secret history sees Justinian as a grand designer, albeit malign), in the legal texts and even in the ecclesiastical texts written by those who experienced persecution at Justinian’s hand: the monophysites shared with those who embraced imperial Christianity the vision of a Christian empire ruled by a Christian emperor.9 It is hardly to be denied that there were moments when Justinian fancied he was fulfilling some such grandiose design. In 536, after reconquering Sicily, Justinian affirmed, ‘we have good hope that God will grant us to rule over the rest of what, subject to the ancient Romans to the limits of both seas, they later lost by their easy-going ways.’10 Whether we should think of Justinian’s reign as the fulfilment of a consciously preconceived grand design is another matter. This raises two interrelated questions: do all the above-mentioned elements fit together into some grand design; and, even if they do, did Justinian really have the means to bring this grand design to fruition? As we shall see, neither of these questions can be answered in the affirmative without heavy qualification. 8 For a cool appraisal of such accounts, see Cameron, Averil (1985), pp. 67–83. See below, p. 277 (Irene), p. 588 (Zoe). 9 Fowden (1993). 10 Justinian, Corpus iuris civilis, novella 30, ed. Krueger et al., III, p. 234; tr. in Honor´ e (1978), p. 19.

108

the earlier empire

Perhaps the most convincing evidence for such a grand design, at least at the beginning of Justinian’s reign, is found in his revision of Roman law. Justinian set this in hand as soon as he could, strikingly fulfilling one of the recognised roles of a ruler: that of ultimate judge and legislator. This was a task especially associated with the Roman emperor, for Romans prided themselves on living under the law, something given signal expression in Priscus’ account of the embassy to the court of Attila in the fifth century.11 Within months of assuming sole rule, Justinian had announced to the senate in a formal legal enactment (a ‘constitution’) his intention of having a new law-code prepared, that would bring matters up to date, reconcile contradictions, winnow out irrelevant legislation, and introduce clarity. He set up a ten-man commission, led by the quaestor Tribonian, which completed its work in little over a year. This code no longer survives, but five and a half years later, in 534, it was issued in revised form, arranged in twelve books and containing constitutions from the intervening period; it is this edition that has survived to exercise such an influence on subsequent European law. By the time of the second edition, there had been a further contribution to the work of legal revision, the publication of the Digest or Pandects, which reduced to order the legal opinions of centuries of Roman lawyers. This was published in December 533. A further part of the legal reform was the publication of the Institutes, a revision of the Commentaries of the second-century jurist, Gaius, which was to be the official textbook for students of law at the two official schools of law, in Constantinople and Beirut. This revision and clarification of Roman law was complemented by the later laws of Justinian, the novellae. Whereas the main body of Tribonian’s work was in Latin, most of the novellae are in Greek, for the reign of Justinian marks a watershed between the Roman empire with Latin as the official language and the so-called Byzantine empire, in which Greek was the principal and eventually the sole language. The purpose of this legal reform should be seen as twofold. It was practical: the code and the novellae provided legal norms to be interpreted by judges with the use of the Digest. It seems, however, that this function was not to continue much beyond the middle of the next century. But its other purpose was to delineate a world-view, enshrining the inheritance of Roman civilisation, the embrace of Christian orthodoxy, and the paramount position of the emperor. This was an enduring legacy, and at its heart was a vision of the complementarity of empire and priesthood, basileia and hierosyn¯e, imperium and sacerdotium. This is expressed nowhere better than in novella 6 (535): The greatest of God’s gifts to men, given from on high in accordance with his loving kindness, are priesthood and empire; the one ministers to things divine, 11

Priscus (fragment 11) in Blockley (ed.), Historians, II, pp. 242–81, esp. pp. 270–3.

1. j ustinian and his legacy (500–600)

109

the other rules and cares for matters human, both proceed from one and the same source, and set in order human life. So nothing is more sought after by kings than the dignity of priests, if they beseech God continually on their behalf. For if the one is always unblemished and has open access to God, while the other rightly and fitly orders the received form of government, then there will be a fair harmony, and everything that is good for the human race will be granted. We therefore have the greatest care for the true dogmas of God, as well as for the dignity of the priests, which we believe cares for them, as through it good gifts are given us from God, so that what we have we possess securely, and what we have not yet attained we shall come to acquire. Thus everything will be done rightly and fitly, if the beginning of everything is proper and acceptable to God. We believe that this will be so, if the observance of the holy canons is preserved, which has been handed down by the apostles, who are rightly praised and venerated as eyewitness and ministers of the word of God, and which has been safeguarded and interpreted by the holy fathers.

Such comprehensive legislative activity can hardly be regarded as other than part of a grand design of imperial rule. The next essential ingredient, reconquest of lost imperial territory, as we have seen above (p. 107), also inspired in Justinian the conviction that he was the divine agent in reconstituting the Roman empire in a Christian form. But was this a settled conviction, or a passing hope? The facts about Justinian’s reconquest of North Africa, Italy and Spain are not in doubt, although we are poorly informed about the Spanish expedition; their interpretation is much more hazardous. In 533, Justinian despatched his general Belisarius to North Africa with an impressive force of 10,000 infantry and 5,000 cavalry.12 However, the reasons for his determination that this enterprise should not fail are perhaps more down-to-earth than the fulfilment of some grand design of imperial restoration. Justinian had only just recovered from the Nika riot, and Emperor Leo’s disastrous attempt in 468 to dislodge the Vandals made it imperative that Belisarius’ expedition should succeed if Justinian’s credibility as emperor were to recover. Even Procopius’ celebratory account seems to depict Belisarius’ swift success as fortuitous. The Italian expedition, which followed up this success, seems to have been a much more modest affair: only 7,000 troops were involved, compared to the 6,000 sent with Narses in the same year to Alexandria, to protect the monophysite patriarch Theodosius (535–6). At this stage it would seem that the expedition was little more than a matter of showing the flag, even if its early successes, following so closely on the victory over the Vandals, conjured up in Justinian’s mind ideas of a grand design, as witness the novellae of the period. In reality, the reconquest of Italy proved to be a long-drawnout affair, during which Italy itself was devastated.13 By 554, however, when Italy was formally restored to Byzantine rule (by a ‘pragmatic sanction’), 12

See below, p. 202; Barbero and Loring (2005), pp. 182–3.

13

See below, pp. 205–9.

ANTES

S

LOMBARDS A

N

AVARS

DI

UN

A

GEPIDS

SL

Ravenna

ALANS ABK

RG

R

BU

F

SUEVI

S NK

H AZ

S AV

IA

LAZES

Danube

SIGOTHS VI

Rome

IBERIANS

Constantinople

Naples

Cordoba

NS

ARMENIANS

Septem

RB

ER

S

LAK

Alexandria G

H

A

BE

Syracuse

S

Carthage

Extent at Justinian's accession Justinian's reconquests 0 0

250

500 250

750

1000 km

500 miles

Map 5 East Rome into west: Justinian’s expansion of imperial borders, and neighbouring peoples

SA

N

ID

HM

S

ID

S

1. j ustinian and his legacy (500–600)

111

most of the Mediterranean littoral belonged to the Roman empire once more. Justinian’s rebuilding programmes likewise fit uneasily into the idea of a grand design. Our principal source for Justinian’s extensive building activity is Procopius’ Buildings, which takes the form of a panegyric and consequently presents the fullest and most splendid account, drawing no distinction between new building work, restoration or even routine maintenance. As we saw earlier, the building of fortresses along the frontier, along the Danube and in Mesopotamia, to which Procopius devotes so much space, should not all be attributed to Justinian himself: as archaeological surveys have shown (and indeed other contemporary historians assert, even Procopius himself in his Wars),14 much of this was begun by Anastasius. And the great wonders with which Procopius begins his account, when describing the reconstruction of the centre of Constantinople, were consequent upon the devastation wrought by the Nika riot of 532, which Justinian can hardly have planned. But however fortuitous the occasion, the buildings erected in the wake of the riot are works of enduring magnificence, none more so than the church of the Holy Wisdom, St Sophia. Contemporary accounts are breathtaking. Procopius says: The church has become a spectacle of marvellous beauty, overwhelming to those who see it, but to those who know it by hearsay altogether incredible. For it soars to a height to match the sky, and as if surging up from amongst the other buildings it stands on high and looks down on the remainder of the city, adorning it, because it is a part of it, but glorying in its own beauty, because, though a part of the city and dominating it, it at the same time towers above it to such a height that the whole city is viewed from there as from a watch-tower.

He speaks too ‘of the huge spherical dome which makes the structure exceptionally beautiful. Yet it seems not to rest on solid masonry, but to cover the space with its golden dome suspended from heaven.’ Contemporaries were struck by the quality of light in the Great Church: ‘it abounds exceedingly in sunlight and in the reflection of the sun’s rays from the marble. Indeed one might say that its interior is not illuminated from without by the sun, but that the radiance comes into being within it, such an abundance of light bathes the shrine.’15 Paul the Silentiary, speaking of the church restored after the collapse of the dome in 558, says ‘even so in the evening men are delighted at the various shafts of light of the radiant, light-bringing house of resplendent choirs. And the calm clear sky of joy lies open to all driving away the dark-veiled mist of the soul. A holy light illuminates all.’16 This 14 15 16

See Cameron, Averil (1985), pp. 104–10. Pr B, I.1.27, 45–6, 29–30, tr. Dewing and Downey, pp. 12–13, 20–1, 16–17. Paul the Silentiary, St Sophia, ll. 902–6, ed. Friedl¨ander, p. 252; tr. Trypanis, p. 418.

112

the earlier empire

Figure 3 The interior of St Sophia, Constantinople

stress on light as an analogy of divinity chimes in well with the vision found in the writings ascribed to Dionysius the Areopagite (commonly known as Pseudo-Dionysius); a fact surely with bearing on the huge popularity these writings were soon to assume.

BLACHERNAE

s

Forum of Constantine Forum of Theodosius

M es e

Hi

CONSTANTINOPLE

Acr op oli s

Va len

1 0.5

1.5

o

CHALCEDON

The Long Wall

CONSTANTINOPLE

Sea of Marmara

2 km

HIERON

Palace of Sophianai (?)

Monastery of St John Stoudios

0.5

HIERIA

Church of St Irene ST SOPHIA (GREAT CHURCH) Chalke (Bronze) Gate Great Palace buildings

Chrysotriklinos Church of Sts Sergius and Bacchus

Sea of Marmara 0

CHRYSOPOLIS

Golden Gate

Golden Gate

s

m e

of

Forum of the Ox Forum of Arcadius

Theodosian Walls

ct

CONSTANTINOPLE

ru

ro

Wa l

Cistern of Mocius

Church of St Polyeuktos

du

od

ls o f Co n

st

ue

pp

e tin an

Aq

(Galata)

Bosp

ls al W

Th

ls

eo

al

do

w Church of the Holy Apostles

SYKAI

rn Ho

(Galata)

n

S

SYKAI

Palace of St Mamas Go l

de

Cistern of Aspar

ea

n

H

rn

sia

en

o

Cistern of Aetius

ld

B o s p o r u s

Ly k

Go

os

rda Da

s nelle

ABYDOS

1 mile

Map 6 Constantinople in the earlier Byzantine period

CHALCEDON

114

the earlier empire

The novel design of the church, with its dome forming an image of the cosmos, was immensely influential: there are many smaller Byzantine imitations of St Sophia, and the suggestion of the church as a mim¯esis of the cosmos influenced later interpretations of the liturgical action taking place within (see the Mystagogia of the seventh-century Maximus the Confessor and the commentary on the liturgy ascribed to the eighth-century patriarch of Constantinople, Germanos).17 But it may not have been novel: recent excavations in Istanbul have revealed the church of St Polyeuktos, built by the noblewoman Anicia Juliana in the late 520s, which seems in many respects to have foreshadowed Justinian’s Great Church.18 Original or not, St Sophia and Justinian’s other buildings in the capital created a public space in which to celebrate a world-view in which the emperor ¯ with the support of the court, ruled the inhabited world (the oikoumene), the prayers of the church and to the acclamation of the people. These buildings included more churches, the restored palace (in front of which, in a kind of piazza, was erected a massive pillar surmounted by a bronze statue of an equestrian Justinian), an orphanage, a home for repentant prostitutes, baths and, finally, a great cistern to secure an adequate water supply in summer. According to Procopius’ description of the mosaic in the great Bronze Gate forming the entrance to the palace, there, amid depictions of Justinian’s victories achieved by his general Belisarius, stood Justinian and Theodora, receiving from the senate ‘honours equal to those of God’.19 justinian’s d rive against pagans and quest f or christian unit y The world-view that Justinian’s achievements – whether part of a grand plan or not – were seen to support set great store by an unblemished priesthood offering pure prayer to the true God, the God of the Christians. Unlike other religions of late antiquity – whether the varieties of what Christians called paganism, Judaism, or even (although yet to evolve) Islam – for Christianity, ‘purity’ or being ‘unblemished’, embraced not just moral (and especially sexual) purity, but also the correctness of a considerably elaborated system of belief. For most Christians of the sixth century, this system of belief had been defined at councils regarded as universal, or ecumenical, although there were differences, as we have seen, as to whether the council 17 Mystagogia in Maximus the Confessor, Opera, PG 91, cols. 657–717. The text of Germanos’ commentary in PG 98, cols. 384–453, is poor, but Meyendorff offers a critical edition with translation: Meyendorff, On the divine liturgy, pp. 56–107. 18 See Harrison (1989). 19 Pr B, I.10.15–19, ed. and tr. Dewing and Downey, pp. 84–7. Rousseau detects irony in Procopius’ account here: Rousseau (1996), p. 19.

1. j ustinian and his legacy (500–600)

115

of Chalcedon was to be regarded as the fourth ecumenical council. Emperor Justin’s embrace of Chalcedonian orthodoxy had healed the long-standing schism between the east and Rome, but left unresolved the disagreement between those who accepted Chalcedon (with whatever refinements) and those who rejected it as a betrayal of Cyril of Alexandria, the ‘seal of the fathers’. But all Christians, whatever their differences, were opposed to what they had come to call the ‘exterior wisdom’, the learning of the classical philosophers. As Romanus the Melodist, the Christian poet who spent most of his life in Constantinople during Justinian’s reign, put it: And why do the fools outside strive for victory? Why do the Greeks puff and buzz? Why are they deceived by Aratos the thrice accursed? Why err like wandering planets to Plato? Why do they love the debilitated Demosthenes? Why do they not consider Homer a chimera? Why do they go on about Pythagoras, who were better muzzled?20

This antipathy had been returned in kind, and some adherents of Neoplatonism, loftily indifferent to the new-fangled teachings of the ‘pale Galilean’, developed a world-view that openly ignored Christianity and through their religious practices sought to revive traditional paganism. A notable Neoplatonist was the deeply learned philosopher Proclus, who lived the life of an ascetic, pagan holy man, with an especial devotion to the sun. For fifty years, until his death in 485, he taught in Athens as head (diadochos) of the Academy that had been founded by Plato in the fourth century bc. Part of Justinian’s commitment to Christian orthodoxy was expressed in his closing of the Academy in 529. The closure, however, did not take place before much of the pagan language and intellectual structures had found Christian expression in the writings ascribed to St Paul’s Athenian disciple, Dionysius the Areopagite; these began to make an impact in the 520s, very shortly, it is thought, after they had been written. The philosophers made their way to Persia in 532, led by Damascius the last diadochos; but they returned after a few years, Damascius going to Emesa where he seems to have continued to teach.21 Neoplatonism continued to thrive in Alexandria for another century, where it was not stridently anti-Christian. Indeed most, if not all, of the Alexandrian philosophers were Christian. But the closure of the Academy meant the end of any institutional expression of intellectual opinion. 20 Kontakion 33 (‘On Pentecost’), stanza 17: Romanus the Melodist, Cantica, ed. Maas and Trypanis, p. 265; Kontakia, tr. Lash, p. 215 (the Greek original is full of untranslatable puns). 21 Cameron, Alan (1969).

116

the earlier empire

Alongside the suppression of pagan Neoplatonism, there was suppression of other forms of heterodoxy. In various parts of the empire we learn of more vigorous attempts to suppress survivals of traditional paganism.22 In the 540s, the monophysite bishop John of Ephesus embarked on a missionary campaign in western Asia Minor with imperial support. He claimed to have converted 70,000 souls there, destroying many temples and founding ninety-six churches and twelve monasteries. In Egypt, too, we know of the destruction of temples. Other forms of heterodox opinion fared no better. The dualist doctrine Manichaeism, whose founder Mani had died in Persia in 276, dogged the Christian church through its years of growing success and was an offence punishable by death. The Samaritans embraced what was perhaps a primitive form of Judaism; their revolt against repression was savagely suppressed in 529. Ancient Christian heresies like Montanism also suffered repression under Justinian. The monophysites, who were both more numerous and closer in belief to the imperial church, are a special case to be dealt with presently. The Jews formed a relatively privileged group of second-class citizens. In contrast to heretics and pagans, who had no rights and no civil status, Jews were allowed to exist and their existence was protected. Jews were allowed to practise circumcision and to observe the Sabbath; their synagogues were protected from violence or desecration, although not always effectively; they kept their Rabbinic courts of law and were not to be molested. But they were to exist as ‘living testimony’ to the truth of Christianity, living testimony to the wretchedness of those who had deliberately rejected their Messiah. So the laws protecting their existence also enshrined the principle that Jews must never enjoy the fruits of office, but only suffer its pains and penalties. They were not to expand, so no new synagogues were to be built, and difficulties were often raised over repairing existing ones. The Jews were to be encouraged to convert, but it was to be from a genuine change of heart; they were not to be coerced. They were thus allowed to exist, with rights and civil status, but in a permanently inferior state.23 In the 530s, in parallel with the furthering of legal reform, reconquista and rebuilding, Justinian sought to achieve a reconciliation between orthodox Chalcedonianism and monophysite anti-Chalcedonianism. The basis for this reconciliation was the doctrine of theopaschism. Brought to Justinian’s attention by the Scythian monks a decade or so earlier, this was now part of a wider theological movement usually known as neo-Chalcedonianism, or Cyrilline Chalcedonianism – after Cyril of Alexandria. This theological movement, which was quite independent of Justinian, seems to 22 It is probably misleading to regard as paganism the continuation of traditional religious practices by people who supposed themselves Christians: see Haldon (1997a), pp. 327–37, with literature cited. 23 Sharf (1971), pp. 19–41.

1. j ustinian and his legacy (500–600)

117

have been inspired by attempts to counter the attack by the great nonChalcedonian theologian Severus, patriarch of Antioch (512–18), on the definition of Chalcedon as being incompatible with the teaching of Cyril. Those eastern Christians who had accepted Chalcedon were by no means a minority and did so believing that it endorsed Cyril’s teaching. Cyrilline Chalcedonianism sought to interpret Chalcedon in the light of Cyril’s teaching, believing (not unreasonably) that this represented the mind of the fathers of the council. It was based on three clarifications of the council’s definition: first, that the ‘one person’ of the Incarnate Christ is the second person of the Trinity; second, consequent acceptance of the theopaschite formula ‘one of the Trinity suffered in the flesh’; and third, agreement that one of Cyril’s favourite ways of describing the Incarnate Christ (‘one incarnate nature of God the Word’) was acceptable and only verbally appeared to contradict the doctrine of one person and two natures. This phrase is the source of the term by which the non-Chalcedonians have come to be called: monophysites, believers in (only) one nature. Notable adherents of Cyrilline Chalcedonianism included John of Caesarea and Leontius of Jerusalem. Justinian was convinced that this provided a way of reconciliation and at a conference held in Constantinople in 532, a large measure of theological agreement was reached; however, discussions faltered over practical arrangements for reinstating non-Chalcedonian bishops.24 Thereafter Justinian resorted to persecution, thwarted by the protection given to the monophysites in the palace itself by Theodora. But he never gave up his attempt to promote Cyrilline Chalcedonianism, which culminated in the fifth ecumenical council, held in Constantinople in 553. The fifth ecumenical council was concerned with two issues: the condemnation of the so-called Three Chapters, and the condemnation of Origenism.25 The condemnation of the Three Chapters was part of Justinian’s attempt to achieve reconciliation between the orthodox and the monophysites, for they were the writings of three bishops who were particularly obnoxious to the monophysites: Theodoret of Cyrrhus; Ibas of Edessa; and Theodore of Mopsuestia, who died in 428. Theodore was regarded as the inspiration behind Nestorius, patriarch of Constantinople (428–31). The emphasis in his teaching about Christ on the separate integrity of his two natures, divine and human, and especially his consequent denial (or at least heavy qualification) of the title Theotokos (‘Mother of God’) of the 24 25

Brock (1980). Because of the silence of western sources about the condemnation of Origenism (including, crucially, the council’s Acta, which only survive in Latin (tr. in Nicene, ed. Wace and Schaff, XIV, pp. 302–16)), some scholars still maintain that Origenism was not dealt with at the council. The arguments of Guillaumont (1962), pp. 133–6, however, seem conclusive. On the western reaction, see also below, p. 213.

118

the earlier empire

Virgin Mary, had provoked the wrath of Cyril of Alexandria, who secured his condemnation at the third ecumenical council, held at Ephesus in 431. Theodoret and Ibas had been condemned at the ‘robber council’ of Ephesus of 449, but reinstated two years later by the council of Chalcedon. There was considerable resistance to the condemnation of the Three Chapters in the west, where it was regarded as an attempt to interfere with Chalcedon, Pope Leo’s council. Pope Vigilius was forcibly summoned to attend the council called by Justinian in Constantinople, where he was held under house arrest until he accepted the condemnation of the Three Chapters, and his successors were required to accept his action, although Pope Gregory the Great only ever speaks of ‘four councils’. But others in the west were not so pliant: the pope was excommunicated by bishops in North Africa and northern Italy, and the schism between Rome and Aquileia was not healed until 700. The condemnation of Origenism has often been regarded as a counterbalance to the condemnation of the Three Chapters, but there seems no reason to accept this. There was nothing monophysite about Origenism: its condemnation really belongs with Justinian’s attack on pagan Neoplatonism, for Origen and the Origenists were regarded as deeply indebted to Platonism. Indeed, Origen had been a disciple of Ammonias Saccas, the master of Plotinus. For this reason, it was an action for which Justinian could count on the applause of most Christians, despite Origenist ideas remaining popular among some more intellectually inclined monks. All these attempts to achieve reconciliation amongst the Christians of the empire achieved nothing. By the time the fifth ecumenical council met, the schism had already become irrevocable. Some ten years earlier, in 542, Theodosius, the exiled monophysite patriarch of Alexandria, had secretly consecrated Jacob Baradaeus in Constantinople as bishop of Edessa; Jacob was responsible for the Ghassanids, an Arab realm allied to the empire (see below, p. 188). Once ordained, he set about ordaining bishops for monophysite congregations throughout the east, thus providing a parallel hierarchy to that of the orthodox church of the empire. Imperial attempts to crush this rival church through persecution met with little success. On the face of it, Justinian’s religious policies look to be a downright failure. This is true, if his endeavours are simply regarded as attempts at healing the schism in the church, especially in the east. But these endeavours can be viewed from another perspective: that of leaving the emperor’s mark on the orthodox church of the empire. From this perspective his success was real. The reception of the council of Chalcedon in the sixth century took place along the lines that Justinian promoted: the Christology of the council was henceforth to be interpreted in the east along the lines of Cyrilline Chalcedonianism, and a theopaschite understanding of the Incarnation became accepted, with implications beyond the narrowly

1. j ustinian and his legacy (500–600)

119

theological. By the ninth century the hymn ‘Only-begotten Son’, ascribed to Justinian, formed a regular part of the eucharistic liturgy. Whether or not the literary composition was Justinian’s, the theopaschite theology of the hymn is certainly his (‘you were crucified, Christ God . . . being One of the Holy Trinity’), and such theopaschite devotion, flanked by the development of angelology and Mariology, found expression in the flourishing iconographic tradition of the eastern church. The answer to the first of the questions raised earlier about understanding Justinian’s reign in terms of a grand design would seem to be negative, although in the first decade of his sole rule Justinian may have entertained some such idea. But when we consider the second question – whether Justinian had the means to implement a grand design – even had its components fitted together as well as has often been maintained (legal reform, reconquest, rebuilding and the furthering of orthodoxy), there are other factors in Justinian’s reign that would have prevented any such grand design from reaching fruition. enemies of justinian and other blows One of the obstacles to any grand design was the Persians, traditional enemies of the Roman empire. After a period of peace in the latter half of the fifth century, war had broken out again in the reign of Anastasius. This led to the building of the fort at Dara shortly after 505 (see above, pp. 104–5 and below, p. 135). It was twenty years before war broke out again between the Roman and Persian empires, partly over Justinian’s decision to reinforce the fort at Dara. The initial battles took place in Lazica, an important buffer zone for the Romans, both against the barbarians north of the Caucasus and against a Persian advance through Iberia. One of the Persian generals on this occasion, Narses, defected to the Romans after having inflicted defeat on them. But the main part of Justinian’s first Persian war took place in Mesopotamia, and this was the theatre in which another of Justinian’s generals, Belisarius, rose to prominence. The Romans held their ground, and the war was concluded with a ‘perpetual peace’, negotiated with Khusro I (531–79), who had become shah after the death of his aged father on 13 September 531. This peace gave Justinian the resources for the North African and Italian campaigns of the 530s. Khusro would reign for nearly fifty years and in Persian historiography he is depicted as one of the greatest of the Sasanian shahs.26 But the ‘perpetual peace’ negotiated at the beginning of Khusro’s reign was not typical of his relations with his western neighbour. In 540 a territorial dispute between two Christian Arab kingdoms, the Nestorian Lakhmids, clients of Persia, 26

See Frye (1983a); see also below, pp. 149–51.

120

the earlier empire

and the monophysite Ghassanids, clients of the Roman empire, provided an opportunity for Khusro to respond to pleas from Witigis, the hardpressed Ostrogothic king of Italy, and from the Armenians, suffering from their incorporation into the Roman empire through the ‘perpetual peace’: Khusro invaded the empire. The war was fought on several fronts – in Syria, Mesopotamia and Lazica – and Antioch was seized by the Persians. A truce was called in 545, but fighting went on in Lazica until 557. In 561 a fiftyyear peace was negotiated, restoring the status quo; the Romans agreed to pay tribute at the rate of 30,000 solidi a year for the whole period.27 Persia was once again a force to reckon with, and would remain so, until it succumbed to the Arabs in the seventh century, together with much of the Roman empire itself. Persia was clearly one obstacle standing in the way of any initiatives undertaken by Justinian. Another constraint on his plans, much harder to assess, is the effect of natural disasters and climate change. The chronicles paint a vivid picture of recurrent earthquake, famine and plague, as well as events recorded as harbingers of disaster, such as eclipses and comets. Malalas, for instance, records ten examples of Justinian making grants for the reconstruction of cities devastated by war or natural disaster.28 Collation of scientific with literary studies suggests that the early years of Justinian’s reign saw extreme climatic conditions,29 whose cause is not yet determined; the years 536–7 saw what is called a ‘dust-veil event’, recorded in the chronicles as a kind of perpetual solar eclipse. One can only speculate about the impact of such phenomena, but it is hard not to think that they led to the disruption of traditional patterns and a growing sense of insecurity, not to mention a drain on finite resources caused by the need for reconstruction. It was in this context that the Nika riot of 532 occurred. Tension between the circus factions, the Blues and the Greens, erupted spectacularly: Justinian was nearly toppled, and much of the palace area, including the churches of St Sophia and St Irene, was destroyed by fire. Popular anger against hate-figures was appeased by the dismissal of the City prefect Eudaemon, the quaestor Tribonian, and the praetorian prefect John of Cappadocia. The riot continued for several days and was only eventually quelled by the massacre of 30,000 people, trapped in the Hippodrome, acclaiming as emperor the unfortunate Hypatius, a general and one of Emperor Anastasius’ nephews. Afterwards Hypatius was executed as a usurper. The reaction of some Christians to the whole sequence of disasters is captured in the kontakion ‘On earthquakes and fires’, composed by Romanus the Melodist. Romanus wrote and performed this kontakion one Lent while 27 29

28 Scott (1996), p. 25, n. 37. Men., 6.1, pp. 60–3, 70–5. Farquharson (1996); Koder (1996); Stathakopoulos (2004), pp. 265–9. See also below, pp. 478–9.

1. j ustinian and his legacy (500–600)

121

Figure 4 Pedestal commissioned for an Egyptian obelisk by Emperor Theodosius I (379–95) in the centre of the Hippodrome of Constantinople, showing the emperor with his family and dignitaries, seated in higher places than the common people, waiting for the start of a race; seating plan and ceremonial were similar in middle Byzantium

the Great Church of St Sophia was being rebuilt (i.e. between February 532 and 27 December 537). It is a call to repentance after three disasters that represent three ‘blows’ by God against sinful humanity: earthquakes (several are recorded in Constantinople and elsewhere between 526 and 530), drought (recorded in Constantinople in September 530), and finally the Nika riot itself in January 532.30 These repeated blows were necessary because of the people’s heedlessness. Repentance and pleas for mercy begin, Romanus makes clear, with the emperor and his consort, Theodora: Those who feared God stretched out their hands to Him, Beseeching Him for mercy and the end of disasters, And along with them, as was fitting, the ruler prayed too, Looking up to the Creator, and with him his wife, ‘Grant to me, Saviour,’ he cried, ‘as to your David To conquer Goliath, for I hope in you. Save your faithful people in your mercy, And grant to them Eternal life.’ 30

For this analysis see Koder (1996), pp. 275–6.

122

the earlier empire When God heard the sound of those who cried out and also of the rulers, He granted his tender pity to the city . . .31

The rebuilt city, and especially the Great Church, is a sign of both the care of the emperor and the mercy of God: In a short time they [the rulers] raised up the whole city So that all the hardships of those who had suffered were forgotten. The very structure of the church Was erected with such excellence As to imitate heaven, the divine throne, Which indeed offers Eternal life.32

This confirms the picture of recurrent adversity, found in the chroniclers and, it is argued, supported by astronomical and archaeological evidence. But it also indicates the way in which religion attempted to meet the needs of those who suffered – a way that evoked and reinforced the Byzantine world-view of a cosmos ruled by God, and the oikoumen¯e ruled, on God’s behalf, by the emperor. But a study of Romanus’ kontakia also reveals the convergence of the public and imperial apparatus of religion, and private recourse to the Incarnate Christ, the Mother of God and the saints; it also reveals the importance of relics of the True Cross and of the saints as touchstones of divine grace. It is in the sixth century, too, that we begin to find increasing evidence of the popularity at both public and private levels of devotion to the Mother of God, and of religious art – icons – as mediating between the divine realm, consisting of God and his court of angels and saints, and the human realm, desperately in need of the grace which flows from that divine realm; icons become both objects of prayer and veneration, and a physical source of healing and reassurance. But if the 530s saw widespread alarm caused by natural and human disasters, the 540s saw the beginning of an epidemic of bubonic plague that was to last rather more than two centuries. According to Procopius it originated in Egypt, but it seems very likely that it travelled from the east along trade routes, perhaps the silk roads. Plague appeared in Constantinople in spring 542 and had reached Antioch and Syria later in the same year. Huge numbers died: in Constantinople, it has been calculated, around 250,000 people died, perhaps a little over half the population. Few who caught the disease survived (one such being, apparently, Justinian himself ); those who died did so quickly, within two or three days. Thereafter the plague seems to have declined somewhat in virulence, but according to the church historian 31 Kontakion 54 (‘On earthquakes and fires’), stanzas 18–19: Romanus the Melodist, Cantica, ed. Maas and Trypanis, pp. 468–9; see also tr. Carpenter, II, pp. 245–6. 32 Kontakion 54 (‘On earthquakes and fires’), stanza 23: Romanus the Melodist, Cantica, ed. Maas and Trypanis, pp. 470–1; see also tr. Carpenter, II, p. 247.

1. j ustinian and his legacy (500–600)

123

Evagrius Scholasticus, there was severe loss of life in the years 553–4, 568–9 and 583–4. Historians disagree about the probable effect of the plague on the economic life of the eastern empire: some take its impact seriously;33 others, following a similar revision in the estimate of the effects of the Black Death in the fourteenth century,34 think that the effect of the plague has been exaggerated.35 In the final months of his life, Justinian himself succumbed to heresy, the so-called Julianist heresy of aphthartodocetism, an extreme form of monophysitism named after Julian, bishop of Halicarnassus who died c. 527, and which Justinian promulgated by an edict. This is recounted both by Theophanes and by Eustratius, in his Life of Eutychius, patriarch of Constantinople, who was deposed for refusing to accept Justinian’s newly found religious inclination, and has been generally accepted by historians. However, it has been questioned by theologians, who cite evidence for Justinian’s continued adherence to a Christology of two natures, together with evidence that he was still seeking reconciliation between divided Christians: not only with the Julianists themselves, which might indeed have led to orthodox suspicion of Julianism on Justinian’s part, but also with the socalled Nestorians of Persia. The question is complex, but seems to be open.36 justinian’s heirs cope with his legacy Justinian died childless on 14 November 565. The succession had been left open. One of his three nephews, called Justin, secured election by the senate and succeeded his uncle; he had long occupied the minor post of cura palatii but he was, perhaps more significantly, married to Sophia, one of Theodora’s nieces. The only serious contender was a second cousin of Justinian’s, also called Justin: one of the magistri militum, he was despatched to Alexandria and murdered, reportedly at the instigation of Sophia. Justin II (565–78) continued, or reinstated, Justinian’s policy of religious orthodoxy, though he – or at least his wife, Sophia – had earlier inclined towards monophysitism. In renewing his uncle’s religious policy, Justin restored religious harmony between east and west, and he affirmed this shared orthodoxy by a gift to the Frankish queen Radegund of a splendid enamelled crucifix containing a relic of the True Cross. This inspired the greatest Latin hymns in honour of the cross, Venantius Fortunatus’ Pange lingua and Vexilla regis. But at the same time Justin sought reconciliation with the monophysites. This attempt at reconciliation ended in 572, with the monophysites’ rejection of Justin’s so-called second Henotikon; this rejection resulted in the 33 35

34 See for example Hatcher (1994). Patlagean (1977); Allen (1979). Whittow (1996a), pp. 66–8. See also Stathakopoulos (2004), pp. 277–94. On the mid-eighthcentury plague, see below, pp. 255, 260. 36 See discussion in Grillmeier (1976–96), II.2, pp. 467–73.

124

the earlier empire

persecution of the monophysites which John of Ephesus recorded in his Church history.37 But Justin is remembered chiefly for his ill-advised foreign policy: by refusing to maintain alliances with barbarian tribes, not least the Avars, or to preserve peace with Persia, he gravely weakened the empire’s position. Throughout the century, the Romans had been concerned for the security of the Danube frontier. Both Anastasius and Justinian had invested a good deal in building a line of forts and fortifying cities close to the frontier. In addition to this, Justinian had established alliances with various of the barbarian groupings – the Antes around 545 and the Avars in 558 – and had used them to check other barbarian peoples north of the Danube. Another set of barbarians, which proved a constant concern, was the Slavs: by the middle of the sixth century they were established along the north bank of the Danube, from where they made raids across the river into Byzantine territory, and from around 560 they began to winter on Byzantine territory. Within a few days of Justin’s accession, an embassy arrived from the Avars, requesting the tribute they had been accustomed to receive from Justinian in return, they claimed, for not invading the empire and even for defending it against other barbarians. Justin haughtily rebuffed them, but since the Avars were more concerned with the Franks at this stage, Justin’s action provoked no immediate response. Two years later, Justin was able to benefit from war between the barbarians. When the Lombards and the Avars formed an alliance to crush the Gepids, another barbarian group who occupied Pannonia Secunda and held the city of Sirmium, he was able to seize Sirmium, and held on to it during the war with the Avars that followed. The fall of the Gepids had further consequences for the empire, as the Lombards, who were occupying the borders of Noricum, now had the Avars as immediate neighbours. To avoid this they headed south and invaded northern Italy, with which many of them were familiar, having served there as allies of Narses in 552.38 Under their king, Alboin, the Lombards took most of Venetia in 568 and most of Liguria in the following year, including Milan; Pavia offered more resistance but it, too, fell in 572. Elsewhere barbarians made inroads. Moorish revolts in North Africa caused the death of a praetorian prefect in 569 and two magistri militum in the next couple of years. In Spain, the Visigoths attacked the Byzantines, taking Asidona in 571 and Cordoba in 572.39 It would therefore seem that 572 was not a propitious year in which to provoke the Persians. However, that was the year when Justin refused the first annual tribute under the fifty-year peace negotiated by Justinian, having evidently paid the three-year tribute due in 569. The Christians 37 39

On this see Cameron, Averil (1976). See Barbero and Loring (2005), p. 183.

38

See below, p. 208.

1. j ustinian and his legacy (500–600)

125

of Persian Armenia had risen in revolt against their governor’s (marzban) attempts to impose Zoroastrianism on them and appealed to Justin. Justin not only refused the tribute due in 572, but also threatened to invade Persia and depose Khusro if attempts to turn the Armenians from Christianity persisted. The Armenian revolt was successful, and they were joined by the Iberian kingdom. Justin ordered an invasion of Persia. His cousin Marcian, appointed magister militum per Orientem in 572, attacked Arzanene on the southern border of Persian Armenia, and the next year attacked Nisibis. The Persian response, once they had overcome their surprise at the Roman attack, was devastating: they invaded Syria and took Apamea, then relieved Nisibis before besieging and capturing the fortress of Dara. News of the fall of Dara drove Justin mad, and his consort Sophia took the reins of power. She negotiated a one-year truce with the Persians for which the Romans paid 45,000 solidi, half as much again as had been due; this was later extended to five years, at the old rate of 30,000 solidi a year. But as a woman, Sophia could not rule as regent herself, and in December 574 she persuaded Justin to promote Tiberius, the count of the excubitors, to the dignity of caesar. Although Justin lived on until 578, government was effectively in the hands of Sophia and Tiberius. Sophia is, in fact, a somewhat neglected Byzantine empress. Though far less famous than her aunt, unlike Theodora she played a direct role in Byzantine politics, securing the succession both of her husband and of Tiberius II (578–82), whom she vainly hoped to make her second husband. She is the first empress to appear on Byzantine coins together with her husband.40 Theophanes the Confessor, who clearly disliked women with pretensions to power, paints an ugly picture of Sophia and her meddling in imperial matters, as he did of Irene, the first Byzantine empress to rule in her own name. It may be significant that he has comparatively little to say about Theodora. Tiberius became emperor in 578, but by then had already effectively been governing for four years. In many respects he was the opposite of his predecessor: whereas Justin was financially cautious to the point of being regarded as miserly, but militarily ambitious, Tiberius bought popularity by reducing taxes, but in military matters exercised caution. He also called a halt to the persecution of the monophysites, on which Justin had embarked. Tiberius quickly realised that the empire did not have the resources to engage with its enemies on all fronts. He thus secured the support of the Avars on the Danube frontier by paying them tribute of 80,000 solidi a year. This gained not just a respite from hostilities, but Avar support against the Slavs: with Byzantine backing, the Avar cavalry devastated the Slavs’ territories on the banks of the Danube. However, this truce with the Avars did not last long. In 580 they attacked Sirmium, and after a lengthy siege 40

For Sophia, see Cameron, Averil (1975).

126

the earlier empire

the city was ceded to the Avars in 582 under an agreement which allowed the garrison and population to withdraw to Roman territory in return for 240,000 solidi, the sum total of the tribute not paid since the Avar attack. During the siege of Sirmium many Slavs crossed the Danube and invaded Thrace, Macedonia and what is now Greece: they would eventually settle throughout the Balkans, although there is no evidence for Slav settlements (called Sklaviniai by the Byzantines) until the next century.41 The attempt to buy off the Avars and secure peace on the Danube frontier was to enable Tiberius to concentrate on the Persian frontier, where his aims seem likewise to have been modest: building up enough strength to re-establish the peace that had been broken by Justin. The one-year truce negotiated by Sophia needed to be extended, but the five-year truce that had later been negotiated seemed to Tiberius too long. On his accession as caesar this truce was set at three years, on the understanding that in the meantime envoys would seek to establish a more enduring peace. At the end of the extended truce, the Byzantine army in the east was in a position to make inroads on the Persians, and had occupied Arzanene; the army was led by Maurice, who had succeeded Tiberius as count of the excubitors on Tiberius’ elevation to caesar. Negotiations were underway for a peace that would restore the fortress of Dara to the Byzantines, but in the course of these negotiations – in 579 – Khusro died. His son Hormizd IV (579–90), who succeeded him, broke off negotiations, and war continued. In August 582 Tiberius himself died, having crowned Maurice augustus the previous day. Maurice (582–602) was an effective general, who had already achieved military success under Tiberius before becoming emperor himself. Even if he is not the author of the military treatise called the Strategikon, such an attribution is not inappropriate. The treatise certainly reflects late sixthcentury Byzantine military practice, with its stress on the importance of cavalry in warfare and provision for campaigning across the Danube.42 Like his predecessor, Maurice initially concentrated his military efforts on the Persian front, and sought to deal with the other threats to the empire by diplomacy and tribute. At the beginning of his reign he paid the Frankish king Childebert II (575–95) to attack the Lombards in northern Italy, which he did in 584, securing the submission of the Lombard dukes. This was repeated in 588 and 589. Maurice had less success on the Danube frontier. Two years after his accession, the Avars demanded an increase in their tribute from 80,000 to 100,000 solidi, and when Maurice refused, they seized Belgrade (then known as Singidunum) and attacked other cities in 41 42

See also, on the Slavs’ numbers, Curta (2001a). Maurice, Strategikon, II, III, IX, XI.4, ed. Dennis and German tr. Gamillscheg, pp. 108–91, 302–35, 370–89; tr. Dennis, pp. 23–51, 93–105, 120–6. See also Haldon (1999a), pp. 139–40, 195–203.

1. j ustinian and his legacy (500–600)

127

the surrounding region. Maurice had to pay the extra 20,000 solidi in order to recover Belgrade and secure peace. But the Avars soon allowed the Slavs to overrun and ravage Thrace; the Slavs reached Adrianople and the Long Wall before they were driven back. After that, the Avars themselves crossed the Danube and made for Constantinople. Having easily defeated a Byzantine force of 10,000 sent against them, the Avars crossed the Haemus mountains, invaded Thrace and besieged Adrianople; they were only defeated in 586 by Droctulft, a Lombard duke, who came to the service of the empire. In the same year Thessalonica was besieged by the Slavs and was only saved, so its citizens believed, by the intervention of their patron saint Demetrius.43 On the Persian front the war dragged on inconclusively. There was a mutiny in the army when Maurice attempted to cut pay by a quarter, to alleviate the drain on the treasury, and Martyropolis, in Arzanene, was taken by the Persians in 590. Soon, however, there was a dramatic change of fortune. The Persian shah, Hormizd IV (579–90), was killed in a rebellion led by one of his satraps, Bahram. His son Khusro fled to the Byzantines and with their help in 591 crushed Bahram’s rebellion and secured the Persian throne. In return for the help of the Byzantine emperor, Khusro II (590–628) gave up his claim to Armenia and Arzanene, and restored Martyropolis and Dara to the empire (see below, pp. 169, 337). After twenty years, there was once again peace between the Byzantine and Persian empires. Maurice now turned his attention to the Danube frontier. In 592 the khagan of the Avars demanded an increase in the tribute paid him. With his troops transferred from the now quiet eastern front, Maurice responded by confronting the Avars, who were obliged to abandon their attempt to occupy Belgrade. This did not stop them from invading Thrace, but they left abruptly under the impression that their homeland in Pannonia was in danger.44 However, the real object of Maurice’s military policy seems to have been the Slavs: in the interests both of preserving resources and of effective military strategy, Maurice ordered the Byzantine troops to engage with the Slavs in their settlements beyond the Danube. The army, accustomed to rest during winter, threatened to mutiny. The next year another measure was introduced, aimed at increasing efficiency and saving money: instead of receiving cash allowances for their military equipment, they were to be issued with it directly. This was deeply unpopular. The Avars made further attacks, being rebuffed in their attack on Belgrade and Dalmatia in 598, 43 Lemerle (ed.), Miracles de saint D´em´etrius, I, pp. 130–65; on St Demetrius, see also below, pp. 856–7. On the emergence of the Slavs in the Byzantine sources, see Kobylinski (2005); Curta (2001a). 44 Theophylact presents this as a cunning Byzantine ruse, but the twelfth-century Syriac chronicler Michael the Syrian invokes fear that the Turks were threatening their homeland: see TS, VI.5.16, ed. de Boor and Wirth, p. 230; tr. Whitby and Whitby, p. 166 and n. 33; MS, X.21, ed. and French tr. Chabot, II, p. 363.

128

the earlier empire

and failing to take Tomi on the west coast of the Black Sea in 599. Later they threatened Constantinople itself, but a bout of plague in the Avar camp led the khagan to withdraw and agree a treaty in which the Danube was recognised as the frontier. Maurice quickly revoked the treaty and in 600 the Byzantine army defeated the Avars. The next year was quiet, but in 602 the Byzantines made successful attacks on the Slavs north of the Danube. Maurice gave orders for a winter campaign in Slav territory. This time there was open mutiny: the commander of the army fled, and under a new commander called Phocas the troops advanced on Constantinople. Maurice, who had made himself unpopular with his economies, found himself defenceless in his capital. After a bungled attempt to seize his son’s father-in-law, Germanus – to whom the troops had offered the crown – Maurice found himself facing a popular riot and the palace of the praetorian prefect of the east was burned down. Maurice fled, and Phocas was proclaimed emperor on 23 November 602. A few days later Maurice was executed, after his sons had been slain before his eyes. The death of Maurice and the accession of the usurper Phocas I (602–10) left the empire in a fragile state: civil war weakened the empire within, and external enemies took advantage of the weakness thus revealed. As the seventh century advanced matters looked very black indeed. fin d e si e` cle: faith, cit y and empire At the end of the sixth century the East Roman empire was, as we know with hindsight, on the brink of dramatic transformation: the rise of Arab power would rob it of its eastern and southern provinces; the settlement of the Slavs in the Balkan peninsula would deprive the eastern empire of those provinces and isolate New Rome from Old Rome; the last vestiges of a traditional city-based society seem to have crumbled in an empire now barely capable of defending its capital, or regenerating itself after natural disaster or epidemic. It is difficult not to see seeds of all this as we survey the history of the sixth century. The idea of an orthodox Christian empire did cause both divisions between Christians in the east, and tensions between the increasingly Greek Christianity of the empire and the Latin Christianity of Rome and the west; the public spaces of the city ceased to be used, and were left to decay or be encroached upon by more private activities. Although all this is true, to think in terms of decline is to look at only part of the picture. The public life of the cities may have declined, but it yielded to the demands of the Christian church for space for its activities: increasingly the urban rituals that expressed such sense of civic identity as survived became Christian rituals. The church buildings themselves became increasingly important as public places and moved from the urban periphery to dominate the centre, while the episcopal offices grew in size, in

1. j ustinian and his legacy (500–600)

129

parallel with the developing role of the bishop. The growth in devotion to icons, for which our evidence increases dramatically in the latter half of the sixth century, has been plausibly attributed to ‘the continuing needs of the ancient city’.45 Such Christianisation is neither a vampirish corollary of decline nor evidence of the success of Christian mission; it is rather evidence for change and needs to be evaluated on its own terms. What was taking place at the level of the city had a parallel in, and may have been inspired by, transformation of imperial ritual. In the latter part of the century, we see a growing tendency to underwrite the imperial structures of authority by appeal to Christian symbols: the court of the emperor is presented as reflecting the heavenly court, Constantine’s labarum is joined by icons of Christ and His Virgin Mother.46 While this transformed society may have come close to disaster in the seventh century, it contained seeds of survival and renewal. What survived was, however, a significantly different society from that of the Roman empire at the beginning of the sixth century. 45 46

Brown, P. (1973), p. 21. For this interpretation see Cameron, Averil (1979a). See also Pentcheva (2002); Speck (2003c).

CHAPTER 2a

EASTERN NEIGHBOURS: PERSIA AND THE SA S A N I A N M O N A R C H Y ( 2 2 4 – 6 5 1 ) ze ev ru b in

romans and sasanians A chapter dealing with Iranian feudalism in a distinguished series dedicated to The rise and fall of the Roman world bears the title ‘Iran, Rome’s greatest enemy.1 This title is more than merely a justification for the inclusion of a chapter on Iran in a work devoted to the history of the East Roman empire. It also reflects a host of fears and prejudices fostered for long centuries in the Roman world, since the trauma of Crassus’ defeat by the Parthians at Carrhae. Not even extended periods of decline and internal disarray within the Parthian monarchy, during which it was repeatedly invaded by the Roman army, could dispel the myth of the uncompromising threat posed by Iran to the Roman order. The replacement of the Parthian Arsacid dynasty by a vigorous new one, based in Fars, namely the Sasanian dynasty, at a time when the Roman empire itself was facing one of its severest crises, only aggravated its inhabitants’ deeply rooted fear of Iran. Ancient writers in the Roman oikoumen¯e passed on this attitude to modern western scholars.2 It is the Sasanian bogeyman which has left a deep imprint in modern historiography. The Sasanian state is widely regarded as a much more centralised and effective political entity than its Parthian counterpart, with a far better army. The great pretensions and aspirations of its monarchs are believed to have been fed by the fervour of religious fanaticism, inspired by the Zoroastrian priesthood, which is commonly depicted as a wellorganised state church. No wonder that such a state posed the gravest threat to its greatest rival – the other great power of late antiquity.3 Each of these accepted beliefs raises a multitude of problems, and a fundamental revision is called for. Only a few of the more salient points can be dealt with here. 1 2

‘Iran, der grosse Gegner Roms’: Widengren (1976). Widengren (1976). In general, see the contributions in Yarshater (ed.) (1983); also Schippmann (1990), Herrmann (1977), Christensen (1944). There are detailed bibliographic essays in Wieseh¨ofer (1996), pp. 282–300. 3 Howard-Johnston (1995a); Lee (1993), pp. 15–25.

130

B L A C K

Ca u c a

S E A

IBERIA

sus

Mo

ta

ALBANIA

S

Edessa Carrhae

Dara Nisibis

O

s

ES

JAN

Merv

TA

M IA

Zagros Gates Ctesiphon

K

H

U

M

ZI

ST

Great Kavir Salt Desert

ES

Dasht-i Lut Salt Desert

AN

HA

N Persepolis

P

Darabjird

e r s i a

100

500 200

300

750 km 400

500 miles

Istakhr FARS

n

250

Nishapur

Zagros Mountains

The ASHURESTAN Sawad

al-Qadisiyya

u ea at Pl

Gandzak GILANT AB AR Ecbatana GURGAN ISTA E l bu Khorasan N rz M I t ran s. Rayy PARTHIA ME ian DI

A

PO

es

Perozshapur Veh Ardashir

TRANSOXIANA

Lake Urmia

KURDISTAN Tigris

M

Chalkis E at hr up

O r o nte

Antioch

IA

AZERBAI

ARMENIA

Amida

Sea

Lake Van

YR

s

an pi as

in

C

Theodosioupolis

Caspian Gates

un

G

u

l f

Map 7 Sasanian Persia

Gedrosian Desert

132

the earlier empire

The Sasanian empire embraced two distinct geographical areas, the very fertile lowlands of Mesopotamia and the Iranian uplands, which were separated from each other by the mighty Zagros chain stretching from the Kurdistan highlands to the fringes of the Persian Gulf in the south.4 Mesopotamia, where a complex irrigation system permitted dense settlement, was the economic heart of the Persian realm. Its rich agriculture generated the largest part of the Sasanian state’s tax revenues and supported a network of major cities: Ctesiphon, the capital; Veh Ardashir, on the west bank of the Tigris opposite Ctesiphon, which was founded by the first Sasanian monarch; Perozshapur on the Euphrates, which commemorated the site of Shapur I’s victory over Gordian and exploited the large number of Roman captives secured then; and Veh Antiok Khusrau, which was a similar foundation by Khusro I to celebrate his capture of Antiochon-the-Orontes and to provide a home for the captives and booty from his successful 540 campaign (see above, p. 120). By contrast, the Iranian plateau was sparsely settled, with its main centres of habitation clustered around the sources of water emerging from the Zagros. Rainfall on the plateau is low and beyond the rivers and qanats (underground water channels) lies desert: the Gedrosian to the south-east, where much of Alexander’s army perished in 324 bc, and to the north the salt desert of the Great Kavir. On the fringes of the Sasanian world were areas of considerable military importance. In the north-west, Iran competed for influence with Rome among the nobilities of Armenia, Lazica, Iberia and Albania, and attempted to control movements across the Caucasian passes. In the wide expanses of Transoxiana, Iran confronted its traditional enemies: the succession of nomadic confederations of the Central Asian steppes. These included the Hephthalites or White Huns, who dominated the frontier in the fourth and fifth centuries; and the Turks, who cooperated with Khusro I in the elimination of their mutual enemy, the Hephthalites, in the 550s, but then rapidly emerged as a much more powerful threat during the rest of the sixth century. The vast barrier of the Zagros restricted communications to a limited number of major passes, so that the structural backbone of the empire was simple: from the economic and political heartland of lower Mesopotamia, routes up the Tigris led to the area of conflict with Rome in the north and north-west; while the road to the east crossed the Zagros into Media and then continued along the southern flanks of the Elburz range, another major defining mountain range, towards Khorasan and the frontier. The Sasanian heartland was located in Fars, the relatively fertile region at the south-western end of the Iranian plateau, where the family combined positions of religious authority (the chief priesthood of the temple 4

Comprehensive discussion of all aspects of Iranian geography in Fisher (ed.) (1968).

2a. persia and the sasanian monarchy

133

of Anahita at Istakhr) and secular power (governorship of Darabjird). After two decades in which a strong local power base was transformed into authority over the Iranian plateau, Ardashir descended to the Mesopotamian lowlands, overthrew the Arsacid monarch and was crowned ‘king of kings’ at Ctesiphon in 226. Military success, and in particular conflict with Rome, was an important mechanism for demonstrating the legitimacy of the new regime. The initial thrusts of the two first Sasanian monarchs, Ardashir I (224–40) and Shapur I (240–70), against the Roman east turned out, in the long run, to be little more than a series of wars of plunder: the Romans were defeated three times in the field, with Emperor Valerian being captured at Edessa in 260; the great cities of Nisibis, Carrhae and Antioch were sacked; and ravaging extended into Cappadocia and Cilicia as well as Syria – but there were no permanent gains.5 Under their immediate successors, the initiative seems to have passed momentarily to the Romans. The conflicts between the two empires at that time brought the problem of Armenia to the fore, and this was to be a major bone of contention for most of the following century (see below, pp. 156–7). The attempt of Shah Narseh (293–302) to regain the upper hand ended in humiliating defeat by the Romans in 297, followed by a no less humiliating treaty. The tide was partly reversed during Shapur II’s long reign (309–79). The wars fought between the two powers at the time were largely over contested frontier lands – first and foremost Armenia and northern Mesopotamia. Stability began to emerge after Julian’s invasion in 363 permitted the Persians to regain Nisibis and other territories in upper Mesopotamia, and this was reinforced by the treaty between Shapur III (383–8) and Theodosius I in 384, which arranged the division of Armenia.6 This ushered in a long period of relative quiet in relations between the empire and Persia, apart from two brief conflicts in 421–2 and 440–1. On the first occasion, the dispute was caused by the Roman reception of Christian fugitives, especially from the Arab tribes allied to Persia. Yazdgard I (399–420) had been favourably disposed towards Christians and other minority religious groups within his kingdom, but energetic Christian missionary activity seems eventually to have forced him to permit persecution; an Arab chief, Aspabad, was instructed to prevent the flight of Christian converts to the Romans, but he proceeded to join the exodus, converted and, now renamed Peter, became bishop for the wandering tribal groups in the desert.7 Persian demands for subsidies towards the cost of defending the Caspian passes (the so-called Gates) caused the second conflict, 5 6

Sources in The Roman eastern frontier, ed. Lieu et al., I, pp. 9–67. Rubin, Z. (1986); Frye (1983), pp. 153–70; Blockley (1992), pp. 39–45; Whitby, Michael (1988), pp. 197–218. 7 Cyril of Scythopolis, Life of Euthymius, ch. 10, in Cyril of Scythopolis, Saints’ lives, ed. Schwartz, pp. 18–21; tr. Price, pp. 14–17. On this see Rubin, Z. (1986), pp. 679–81; Blockley (1992), p. 199, n. 28.

134

the earlier empire

when Yazdgard II (438–57) attempted to exploit Theodosius’ concern over the Vandal capture of Carthage. On each occasion Roman armies checked Persian attacks and peace was rapidly restored, with renewed treaties that contained clauses to regulate the alleged origins of the war.8 A plausible explanation for the change from persistent warfare in the third and fourth centuries to peaceful relations in the fifth is provided by the other external problems which faced successive rulers. Developments in the west and the Balkans, as well as internal problems in Isauria, commanded the attention of the emperor at Constantinople, while Sasanian shahs had to contend with the equally serious threat posed by the Hephthalites on their north-east frontier. This Sasanian problem is not regularly reported in our sources. The succession of Greek classicising historians from Priscus of Panium through to Theophylact Simocatta narrate diplomacy and warfare that involved Romans and Sasanians, but seldom extend their horizons further east.9 Sasanian sources are mostly preserved for us through compilations from the Islamic period, of which the most important are the Ta’rikh of al-Tabari in Arabic and the Shahnama (Book of kings) of Firdausi in New Persian. Both date from the tenth century and depend on lost Iranian sources, in which anecdotal material had substantially ousted reliable information, so that the resulting narratives are dominated by charming and exotic stories. Though al-Tabari attempted to cut his way through the more sensational of his source materials and to produce a sober historical narrative, he still incorporated two parallel versions of Sasanian history: it is not safe to trust his information uncritically.10 Furthermore, these Iranian sources are more informative for the royal court and internal affairs and, like their Roman counterparts, are silent about a difficult frontier relationship in which the Persians were often at a disadvantage. Only for the reign of Peroz (459–84) is there substantial information about PersoHephthalite relations, partly because Peroz was defeated in 464–5 when the Roman ambassador Eusebius was accompanying the royal army, and partly because two decades later Peroz perished with much of his army in a catastrophic attempt to reverse the previous humiliation.11 The death of Peroz was followed by a period of dynastic weakness in Iran. Peroz’s brother Valash ruled for four years (484–8) before being overthrown by Peroz’s son Kavad I (488–96), who relied on Hephthalite support. Kavad, however, was in turn ousted by the nobility and replaced by his brother Zamaspes (Jamasp); but he was returned to power (498–531) with Hephthalite assistance, after marrying their ruler’s daughter. Kavad’s reign 8 9

Blockley (1992), pp. 56–61; Frye (1984), pp. 320–1. Discussions in Blockley, Historians; Cameron, Averil (1969–70); Cameron, Averil (1985); Men., pp. 1–30 (introduction); Whitby, Michael (1988). 10 Howard-Johnston (1995a), pp. 169–72. 11 Pr W, I.3–4, ed. and tr. Dewing, I, pp. 12–31.

2a. persia and the sasanian monarchy

135

witnessed the rise of the Mazdakite ‘movement’ (see p. 149 below), which advocated communal rights over property, and perhaps also women. It appears to have received some support from the shah, and can be interpreted as an attempt to undermine the entrenched power of the hereditary aristocracy. An indirect consequence of Kavad’s dynastic problems was resurgence of warfare with Rome: Kavad undoubtedly needed money to repay the Hephthalites and to enhance his position as supreme patron within Persia, and this led him to ask the Romans for contributions towards the costs of defending the Caspian Gates. Anastasius’ refusal provided a pretext for war (502–5), and although Kavad’s first campaign secured considerable prestige and booty – with the capture of both Theodosioupolis and Amida – the Roman generals gradually stabilised matters after that.12 Sixth-century Romano-Persian relations are characterised by two opposing tendencies: a recollection of the relatively harmonious fifth century, when elaborate diplomatic practices for managing relations had emerged; and international rivalry, caused both by weakness in the Persian shah’s position and by mutual suspicion of each other’s intentions. In 527, towards the end of Kavad’s reign, war broke out again (see above, p. 119). Tension had risen as the empires competed for the allegiance of the principalities around the Caucasus, where acceptance of Christianity by local rulers threatened to weaken loyalties to Persia. However, the flashpoint came when Justin I (518–27) refused to cooperate with Kavad’s plans to ensure the succession of his third son, Khusro. Although the Persians took the offensive, a series of invasions failed to capture any major Roman city, and two pitched battles – at Dara in 530 and at Callinicum in the following year – resulted in a victory apiece. Hostilities were concluded with the ‘perpetual peace’ of 532, when the new Persian shah, Khusro I (531–79) accepted a lump sum of 11,000 pounds of gold in lieu of regular contributions for the defence of the Caucasus.13 Peace did not last. Justinian (527–65) exploited the quiet on his eastern frontier to launch the reconquest of Africa and Italy, but his startling victories were brought to Khusro’s attention; jealousy fuelled suspicions about Justinian’s long-term intentions, and Khusro exploited a dispute between client Arab tribes to attack in 540. After spectacular Persian successes in this first campaign, the Romans organised their defences and a truce confined fighting to Lazica after 545. However, their Arab allies went on fighting (see below, p. 188). This ended with a decisive victory for the Ghassanid allies of 12 Joshua the Stylite, Chronicle, chs. 54–83, ed. and tr. Wright, pp. 51–78, 43–66; tr. Trombley and Watt, pp. 63–101; Pr W, I.7–10, ed. and tr. Dewing, I, pp. 48–83; Theoph., ed. de Boor, I, pp. 144–9; tr. Mango and Scott, pp. 222–8; Zacharias of Mytilene, Chronicle, VII.3–5, tr. Hamilton and Brooks, pp. 151–64; Blockley (1992), pp. 89–96. 13 Pr W, I.11–22, ed. and tr. Dewing, I, pp. 82–209; John Malalas, Chronicle, XVIII.4–69, ed. Thurn, pp. 355–94; tr. Jeffreys et al., pp. 246–74.

136

the earlier empire

Byzantium in 554 near Chalkis, when the Lakhmid ruler al-Mundhir III – scourge of imperial provinces for the previous half-century – was killed. Peace finally came in 562 with an agreement that was intended to last for fifty years; the detailed terms illustrate the range of disputed issues that could provoke conflict, and are preserved in an important Fragmentum of Menander the Guardsman.14 Peace lasted for a decade, but on this occasion the Byzantines were the aggressors: Justin II (565–78) objected to paying for peace (at the rate of 30,000 solidi per year) and believed that he could count on the support of the Turkish confederation in Central Asia, which had replaced the Hephthalites as Persia’s north-eastern neighbours, to crush their common enemy. Two decades of fighting ended when Khusro I’s son and successor, Hormizd IV (579–90), was overthrown in a palace coup; Hormizd’s son, Khusro II (590, 591–628), was almost immediately challenged by Bahram Chobin, who had gained great glory from defeating the Turks and was the first non-Sasanian to seize the throne (590–1). Khusro sought assistance from Emperor Maurice (582–602), was reinstated by a Roman army in 591, and peace was again arranged.15 The final conflict of the two great rivals of the ancient world broke out in 602, when Khusro took advantage of the murder of his benefactor Maurice and the arrival in Persia of Maurice’s eldest son Theodosius (or at least a plausible impersonator); Khusro could shed the image of imperial client, present himself as the supporter of international ties of gratitude and friendship, and obtain significant booty and military glory into the bargain. For twenty-five years the conflict ranged across the entire Middle East, from Chalcedon on the Bosporus to Gandzak on the Iranian plateau, until a daring counter-offensive by Heraclius (610–41) prompted the Persian nobility to overthrow Khusro in 628.16 Once more peace was restored, but the defeated Sasanian dynasty lapsed into a rapid turnover of rulers (eight within five years, including, for forty days, the Christian and nonSasanian Shahrvaraz). The last Sasanian ruler Yazdgard III (633–51) had only just ascended the throne when he had to confront Islamic attacks; the diminution of royal prestige and the weakness of his armies after a quarter of a century of unsuccessful warfare against Byzantium made Persia particularly vulnerable, and Yazdgard was forced to flee to the north-east, where he was eventually killed. Wars and animosity loom large in the record of the relations between Byzantium and Persia, both of which laid claim to universal ascendancy. The imprint they have left on the Byzantine sources tends to obscure the fact that both sides could also exploit a rhetoric of peace and co-operative 14 15

Pr W, II, ed. and tr. Dewing, I, pp. 260–557; Men., 6.1, pp. 70–3. TS, I–V, ed. de Boor and Wirth, pp. 36–220; tr. Whitby and Whitby, pp. 17–157; Whitby, Michael (1988), pp. 250–304. See also above, p. 127. 16 Howard-Johnston (1994); Stratos (1968–80), I.

2a. persia and the sasanian monarchy

137

relations. The Sasanians, who had to contend with a succession of nomadic and semi-nomadic powers along their extensive frontiers, tried to impress on the Byzantines that they were defending these frontiers for their mutual benefit. This claim justified repeated demands for diplomatic subsidies, but Sasanian internal propaganda depicted these as tribute, which aggravated imperial resistance to paying up:17 international prestige was one of the factors that individual Sasanian monarchs used in order to balance the divergent constituencies within their realm and preserve their own supreme position. royal legitimation The best evidence about Sasanian royal ideology comes from the first century or so of the dynasty, and although it is possible to detect developments thereafter, the basic principles apply throughout the regime’s history. Shapur I was the first to claim the title ‘king of kings of Iran and non-Iran’, whereas his father, Ardashir, had contented himself with the title ‘king of kings of Iran’ only. The legitimation of the new royal dynasty in its own realm was the immediate task the early Sasanians had to face. The great official state inscriptions from the early Sasanian period do not conceal the newness of the dynasty. The Res gestae divi Saporis is a list of the exploits of Shah Shapur I on the so-called Ka‘ba of Zardusht,18 and traces the royal genealogy back three generations, through his father Ardashir to his grandfather Papak. On the Paikuli inscription, set up by Shah Narseh to commemorate his successful bid for supreme power and his victory over his nephew Bahram III (293), there is only one significant addition. The dynasty is called ‘the seed of the Sasanids’, elucidating to some extent the role of ‘the lord Sasan’, mentioned in the Res gestae divi Saporis as recipient of an honorary cult, but not explicitly as a forebear of the dynasty. None of the other remaining six inscriptions that allude to the genealogy of the Sasanian shahs adds anything of significance.19 The great pictures that accompany many of these inscriptions present the key elements of legitimate royal authority. In some, the shah and his entourage unseat their rivals in a dramatic joust; or foreign enemies demonstrate their submission – including in some scenes the Roman emperor, who arrives at speed to acknowledge Sasanian mastery, kneels before his conqueror or lies prostrate at his feet. The proper transfer of power at each accession is symbolised by grand ceremonies involving shah and court; and 17 18

Rubin, Z. (1986); Braund (1994), pp. 270–1. The Ka‘ba of Zardusht (Cube of Zoroaster) was an Achaemenid tower at Naqsh-i Rustam, a royal burial-ground near the ancient capital of Persepolis. See below, p. 139. 19 On the Res gestae, see Back (1978), pp. 284–371; Huyse, Dreispr¨ ˇ achige Inschrift Saburs I.; on the Paikuli inscription, see Humbach and Skjaervø (1978–83), III.1.

138

the earlier empire

Figure 5 Portrayal of Shah Peroz being invested with two diadems by a god and goddess, Taq-i-Bustan, Iran

in some pictures, divine investiture is symbolised by the figure of Ahura Mazda or of Anahita handing over a diadem to the shah.20 The monuments present a self-fulfilling legitimation. Supernatural sanction for the Sasanian house is demonstrated by the sequence of royal victories through which the Sasanians have achieved power; royal gratitude for this divine support is displayed by the establishment of a series of ritual fires. No attempt is made to conceal the shah’s bellicosity, and this self-glorification in divinely sponsored aggression is repeated three times in the Res gestae divi Saporis. According to the ideology enunciated in this document, wars of conquest are the duty of a good shah and military success proves legitimacy.21 20

Pictures in Ghirshman (1962), pp. 135–201.

21

Whitby, Michael (1994).

2a. persia and the sasanian monarchy

139

Externally, or at least with regard to the Roman empire, the only area for which we have evidence, Sasanian strategies for legitimation were slightly more complex. Victory was still crucial, but warfare ought to have some justification. In his Res gestae, two of Shapur’s three expeditions against the Romans are presented as responses to Roman aggression; one of the three versions of the inscription is in Greek, and its contents were probably proclaimed to the inhabitants of the Roman empire, or to its former inhabitants resettled in Iran.22 More significantly, three historians writing in the Roman empire – Cassius Dio (LXXX.3.3) and Herodian (VI.2.1–5) from the third century, Ammianus Marcellinus (XVII.5.3–8) from the fourth – record how Sasanian envoys presented territorial demands on the Romans in terms of the revival of the old Achaemenid empire.23 The repeated Roman refusal to return what rightfully belonged to the new dynasty was sufficient justification for war. If the Achaemenid heritage was important in their western diplomatic dealings, there is no evidence that it was significant for internal legitimation. Although Ardashir and Shapur I chose to glorify themselves at Naqsh-i Rustam, near Persepolis, a site rich in Achaemenid associations,24 the possible connection is not voiced in their public inscriptions. The site was chosen for its monumental and awe-inspiring nature; there is no evidence that those who beheld these monuments were aware of their specific Achaemenid associations, or indeed of the pristine greatness of the Achaemenids themselves. The modern name of the site, Naqsh-i Rustam, with its reference to the hero of Iranian epic tradition, indicates the extent to which folk memory can misrepresent the true nature of such sites. When Shapur I refers to his ancestors’ domain in his Res gestae, this is merely to state that exiles from the Roman empire were settled in Iran on crown lands – in Fars, Khuzistan and Ashurestan. Again, this is neither evocation of the Achaemenid empire nor a claim to legitimation as their heirs.25 It has been alternatively suggested that the Sasanians’ claims to legitimation harked back not to the Achaemenids but to the Kayanids, the heroic mythical rulers of Iran long before the historical Achaemenids.26 However, this hypothesis is not supported in the inscriptions: Shapur I only traced his genealogy back to his grandfather Papak, and did not claim universal kingship before his own reign (he is the first ‘king of kings of Iranians and non-Iranians’). More striking is the absence of any allusion to the dynasty’s 22 English translation, based mainly on the Parthian and Middle Persian versions, in Frye (1984), pp. 371–3. 23 Whitby, Michael (1988); Potter (1990), p. 373 argues that Persian demands were reshaped to fit the presuppositions of Roman historiographical traditions. 24 Wieseh¨ ofer (1996), pp. 27–8, 154–5; Lee (1993) pp. 21–2. 25 As suggested by Wieseh¨ ofer (1996), pp. 155, 223; see also Lukonin (1961), p. 23 for a less extravagant interpretation of this passage. 26 For a full development of this hypothesis, see Yarshater (1971).

140

the earlier empire

Kayanid origin in Narseh’s Paikuli inscription, precisely the context where self-designation as ‘the seed of the Sasanians’ invited a link with a more glorious house. Kayanid names such as Kavad and Khusro only enter royal nomenclature in the late fifth century and probably reflect a change at that time in strategies for dynastic legitimation. Furthermore, it is the mythological Kayanid link which eventually introduces into royal genealogies an Achaemenid element that had not been present before. This Achaemenid link was clearly derived from the Alexander romance, which became popular at the Sasanian court in the first half of the sixth century. The Sasanian genealogies relayed through Arabic and New Persian sources deriving from lost Pahlavi historiography reflect, as often, the conditions and traditions of the last century of Sasanian rule; little genuine knowledge was preserved.27 sasan i an shahs and the zoroast rian priests Divine sanction was an important part of royal legitimation, and one must therefore investigate the relations between monarchs and the Zoroastrian priesthood, the repository of pristine mythological traditions. The established view that the Sasanian shahs relied on the Zoroastrian priesthood’s support, and as a consequence actively encouraged their beliefs and enhanced their power, has been largely modified in recent decades.28 Although the term mazdesn (Mazda-worshipping) recurs frequently on Sasanian monuments as a royal epithet, this need not imply automatic recognition of one organised priesthood as sole exponent of this deity’s cult. Shahs could perhaps best consolidate royal power by fostering variety, both inside the Zoroastrian church and between different religions. The traditional view encounters difficulties even with the dynasty’s founder, Ardashir I. According to the Denkard – the post-Sasanian Zoroastrian encyclopedia – Ardashir should be considered as the great restorer of the Zoroastrian faith: it was under his aegis that the priest Tansar allegedly collected the scattered remnants of the Avestan books, which had survived since Alexander’s conquests.29 However, the picture that emerges from the Res gestae divi Saporis is rather different: it makes no mention of Tansar or any member of the Zoroastrian priesthood other than Kirder, whose appearance is rather muted. Ardashir himself can reliably be described as a worshipper of Anahita of Stakhr, whereas evidence of his attachment to Ahura Mazda is more equivocal. As worshipped by the early Sasanians, Anahita was the goddess of victory at whose shrine the severed heads of vanquished enemies were habitually dedicated. If the devotion of Ardashir and his immediate successors to Anahita can be considered as part and 27 28

N¨oldeke (1887b), pp. 87–8; N¨oldeke (1920), p. 13. For a survey of views, see Schippmann (1990), pp. 92–102.

29

Shaki (1981).

2a. persia and the sasanian monarchy

141

parcel of a Zoroastrian orthodoxy, then this orthodoxy must have been entirely different from the kind of orthodoxy assumed in his glorification in the Denkard.30 The absence of any clear reference to an organised clergy in the Res gestae divi Saporis is at odds with the role ascribed by modern scholars to a ‘Zoroastrian church’, at least under the early Sasanians. This gap is not filled by the far-reaching claims made in four inscriptions celebrating the career of Kirder, the one priestly character who does figure on Shapur’s monument. Kirder was promoted within the Zoroastrian priesthood from a mere herbed under Shapur I to the rank of a mobed (chief magus) under his immediate successors, Hormizd I (270–1), Bahram I (271–4) and Bahram II (274–93). Bahram II bestowed additional honours and supposedly authorised Kirder to enforce Zoroastrianism and persecute heresies and other religions. This only indicates that this shah was attached to the kind of Zoroastrianism preached by Kirder, which is more than can be said of Shapur I.31 The extent of Shapur I’s Zoroastrian piety as it emerges from his own Res gestae is not entirely clear. He was indeed the founder of many fire-temples throughout his realm, according to his own testimony as well as to Kirder’s. Yet fire-temples were sacred not only to Ahura Mazda but also to Anahita, and Shapur’s favourable attitude to Zoroastrianism should be conceived in the framework of a religious eclecticism that could also accommodate Manichaeism.32 Furthermore, the fact that he granted Kirder sweeping powers to conduct religious affairs, without matching these powers with the appropriate title – whatever its meaning, herbed appears to be a rather modest rank – suggests that Kirder was more a court priest than the designated head of a powerful church. We cannot rule out a degree of tension between Kirder in this function and some of his brethren. Reiterated as a refrain on his inscriptions, Kirder’s statement that under his leadership many of the magi (not all of them) were happy and prosperous implies an attempt to mute some opposition voices. The early Sasanian monarchs, far from depending on an already powerful organisation for vital support, may rather have helped Zoroastrian clergy to improve their position in a fluid and competitive religious milieu. It is usually assumed that under Narseh the influence of the Zoroastrian priesthood declined, but that it regained much of the lost ground under Shapur II. The figure of Aturpat, son of Mahrspand, looms large in postSasanian Zoroastrian literature: he is depicted as a model of Zoroastrian orthodoxy who submitted himself to the ordeal of molten metal to refute heretics whose precise doctrine is disputed. It is natural enough to suppose 30 31 32

Chaumont (1958); Duchesne-Guillemin (1983), pp. 874–97. Back (1978), pp. 384–488; Duchesne-Guillemin (1983), pp. 878–84. Wikander (1946), pp. 52–124; Chaumont (1958), pp. 162–3. For Manichaeism in the Sasanian empire: Lieu (1994), pp. 24–5, 35–6.

142

the earlier empire

that Aturpat stood at the head of a mighty Zoroastrian hierarchy, authorised by the shah himself to administer the institutions of the only fully recognised official state religion. However, the hierarchy of what tends to be conceived of as ‘the Zoroastrian church’ did not in all probability become fully established until much later. It is only under Yazdgard II (438–57) that the high priest Mihr-Shapur, who had already distinguished himself under previous reigns as a persecutor of Christians, is called modaban mobad, the earliest reliable attestation of this title. But even then the relative position of mobeds and herbeds in the organisation of Zoroastrian clergy is not entirely clear. The title herbedan herbed, conferred upon Zurvandad, the son of Yazdgard’s powerful prime minister, Mihr-Narseh, has been interpreted as evidence for a hierarchy distinct from that of the mobeds within the Zoroastrian church. The Zoroastrian priesthood appears to have gained a truly undisputed position as the sole representative of the one and only state religion in the course of the fifth century. It is precisely at this time that Avestan names suddenly proliferate among members of the royal house, and the title kavi or kay appears on its coins, marking a crucial stage in the fabrication of the Kayanid genealogy as a source of legitimation of the Sasanian dynasty. Yet the Zoroastrian priesthood was soon to suffer a severe blow under Kavad I (488–96), during the Mazdakite revolt (see below, p. 149). The reign of Khusro I (531–79) appears to have been a period of harmony between the monarchy and the Zoroastrian priesthood, but it was a priesthood restored by the shah following the Mazdakite debacle, and consequently more dependent on the shah than before. Under Khusro’s successors, Zoroastrian influence seems to have declined. Khusro II (590– 628), rather than follow his predecessors in the large-scale establishment of fire-temples staffed with a vast multitude of herbedan, relied heavily on Christians, including his favourite wife, his finance officer and his chief general (see below, p. 144); Zoroastrian tradition, as reflected in the apocalyptic composition Jamasp namagh, branded him an unjust and tyrannical shah.33 The figure of Mihr-Narseh, Yazdgard II’s prime minister, illustrates the problem of Zoroastrian orthodoxy and heterodoxy in the Sasanian period. From Armenian sources recounting the persecution he launched against the Christians in Armenia, it is clear that Mihr-Narseh was an adherent of Zurvanism (belief in Zurvan i Akanarag or Infinite Time).34 His son Zurvandad bore a name celebrating this rather shadowy divine personification, 33 Text in Ayatkar i zamaspik, XVI.30, ed. and Italian tr. Messina, pp. 70–1 (text), p. 115 (tr.), where aparvez (‘the victorious’) is an unmistakable allusion to Khusro II; see also Duchesne-Guillemin (1983), p. 896; Boyce (1983), p. 1160. 34 Elishe, History, tr. Thomson, pp. 77–80; French tr. in Langlois, Historiens de l’Arm´enie, II, pp. 190–1; see also Eznik of Kolb in Boyce, Zoroastrianism, pp. 97–8.

2a. persia and the sasanian monarchy

143

and such names seem to have been common among Iranian nobles under the Sasanians. The role of Zurvan in the Zoroastrian pantheon is much disputed, but it represents a trend in Zoroastrianism which sought to provide a unifying monistic framework for its fundamentally dualist theology: Ohurmazd, the good principle, and Ahriman, the evil principle, were depicted as the twin sons of Infinite Time. However, there is little reliable information. Whereas contemporary non-Sasanian and non-Zoroastrian sources suggest that this monistic doctrine was the orthodoxy endorsed by the Sasanian shahs, the Pahlavi Zoroastrian literature of the post-Sasanian era is virtually silent on this.35 Various attempts have been made to explain this discrepancy. One suggestion is that the dualist orthodoxy reflected in the surviving Zoroastrian literature only triumphed after the collapse of the Sasanian monarchy: that the former monistic orthodoxy was deliberately suppressed by supporters of the old national religion, in the face of the new Islamic monotheism.36 According to another view, the story of Zurvanism is one of intermittent success: whereas under some shahs it was indeed the accepted orthodoxy, under others the pendulum swung in the opposite direction and the dualist trend became dominant. Dualism was finally triumphant in the mid-sixth century under Khusro I, whose reign also constitutes a decisive stage in the establishment of a canon of the Zoroastrian scriptures, the Avesta, and in the development of Zoroastrian theological literature. Attempts have also been made to play down the significance of Zurvanism, either as a fad entertained by the upper classes or as a popular version of Zoroastrianism: nothing tantamount to a heresy in its familiar Christian sense.37 Perhaps the best way of approaching a solution is to get rid of the notion of a Sasanian Zoroastrian church, analogous in its position to that of the Christian church in the late Roman empire and intent upon using secular support to impose a uniform doctrine within its ranks. The truth may well have been that although the early Sasanian shahs found Zoroastrianism, as represented and propounded by the estate of the magi, the most potent religious factor in many of their domains, they were not always prepared to allow it to become the sole officially dominant state religion. Thus, for example, Anahita, who seemingly fades out after the reign of Narseh, springs again into prominence under the last Sasanians, from Khusro II to Yazdgard III.38 Furthermore, the fact that some Sasanian shahs, like Shapur I, were prepared to unleash the Zoroastrian priesthood against the Christians in 35 36 37

Christensen (1944), pp. 149–54; Boyce (1979), pp. 112–13, 160–1. Boyce (1979), pp. 160–1; see also Boyce, Zoroastrianism, pp. 96–9. Zaehner (1955); reaction in Boyce (1957); Boyce (1990); Frye (1959); Frye (1984), p. 321 with n. 27; Denkard VI, tr. Shaked, p. xxxiv (introduction). 38 Wikander (1946), pp. 55–6; Duchesne-Guillemin (1983), p. 897.

144

the earlier empire

the service of their own policies does not mean that they themselves subscribed to any version of Zoroastrianism as the binding orthodoxy. Attitudes towards this religion appear to have varied according to circumstances and the tempers of individual rulers. A sober monarch like Shapur I was quite capable of striking an alliance of convenience with the Zoroastrian clergy, while keeping his options open by toying with Manichaeism. Shapur II, a notorious persecutor of the Christians, may well have played off dualism against Zurvanism precisely in order to check the growth of an excessively strong, unified priestly caste. Yazdgard I was favourably inclined towards Christianity and Judaism for most of his reign.39 On the other hand, such shahs as Bahram I and Bahram II may be described as truly pious followers of the form of Zoroastrianism propounded by Kirder: probably, but not certainly, dualism. The Sasanian monarchs’ attitude towards Nestorian Christianity is another consideration against interpreting their religious policy exclusively in terms of their Zoroastrian piety. After this creed had been condemned as a heresy at the council of Ephesus in 431, believers found a relatively safe haven in the Sasanian empire. In 457, a Nestorian school was founded in Nisibis by Bar Sauma and Narsai, fugitive Nestorian teachers from Edessa; it flourished there, particularly under Shah Peroz (459–84), when the Zoroastrian priesthood appears to have been at the peak of its power. There was no danger in a policy of toleration towards a religious sect now banned within the Byzantine empire, whose rulers were either Chalcedonian or inclined to monophysitism. However, even a shah such as Khusro I – who could afford to be tolerant without marring his relations with a Zoroastrian priesthood firmly under his control – could or would not prevent persecution, even of Nestorians, after war against Byzantium flared up in 540. Khusro II is often described as sympathetic to the Christians, but the picture is more complex: he astutely played off monophysites (whose cause was advocated at court by his favourite wife, Shirin, and her influential physician, Gabriel) against Nestorians (who found a faithful champion in his powerful finance minister, Yazdin). Towards the end of his reign, when his empire succumbed to a Byzantine invasion, Khusro reversed his policy of general toleration and threatened a wave of persecutions.40 shahs and nobles The Sasanian monarchy has a reputation for being better organised and more centralised than its Arsacid predecessor. But the notion that the 39 40

Widengren (1961), pp. 139–42; Rubin, Z. (1986), pp. 679–81. Duchesne-Guillemin (1983), pp. 889–90; Khuzistan chronicle, German tr. N¨oldeke, pp. 9–13, 18–22, 28; tr. Lieu et al., pp. 230–1, 232–4, 236.

2a. persia and the sasanian monarchy

145

Arsacid kingdom was in essence a cluster of largely independent political entities, held together in little more than a semblance of formal allegiance to a shadowy central royal authority, may have its roots in tendentious Sasanian traditions. These treat the whole of the Seleucid and Arsacid periods as that of the ‘petty shahs’ or ‘tribal shahs’ (m¯ul¯uk al-taw¯a’if ) and, in sharp contrast, depict the monarchy established by Ardashir as a coherent and effective political and military power. In the Sasanian sixth-century historical romance, the Karnamag Ardasher i Papakan, the fragmentation of Alexander’s empire into 240 small states is the foil to Ardashir’s expoits; the impression produced by the Khwaday-namag tradition of national historical writing, as reflected principally by al-Tabari, is that Ardashir’s rise to power was in effect a long succession of wars for the unification of Iran.41 Greek and Latin sources give the point of view of contemporary outside observers and help to modify this distorted picture, especially with regard to the Parthian empire. However, even these sources suggest that the establishment of the Sasanian monarchy was a dramatic development, for the drive of a rising new power is all too easily contrasted with the lethargy of the ancien r´egime. The result is a widespread consensus among modern scholars that the Sasanian state was more highly centralised and advanced than its Arsacid predecessor. A more balanced picture emerges from an examination of Sasanian institutions, allowing for the distorting vein of propaganda that runs through many of our surviving sources: the dynasty was new, but many of its structures were inherited. Careful analysis of the epigraphic monuments reveals a strong Parthian inheritance, notably an indomitable nobility whose power was only inadequately matched by a somewhat flimsy central administration. Even the question of the genesis of so monumental an inscription as the Res gestae divi Saporis can be misrepresented when coloured by the presupposition of a central royal government controlling every aspect of its erection. A more realistic view would allow for the employment of the remnants of a Parthian chancellery whose execution of the shah’s instructions was not always in perfect accord with his intentions.42 The territorial extent of the Sasanian empire was vast, but the control exercised by central government was not uniformly effective.43 Evidence for the foundation of cities by the Sasanian monarchs after Ardashir, based 41 Karnamak Artakhshir-i Papakan, ch. 1, ed. and tr. Antia, pp. 1–5; German tr. N¨ oldeke, pp. 35–8; al-Tabari, Ta’rikh, ed. de Goeje et al., I, pp. 813–21; tr. Bosworth, V, pp. 1–18; German tr. N¨oldeke, pp. 1–22. 42 On the genesis of the Res gestae divi Saporis, see Rubin, Z. (2002), pp. 291–7. 43 The efficacy of royal control is stressed by Howard-Johnston (1995a), but his model is based on a hypothetical interpretation of archaeological finds rather than the more explicit literary evidence. Limitations on ability to tax: Altheim and Stiehl (eds.) (1954), pp. 47–8.

146

the earlier empire

chiefly on the detailed data preserved by al-Tabari, indicates that their activity was confined to a fairly limited area – the provinces of Fars, Meshan, the Sawad and Media – which were basically the territories conquered by Ardashir I during his wars against the Arsacids and the m¯ul¯uk al-taw¯a’if under their aegis. As a general rule, the Sasanian shahs did not encroach on those territories held by the great lords of the realm, some of whose lineages reached far back into the Parthian era. The one exception to this rule was the occasional establishment of cities in newly acquired border zones, where the shah’s lordship by right of conquest could not be contested; or in remote provinces where royal authority was being re-established. Examples of this exception are the cities founded by Peroz following his war against the Hepthalites: Ram Peroz in the region of Rayy; Roshan Peroz on the border of Gurgan and the Gates of Sul; and Shahram Peroz in Azerbaijan.44 Foundation of a city represented a substantial investment of manpower and resources, and shahs only undertook this in places where it would benefit them, and not one of their overmighty nobles. The picture of a well-ordered hierarchical society, controlled and regulated by a strong monarchy, needs to be reassessed. It emerges from later literary sources of the Islamic period, such as al-Tabari, al-Mas‘udi, Pseudoal-Jahiz45 and The letter of Tansar. The latter is attributed to the powerful third-century herbed Tansar, but was probably composed three centuries later and is preserved in Ibn Isfandyar’s Ta’rikh-i Tabaristan, a problematic source.46 However, these complex issues can be avoided, as the epigraphic sources from the earlier Sasanian period – notably the third century and first half of the fourth – anticipate and corroborate our later literary sources. The inscriptions suggest that the framework of a social hierarchy had already been formally established under Shapur I. The highest rung, immediately below the ‘king of kings’, was that of the shahrdaran. These virtually independent shahs, whose numbers seem to have been much lower under the Sasanians than the Arsacids, tended to be senior members of the royal dynasty and officially ruled their kingdoms as royal appanages. Below them ranked the vaspuhragan, apparently princes of the royal family who held no official post in the royal court. Third in rank were the vuzurgan: members of the great noble houses, including Suren, Karin and the Lords of Undigan, among others. As late as c. 500, the unruly heads of these houses admitted only a nominal allegiance to the central power, and were virtually independent in their hereditary territorial 44 Altheim and Stiehl (eds.) (1954), pp. 12–18; see also Lukonin (1961), pp. 12–19, specifically on the foundations of Ardashir I and Shapur I. 45 al-Tabari, Ta’rikh, ed. de Goeje et al., I, p. 821; tr. Bosworth, V, pp. 18–20; German tr. N¨ oldeke, p. 22; al-Mas‘udi, Muruj al-dhahab, chs. 581–8, ed. Pellat, I, pp. 287–91; rev. French tr. Pellat, I, pp. 218– 20; Pseudo-al-Jahiz, al-Taj, ed. Zaki, pp. 21–8; tr. Pellat, pp. 51–6. 46 Letter of Tansar, tr. Boyce.

2a. persia and the sasanian monarchy

147

domains. The fourth and the lowest rung documented in the inscriptions was the azadan, minor gentry of free status, and distinct from the other nobility, but probably also dependent on them in many cases. From this lesser nobility were recruited the mounted warriors, asavaran, who made up the core of the Sasanian army.47 These may be identical with another category, that of the kadxvadayan, who occupy a place above the azadan and below the vuzurgan in the lists of the Paikuli inscription. On the other hand, they may be an especially favoured group among the asavaran, perhaps akin to enfeoffed ‘knights’ in medieval Europe.48 The stratification that emerges from the later literary sources is more general and reflects the (post-Sasanian) Avestan concept of social stratification. The priests (asronan) appear at the top of the ladder. They are followed by the military estate (artestaran). The third estate is that of the royal bureaucracy (dibiran, i.e. scribes). Finally, the commoners are enumerated, subdivided into peasants (vastaryoshan) and artisans (hutuxshan). If the two hierarchies, inscriptional and literary, are to be amalgamated, the inscriptional hierarchy of nobility should be seen as an expansion of the second estate in the literary sources; on the other hand, the literary hierarchy may not be contemporaneous, since there is no evidence for a separate priestly caste in the early period. Royal power and influence depended to a large degree on effective control of the shahrdaran, as well as on the active support of the majority of the vuzurgan, or equivalent groups, whatever their names in later periods. Their cooperation would be needed for the recruitment of the asavaran who owed them allegiance, and their consent would be required for the imposition of royal taxation within their domains. Sasanian military organisation has been described as feudal, basically similar to its Arsacid predecessor, and this definition may help us to understand how the Sasanian regime worked. From our meagre information about remuneration for the professional core of soldiery, we may conclude that it was supported through land-grants rather than paid in money or kind. Thus it is tempting to accept the notion of enfeoffment, which by its very nature entails bonds of trust and dependence that may be described as ties of vassalage. Yet, if this picture provides a fairly accurate idea of the relationship between the shah and warriors conscripted in his own domain, as well as of that between the grandees and their own warriors, it does not reveal the realities of the links between shah and grandees. The grandees’ domains might have been deemed fiefs granted by the shah, but in most cases this status would only have been theoretical, since forfeiture of such fiefs to the crown could hardly be enforced by means of a simple legal procedure, without recourse to arms: 47 48

For these ranks, see Schippmann (1990), p. 82; see also Wieseh¨ofer (1996), pp. 171–82. See Humbach and Skjaervø (1978–83), III.1, pp. 33–4.

148

the earlier empire

as in any feudal monarchy, there was no guarantee that every Sasanian shah could control all the grandees all the time. There are clear signs that the great nobles of the Sasanian kingdom developed their own concept of legitimation. It was one of basic loyalty to the royal dynasty, but this by no means entailed unconditional loyalty to the individual seated on the throne at any given moment. The shah in power might be replaced by another member of the dynasty if a significant body of nobles found his reign unjust and tyrannical. The nobles likewise did not consider themselves utterly bound to abide by a reigning shah’s own choice of successor. A more suitable candidate might be substituted for his appointee, provided that he came from among the members of the royal house.49 tax ation and military organisation In an empire which minted a stable silver coinage, the drahm, throughout most of its history, the continuing resort to land-grants in return for military service calls for an explanation. The drahm was the only denomination in constant circulation, raising the question whether such a simple economic system can be described as a truly advanced monetary economy. Gold dinars were issued occasionally – not, it seems, for purposes of monetary circulation, but rather in commemoration of solemn events. Bronze change seems to have been issued only very intermittently, perhaps in response to specific demands, as at Merv; the volume progressively decreased, posing problems for the mechanics of everyday economic exchanges.50 The assumption that Arsacid copper coinage was still used in many parts of the Sasanian kingdom is unconvincing,51 and the conclusion must be that much economic activity was based on barter. This situation explains a good deal about the Sasanian system of taxation before the beginning of the sixth century. It was based on crop-sharing, the exaction of agricultural produce proportionate to annual yield, as assessed by royal tax-collectors on the spot, and levied in kind. In addition, a poll tax was imposed on most subjects, which may largely have been paid in money, though part was perhaps commuted to goods. The system was inefficient and wasteful, especially with regard to the land tax; it was subject to frequent fluctuations, and allowed little scope for advance financial planning. The necessity of waiting for the tax-collector with the crops untouched in the field or on the tree meant that some might be damaged 49 50

Rubin, Z. (2004), esp. pp. 263–72. See G¨obl (1954), pp. 96–9; see also G¨obl (1971), pp. 25–30; G¨obl (1983), pp. 328–9. On Merv, see Loginov and Nikitin (1993a); Loginov and Nikitin (1993b); Loginov and Nikitin (1993c). 51 G¨ obl (1954), p. 98; also G¨obl (1971), where continued circulation is suggested only for the earlier period, with no explanation offered for the subsequent mechanics of exchange.

2a. persia and the sasanian monarchy

149

or destroyed before being enjoyed by farmers or the shah. Only lands held directly by the shah could be taxed in this manner effectively, but even on royal domains the avarice of corrupt tax-assessors will have hampered collection.52 Towards the end of the fifth century, the burden of taxation on the peasantry seems to have become increasingly oppressive: the complex relations with the Hephthalite khanate, looming in the east, resulted in heavy demands at a time when recurrent famines compelled shahs to grant occasional – and somewhat measly – tax relief. This oppression contributed significantly to the popularity of Mazdak, a heretical Zoroastrian priest, who advocated the economic equality of all human beings and regarded the higher classes of the Sasanian kingdom as the worst enemies of his doctrines. For some time he managed to enlist the support of Shah Kavad I himself: Kavad appears to have used this movement precisely in order to humble his recalcitrant nobility.53 When eventually he turned his back on the movement and allowed his son to put it down, the battered nobles needed royal support to recuperate and regain a fraction of their former grandeur. They were obviously in no position to form a viable opposition to the one serious attempt to introduce a tax reform in the Sasanian realm, begun apparently towards the end of Kavad’s reign (531) and continued by his son Khusro I.54 On the basis of a general land survey, a new system for exacting the land tax was devised. Fixed rates of tax were imposed on agricultural land according to its size and according to the kind of crops raised. The tax was calculated in drahms, although at least some was probably still levied in kind, calculated according to the current value of the produce in drahms. This new system, efficiently applied, would enable a monarch to anticipate incomes and budget expenses. It might be seen as harsh on the peasantry, primarily because the fixed drahm rates apparently disregarded fluctuations in agricultural yield caused by drought, other natural calamities or war. But this is to ignore the best testimony about the reform: if a distinction is drawn between the reform’s institution and operation in Khusro’s reign, and the way it subsequently worked, the system appears reasonably efficient and fair. It considerably augmented crown revenues, but also included a mechanism for constant revision, making tax rebates and remissions possible when and where necessary. The fiscal reform was accompanied by agricultural reform. Dispossessed farmers were restored to their lands, financial help was available to enable them to restart cultivation, and a mechanism was instituted to assist farms 52 For a very different picture, see Howard-Johnston (1995a), who postulates an efficient tax-raising system not unlike that in the Roman empire. 53 For summary and bibliography on Mazdak, see Guidi (1991); Crone (1991). 54 For more detailed discussion of sources, see Rubin, Z. (1995).

150

the earlier empire

affected by natural disasters. The overall result should have been to maintain a system of small farms that could be taxed easily, and to prevent the growth of huge estates whose powerful owners might accumulate privileges and immunities, and obstruct effective taxation. Khusro’s reform was meant to have a lasting impact on Sasanian military organisation by providing the shah with a standing army of crack units of horsemen (asavaran), under his direct command and permanently at his disposal, who received a salary, at least while on foreign campaign. This body of palace guards was recruited from among young nobles, as well as the country gentry who wished to start a military career. On the frontiers, troops recruited from the nomadic periphery, such as Turks, as well as from semi-independent enclaves within the empire – for example, Daylam in the mountainous region of Gilan – might be employed to repel invasions or hold them up until the mobile crack units arrived. Khusro’s system appears to have enjoyed moderate success for a few decades, until the difficulties that beset the Sasanian monarchy exposed its weaknesses. In the fiscal area, its proper functioning depended on internal stability, external security and continuing financial prosperity, backed up by revenues other than the land and poll taxes – such as taxes on international trade, especially the silk trade, booty from foreign wars, tribute and diplomatic subsidies. These supplementary sources of income were necessary to ensure the smooth running of the control mechanism that was integral to Khusro’s system. However, its stability as a whole depended too much on a delicate balance which only a very powerful monarch could maintain at the best of times, and in the vast Sasanian monarchy, with its long frontiers, it was exposed to the dangers that threatened the empire itself. Growing military commitments increased the financial demands and pressure on tax-payers, thereby threatening the system; if central government lost effective control, abuse and corruption might swamp arrangements. A neglected source which appears reliable on this issue – the Sirat Anushirwan, embedded in Ibn Misqawayh’s Tajarib al-umam – indicates that towards the end of his reign, Khusro struggled to keep his system functioning.55 The control mechanism proved to be as susceptible to corruption as the taxation machinery it was supposed to regulate. Furthermore, the strained relations between soldier and civilian, especially in the remoter zones, took their toll. In effect, the shah could restrain only those soldiers under his direct command from despoiling the rural tax-payers, as is shown by the restrictions imposed by Hormizd IV on a journey to Media. It is probable, however, that even during the last days of his father many of the cavalrymen no longer owed direct allegiance to the shah, and had reverted to 55

Discussion in Rubin, Z. (1995), pp. 237–9, 279–84.

2a. persia and the sasanian monarchy

151

being retainers of the great, virtually independent landlords. A brief glance at the aftermath of Khusro’s military reforms may help us to understand what happened. The fragility of the financial arrangements underpinning the standing army militated against enduring success for Khusro’s reforms. If, as suggested above, the Sasanian economy was never fully monetarised, the need to provide for the army’s everyday needs, at times mostly in goods, will have encouraged the reintroduction of enfeoffment as the standard military contract, even among the lower ranks. Following a short period when Khusro made serious attempts to sustain his new standing army, even in his own lifetime the asavaran increasingly reverted to an enfeoffed estate, despite such fiefs’ tendency to become hereditary and the consequent problems of alienation.56 Khusro’s reforms were, at best, of such limited duration and impact that their scope and intent might be questioned. From the royal perspective, the higher nobility posed even more serious problems than the cavalrymen. The Mazdakite revolt and its aftermath made possible a feudal system more directly dependent on the shah than ever before. The nobility restored by Khusro was firmly beholden to the shah, so there could be no doubts about the origin of its estates or the nature of the services it owed the crown. But the nobility soon returned to its former position of power. The notion that the supreme military commanders and ministers of state were now salaried civil servants is contradicted by the limited evidence available. Thus, for example, Khusro’s nominees as spahbads – the four supreme military commands he created to supersede the old office of the artestaransalar – can only have been mighty territorial lords from the start, as the very territorial nature of their command suggests. The same goes for the marzbans, the commanders of the frontier provinces. The supposition that direct dependence on Khusro as restorer and benefactor would make his nobility more tractable and obedient to the shah in the long term is not sustainable, in view of the role played by the nobility under subsequent reigns, quite apart from the revolts in Khusro’s first decade. Bahram Chobin of the noble house of Mihran, the first serious pretender outside the royal house since the establishment of the Sasanian dynasty, was supported by many disgruntled nobles. Khusro II overcame him in 591 with great difficulty, and only with the expensive support of the Byzantine emperor Maurice.57 Later, the Sasanian monarchy was rocked 56 TS (III.15.4, ed. de Boor and Wirth, p. 141; tr. Whitby and Whitby, p. 96) states that Persian troops did not receive a proper salary during service within the kingdom’s borders, but had to rely on ‘customary distributions’ from the shah. This contradicts the hypothesis of a salaried standing army in Howard-Johnston (1995a). 57 TS, IV–V, ed. de Boor and Wirth, pp. 149–220; tr. Whitby and Whitby pp. 103–57; Whitby, Michael (1988), pp. 276–308.

152

the earlier empire

by other major revolts, such as those of Bistam and Bindoe – Khusro’s relatives and allies turned foes – and of his powerful general, Shahrvaraz, who was to depose his grandson Ardashir III (628–9) and claim the throne.58 By the time of the Arab conquest local rulers, especially in the east and in the Caspian provinces, had become virtually independent. The same is indicated by the confused Arabic traditions concerning Yemen after its conquest by the Persians in the last decade of Khusro I’s reign. The growing independence of the great landlords meant that sooner or later they would inevitably control not only their own retinues of fighting men but also independent taxation in their domains. Thus, for example, according to Dinawari, the future rebel Bistam, upon his nomination as governor by Khusro, instituted taxation in the territories under his rule (Khorasan Qumis, Gurgan and Tabaristan) and in the process remitted half of the tax.59 Other potentates, not in direct or prospective revolt against the shah, may have acted less openly but may not have been impelled by the requirements of war propaganda to be so generous. Under Khusro II, oriental sources record impressive data about royal revenues, which might suggest that the machinery devised by Khusro I was still operating smoothly, and that Khusro II made even better use of it than his grandfather.60 But the full narrative of al-Tabari gives a different impression: the revenues were not the product of regular taxation and should be explained in part by the influx of booty from Byzantine territories (the rich spoils of Alexandria and Jerusalem), and in part by extreme measures of extortion.61 It was primarily as an efficient operator of the taxation machinery that Khusro’s Nestorian finance minister (vastaryoshansalar), Yazdin, endeared himself to his lord; the favourable Khuzistan chronicle insists on the vast amounts of money that he sent to the treasury from the sunrise of one day to the sunrise of the next.62 Such extortions seem to have involved not only an unbearable burden on tax-payers in the royal domain but also an attempt to reintroduce direct royal taxation in the domains of grandees, who had by now come to regard this as a blatant encroachment upon their privileges: the nobles proved ultimately to be his downfall. Thus Khusro II’s riches cannot be attributed to the tax reforms of Khusro I. 58 60 61

59 al-Dinawari, al-Akhbar, I, p. 102. Whitby, Michael (1994), pp. 252–3. Altheim and Stiehl (eds.) (1954), pp. 41–2; Altheim and Stiehl (1957), pp. 52–3. al-Tabari, Ta’rikh, ed. de Goeje et al., I, p. 1042; tr. Bosworth, V, p. 377; German tr. N¨oldeke, pp. 354–5. Al-Tabari might give the impression that the enormous sums referred to derived exclusively from taxation; this holds true only if one ignores the other sources of income which he mentions and if the words concerning spoils of war as a source of income subsequently put into Khusro’s own mouth are disregarded: al-Tabari, Ta’rikh, ed. de Goeje et al., I, p. 1056; tr. Bosworth, V, pp. 392–3; German tr. N¨oldeke, pp. 376–7. 62 Khuzistan chronicle, German tr. N¨ oldeke, p. 22; tr. Lieu et al., p. 234; al-Tabari, Ta’rikh, ed. de Goeje et al., I, pp. 1041–3; tr. Bosworth, V, pp. 375–8; German tr. N¨oldeke, pp. 351–6.

2a. persia and the sasanian monarchy

153

sasanian coll apse The last decades of the Sasanian dynasty are the story of a chain of violent upheavals, exposing all the inherent weaknesses of the huge empire. The reforms of Khusro I did constitute a serious attempt to cope with these weaknesses and to re-establish the shah’s position on a firmer basis. They failed in the long run because they strove to superimpose the framework of a fully centralised state, with a salaried civil bureaucracy and army, financed by an efficient and easily manageable taxation apparatus, on a realm which proved too weak to bear these heavy burdens. The political and military organisation of its vast territories was too flimsy, the economic infrastructure too primitive, and the social structure hidebound by traditions that could not be easily transformed. Khusro’s own conservatism was a characteristic reflection of these traditions, for it was Khusro who did much to restore the battered nobility to its traditional powers after the Mazdakite interlude. Warfare had always been the primary activity of the Sasanian state, but even by its own standards the last century of its existence witnessed a sustained intensity of campaigning that may have weakened the structures of society. After war broke out against Justin I in 527, there were only twenty-eight years of formal peace with Byzantium until the conclusive victory of Heraclius in 628 – and this is to ignore the recurrent tensions enmeshing the Arab satellites of the rival empires, Sasanian involvement in the affairs of the Arabian peninsula and the struggle to maintain control in Caucasian principalities such as Suania and Albania. We know much less about the sequence of campaigns on the north-eastern frontier, but these were probably more debilitating. Khusro’s apparent triumph over the Hephthalites in the 550s was only achieved through alliance with the rising Turkish confederation, which now replaced the Hephthalites as Persia’s neighbours and soon constituted a far more powerful threat during the 570s and 580s.63 No less than Justinian, Khusro was repeatedly involved in wars on more than one front, and the expenses of eastern campaigning probably proved much heavier than the gains from spoils, ransoms and payments stipulated in his treaties with Byzantium. The success of the state depended ultimately on the character and reputation of the shah, and there was a recurrent danger that such a personal monarchy would experience bouts of severe dynastic competition: thus, the long reigns of Shapur I and Shapur II were both followed by shorter periods of instability. This danger may have been increased in the sixth century by the withdrawal of Persian shahs from regular active participation in warfare, a move which fundamentally changed the nature of royal legitimation. 63 Men., 10.3, 13.5, pp. 116–23, 146–7; TS, III.6.9–14, ed. de Boor and Wirth, pp. 121–2; tr. Whitby and Whitby, pp. 80–1.

154

the earlier empire

Early rulers from the house of Sasan had demonstrated divine favour for their rule through personal victories, but the successors of Khusro I relied on others to win their wars.64 From the royal perspective, legitimacy ran in the family, but the nobility and armies might prefer to give their loyalty to a successful commander such as the non-Sasanian Bahram Chobin or Shahrvaraz. The existence of substantial minority religious groups, Jews as well as Christians, allowed an established ruler to secure his position by balancing their different claims against the majority Zoroastrians. But it also meant that a rival could promote himself by seeking the support of one particular group: Bahram Chobin is known for his links with the Jews. In spite of the attempted reforms of Khusro, the Sasanian state remained a fairly simple structure in which much economic and military power rested with the feudal nobility. Royal authority was bolstered by a supremacy of patronage, but this presupposed regular inflows of wealth for redistribution. Wars against the empire provided considerable short-term gains, and Byzantine peace payments under the ‘perpetual peace’ (532) and the fifty-year peace (562) were also important, but it is impossible to calculate how much of this wealth drained eastwards, almost immediately, to the Hephthalites or the Turks. The monetarised heartland of the Sasanian state (as of its Achaemenid antecedent) lay in the rich agricultural lands of Mesopotamia and lower Iraq, areas susceptible to attack from the west, and it seems to have been impossible to increase their tax revenues in the long term. It is ironic that the most successful Sasanian conqueror, Khusro II, must also bear responsibility for the monarchy’s subsequent rapid collapse. In the first decade of his reign, his status as a virtual puppet of Constantinople must have contributed to support for the long-running rebellion of Bistam in the east.65 The overthrow of his patron Maurice in 602 gave Khusro an opportunity to assert his independence, and the disorganisation of Byzantine defences, particularly during the civil war between supporters of Phocas and Heraclius in 609–11, permitted Khusro to transform a sequence of traditional lucrative frontier campaigns into a massive expansionist thrust towards the west. But whereas a war of pillage replenished royal coffers, the annexation of territories reduced the inflow of funds and meant that the newly acquired resources had to be devoted to maintaining troops in remote regions. Furthermore, Khusro’s successful armies had little direct contact with their distant monarch, being tied more closely to their victorious commanders; as a result, the soldiers of Shahrvaraz supported 64 65

Whitby, Michael (1994). Nihayat al-arab, ed. Danish-Pazhuh, p. xx; summarised in Browne (1900), p. 240; Firdausi, Shahnama, lines 2791–6, ed. Nafisi, IX, pp. 2791–6; French tr. Mohl, VII, pp. 143–50; tr. Warner and Warner, VIII, pp. 306–11.

2a. persia and the sasanian monarchy

155

their general when he was threatened by the shah. In the 620s, Heraclius’ campaigns into the heart of Persia exposed the fragility of Khusro’s achievements, prompting a palace coup that introduced the most severe bout of dynastic instability the Sasanian state had ever known. The return of booty to the Byzantines together with the destruction caused by campaigns in Mesopotamia left the monarchy short of wealth and prestige at the very moment when the Arabs started to raid across the Euphrates. Yazdgard III was forced to abandon Iraq in 638–9 and thereafter lacked the resources and reputation to challenge the new Islamic superpower. The Iranian nobility abandoned the Sasanians and transferred their allegiance to the Muslim rulers, who offered stability, while the rural majority went on paying their taxes – to support a new elite.

CHAPTER 2b

EASTERN NEIGHBOURS: ARMENIA (400–600) r. w. t homs on

introduction Armenia has always had an ambiguous place between the major powers, be they the East Roman empire and Sasanian Iran, the Byzantine empire and the caliphate, or the Ottoman empire and the Safavids. Armenian loyalties have not been consistent, either in support of a coherent internal policy or with regard to external diplomacy. The very definition of Armenia highlights the problem. Does the term refer to a geographical entity – and if so, what are its borders? Or does it refer to a people with common bonds – and if so, are those bonds linguistic, religious, cultural or political?1 1 The emphasis in this chapter will be on Armenian reactions to events as expressed by the native historians. The principal Armenian sources for the period are:

Agathangelos: an anonymous history, written at the end of the fifth century, which gives the traditional account of the conversion of King Tiridates and the missionary activity of Gregory the Illuminator at the beginning of the previous century. Although replete with legendary tales and hagiographical commonplaces, it is important for the Armenian Arsacid reaction to the overthrow of the Parthian Arsacid dynasty by the Sasanians. The Buzandaran: this traces the history of Armenia from the death of King Tiridates c. 330 to the division of the country into Roman and Iranian spheres c. 387. The author is unknown. The work is a compilation of epic tales describing the feats of the Arsacid dynasty, the noble house of the Mamikonean family (which played the leading role in the fourth and fifth centuries), and the descendants of Gregory in the office of catholicos. It is the last witness to the disappearing Iranian traditions of Armenia, although the Christian author did not himself comprehend the original significance of all the aspects of social and political life which he described. Koriwn: a disciple of the inventor of the Armenian script, Mashtots‘. His biography of the master is probably the earliest original composition in Armenian. Moses of Khoren: author of a history of Armenia from the days of Noah to the death of Mashtots‘, whose pupil he claims to be. Very important as the first account of Armenian origins, in which oral traditions are integrated into the schema of Eusebius of Caesarea’s Chronicle, it is the most learned of early Armenian histories. Moses used many Greek, Jewish and Syriac sources (via Armenian translations). But his strong pro-Bagratuni bias and his clear distortions of previous writers suggest a later authorship than that claimed. The date and authenticity are hotly contested. But it is significant that the Bagratuni did not gain their ultimate prominence until the eighth century, and Moses’ history is not quoted until the tenth. Elishe: unknown author of a history describing the revolt against Shah Yazdgard II in 450–1, the defeat of the Armenian army led by Vardan Mamikonean at Avarayr, and the ensuing imprisonment of surviving Armenian nobles. This is probably not an eyewitness account as claimed, but a rewriting of the shorter version of these events in Lazar. Its great importance is the adaptation of the story of the Maccabees to the Armenian situation, and the identification of Christian with patriotic

156

2 b. armenia (400– 600)

157

Despite the conversion of King Tiridates IV (c. 283–330) to Christianity, probably in 314,2 and the establishment of an organised church, the continuing strength of Iranian traditions and the cultural and kinship ties of the Armenian nobility to Iran made Armenia an uncertain ally for the Romans. Yet since the Armenian monarchy was a branch of the Arsacid dynasty which had been overthrown by the Sasanians in 224, relations between Armenia and Iran were already strained. Tiridates’ conversion compounded an already difficult situation, for the shahs naturally became suspicious of the future loyalty of Armenians to their Iranian heritage.3 In the fifth century, attempts by the shahs to impose Zoroastrianism led to armed conflict – while to the west, the Armenians found their relationship with fellow Christians increasingly marred by their involvement in the struggles over orthodoxy. The division of Armenia c. 387 into two monarchies and two spheres of influence – a large Iranian sector east of a line running from Sper to Martyropolis (see map 8), and a much smaller Roman sector west of that line up to the Euphrates – did not solve ‘the Armenian question’.4 Both powers were to find Armenia a difficult neighbour. christian armenia bet ween persia and byz antium The ruler of the Roman sector (Inner Armenia) King Arshak III, died in 390. His subjects were immediately placed under direct imperial rule through a comes Armeniae; on the other hand, the traditional rights of the Armenian princes in that area were not abrogated.5 They enjoyed immunity from virtues. A sophisticated literary work, it shaped Armenian attitudes towards the interaction of religion and politics down to the present time. Lazar of P‘arp: author of a history of Armenia from 387, picking up where the Buzandaran ends, to the appointment of Vahan Mamikonean as governor of Persian Armenia in 484. His history is an encomium of the Mamikonean family. Despite its bias, it is valuable as an account by someone who knew the major participants, as most other early Armenian histories are by unknown authors and of uncertain date. Sebeos: a ‘History of Heraclius’ by a bishop Sebeos is mentioned by Armenian authors of the tenth and later centuries, but their quotations do not match the untitled text discovered in the early nineteenth century and published as the work of Sebeos. This anonymous work is important, nonetheless, as a product of the seventh century by an author familiar with events in the milieu of the Armenian catholicos. The emphasis is on Armenia in the context of Byzantino-Iranian rivalry from the time of Maurice (582–602) to the accession of Mu‘awiya as caliph in 661. Book of letters: a compilation of documents dealing with ecclesiastical matters from the fifth to the thirteenth century. Of particular importance are the letters exchanged between official representatives of the Armenian church and foreign dignitaries of the Greek-speaking imperial church, the Syriac-speaking church in Iran and the church in Georgia. 2 This is the usually accepted date for the consecration of Gregory at Caesarea, which marks the beginning of the formal organisation of the church in Armenia. For the origins of Christianity in Armenia, see Thomson (1988–9); Seibt (ed.) (2002). 3 For the Iranian heritage in Armenia: Garso¨ıan (1976); for the religious background: Russell (1987). 4 On this division and the geographical setting: Adontz (1970), pp. 7–24; Hewsen (2001), map no. 69, p. 90. 5 Toumanoff (1963), pp. 133–4.

100

IC

U

IN

A

T A

I

Caspian Gates C A (Darial Pass) S U S M O U K ‘ A RT ‘ L I N

P

S E A

S

N

Ruled by five Armenian dynastic families Lesser Armenia

GUGARK

Trebizond

E S P

R

TA O BASEAN

Lake Sevan

Aza

t

Dvin Avan REVAND G BA Bagavan

S

Theodosioupolis

es

ax

Ar

E

A

FOURTH ARMENIA Martyropolis

A

S

B L A C K Martyropolis

Trebizond FIRS TA RM EN Sebasteia IA Theodosioupolis SECOND ARMENIA

THIRD ARMENIA Melitene

Z

A

Byzantine Armenia 387–c. 528

C

LA

A

ARMENIA II Melitene

Theodosioupolis INNER ARMENIA

400 km

200 miles

C

ARMENIA I

100

azd an

Sebasteia

300

200

Hr

Trebizond

Melitene

hr

at

es

Justinian’s reforms c. 528–536

E

up

Edessa

Avarayr

TARON Martyropolis ARZANENE

Van Lake Van

Lake Urmia

Ti g r i s

Dara

Map 8 The Armenian lands in the earlier Byzantine period

Gandzak

2 b. armenia (400– 600)

159

taxation, and no military garrisons were imposed. Procopius claims that it was this military weakness that later led Justinian to tighten his control. He observed that ‘Armenia was always in a state of disorder, and for this reason an easy prey for the barbarians.’6 He might have added that social, religious and cultural ties with their kinsmen across the border could not enhance security. Not until the sixth century did Justinian do away with the traditional rights of the Armenian princes in a series of moves between 528 and 535. Armenian lands west of the border with Iran were then fully integrated into the empire as the four provinces of Armenia (see below pp. 167–8). It was in eastern Armenia – the sector under Persian suzerainty, which composed about four-fifths of the earlier kingdom – that the major cultural and religious developments of this period had their origin. Yet the border between the two sectors was no solid wall. Although Armenian writers rarely refer explicitly to the border, through the communities in the west contacts between the imperial capital and Persian Armenia were promoted and sustained. In eastern Armenia the centrifugal tendencies of the leading princely families rapidly overcame the weakened monarchy. The rights and privileges of the noble families, jealously guarded over generations and considered more fundamental than royal authority, had been recognised by the Arsacids and legitimised. The office of chief military officer (sparapet), for example, was the perquisite of the Mamikonean family, which played the leading role in politics during the fifth and sixth centuries. Their principal rivals, the Bagratuni – who did not attain the leading role until the eighth century – held the hereditary right to crown Armenian rulers (see below, p. 348). In Arsacid Armenia there were some fifty noble families of varied size and power, each with its own military forces.7 Cities played little political or cultural role, despite their economic significance.8 The focus of noble life was the family holdings. The Mamikonean territories were in Tao, Bagrevand and Taron – i.e. much of north-central Armenia. The Bagratuni homeland was in Sper, but they gradually acquired territories to the southwest. A branch of this family was established in eastern Georgia (‘Iberia’ to the Greeks, ‘K’art’li’ to the Georgians). To the south-east of Lake Van another family, the Artsruni, were settled. They acquired land between Lake Van and the Araxes, and were later to become the principal rivals of the Bagratuni. After the demise of the royal line, these families pursued their own interests with regard to Rome or Persia, conducting, as it were, an individual foreign policy. Eastern Armenia was thus not a stable unity. 6 7 8

Pr B, III.1.16, ed. and tr. Dewing and Downey, pp. 182–3. Toumanoff (1963), pp. 147–259. Garso¨ıan (1984–85). For the economic situation: Manandian (1965), pp. 67–127.

160

the earlier empire

This traditional pattern of society was reinforced by the growth of an organised Armenian church. Armenian historians of the fifth and sixth century often stress the ‘national’ role of the church and the leadership of the catholicos. But they do not explain that the bishoprics were established within the princely families, reinforcing the authority of the princes. This Armenian pattern, reflecting Armenian society of the time, was very different from that in the empire, which was based on the relative importance of the cities where the bishops resided. Furthermore, the Armenian office of catholicos until the death of Sahak in 439 was itself regarded as a hereditary perquisite of the Pahlavuni family, just as were other offices of state in other families.9 Unhappy with any diminution of their privileges, the magnates of eastern Armenia quarrelled with their king Khusro IV (who had been installed in 387 when Arshak III moved to western Armenia) and succeeded in having him deposed in favour of his son Vramshapur (389–417). The weaker the monarchy – from their point of view – the better, and soon the princes came to regard the Persian shah himself as their immediate sovereign. On Vramshapur’s death his father was briefly reinstated; then Shah Yazdgard I (399–420) appointed his own son, Shapur. On Yazdgard’s death Shapur failed to win the succession to the Sasanian throne. Shah Bahram V (420–38) permitted Vramshapur’s son Ardashir to reign, but he too was unpopular. In 428 Bahram agreed to accept the direct submission of the Armenian princes. The monarchy was abolished, and a Persian governor, the marzban, installed at Dvin.10 The marzban was responsible for collecting taxes; the princes provided military service to the shah in person with their private armies. In their own lands they were autonomous. In this way, the shah took advantage of age-long Armenian practices to increase Persian control of Armenia. Recognising the importance of the church in that valuable province, he attempted to strengthen his hand even more by deposing the catholicos Sahak (c. 390–438), who represented continuity with the past through his descent from Gregory the Illuminator and whose outlook allied him to Greek cultural interests. Sahak was replaced by an insignificant appointee, to be succeeded by two Syrians.11 Bahram’s policy with regard to the political administration of Armenia was moderate and successful. But his interference in ecclesiastical affairs was less well received. And his successor’s harsher measures, aimed at integrating Armenia more closely into the Sasanian empire, eventually sparked outright rebellion. The passion of those who resisted – and resistance was by no means unanimous – reflects the increased Armenian allegiance to the church and to Christianity as their birthright. The terms ‘patrimonial’ or 9 11

10 On this office: Christensen (1944), pp. 131–9. Garso¨ıan (1983), pp. 233–5. On these three – Surmak, Brkisho and Shmuel: Narratio de rebus Armeniae, ed. Garitte, pp. 99–102 (commentary).

2 b. armenia (400– 600)

161

‘ancestral way of life’ originally used for the secular realm – where they applied to personal estates or the monarchy – were adapted by early Armenian historians to the religious sphere, where they now defined Christianity and the church within an Armenian context. Yet Christianity was hardly ‘an ancestral way of life’ in fifth-century Armenia. The anonymous historian known as Agathangelos, who gives the standard account of the conversion of the country, claims that Gregory visited the whole Caucasus, baptised millions of Armenians and established hundreds of bishoprics.12 Agathangelos was too optimistic. The process of conversion took many generations, and the church met with opposition on many fronts. The Buzandaran paints a vivid picture of the pro-Iranian tendencies of many noble families, whose allegiance to the shah was strengthened by acceptance of Zoroastrianism. For many the Christian message, which reached Armenia from Syria in the south and from Asia Minor to the west, was a foreign faith.13 The fact that no written medium for the Armenian language existed in the fourth century added to the difficulty of strengthening the church’s position and overcoming resistance to this alien innovation. So the invention by Mashtots‘ of a script for the native tongue c. 400 marked a very significant stage in the conversion of Armenia to Christianity, though it was not in itself the last step in that process (see fig. 6 below). Mashtots‘ had received a Greek education and rose to a prominent position in the royal chancellery, but withdrew in order to lead a hermit’s life. In due course he attracted disciples and, with support from Catholicos Sahak and King Vramshapur, formed a script based on the Greek model – i.e. a fully alphabetical script with separate characters for each consonant and vowel. With only minor modifications, it has remained in continuous use down to the present day. His disciples were sent to Syria and Asia Minor to learn Syriac and Greek and to make translations of books needed for the church. Rapidly a corpus of biblical, liturgical, theological and historical texts was made available. The circle around Mashtots‘ began to create original works as well, and their interests soon extended to secular studies as pursued in the contemporary schools and universities of the eastern Mediterranean – they produced works of philosophy, grammar and rhetoric, and of scientific enquiry.14 The development of a specifically Armenian literature – in the broadest sense of the term – brought several consequences: an increasing sense of solidarity among Armenians on either side of the Byzantine–Iranian border, a 12 For a comparison of the various recensions and versions of this history see Agathangelos, History, tr. Thomson, pp. xxi–xcvii (introduction). 13 There is no general study in a western language of the impact of the Syrian strain in Armenian Christianity more recent than Ter-Minassiantz (1904). Aspects of Syrian liturgical influence are brought out by Winkler (1982) and Winkler (2000), with good bibliographies. 14 Renoux (1993). Survey of the early period: Thomson (1982); detailed bibliography of sources: Thomson (1995); Thomson (2007).

162

the earlier empire

Figure 6 Inscription dated to 668 from the church of St Gregory at Aruch; a fine example of the uncial script invented by Mashtots‘ c. 400 ad

stronger voice in national affairs for church authorities as a body that spoke for interests broader than those of individual families, and greater involvement in the ecclesiastical questions that were shaking the East Roman empire. Armenia’s liturgical practice was greatly influenced by Jerusalem. Many from Armenia and Georgia made pilgrimages to the Holy Land, some staying on as monks. The theological exegesis of Syria made a great impact, and the Greek fathers of the fourth century were well-known. As the Armenians forged their own traditions in matters of practice, their attitudes with regard to matters of faith were sharpened by involvement in the burning issues of the day. This heightened sense of commitment to a faith associated with the empire could only be regarded with concern by the rulers of the Sasanian world in which most of Armenia lay. The attempt of Yazdgard II (438–57) to impose a form of Zoroastrianism by force in 450 prompted immediate resistance by the church authorities; popular resentment coalesced around the prince of the Mamikonean family, Vardan. He was related by marriage to Catholicos Sahak, whose daughter his father had married, and his family played the leading role in contemporary Armenian politics. Like many other Armenian princes, Vardan had earlier temporised by submitting to Zoroastrianism when summoned to court. But he agreed to lead the revolt, and one of his brothers went with a delegation to seek aid from Theodosius II. The latter died in July, and Marcian refused

2 b. armenia (400– 600)

163

to become involved in Armenia, having many distractions closer to home in the Balkans.15 For that first year the Armenians held off the Persian forces. But faced with dissension in their own ranks, they could not resist a large Persian army sent to Armenia in 451. In June, Vardan and many nobles met their deaths on the field of Avarayr in eastern central Armenia; other leaders, both clerical and lay, were taken in captivity to the region of Nishapur. Resistance in a military sense was thus ended. But Armenia was a valuable asset to the Sasanian empire, and calmer views prevailed. Forced conversion to Zoroastrianism was dropped, and an uneasy peace marked the next generation. During the reign of Peroz (459–84) the imprisoned leaders of the rebellion were released. The close ties between Armenia and Georgia were the indirect cause of the next attempt to loosen Iranian control. A daughter of Vardan Mamikonean, Shushanik, had married Vazgen, governor of the neighbouring province of Gugark‘. But he accepted Zoroastrianism, in return for which he was given a royal princess to wife. His first wife, Shushanik, died of subsequent illtreatment, and was to become a martyred saint revered on both sides of the Armenian–Georgian border.16 Her Life is the first original composition in Georgian. The Georgian king, Vakhtang-Gorgasal, eventually put Vazgen to death in 482, thereby incurring the immediate wrath of his lord, Shah Peroz. In this emergency Vakhtang sought aid from Huns beyond the Caucasus and from his Christian neighbours to the south. Vardan’s nephew, Vahan Mamikonean, now the leading prince of that family, thus found himself at the head of the Armenian forces engaged in another rebellion, thirty-one years after his uncle’s death. Military success was no more possible now than it had been earlier. Armenian–Georgian co-operation was marred by mutual antagonisms, brought out clearly by the historian Lazar, who describes this period in detail – Vahan being the patron and hero of his History. The Armenian troops were forced to withdraw to the mountains of north-western Armenia. They were rent by internal dissensions, the Persians always finding supporters among the Armenian nobility. On the other hand, the Persian forces were not at full strength, since Peroz had taken a large army to attack the Hephthalites (see above, p. 134). His unexpected defeat and death on the battlefield in 484 changed the situation entirely. Anxious to placate their fractious subjects, whose Christian ties to the empire were a potential source of danger, the Persians removed their governor. In his place, the prince of the most prominent local noble family was appointed marzban. 15 This revolt is not mentioned in contemporary Greek sources. For the date of Elishe’s classic description and its relation to the version of Lazar: Elishe, History, tr. Thomson, pp. 23–9 (introduction). 16 For discussion of the original text by Jacob of Tsurtav and later versions in their historical setting: Peeters (1935).

164

the earlier empire

Thus Vahan Mamikonean gained the measure of internal autonomy for which his uncle Vardan had died in 451. The attention of Armenian historians moves rapidly from Vahan’s success to the involvement of Armenia in the Byzantine–Persian wars of the late sixth and early seventh centuries. In doing so, they ignore the growing estrangement of the Armenian from the imperial church – a rift with cultural and political consequences of the first magnitude. the arm en ian church as rallying-point and rel at ions with the imperial church The increasing importance of the church as a cultural institution following the abolition of the monarchy in 428 is not of itself surprising. It was the only institution that cut across factional lines, and it was the only medium through which literary and artistic endeavours could be realised on any meaningful scale. Individuals with financial backing would still attend the universities of the eastern Mediterranean; Greek and Syriac as well as Armenian sources attest the presence of Armenian students in Antioch, Beirut, Alexandria, Athens and elsewhere. But government service as a career for the educated was no longer an option after 428; the only major patron of education and learning was the church, and only the church could offer advancement for the ambitious and a haven for the studious. The complaints of Anania of Shirak in the seventh century that his fellow countrymen did not admire learning suggest that without patronage a teaching career was difficult.17 There were cities in Armenia, but they did not play the cultural role of an Antioch or an Athens, with organised schools and subsidised professorial chairs. The relationship of the Armenian church to the larger Greek-speaking world was thus of importance. Armenians were always admirers of Greek learning, but their attitude to Constantinople was ambivalent. In part, such an attitude reflected the political situation; a pro-Greek attitude could arouse suspicions of disloyalty to the shah. Some part was played by the very different backgrounds of Armenians and Greeks – and, not least, the strong Syrian strain in Armenian ecclesiastical life, church ritual and theological exegesis prevented any automatic acceptance of things Greek. An official break between the churches was long in coming. But the steps leading to that eventual rupture deserve a brief review. Luckily, the Armenian reaction to the theological questions that divided the Greek oikoumen¯e – debates which gave the Armenians an opportunity to define more carefully their own position – is well documented in the Book of letters. The first three sections of this unique collection of official documents 17

His short Autobiography is a unique document in early Armenian literature. See Berb´erian (1964).

2 b. armenia (400– 600)

165

comprise exchanges of letters between Armenian ecclesiastical authorities and members of the Greek-speaking imperial church, representatives of the Syriac-speaking church in Iran, and ecclesiastics in Georgia, covering the fifth, sixth and seventh centuries. The earliest is a letter by Acacius, bishop of Melitene, written soon after the council of Ephesus, held in 431.18 Melitene had been one of the cities where the pupils of Mashtots‘ pursued their study of Greek. Acacius had met Mashtots‘ on the latter’s travels in Roman territory, and was well informed of events in Armenia. He had recently played a significant role in the council of Ephesus, where Nestorius, patriarch of Constantinople (428–31) and other Antiochene theologians had been condemned. So he took alarm when he heard that works by Theodore of Mopsuestia were being read in Armenia. For Theodore was a prominent biblical exponent of the Antiochene school, whose interpretation of the Incarnation had been rejected at Ephesus. But Armenian interest in Theodore was not surprising, since the tradition he represented had been strong in Edessa, the centre of Syriac-speaking Christian culture. It was to Edessa that Mashtots‘ had gone in his search for an Armenian script, and it was in Edessa that many of his pupils studied. The reply to Acacius’ letter, signed by Sahak as head of the Armenian church, was polite but guarded, denying any Armenian involvement in heresy yet not specifying any heresy by name. A second letter was sent by Acacius to the secular authorities of Armenia. It had been prompted by Syrian priests who reported that the influence of Nestorian ideas in Armenia was continuing. But it passed without response. Of greater impact was a letter from the patriarch of Constantinople, Proclus (434–46). This time it was not foreign Syrians, but two pupils of Mashtots‘ who had taken the initiative. While in the capital to translate Greek texts, they approached the patriarch for an authoritative interpretation of the doctrine of the Incarnation. That this was not an official solicitation by the Armenian authorities is clear from an apology by a third Armenian disciple, Eznik, who had studied in Edessa before going to Constantinople. Proclus responded by addressing a detailed exposition of the matter to the bishops of Armenia. The Armenian reply was signed by both Sahak and Mashtots‘. After defining their own faith, they assured the patriarch that no heretical ideas attributable to Theodore were circulating in Armenia. The letter of Proclus, however, was to remain a keystone of Armenian orthdoxy, and this early emphasis on the council of Ephesus had a profound impact. Ephesus, rather than the council of Chalcedon, held twenty years later, would be the rallying-cry of Armenian theologians. 18 For the Armenian correspondence with Acacius and Patriarch Proclus: Book of letters, French tr. Tallon, pp. 29–44, 53–77; for the Armenian reaction to the theological disputes: Sarkissian (1975); and in much greater detail with French translations of the documents: Garso¨ıan (1999a); for the debates within the Eastern Roman empire: Grillmeier (1975–96).

166

the earlier empire

The fourth ecumenical council – held beside the Bosporus in Chalcedon in 451 – did not bring peace to the warring parties or solve the theological question of defining the Incarnation in a manner satisfactory to all. The catholicos of Armenia was not represented at Chalcedon, though bishops from Armenian provinces on the Roman side of the frontier were in attendance.19 Somewhat surprisingly, the early Armenian historians pass over both the second (‘robber’) council of Ephesus in 449 and that of Chalcedon in 451. It was the Henotikon of Emperor Zeno (474–91), promulgated in 482, which Armenians emphasised as orthodox.20 Bypassing the recent divisive council of Chalcedon, in their official pronouncements the Armenians were happy to pledge their allegiance to the councils of Nicaea (325) and Ephesus. As they developed their own traditions in ecclesiastical architecture and moulded an individual Armenian literature, they were not at the turn of the century acting in deliberate opposition to what was then the orthodoxy of the empire. At a council held in 505–6 in Dvin, the residence of the marzban and the main city of Persian Armenia, a group of Syrians from the Persian empire appeared, requesting episcopal consecration for one of their monks, Symeon. These Syrians were not members of the church in Persia which enjoyed the shah’s official recognition, but were monophysites. The Armenian bishops consecrated Symeon and recognised the orthodoxy of these Syrians as being in conformity with their own faith and that of the Greeks. But the zealous Symeon, an opponent of the official church in Persia, persuaded the Armenians to anathematise the council of Chalcedon as expressing the views of Nestorius.21 The Armenians did not anathematise the imperial church as such; the Henotikon of Zeno was still in force, and he was regarded by the Armenians as ‘the blessed emperor’.22 But this apparent unanimity of the imperial and Armenian churches was short-lived. Zeno’s policy of compromise with the opponents of Chalcedon was reversed on the accession of Justin I (518–27). After 518 the imperial church of Constantinople made peace with Rome and stood firmly behind the definitions of Chalcedon. As the sixth century progressed, the monophysites in Syria and Egypt became more coherently organised, thanks mainly to the labours of Jacob Baradaeus (see above, p. 118), while their 19 20

Garso¨ıan (1988); Garso¨ıan (1999a), pp. 127–9. See also above, p. 102. As noted above (p. 163), Vahan Mamikonean was then engaged in open rebellion against the shah, a situation resolved by his eventual appointment as marzban. 21 On Symeon’s career: Shahid (ed. and tr.), Martyrs of the Najran, pp. 159–79. 22 In the Book of letters, Zeno is called ‘blessed’ (ed. Izmireants‘, pp. 49, 140, 268; ed. Pogharean, pp. 159, 284, 112; French tr. Garso¨ıan (1999a), pp. 448 (but Garso¨ıan here translates as ‘pieux’), 539); ‘pious’ (ed. Izmireants‘, pp. 141, 142, 328; ed. Pogharean, pp. 286, 504; French tr. Garso¨ıan (1999a), pp. 540, 541); ‘orthodox’ (ed. Izmireants‘, pp. 126, 262, 277; ed. Pogharean, pp. 266, 105, 144); ‘benevolent’ (ed. Izmireants‘, pp. 266, 267, 269; ed. Pogharean, pp. 109, 111, 114). On this council in Dvin: Sarkissian (1965), pp. 196–213; Garso¨ıan (1999b).

2 b. armenia (400– 600)

167

theology found definite expression in the works of Severus of Antioch. The differences apparent at the time of Chalcedon had now become quite clear-cut, and compromise was increasingly difficult.23 rel ations with the syrian s, just inian and his successors The Armenians do not seem to have taken any definite steps to repudiate the Byzantine return to Chalcedonian orthodoxy until they were prompted to do so by another Syrian delegation from Persia, which appeared at another council held in Dvin in 555, again requesting consecration for one of their company. These Syrians were members of a splinter group within the monophysite church, the Julianists, who held that Christ’s body had remained ‘incorruptible’.24 The Armenian catholicos Nerses II (548–57) and his bishops found the Syrians’ profession of faith orthodox and consecrated Abdisho. The impact of Julianist ideas was not the most important result of this encounter in 555; in later years there was no unanimity among Armenian theologians on that issue. The significant fact was that the Armenians not only rejected Chalcedon again; they also, for the first time, specifically anathematised the imperial church for upholding that council – which to Armenian eyes had approved the ideas of Nestorius.25 Despite these important developments, whose significance was perhaps not obvious at the time, Armenian historians have remarkably little to say about Armenian affairs during the reigns of Justin and Justinian (527– 65). The first Persian war, which ended in 532, brought no change to the frontiers or the status of the divided country. Even the reorganisation of the Armenian territories within the empire by Justinian is passed over by Armenian sources. In 528 the right of Armenian princes to maintain their private military forces was abrogated when the office of magister militum per Armeniam was created. The civil standing of the princes was diminished when their traditional rights of inheritance were brought into line with imperial practice. In 536 Armenian territory was reorganised into First, Second, Third and Fourth Armenia at the expense of neighbouring land in Cappadocia. The use of the name ‘Armenia’ is an indication of the strongly Armenian presence west of the Euphrates, which had been increasing rather than diminishing. Now, not only were the Armenians inside the imperial borders deprived of their long-standing rights and governance by 23 Frend (1972), pp. 201–20 (on Severus); pp. 284–7 (on Jacob Baradaeus). See also Kennedy (2000), p. 594 and above, p. 118. 24 For this controversy: Draguet (1924). 25 On this second council of Dvin and the correspondence in the Book of letters: Narratio de rebus Armeniae, chs. 60–76, ed. Garitte, pp. 34–6 (text), pp. 130–75 (commentary); French tr. Mah´e, pp. 433–4.

168

the earlier empire

traditional princely families (which had been guaranteed in the original treaty), but this significant portion of the total Armenian population was lost to Armenia proper. Imperial authorities did not speak Armenian or encourage allegiance to the Armenian church, as Justinian attempted to impose imperial orthodoxy on his realms. Armenians were useful to the empire in many ways, especially in the army. But an individual Armenian culture flourished henceforth only on the Persian side of the frontier. Justinian’s treatment of his Armenian nobles led to complaints to the shah26 and Armenian involvement in war plans against the emperor.27 In 540 hostilities between Byzantium and Persia reopened. Antioch was captured, but Dara resisted the invading Persians. Military operations were confined to Mesopotamia and Lazica during the war, save for an encounter at Dvin in 543. The peace of 545 was one of many made during the long confrontation, which continued into the following century (see above, pp. 120, 135–6). There was no overt sign of unrest in Persian Armenia until the latter part of the sixth century. When trouble did break out, it seems to have been caused by the attitude of the Persian marzban of the time, Suren, not by the official policy of the shah. In 571 Suren set up a fire-temple in Dvin and attempted to impose Zoroastrianism on the country. The reaction was parallel to that of 450. Led by Vardan, prince of the Mamikoneans (not to be confused with the leader of the fifth-century revolt), the Armenians rebelled. When Suren returned the following year with reinforcements, he perished in the encounter. However, the Persians retook Dvin, and Vardan fled to Constantinople. Now, for the first time, the consequences of the religious differences became clear. Vardan had to accept communion with the imperial church, while Catholicos John II (557–74), who had fled with him, remained at Constantinople under the cloud of submission to Chalcedon until his death in 574.28 Justin II (565–78) gave Vardan military forces, and Dvin was retaken. But Byzantine success was not lasting. In 576 Persian forces under Khusro I (531–79) crossed Armenia but failed to capture Theodosioupolis. After advancing as far as Sebasteia, Khusro withdrew and sacked Melitene, but after a confrontation there, he fled back to Persia in confusion. During negotiations the following year, the Byzantine general Justinian was defeated by Khusro in Basean and Bagrevand,29 and the Persians retained the frontier fortress of Dara, which they had captured in 573.30 26 27 28

Toumanoff (1963), p. 175; Pr W, II.3.31–3, ed. and tr. Dewing, I, pp. 278–9. Adontz (1970), pp. 160–1; Pr W, II.3.53, ed. and tr. Dewing, I, pp. 284–5. On this rebellion and the ‘union’ of 572: Narratio de rebus Armeniae, ed. Garitte, pp. 183–225 (commentary). 29 Seb., ch. 71, tr. and comm. Thomson and Howard-Johnston, I, p. 11. 30 Whitby, Michael (1988), pp. 264–7.

2 b. armenia (400– 600)

169

Imperial fortunes revived in 590 when the general Bahram Chobin seized the Sasanian throne upon the murder of Shah Hormizd IV (579–90). The legitimate heir, Hormizd’s son Khusro II, appealed to Emperor Maurice for help, promising in return to cede to the empire all Armenia as far as Lake Van and Dvin, plus part of Georgia. The offer was accepted, and the Armenians under Mushegh Mamikonean sided with Khusro and the Byzantines. Their combined forces defeated Bahram the following year at Gandzak in eastern Armenia. Installed as ruler of Persia, Khusro II (591– 628) fulfilled his promise: Armenia west of the Hrazdan and Azat rivers passed to Byzantium (see above, pp. 127, 136). This success for the Roman empire was fraught with a number of consequences for the Armenians. Maurice attempted to integrate Armenia more securely into the empire. He deported significant numbers of Armenians to the Balkans to strengthen his borders there and weaken resistance to imperial rule among Armenians now incorporated into the empire. The Armenian general Mushegh Mamikonean was killed in Thrace.31 But Maurice sometimes encountered resistance by Armenian soldiers. The Bagratuni prince Smbat rebelled and was condemned to the arena. Saved by his strength, according to the Armenian historian (by the clemency of the empress, according to a Greek source), he was exiled to Africa.32 But it was not long before he was back east, serving the shah. The plight of the Armenians between shah and emperor is well expressed in an apocryphal letter which the Armenian historian known as Sebeos claims was sent by Maurice to Khusro: They are a perverse and disobedient nation, who stand between us and disturb us. I shall gather mine and send them off to Thrace. You gather yours and order them to be sent to the east. If they die, it is our enemies who die. If they kill, they kill our enemies. Then we shall live in peace. For if they remain in their own land, there will be no repose for us.33

But the most significant aspect of his policy was the attempt to enforce imperial orthodoxy in the newly acquired territories. The Armenian catholicos was summoned to a synod where the union of the churches might be effected – that is, where the Armenians would accept Chalcedon and take communion with the Byzantines. Catholicos Moses II (574–604) refused to go and remained in Dvin, just across the border. On this occasion he is credited with a riposte that clearly expressed Armenian resistance to assimilation. It is preserved in a rare pro-Chalcedonian document of Armenian 31 Seb., chs. 90–1, tr. and comm. Thomson and Howard-Johnston, I, pp. 35–6; Whitby, Michael (1988), pp. 127–8, notes that Sebeos’ account seems to conflate several campaigns. 32 Seb., ch. 93, tr. and comm. Thomson and Howard-Johnston, I, pp. 39–40; Whitby, Michael (1988), p. 127. 33 Seb., ch. 86, tr. and comm. Thomson and Howard-Johnston, I, p. 31.

170

the earlier empire

origin: ‘I shall not cross the Azat; I shall not eat bread baked [in the oven]; I shall not drink warm water.’ The Azat was the river marking the border and is a pun, the word meaning ‘free’. The other two comments refer to the differing practices of the liturgy, since Armenians used unleavened bread and did not mix warm water with the wine.34 Matters of doctrine may figure more prominently in the written records of historians and theologians, but the development of different rituals was no less potent a factor in the estrangement of the churches. Nevertheless, the Armenian bishops in Byzantine territory did go to Constantinople and accept communion, thus causing a schism in the Armenian church. But once Byzantine forces withdrew, then Armenian unity was restored. This pattern recurred in the time of Heraclius (610–41) and again under Justinian II (685–95), but proved no more lasting than under Maurice. Despite the fact that many sympathised with the position of the imperial church – and significant groups of Chalcedonian Armenians existed in the succeeding centuries35 – reunion between the Byzantine and Armenian churches was never achieved. Yet the time of Maurice was remembered as a time of peace. The curious text known as ‘Pseudo-Shapuh’ – a medley of tales dating from the ninth to the twelfth century, and not the lost work of the ninth-century historian – refers to the proverb: ‘as in the time of Maurice, when one lived untroubled’. It also reports that when Maurice summoned his father David, who lived in Armenia, the latter said: ‘I cannot come. I prefer my small garden to the Roman empire.’ But by cutting off the heads of the largest beetroots in his garden, he indicated to his son’s messengers how Maurice should treat his magnates.36 Just as Maurice used Armenian arms in the Balkans, so did those Armenian princes on the Persian side of the border continue to provide military service to the shah. The most notable example is the career of Smbat, prince of the Bagratuni, who served at different times both emperor and shah – Armenian loyalties being rarely unequivocal and permanent. Just as Maurice settled colonies of Armenians in the west, so did Smbat find Armenians, Greeks and Syrians deported to Hyrcania when he was serving as governor there for Khusro II. Sebeos notes that the Armenians had even forgotten their own language, and that Smbat remedied this by arranging for the ministry of a priest.37 The role of language and religion as a means of

34 On the ‘union’ of 591 and Moses II’s comments: Narratio de rebus Armeniae, ed. Garitte, pp. 225–54 (commentary); for the border running between Dvin and Avan: Hewsen (2001), p. 90. 35 For Chalcedonian Armenians in later centuries: Arutiunova-Fidanian (1980). See also below, pp. 333–64. 36 Pseudo-Shapuh, chs. 49, 51, tr. Thomson, p. 185; see also Adontz (1934), pp. 1–9. 37 Seb., chs. 96–7, tr. and comm. Thomson and Howard-Johnston, I, pp. 43–4.

2 b. armenia (400– 600)

171

preserving Armenian identity in colonies outside the homeland was already clear. At the same time, the Armenians were estranged from their northern neighbours. The Georgians under their catholicos Kyrion disavowed the Armenian rejection of Chalcedon and henceforth remained firmly committed to the orthodoxy of Constantinople. The final rupture occurred after a series of bitter exchanges. At another council held in Dvin in 608, the Armenians excommunicated the Georgians.38 But contacts between the two peoples could not be stopped by fiat, not least because of the extensive bonds of consanguinity linking noble families on both sides of the frontier. Pro-Chalcedonian Armenians were particularly numerous in Tao and Gugark‘, where the two peoples mingled. Maurice’s downfall in 602 gave Khusro II an opportunity to recover the Armenian lands ceded to his earlier supporter. The reign of Heraclius would see the final defeat of Sasanian Persia and the rise of a new power in the Middle East. But already by the turn of the sixth century the building-blocks of an independent Armenian culture had been formed.39 conclusion Many years earlier in his Annals (II.56; XIII.34) Tacitus had referred to the ambivalent role of Armenia and the Armenians between Rome and Parthia: ‘a people from the earliest times of equal ambiguity in character and geography . . . placed between two great empires, with which they differ frequently’. He described their dealings with both sides, and he knew that fundamentally the Armenians were closer to Iran than to Rome. In Sasanian times as well, the value of Armenia as a vassal state was recognised by the two sides: the East Roman empire and Sasanian Iran both sought to control Armenia, to engage its troops and to profit from its gold mines and other natural resources. After the division of the country and the abolition of the monarchy, attempted control became attempted integration – more successful in Roman Armenia than in the much larger eastern sector. The conversion of the Armenians to Christianity gradually changed their relationship with Iran, but slowly and painfully. The various strands of Christian practice from Jerusalem, Syria and Asia Minor were moulded into a national tradition. But their faith and practice kept the Armenians apart from the imperial church of Constantinople. Armenian scholars created a national literature that was overtly patterned on the Christian literatures 38 The Armenian–Georgian correspondence in the Book of letters has been translated into French in Garso¨ıan (1999a), pp. 516–83. Many of the documents are also quoted by the tenth-century historian Ukhtanes, History of Armenia. 39 Mah´ e (1997); Garso¨ıan and Mah´e (1997). See also below, pp. 333, 335–6.

172

the earlier empire

in Syriac and Greek, reflecting also the influence of late antique culture which Armenians of the fourth and later centuries absorbed in the schools of the eastern Mediterranean. But the Iranian background was not easily shaken off, and Persian motifs reappeared throughout the centuries. Many Armenians found fame and fortune in the Byzantine empire,40 but Armenia as a whole was never integrated into the Greek-speaking empire. When Armenians later reflected on their individuality and the formation of their unique culture, they concentrated on a few specific episodes: the conversion of King Tiridates, the invention of the Armenian script and beginnings of a literature in the vernacular, and the heroic resistance to Sasanian attempts to impose Zoroastrianism. The interpreters of those events, no matter how far removed or tendentious, became the classic authors par excellence. And the images of those events as expressed in the classic histories gave meaning to succeeding generations who sought to understand the role and fate of Armenia in an unfriendly world. Armenia may have played a larger role in the politics of the Middle East in the time of Tigran the Great, as Moses of Khoren rightly stated: ‘He extended the borders of our territory, and established them at their extreme limits in antiquity. He was envied by all who lived in his time, while he and his epoch were admired by posterity.’41 Yet Tigran and military success were not the typical models in terms of which Armenians thought of their present and future. Imagery of a ‘golden age’ described the harmony of King Tiridates and Gregory the Illuminator, while wishful prophecies foresaw the eradication of present woes by the restoration of the descendants of the one to the Arsacid throne and of the descendants of the other to the office of catholicos. More powerful than the memory of the heroic Tigran was the model of the Maccabees, whose defence of ancestral customs and an individual religious culture evoked a strong response in Armenian minds.42 So in the fifth and sixth centuries the image of an Armenian ‘classical’ age was created. Perhaps exaggerated in retrospect, it nonetheless depicted a people who could not be assimilated into either of the imperial powers. 40 41

Charanis (1961); Kazhdan (1975) for middle Byzantium. Moses of Khoren, History, I.24, tr. Thomson, p. 113 – though he has dated this Tigran far too

early. 42 See Elishe, History, tr.Thomson, pp. 11–18 (introduction). Such imagery was applied by the tenthcentury Thomas Artsruni to the Muslim rulers, and is frequently found in later Armenian writers. See also below, p. 336.

CHAPTER 2c

EASTERN NEIGHBOURS: THE ARABS TO THE TIME OF THE PROPHET law re n ce i. con ra d ∗

introd uction: the question of sources In the present state of our knowledge it is not difficult to describe the physical setting for pre-Islamic Arabian history, and new archaeological discoveries in Saudi Arabia, Yemen, Jordan and the Gulf are producing much valuable evidence. Over the past century a vast body of epigraphical material – some 50,000 north and south Arabian inscriptions and the inscribed sticks now emerging by the hundreds in northern Yemen – has provided a wealth of information on the societies of the peninsula, especially the bedouins.1 But all this seldom provides a coherent picture of the course of events, as opposed to vignettes and bare details, and thus does not replace a literary historical tradition. There are external epigraphic records of the Arabs and Arabia, and historical sources – especially in Greek and Syriac – are often helpful.2 But this information too is profoundly discontinuous, and in any case represents the perspective of outsiders who regarded the Arabs as barbarian marauders and most of Arabia as a menacing wasteland.3 There is voluminous material on the subject in the Arabic sources, but herein lies the problem.4 The relevant accounts include a vast bulk of poetry and are frequently attributed to the pre-Islamic period, or are presented as describing events and conditions of that time; but – apart from the Koran – the sources containing these accounts date from at least two centuries later. In times past it seemed reasonable simply to compare the various accounts to determine which seemed most likely to be true. More recently, however, it has become clear that the Arabic sources on the Arabs in pre-Islamic Arabia – and indeed, on the first century of Islamic history as a whole – represent a fluid corpus that adopted a range of argumentative views on issues important at the time when the accounts were being transmitted and * I would like to thank Fidelity and William Lancaster and Michael Macdonald for their valuable comments and suggestions. 1 See, e.g., Robin (1991); Macdonald (1995a). 2 Papathomopoulos (1984); Segal (1984); MacAdam (1989). 3 On the distorted image of bedouins among settled folk: Shaw (1982–3). 4 Two still valuable overviews are Olinder (1927), pp. 11–19; Caskel (1927–30).

173

174

the earlier empire

the sources compiled; the result was the colouring and reshaping of much early and possibly genuine material and the creation of many new accounts.5 Most importantly, pre-Islamic Arabia played an important role in early Islamic preaching of the Word. In explaining the success of Islam and the Arab conquerors, scholars and commentators interpreted Islam’s emergence from Arabia as part of God’s divine plan.6 This involved presenting the pre-Islamic Arabs as naive barbarians – ragged ignorant nomads and eaters of snakes and lizards – and Arabia as a quintessential wasteland. This was in sharp contrast with the powerful, sophisticated peoples of the empires to the north and the richness and fertility of their lands: clearly, Arab victories against such formidable foes could only have been won with God’s permission and as part of his plan for mankind.7 This paradigm manifestly proclaims a religious truth, and while it may at various points correspond to historical reality, it does not spring from that reality. In each case, then, we must judge – often on insecure grounds – the extent to which the motifs and stereotypes of this story of the spreading of the Word have affected our sources.8 the arabs in l ate antiquit y 9 Extant references to ‘Arabs’ begin in the ninth century bc,10 and in ensuing centuries attest their presence in Arabia, Syria and Iraq, and their interaction with the peoples of adjacent lands. This interaction was encouraged in part by the Roman and Persian policy of using Arab groupings to protect their desert flanks and to perform military functions as confederates and auxiliaries. In Syria, an Arab presence was prominent all along the fringe between the desert and the sown,11 and inscriptions and literary sources confirm that many Arabs took up settled life in rural villages.12 The hinterlands of inland Syrian cities were partly populated by Arabs, and major cities such as Damascus and Aleppo had significant Arab populations. In such situations Arabs certainly knew Greek or Syriac – possibly both – and 5 Ahlwardt (1872); Husayn (1927), pp. 171–86; Caskel (1930); Blach` ere (1952–66), I, pp. 85–127, 166– 86; Birkeland (1956); Arafat (1958); Caskel (1966), I, pp. 1–71 (with the review in Henninger (1966)); Crone (1987), pp. 203–30. 6 See the discussion in Conrad (2002). 7 Conrad (1987b), pp. 39–40 and n. 46; Conrad (1998), p. 238. 8 The gravity of the source-critical problems is stressed in Whittow (1999), a detailed critique of the volumes on Byzantium and the Arabs by Irfan Shahid (specifically Shahid (1995)) which, though full of valuable information, pose serious problems and need always to be used with caution. See also Shahid (2000). 9 For recent research on this, see Hoyland (2001); Rets¨ o (2003). 10 Eph‘al (1982), pp. 75–7; Macdonald (1995a). 11 Dussaud (1955), pp. 51–161; Mayerson (1963); Sartre (1982). 12 MacAdam (1983); Millar (1993b), pp. 428–36.

2c. the arabs to the time of the prophet

175

perhaps as their first languages.13 Arabs were also to be found throughout the pastoral steppe lands of northern Mesopotamia, where monks in the Jacobite and Nestorian monasteries occasionally mention them.14 In Iraq there were large groupings of Arabs; settled Arabs lived as both peasants and townsmen along the western fringes, and al-Hira, the focus of Arab sedentary life in the area, was deemed an Arab town (see map 9). Most were converts to Christianity, many spoke Aramaic and Persian, and they were largely assimilated into Sasanian culture.15 The sources referring to the Arabs describe them in various ways. In Greek and Syriac they were most usually called Sarak¯enoi and .tayy¯ay¯e, terms which refer to their tribal origin or to their character as travellers to the inner desert.16 In Arabic, interestingly enough, the terms ‘arab and its plural a‘r¯ab are generally used to refer to tribal nomads. Although the settled folk of Arabia shared much in common with the nomads, they nevertheless drew a sharp distinction between themselves and the bedouins; and rightly so, for a tribesman is not necessarily a nomad. It is true that by the sixth century the Arabic language had spread through most of Arabia – if not so much in the south – and engendered a common oral culture based largely on poetry of often exceptional quality.17 But in none of this should one see evidence of a supposed archetype for Arab unity in any ethnic, geographical or political sense. The basis for Arab social organisation was the tribe.18 Genealogical studies in early Islamic times were already elaborating the lineages and interrelationships of the tribes in great detail. The Arabs comprised two great groupings, northern and southern; the former were traced to an eponymous founder named ‘Adnan and the latter to a similar figure called Qahtan, and both were further divided into smaller sections and sub-groupings. Ancient Arab history is routinely presented in the sources as determined by these tribal considerations,19 but modern anthropology has cast doubt on this and has raised the question of whether such a thing as a ‘tribe’ even exists. While the term is problematic, it seems excessive to resolve a conceptual difficulty by denying the existence of its object.20 The notion of the tribe, however ambiguous, has always been important in traditional Arab society; 13 14 15 16

Nau (1933), pp. 19–24; Trimingham (1979), pp. 116–24; Shahid (1989), pp. 134–45. Nau (1933), pp. 15–18, 24–6; Charles (1936), pp. 64–70; Trimingham (1979), pp. 145–58. Charles (1936), pp. 55–61; Morony (1984), pp. 214–23. Macdonald (1995b), pp. 95–6. Other views: Christides (1972); Graf and O’Connor (1977); O’Connor (1986). 17 F¨ uck (1950), pp. 1–28; Blach`ere (1952–66), I, pp. 66–82; Gabrieli (1959b); von Grunebaum (1963). 18 See Caskel (1962); and for modern parallels: Musil (1928), pp. 44–60; Jabbur (1995), pp. 261–8, 286–306. 19 Caskel (1966), I, pp. 1–71. 20 Inter alios, Schneider (1984). Discussion in Crone (1986), pp. 48–55; Crone (1993), pp. 354–63; Tapper (1990), pp. 60–4.

176

the earlier empire

in pre-Islamic Arabia there can be no doubt that kinship determined social organisation.21 The problem can perhaps best be formulated as revolving around the questions of how far back this was meaningfully traced, and how stable perceptions of kinship were. Individuals were very often aware of their primordial tribal affiliations, and took pride in the achievements, glories and victories of their ancestors. Similarly, personal enemies often vilified the individual by calling into question his tribe as a whole. In practice, however, the vast tribal coalitions rarely acted as a unified whole, and the socially meaningful unit was the small tenting or village group tracing its origins back four or five generations at most. The perception of common descent was not unimportant to the cohesion of such groups, but even more vital were considerations of common interest. In order to maintain itself, the group had to be able to defend its pasturing grounds, water supplies and other resources from intruders, and its members from injury or harm from outsiders. Dramatic changes in kinship affiliations could occur when, for example, the requirements of contemporary alliances or client relationships dictated a reformulation of historical genealogical affinities.22 Such shifts could even occur at the level of the great tribal confederations,23 and were facilitated by the fact that no loss of personal or legal autonomy was involved – a ‘client’ tribe was not in the state of subservience implied by the western sense of the term.24 Through most of Arabia, the welfare of the individual was secured by customary law and the ability of his kin or patron to protect him. If a member of a group were molested or killed, this dishonoured the group as a whole and required either retaliation or compensation. Individuals thus adhered to at least the minimum standards required to remain a member of their group, since an outcast could be killed with impunity.25 This system provided security and guaranteed the status of tradition and custom.26 Violence in the form of warfare, feuding and raiding did occur, but the last of these has given rise to much confusion, and its scope and scale have often been exaggerated:27 there was no glory in raiding a weak tribe or ravaging a defenceless village, and fatalities on either side posed the immediate risk 21 Even with respect to Arabs from south Arabia, where Dostal’s hypotheses (1984) would lead us to expect social organisation along other lines. Note that in all three of the early Arab urban foundations in Egypt and Iraq, the Arab conquerors – even Yemeni contingents – organised themselves according to tribe. See Pellat (1953), pp. 22–34; Djait (1986), pp. 73–135; Kubiak (1987), pp. 58–75. 22 Ibn Khaldun, Muqaddimah, ed. Quatrem` ere, I, p. 238; tr. Rosenthal, I, pp. 266–7. 23 Goldziher (1967–71), I, pp. 92, 96; Caskel (1953), pp. 8, 15; Caskel (1966), I, pp. 31–2, 43–4; II, pp. 22–3, 72, 448; Lancaster (1997), pp. 16–23, 32–4, 151–7. See also Gellner (1973). 24 Lancaster (1997), pp. ix, 73, 128–9. 25 Musil (1928), pp. 426–70, 489–503; Far` es (1932), pp. 44–101; Chelhod (1971), pp. 231–341; Stewart (1994), pp. 130–44. 26 See Stewart (1994), pp. 139–43. 27 Most notoriously in Lammens (1928), pp. 181–236; see also Meeker (1979), pp. 111–50.

2c. the arabs to the time of the prophet

177

of a blood feud. Prowess in battle was without doubt a highly esteemed virtue, and Arabian society was imbued with a martial spirit that elevated the raid (ghazw) to the level of an institution.28 Still, this usually involved one powerful tribe raiding another for their animals,29 and the violence involved was limited by considerations of honour, by the ordinarily small size of raiding parties, and – where weaker groups were concerned – by networks of formal arrangements for protection. Headship of a tribal unit was vested in a sheikh (‘chief’ or ‘elder’, although other terms were also used), but the powers of this office were seriously limited, and the sheikh remained in power as long as the tribe felt this was to their benefit. He was expected to lead the tribe, protect its prerogatives and interests, mediate among its members and with other tribes, and serve as an exponent of muruwwa, an ethic of masculine virtue bound up in such traits as courage, strength, wisdom, generosity and leadership.30 While the chief had no power to enforce his decisions, it was not in the group’s interest to maintain a leader in power and yet regularly defy his decisions. The sheikh led by example and by exercise of a quality of shrewd opportunistic forbearance (h.ilm): he was a mouthpiece of group consensus whose reputation required assent to his judgement.31 The exception to all this was the south, where plentiful rainfall, carried by monsoon winds, allowed for levels of agriculture, population and sedentary development not possible elsewhere. The numerous small towns of the region thrived on the spice trade and enjoyed the stability of a highly developed agrarian economy with extensive terrace farming and irrigation. The towns were closely spaced settlements of tall tower-dwellings, often with a distinct ‘centre’, and their organisation tended to promote commercial and professional bonds at the expense of large-scale kinship ties. Out of this stability there arose a number of coherent regimes with identifiable political centres: Ma‘in, Saba’, Qataban and Hadramawt, based respectively at Qarnaw, Ma’rib, Tamna‘ and Shabwa. The most dynamic of these was Saba’, which by the third and fourth centuries had managed to annex the territories of all the others. The early south Arabian entities were ruled by figures called ‘federators’ (mukarribs). It has long been held that this office was hereditary and had a distinctly religious function, but this now seems unlikely.32 Not unexpectedly, social differentiation reached levels unknown in lands to the north. The sedentary tribes were led by powerful chieftains known as qayls, and at the other end of the spectrum both serfdom and slavery were wellestablished institutions. Nomads were held in check by granting them lands 28 29 31

Musil (1928), pp. 504–661; Jabbur (1995), pp. 348–55; Lancaster (1997), pp. 140–5. 30 Goldziher (1967–71), I, pp. 11–44. Sweet (1965), pp. 1138–41. 32 Robin (1991), pp. 52, 55. See Pellat (1962–3); Pellat (1973); Lancaster (1997), pp. 87–9.

178

the earlier empire

in exchange for military services, thus rendering them dependent upon the regime. arabian religious tradit ions The social organisation of pre-Islamic Arabia was closely bound up with considerations of religion, and it is in this area that problems of methodology and source criticism are most acute. Issues such as borrowing from more advanced civilisations, the starting-points and relative antiquity of religious forms, the roles of animism and totemism, and differences between sedentary and nomadic peoples have been and remain highly controversial. In many cases important arguments involve value judgements about nomads and, similarly, supposed distinctions between ‘high’ and ‘low’ forms of religious expression. There is also the problem that the Arabic sources, where the vast bulk of our source material is to be found, can hardly be said to offer an objective view of pre-Islamic religion. The folly of idolworship and the credulity of its adherents are routinely stressed in stereotyped ways. One tale describes how a tribe fashioned an idol out of dried curd mixed with dates and clarified butter (h.ays) and worshipped it for a time, but eventually devoured it during a famine, leading to a poet’s wry comment: The tribe of Hanifa ate their lord When dearth and hunger swept the land, Fearing naught for consequences From their lord’s avenging hand.33

Inspired by Koranic criticisms,34 Arabic sources also present bedouins as indifferent to matters of faith.35 Arabian polytheism took several forms,36 one of which was stoneworship. Greek and Syriac sources presented this as adoration of lifeless rocks, but such objects were not deities in themselves, but their dwellingplaces or the focus of the rituals of the cult. Offerings were made at the site, and ritual observances included circumambulation of the stone. The best-known example is of course the Ka‘ba in Mecca, but we are told that other places had such cultic foci.37 These foci were often surrounded by a sacred territory, usually called h.aram in the north and h.awa in the south. These were precincts associated with the sanctity of worship and sacrifice; 33 34

Ibn Qutayba, Ma‘arif, p. 621. Koran, Surat al-Tawba, IX.90, tr. Arberry, pp. 189–90 (on procrastinators, liars, malingerers); IX.97–8, 101, tr. Arberry, pp. 190–1 (on hypocrites, stubborn in unbelief, opportunists); Surat al-Fath, XLVIII.11, tr. Arberry, p. 532 (on dissemblers, malicious, corrupt); Surat al-Hujurat XLIX.14, tr. Arberry, p. 538 (on superficial in belief ). 35 Bashear (1997), pp. 7–14. 36 Arafat (1968). 37 This is made especially clear in Lughdah al-Isfahani, Bilad al-‘arab, p. 32.

2c. the arabs to the time of the prophet

179

violence and killing, including hunting, were forbidden there. Holy men were in charge of these precincts, and their descendants enjoyed special religious esteem.38 Also prominent was religious observance revolving around idols – again, with the idol probably representing the deity being worshipped. The names of many idols are known from ancient poetry and from later prose works drawing on this verse. Important new details pertaining to Yathrib (Medina) may be indicative of a more general pattern: there, clans each had an idol in a room belonging to the whole clan, where the idol was venerated and sacrifices made to it. People also had wooden idols in their homes, making similar observances at that level. To offend the idol was an offence against the honour of the head of the house and a matter for retaliation, and there is some evidence that these idols were intended to be figures of ancestors. There was thus a hierarchy of idols, corresponding to the social status of their owners.39 There is good evidence of star-worship and astral divinities as well. The widely venerated al-Lat (a sky goddess) and al-‘Uzza (possibly the morning star) may have been representations of Venus, and Byzantine polemics against Islam claim that the Islamic slogan All¯ahu akbar (‘God is great’) has as its origin a cry of devotion in astral religion.40 The worship of astral divinities has also been connected with the veneration of idols. The attitude of the ancient Arabs towards their gods was entirely empirical and pragmatic. Although they did consider problems of human existence and the meaning of life,41 they did not look to their deities for the answers. They regarded their gods as the ultimate sources of worldly phenomena beyond human control, such as disease, rain, fertility, and personal and communal adversities of various kinds; they worshipped the gods in expectation of their assistance, but they did not revere them or consider that they owed unwavering commitment to them.42 Monotheistic religion was also known in Arabia from an early date. The influx of Jews into Arabia is difficult to trace, but probably had much to do with the failure of the Jewish revolt and the destruction of the Temple in ad 70, and the gradual spread of Christianity over the next three centuries. In south Arabia, Judaism enjoyed considerable success in the fifth and early sixth centuries, and to the north there were various important Jewish communities, notably at Yathrib. Judaism seems to have had deep and powerful roots there, judging from reports that in pre-Islamic times the Jews there had three times as many fortified compounds (qus.u¯ r) as all the 38 40 41

39 Lecker (1993). Serjeant (1962). On this see Rotter (1993); Hoyland (1997), pp. 105–7. For example, the ephemeral joys of youth and the ultimate fate of either death or senility: Zuhayr, Sharh Diwan, p. 29; al-‘Askari, al-‘Awa’il, I, p. 57. 42 Wellhausen (1897), pp. 213–14; Crone (1987), pp. 237–41.

180

the earlier empire

other non-Jewish clans combined,43 and that in the latter half of the sixth century the Jewish clans of Qurayza and al-Nadir collected taxes from the other tribes.44 The question of Jewish influences in Arabia and on Islam has become highly sensitive in modern scholarship, but there can be no doubt that such influences were profoundly important; the Koran itself contains many tales and accounts of Jewish origin, as also do early Islamic religious lore and scholarship.45 The Christianisation of the Roman empire in the fourth century opened the way for the large-scale spread of the faith along and beyond the empire’s frontiers, including Arabia.46 Along the Syrian desert fringe from the Red Sea to the Euphrates, it spread to the Arab tribes via monasteries and wandering missionaries, primarily monophysite. In some cases, as with the Banu Taghlib and Ghassanids, entire tribes converted; some tribal settlements such as al-Jabiya and Jasim, south of Damascus, also became ecclesiastical centres. These tribes were familiar with at least basic observances, yet remained completely within Arab tribal culture as well.47 Along the Iraqi frontier the spread of Christianity was somewhat slower, perhaps because a network of Nestorian monasteries in the area took longer to appear than had been the case among the monophysites.48 Still, the Lakhmid base of al-Hira was the seat of a bishopric by 410.49 Further south, there were major Christian communities at such centres as Najran and Sanaa, and small ecclesiastical outposts along the Arabian coast of the Persian Gulf. Specifically monophysite or Nestorian forms of Christianity were practised in such centres, but elsewhere the Arab tribesman’s main contact with the faith was through individual monks and hermits, and there confessional boundaries may have been less sharply drawn.50 Two other beliefs – which were influenced by Judaism and Christianity, yet remained distinct from both – revolved around a ‘high god’ and around h.an¯ıf¯ıya. Little can be said about belief in a ‘high god’ in ancient Arabia, apart from the fact that, as elsewhere in the Middle East,51 some held that a god called Allah had a certain dignity and status above the other deities of 43 44

Ibn al-Najjar, Al-Durrah al-thaminah, II, p. 325; Conrad (1981), p. 22. Ibn Khurradadhbih, al-Masalik, ed. and French tr. de Goeje, p. 128 (text), p. 98 (tr.); Yaqut, Mu‘jam al-buldan, IV, p. 460; Kister (1968), pp. 145–7. 45 On the Jews of pre-Islamic Arabia, see Newby (1988), pp. 14–77; and on influences, Geiger (1833); Rosenthal (1961), pp. 3–46; Nagel (1967); Rubin, U. (1995), pp. 32, 217–25. 46 For an overview, see Charles (1936); Trimingham (1979). 47 For a valuable anthology of the verse of early Arab Christian poets, see Cheikho, Shu‘ara’ alNasraniya; also Conrad (1994), pp. 30, 31, 51. 48 Brock (1982). 49 Synodicon orientale, ed. and French tr. Chabot, p. 36. 50 On the Koranic evidence, see Ahrens (1930); Michaud (1960); Parrinder (1965); Bowman (1967); Robinson (1991). The relevant Koranic verses, with the commentaries from many tafs¯ırs, are assembled in Abu Wandi et al. (1996). 51 Teixidor (1977), pp. 17, 161–2.

2c. the arabs to the time of the prophet

181

Map 9 Pre-Islamic Arabia and its northern neighbours

the Arabian pantheon and was extolled as a god to whom one could turn in case of particular need.52 On h.anfya there is more information.53 The Koran makes it the religion of Abraham and associates it, on the one hand, with belief in a single God and, on the other, with rejection of idolatry and repudiation of worship of the sun, moon and stars. In particular, and most importantly, h.an¯ıf¯ıya reflects not the pragmatic attitude towards religion described above – in which the god(s) were worshipped in expectation 52 53

Watt (1979); Welch (1979); Rubin, U. (1984). For differing interpretations, see Gibb (1962); Rubin, U. (1990); Rippin (1991).

182

the earlier empire

of help with worldly needs beyond an individual’s control – but rather a submissive devotion to and faith in God for his own sake. Nevertheless h.anfya is distinct from Judaism and Christianity: in several passages of the Koran, its adherent (a h.anf ) is equated with a Muslim, and in one variant to the Koranic text, h.anfya replaces Islam as the ‘true religion’.54 Other sources suggest that there were h.anfs in various parts of Arabia, that the movement was one of individuals rather than religious communities, and that Mecca was important to its adherents. Other details are less reliable, and there is no evidence to link h.anfya with south Arabian inscriptions attesting to worship of a god called al-Rahman, ‘the Merciful’, one of the Islamic names for God. But the fact that the tradition on the h.anfs makes some of them doubters or enemies of Muhammad suggests that it should not be dismissed entirely as later prophetic annunciation or the tidying up of a pagan past. Of interest in this respect is the testimony of Sozomen, who died before 448; writing from the vantage-point of Gaza in southern Palestine, he offered the following comments on Arab religion: It seems that the Saracens were descended from Ishmael, son of Abraham, and hence were originally called Ishmaelites. Their mother Hagar was a slave, so in order to hide the shame of their origin they took the name of Saracens, pretending to be descended from Sarah, the wife of Abraham. As such is their descent, they practise circumcision like the Jews, abstain from eating pork, and adhere to numerous other Jewish observances and practices. In so far as they in any sense diverge from the observances of that people, this arises from the passage of time and their contacts with other neighbouring peoples . . . It seems likely that with the passage of time their ancient customs fell into disuse as they gradually took to observing the customs of other peoples. Eventually, when some of their tribe came into contact with the Jews, they learned from them the facts of their true origin and returned to observance of Hebrew custom and law. In fact, even at the present time there are some of them who live their lives in accordance with the Jewish law.55

The connection with Judaism may reflect an inclination to associate false belief with the machinations of Jews.56 As to the Abrahamic religion attested in the text, while the connection is circumstantial and Sozomen wrote long before the testimony of the Koran, the Islamic scripture may refer to continuing monotheistic trends in Arabia that it wishes to distance from earlier monotheistic faiths now viewed as rivals. 54 55

Jeffery, Text of the Qur’an, p. 32; Koran, Surat al-‘Imran, III.19, tr. Arberry, p. 47. Sozomen, Church history, VI.38.10–14, ed. Bidez and Hansen, p. 299. See Cook (1983), p. 81; Millar (1993a), pp. 42–4. 56 On scheming Jews as a clich´ e, see Schafer (1997).

2c. the arabs to the time of the prophet

183

economic life in arabia It is difficult to generalise on the notion of an Arabian ‘economy’, since the internal economic situation in the peninsula varied from place to place and depended on whether a community was settled or nomadic. As noted above, the south had a lively village economy based on terraced farming and irrigation; but even here, production was primarily limited to foodstuffs and use-value goods. South Arabian spices and incense were much soughtafter items for centuries, and undoubtedly fortunes were made from trade in them,57 but overland trade in such goods appears to have collapsed by the first or second century ad.58 In the rest of the peninsula the economy was far more rudimentary. The interior of the peninsula consists of various types of steppe lands where lack of water makes major cultivation unsustainable in most years. Reliable water supplies come from wells and oasis springs, and it was around these that Arabia’s towns developed. The date palm dominated agriculture in many places, and this and other crops were often cultivated in large walled gardens (h.aw¯a’i.t) scattered over whatever patches of arable land there were in or around a settlement. Goats and sheep were kept, and items produced for sale included hides and leather, wool, cloth, dairy products, raisins, dates, wine, and utensils and weapons of various kinds. Gold and silver were mined, but often figured as a replacement for currency rather than as an export item; perfume was produced, especially in Aden and Najran, but beyond the Arabian and Syrian markets it could not compete with the cheaper products of Byzantine centres such as Alexandria.59 Arabian traders in late antiquity were thus known to their neighbours – in Palestine, for example – as bearers not of costly luxury items, but rather of animals, wool, hides, oil and grains.60 Bedouins, on the other hand, were largely herders and pastoralists, though members of many tribes settled for varying periods of time and others engaged in opportunistic agriculture – for example by sowing on a fertile watered plot on their way somewhere else, and then reaping when they returned. Tenting groups travelled in recognised tribal territories, their schedules and movements (and willingness to encroach on the lands of other tribes) largely dictated by the needs of their animals. Those who lived along the desert fringes tended sheep and goats, as well as the singlehumped dromedary camel; groups venturing into the depths of the Arabian 57 58 59 60

Groom (1981). The last reference to it is in the Periplus Maris Erythraei, ch. 27, ed. and tr. Casson, pp. 66–7. Dunlop (1957), pp. 37–40; Crone (1987), pp. 87–97. Krauss (1916), pp. 335–6; Excavations at Nessana, ed. Kraemer, pp. 251–60 (no. 89). The Palestinian church at Dayr al-‘Adasa, dedicated in 621, has a mosaic floor bearing various rural scenes, including one of a caravan of camels carrying oil or wine jars: Balty (1989), pp. 149–51.

184

the earlier empire

steppe lands did best with camels, but on occasion are known to have taken goats and sheep as well. For barter or sale, nomads could offer such animal products as hides, leather, wool and dairy products. The symbiosis between village-dwellers and nomads was important to the whole economic structure of Arabia. Leather, for example, was an extremely important product and was the plastic of its day; everything was made from it, from buckets to items of clothing, and agriculture could not have been maintained without huge supplies of leather for ropes, irrigation equipment, harness and so forth. Apart from often quite complex exchanges of goods and services, bedouins played a major role in economic development. There is evidence, for example, that parts of different tribes concluded share-cropping agreements and worked together to promote and protect agriculture.61 Certain villages also specialised in serving the needs of nomads, and oases and springs where herds could be watered attracted settlements that thrived on trade with the nomads. Relations were further dictated by the need of settled merchants to move their goods through lands controlled by nomads, and hence to remain on good terms with the tribes.62 Arabian domestic trade thus consisted of caravans of camels organised by settled merchants and protected and guided by bedouins who controlled the lands through which the caravans passed. Seasonal fairs were often held, especially around religious shrines, and security at such important times was guaranteed by the declaration of sacred periods during which no raiding or fighting was to occur.63 The goods being traded were for the most part not costly items, but rather the basic goods and commodities that people needed to live. This in turn limited the distance and duration that the caravans could travel, since the longer the journey, the more expensive the goods would be at their destination;64 that is, the longer the contemplated journey was in both distance and time, the more precious the goods being carried would have to be in order to generate sufficient income to make the journey economically feasible. The internal trade of Arabia thus seems to have involved the transport of goods on short or medium-length journeys, and it is probably this factor that accounts for the proliferation of market centres. The sources present a picture of lively markets dotting the steppe landscape of the peninsula; wells, springs and small villages were all attractive sites for established market activities, though the scale of such 61 al-Bakri, Mu‘jam ma sta‘jam, I, pp. 77–8. See also Kister (1979), p. 70 on similar arrangements at the time of the Prophet. The same system is still widespread today. 62 See Simon (1989), pp. 78–86; Morony (1984), pp. 218–19; Donner (1989), pp. 77–8; and for modern examples, Jabbur (1995), pp. 1–2, 5–8, 32–8, 250. See also Nelson (1973). 63 Wellhausen (1897), pp. 84–94; Brunschvig (1976a), pp. 113–18; Crone (1987), pp. 87–108. 64 Jones (1955), p. 164; Hendy (1985), pp. 556–7; Crone (1987), p. 7. Not all trade was profit-driven, however: Villiers (1940).

2c. the arabs to the time of the prophet

185

operations was probably small.65 In some cases, commerce was encouraged by banning private land ownership within the market precinct, thus preventing dominance by a few successful merchants, and suspending taxes and fees on traders and visitors.66

i m p e r i u m and imperial polit ics It will be seen from the above that there was little in Arabia to attract the attention of the great powers of late antiquity, and at first it was only Arabia’s role athwart the route to the east that lent it any importance to them. This factor alone was sufficient to make Arabia a focus of imperial manoeuvring and power politics, but trade operated in conjunction with other factors as well. The spread of Christianity and to a lesser extent Judaism in Arabia reflects the interest of external powers from an early date. In fact, it was the great triad of politics, trade and religion that determined the course of events there from late antiquity onwards, with trade providing an imperial momentum later transferred to the other two factors. All around the peripheries of Arabia the impact of imperium was being felt. Behind the Roman presence advancing in the north came Roman roads, way-stations and forts, reflecting an increasing interest in control of what lay beyond. Far more vigorous, however, were the inroads by the Sasanians: they had a more immediate stake in Arabia, with their capital at Ctesiphon, the rich agricultural alluvium of Iraq, and the Persian Gulf trade to consider. Settlements were founded up and down the Gulf, and Oman was annexed by Shapur I (240–70). In the fourth century, Arab raids provoked a punitive expedition that reached as far as the Hijaz. Discovery of silver and copper in the Najd led to the foundation of a Sasanian outpost at Shamam.67 Several factors exacerbated the rivalry between the two imperial powers. The establishment under Constantine of a Christian empire based at Constantinople made competition with Persia more immediate and provided yet another arena for intrigue and dispute. But more important by far was the evolution of the rival polities themselves. From largely decentralised and culturally diverse empires, tolerant of a broad range of contradictory ideologies and traditions, both developed into world powers; they used political, economic and military strength to pursue imperial aims that were justified by elitist ideologies, spurred by aspirations to universal dominion, 65 Lughdah al-Isfahani, Bilad al-‘arab, for example pp. 224, 227, 243, 333–4, 335, 345, 358, 361, 397; Muhammad ibn Habib, al-Muhabbar, pp. 263–8; al-Marzuqi, al-Azminah, pp. 161–70. See also alAfghani (1960); Hammur (1979). 66 Kister (1965); Dostal (1979); Lecker (1986). 67 al-Tabari, Ta’rikh, ed. de Goeje et al., I, pp. 838–9; German tr. N¨ oldeke, pp. 54–7; tr. Bosworth, V, pp. 54–5; Dunlop (1957), p. 40; Crone (1987), p. 46.

186

the earlier empire

and increasingly dictated from the capital. The Byzantine and Sasanian empires competed for control of western Asia and adopted more global strategies in efforts to promote their own interests and undermine those of their rival.68 Thus, while the rise of Christianity led to the collapse of the market for the incense consumed so massively and ostentatiously by pagan Rome,69 the demise of this formerly crucial aspect of the eastern trade was more than replaced by new rivalries of unprecedented intensity. The new level of conflict generated by escalating competition between the two great powers manifested itself in several ways where Arabia and the Arabs were concerned. Firstly a pronounced religious element was introduced into the struggle, primarily in the southern part of the peninsula and surrounding lands. Monophysite missionary activity70 led to the conversion of Ethiopia to Christianity in the fourth century and the spread of the faith in Yemen and elsewhere in south Arabia. The Christian presence noted frequently in the Koran was probably the result of commercial contacts with Syria. The Sasanians, on the other hand, supported the spread of the rival confession of the Nestorians and also encouraged the Himyarites, a predominantly Jewish regime which ruled most of south Arabia and had influence elsewhere. Religious rivalries played an instrumental role in an Ethiopian invasion of Yemen in about 518 and shortly thereafter in a Himyarite civil war between Christian and Jewish factions. This struggle led to a persecution of Christians in south Arabia under the last Himyarite ruler Dhu Nuwas, culminating in the 520s with the massacre of the Christians of Najran. Ethiopia responded with a second invasion, killing Dhu Nuwas and once again installing a puppet regime in Yemen. The power of the Ethiopian governor, however, was soon usurped by a certain Abraha, who established himself as the paramount authority in the south; the Meccans viewed his expedition of 552 as directed against themselves, but it was in fact a move against tribal forces to the east. Secondly, external forces gradually encircled and penetrated the peninsula. The Sasanians established trading posts beyond the Straits of Hormuz as far as Aden and in the sixth century occupied Yemen. Persian authority extended as far as Yathrib, where taxes collected by the Jewish tribes of Qurayza and al-Nadir were sent on in part to a Persian ‘governor of the desert’ (marzub¯an al-b¯adiya).71 Byzantium, on the other hand, still had trade through Clysma and Ayla to protect,72 and sought a sea route to the 68 Fowden (1993), pp. 24–36, 80–137, though the focus on monotheism and the stress on premeditated planning from the centre seem overstated. See also Crone (1987), p. 47; above pp. 135–7. 69 M¨ uller (1978), pp. 733–64; Groom (1981), p. 162; Crone (1987), p. 27. 70 Altheim and Stiehl (1971–3), I, pp. 393–431; Shahid (ed. and tr.), Martyrs of Najran, pp. 252–60. See below, p. 308. 71 Christensen (1944), pp. 373–4; Altheim and Stiehl (1957), pp. 149–50; Whitehouse and Williamson (1973); Frye (1983a). 72 See, for example, Jerusalem pilgrims, tr. Wilkinson, pp. 147–8.

2c. the arabs to the time of the prophet

187

east that would not be subject to Persian taxes and interference. It thus tried to extend its influence down the Red Sea and battled against pirates and adventurers to maintain control of ports and customs stations; epigraphical evidence places Byzantine forces nearly a thousand kilometres south of Damascus in the mid-sixth century.73 It also used its new Christian ally, Ethiopia, to pursue its economic interests and intervene militarily in the affairs of the south, encouraging the Himyarites to attack Persian interests.74 Thirdly, both powers used tribal allies in Arabia to further their own interests, protect their Arabian frontier zones, and confront the tribal forces of the other side. Such a tactic was not new. Rome and Persia had routinely used tribal auxiliaries in various capacities,75 and in the late fifth and early sixth centuries the Himyarites in Yemen coopted the great north Arabian tribal confederation of Kinda into acting in their interest and controlling caravan traffic along the routes from Yemen to Syria and Iraq. Kinda eventually extended its control across central Arabia, as well as part of the Hijaz and areas along the Persian Gulf coast, and in the early sixth century it was attacking both Byzantine and Sasanian targets along the desert fringes of Syria and Iraq. Seeking to avoid further incursions and to gain a strong tribal ally against forces acting for the Sasanians, the Byzantines reached an understanding with the confederation and on several occasions sent embassies to promote good relations. Kinda thus became an ally of Byzantium; turning against the Sasanians, it gained considerable authority in the hinterlands of south-western Iraq and even occupied al-Hira for a time.76 However, its primary sponsors remained the Himyarites in Yemen, and as this regime declined, so did the fortunes of Kinda. The Sasanians’ main tribal ally was the Lakhmids, a tribe that had established itself in north-eastern Arabia by the fourth century and founded a stable base at al-Hira. There had been contacts and relations between the two sides in the past, but the combination of deteriorating relations with Byzantium and the spectre of powerful Kinda forces allied to Byzantium and positioned within easy striking distance of Ctesiphon and the agricultural plains of Iraq led the Sasanians to support and encourage the Lakhmids with renewed vigour. The latter had long been subordinate to Kinda, and double marriages between them had been arranged at least twice in the past. Nevertheless, by about 504 the new Lakhmid chieftain, al-Mundhir III (504–54), was able to rid himself of Kinda suzerainty and launch operations against the confederation with a well-organised army.77 73 75

74 Smith (1954), p. 427. See Abel (1938); Seyrig (1941); Simon (1989), p. 34. On Rome, see Shahid (1984), pp. 52–63; and on the fifth-century Salihids in particular, see Shahid (1958); Shahid (1989). 76 Olinder (1927), pp. 32–93; Simon (1989), pp. 42–6; Lecker (1994); Shahid (1995), I, pp. 148–60. 77 Rothstein (1899), pp. 134–8; Altheim and Stiehl (1957), pp. 117–23; Kister (1968), pp. 165–7.

188

the earlier empire

Fighting over the next two decades ended with the utter disintegration of Kinda and the extension of Lakhmid authority over their rival’s former clients among the Arab tribes. By the 540s the Lakhmids held sway over many of the tribes of central Arabia and over towns as far west as Mecca.78 Byzantium was thus forced to turn to other Arab clients for the protection of its position and interests. Its choice fell on the Ghassanids, a south Arabian tribe closely related to Kinda, that had migrated to northern Arabia and Syria in the fifth century and established itself as the pre-eminent power on the desert fringe there. The Ghassanids were a more nomadic group than the Lakhmids; although they were often associated with the camping-ground called al-Jabiya 65 kilometres south-west of Damascus, they had no real fixed centre comparable to that of the Lakhmids at alHira. Their influence was not as broad-ranging as that of the Lakhmids, and although they had trading connections with Iraq through Nisibis and Dara, their control over the relevant routes was tenuous. Nevertheless, Byzantium granted the Ghassanid sheikh the title of phylarch and showered him with honours, privileges and money. In return, it was expected that the chieftain would keep his own tribe under control and protect imperial interests from other tribes as well.79 The Ghassanids and Lakhmids, confronting one another across the Syrian desert, were thus drawn into the series of great Byzantine–Persian wars that began in 502 and ended with a decisive Byzantine victory in 628 (see above, pp. 119–20, 124–7, 135–6). Significant fighting between them began in the 520s and continued sporadically for sixty years, with dire consequences for the agricultural infrastructure of both Syria and Iraq. Several observers describe the destruction in Syria,80 and whatever survived the passage of raiding parties and military expeditions was exposed to the brigands and outlaws hovering around such forces.81 This military conflict tends to overshadow other developments in which the two sides were variously involved. The Ghassanids were responsible for the establishment of several small towns in the hinterlands south of Damascus and perhaps also for some of the so-called ‘desert palaces’ of the Syrian steppe.82 Sponsors of monophysite Christianity, they also erected 78 Rothstein (1899); Simon (1967); Simon (1989), pp. 27–30, 42–6, 55–8, 149–52; ‘Abd al-Ghani (1993), pp. 11–23. 79 N¨ oldeke (1887a); Simon (1989), pp. 27–32, 55–8; Sartre (1982); Peters (1984). On the term phylarch, which originated as a post in the provincial administration, not necessarily relating to nomads, see Macdonald (1993), pp. 368–77. 80 John Moschus, Pratum spirituale, chs. 99, 133, 155, cols. 2957–8, 2995–8, 3023–4; Life of John the Almsgiver, ch. 9, ed. Delehaye, pp. 23–4; tr. Dawes and Baynes, pp. 203–4; al-Tabari, Ta’rikh, ed. de Goeje et al., I, p. 1007; German tr. N¨oldeke, p. 299 and n. 4; tr. Bosworth, V, pp. 326–8; al-Washsha’, ‘Kitab al-fadil’, fol. 105r. See also Foss (1975); Foss (1977b), pp. 68–71; Schick (1995), pp. 25, 31–3. 81 Abu al-Baqa’, al-Manaqib, I, pp. 105–6. Early Islamic works on jihad also mention the problems posed by these elements. 82 Gaube (1984).

2c. the arabs to the time of the prophet

189

numerous churches and monasteries. In Iraq, al-Hira grew from a camp (which is what the name means in Arabic) into a lively Arab town, noted for its churches and monasteries, impressive residential compounds and taverns. Persian Gulf shipping could sail up the Euphrates as far as al-Hira, and Lakhmid income included proceeds not only from raids but also from agricultural rents and produce, trade, and taxes from tribes they controlled. There also seems to have been a nascent literary tradition emerging there.83 Both sides, especially the Lakhmids, were also major patrons of Arab oral culture, and some of the most important poets of pre-Islamic times gained generous support from Ghassanid or Lakhmid sheikhs.84 The history of the Arab client regimes is important, but they were not central in the imperial planning of either Byzantium or Persia, in which they figured mainly as threats that had to be countered.85 Little is known from the Lakhmid and Persian side, but Byzantine emperors, political strategists and historians such as Procopius certainly held the Ghassanids in low esteem. The Byzantines had little faith in the abilities, motives or intentions of their Arab allies. The treaty of 561, for example, expresses dissatisfaction with Saracen adherence to treaty terms in the past, comes close to calling them smugglers and traitors, and warns of harsh punishment for lawbreakers.86 When Ghassanid phylarchs refused to adhere to Chalcedonian orthodoxy, they were exiled. Byzantium made overtures to the Lakhmids when it was expedient, and the lack of trust and commitment worked both ways: the capture of Dara by Khusro I probably involved some negotiations with the Ghassanid phylarch al-Mundhir (569–82).87 Neither side survived the manoeuvrings of their patrons or the broader conflict which engulfed the Middle East in the sixth century. In 581 alMundhir was arrested by Emperor Tiberius I (578–82) and exiled to Sicily in a religious dispute, and in 584 his son and successor al-Nu‘man joined him. The Ghassanid phylarchate rapidly fell apart, fragmented by Emperor Maurice (582–602) into a host of smaller entities and riven with dissension and conflict over the deposition of two leaders within four years. Forces from the tribe are mentioned in accounts of the Arab conquest of Syria, but not in a leading role.88 The Lakhmids survived a while longer, but during the reign of Shah Khusro II (591–628) they were displaced in favour of a similarly decentralised system. The Sasanians also promoted the position 83 84

Much valuable material is collected in ‘Abd al-Ghani (1993), pp. 25–138. Nicholson (1907), pp. 38–54; Blach`ere (1952–66), II, pp. 293–356; ‘Abd al-Ghani (1993), pp. 365–

469. 85 86 87

Important discussion in Whitby, Michael (1992). Men., 6.1, pp. 70–3. See also above, p. 124. Whitby, Michael (1988), pp. 257–8. The Nemara inscription of ad 328 has Arabs in the eastern Hawran in contact with both the Romans and the Persians: Bowersock (1983), pp. 138–47; Bellamy (1985). 88 N¨ oldeke (1887a), pp. 33–45; Shahid (1995), I, pp. 455–71, 634–41, 648–51.

190

the earlier empire

of the Banu Hanifa, who roamed in the desert on their southern flank.89 Later, when a force of Persian troops and Arab auxiliaries sought to quell a desert revolt in about 610, their army was beaten at Dhu Qar; this was the first time the tribes had been able to defeat the Sasanians in battle.90 It also illustrates how the demise of the Arab client regimes marked not the shift from one system of frontier defence to another, but rather the opening of a great power vacuum extending from the desert fringes of Syria and Iraq all the way to central Arabia. Inhabitants of the peninsula remembered that they had once been ‘trapped on top of a rock between the two lions, Persia and Byzantium’.91 But as the next decade was to reveal, those days were gone forever and the Persian setback at Dhu Qar was but a hint of things to come. m ecca, muhammad and the rise of isl am In about 55292 a boy named Muhammad bin ‘Abd Allah was born into a minor clan of the tribe of Quraysh, which was settled in and around the shrine centre of Mecca in the Hijaz, about 900 kilometres south of Syria. A trader by profession, he participated in the caravan trade of Arabia and visited Syria on several occasions. In about 610 he began to preach a monotheistic faith called ‘submission to God’, or Islam, and summoned his fellow Meccans to prepare for the Last Judgement. By 622 difficulties in Mecca and the erosion of vital support had reached the point where he was obliged to move to Yathrib, 300 kilometres to the north. This migration (the hijra)93 proved to be of crucial importance: for in Yathrib, henceforth called Medina,94 the ranks of his followers increased dramatically. Raids on enemy caravans, camps and villages met with success and further expanded his support. Muhammad returned to Mecca in triumph in 630, and by the time of his death two years later his authority extended over much of Arabia. The rest was brought under control by the first caliph, Abu Bakr (632–4), and Muslim forces went on to campaigns of conquest that, in less than a century, created an empire extending from Spain to Central Asia. How all this occurred and why it focused on Muhammad, Mecca and the late sixth century are questions that early Muslims took up themselves,95 and they are a major concern of modern historical research. In the 1950s William Montgomery Watt proposed a socio-economic solution. Mecca 89 90

al-A‘sha, Diwan, pp. 72–87, no. 13, esp. 86 vv. 47–9; al-Isfahani, al-Aghani, XVII, pp. 318–22. al-Tabari, Ta’rikh, ed. de Goeje et al., I, pp. 1029–31; German tr. N¨oldeke, pp. 332–5; tr. Bosworth, V, pp. 358–61; Rothstein (1899), pp. 120–3. 91 Qatada (died 735) in al-Tabari, Tafsir, ed. ‘al-Ghumrawi et al., IX, p. 145; Koran, Surat al-Anfal, VIII.26, tr. Arberry, p. 172. See also Kister (1968), pp. 143–4. 92 For the date, see Conrad (1987a). 93 Crone (1994). 94 al-Madina, probably referring not to ‘the city’, but to the Prophet’s house. 95 But not immediately: Donner (1998), pp. 75–85.

2c. the arabs to the time of the prophet

191

was a major centre for overland caravan trade, and its merchants and others grew wealthy on the profits from commerce in such precious items as incense, spices, gemstones and gold. This widened the gap between rich and poor and led to social malaise as crass materialism eroded traditional values. Muhammad’s message was essentially a response to this crisis.96 More recently, however, serious challenges have been made to the notions of a lucrative Arabian trade in luxury items, of Mecca as an important entrepot, and hence of some serious crisis provoking a religious response.97 Mecca is not mentioned in any non-Arabic source of the pre-Islamic period, and does not lie on the main communication routes in western Arabia. The site itself is barren, inhospitable and incapable of sustaining agriculture for more than a minuscule population. Even had there been a lucrative international trade passing through the Hijaz in the sixth century, it would not have found an attractive or logical stopping-point at Mecca, which owed its success to its status as a shrine and pilgrimage centre. As at certain other shrines in Arabia, pilgrims came to circumambulate a rock – in this case associated with an unroofed building called the Ka‘ba – and to perform religious rituals with strong affinities to those of Judaism: these included offerings and animal sacrifice, washing and concern for ritual purity, prayer and recitation of fixed liturgies.98 There are indications that, early on, few people were resident at the site: ‘People would perform the pilgrimage and then disperse, leaving Mecca empty with no one living in it.’99 The success and expansion of Mecca were due to the administrative and political skills of its keepers, the tribe of Quraysh. The Ka‘ba seems to have been a shrine of the god Hubal,100 but in the religiously pluralistic milieu of pagan Arabia it must not have been difficult to promote it as a place where other deities could be worshipped, too. A greater achievement was convincing other tribes to honour the sanctity of the h.aram of Mecca and to suspend raiding during the sacred months when pilgrims came. As agriculture was not possible at Mecca, Quraysh had to bring in food from elsewhere and so was at the mercy of nearby tribes in any case. The very fact that Mecca survived, much less prospered, thus reflects the diplomatic skills of Quraysh. The Islamic tradition, of course, makes much of the a priori importance of Quraysh, but this is surely something that emerged within the paradigm of a sedentary tribe seeking to protect and promote its 96 Watt (1953), pp. 1–29 and in numerous publications of his thereafter. See the review by Bousquet (1954). 97 Simon (1989); Peters (1988); Crone (1987). See review of Crone in Serjeant (1990) and reply in Crone (1992). 98 Hawting (1982); Rubin, U. (1986). 99 al-Bakri, Mu‘jam ma sta‘jam, I, p. 89, citing al-Kalbi (died 763). 100 Wellhausen (1897), pp. 75–6; Crone (1987), pp. 187–95.

192

the earlier empire

interests through skilful manipulation of relations with the nomadic tribes around it. There was mutual advantage in the prosperity of Mecca: trade with pilgrims, import and marketing of foodstuffs and other necessities, and collection and distribution of taxes levied in kind for feeding and watering pilgrims.101 It may even be that Quraysh was able to organise a profitable trade with Syria, perhaps as a result of disruption to the agricultural productivity of the Levant caused by the destruction of the Persian wars, numerous droughts in Syria,102 and the repeated visitations of bubonic plague after 541.103 The message that Muhammad preached in the milieu of a prosperous Mecca was in many ways a familiar one, and in others quite a novelty.104 His summons to the worship of one God recalled the notion of a ‘high god’, and his identification of Islam as the religion of Abraham had important associations with the doctrines of h.anfya. As can be seen from the testimony of Sozomen, his call for the restoration of a pristine faith, free from the corruptions that had crept into it, was already a time-honoured tradition in Arabia. The observances he advocated were also well known from either pagan Arabian or Jewish practice: prayer and Friday worship, fasting, pilgrimage, ritual purity, almsgiving, circumcision and dietary laws.105 Where Muhammad broke with tradition was in his insistence on absolute monotheism and his advocacy of a relationship with God that abandoned traditional pragmatic views of religion and summoned man to unconditional commitment and faith in response to God’s creative munificence and continuing solicitude. The rejection of pagan eclecticism, however, threatened the entire social and economic position of Quraysh and thus earned him the enmity of their leaders. Among the public at large his message – with its corollaries of reward and punishment in the hereafter – seemed extreme and delusory and evoked little positive response.106 In order to gain support Muhammad had to prove that his God was a winner, and this he achieved by moving to Medina, where he used his expanding following to disrupt Meccan commerce and food supplies.107 His military success made him a force to be reckoned with: the tribal arrangements so carefully nurtured by Mecca over the years soon fell apart in the face of this challenge, while the victories of the new religion provided the worldly success which Arabs demanded of their gods and also appealed to the Arabs’ warrior ethic. Islam also had a broad appeal on other grounds. The Koran presented itself as a universal scripture ‘in clear Arabic speech’,108 and thus took advantage 101 102 104 105 106 108

For example, Ibn Hisham, Sirat Rasul Allah, ed. W¨ustenfeld, I.1, p. 83; tr. Guillaume, pp. 55–6. 103 Conrad (1994); Conrad (1996b). Butzer (1957), p. 362. Cook (1983), pp. 25–60. Goitein (1966), pp. 73–125; Bashear (1984), pp. 441–514; Rippin (2005), pp. 103–17. 107 Discussion in Donner (1977). See Izutsu (2002), pp. 45–54. Koran, Surat al-Nahl, XVI.103, tr. Arberry, p. 270; Surat al-Shu‘ara’, XXVI.195, tr. Arberry, p. 379. See also Surat Ibrahim, XIV.4, tr. Arberry, p. 246.

2c. the arabs to the time of the prophet

193

of the position of the Arabic language as the common cultural tongue of Arabia and a basis for common action.109 Arabs could also identify with one another, despite their tribal distinctions, on the basis of a shared participation in Arabian tribal organisation and custom, a heritage of similar cultural and religious experience in pagan systems and folklore, and a long history of trade and commerce, revolving around fairs and religious shrines, that engendered a certain feeling of familiarity around the peninsula. It has often been asserted that the Arab conquests were of essentially Islamic inspiration. The Islamic tradition of spreading the Word sees things this way, and the Armenian chronicle, written in the seventh century and attributed to Bishop Sebeos, also has Muhammad urging his followers to advance and claim the land promised to them by God as the descendants of Abraham.110 It therefore seems probable that there was a religious agenda to the conquests from the start, and it is certainly true that without the unifying factor of Islam there would probably have been no conquest at all. But the arguments of leaders and advocates are one thing, and the response of the fighters themselves is another. Even in Mecca and Medina the teachings of Muhammad and the text of the Koran were still known in only fragmentary fashion, and it is difficult to see how most tribesmen elsewhere could have had more than a vague and trivial knowledge of either so soon after the Prophet’s death. Many warriors who joined the conquest forces had only recently fought against the Prophet himself, or had resisted the efforts of the first two caliphs to bring Arabia under their control. It is also implausible that tribal warriors all over Arabia could so quickly have abandoned the pragmatic and worldly attitude towards religion that had prevailed for centuries, in favour of one that expected genuine commitment to the one God. There is, in fact, good evidence on the conquests showing that this was not the case at all.111 This is not to detract from the centrality of the message of Islam to Muhammad’s own sense of mission and purpose, and probably to that of others around him. One may also concede that Islam enabled the Muslim leadership to mobilise warriors in a way that transcended important differences, and it is likely that Islamic slogans and admonitions of various kinds were often inspiring to fighters on the ground. But if the faith played 109 Blach` ere (1952–66), II, pp. 230–41; Blach`ere (1956); von Grunebaum (1963); Bashear (1997), pp. 54–5. 110 Seb., ch. 135, tr. Thomson and Howard-Johnston, I, pp. 95–6. See also the quotations from Dionysios of Tell-Mahre in MS, XI.2, ed. and French tr. Chabot, II, pp. 403–5 (tr.); IV, pp. 404–8 (text); Chronicon ad 1234, ed. Chabot, I, pp. 227–30 (text); I, pp. 178–80 (tr.); tr. in Syrian chronicles, tr. Palmer et al., pp. 129–32. Discussion in Crone and Cook (1977), pp. 8–10; Hoyland (1997), pp. 124–30. 111 For example al-Walid ibn Muslim (died 810) in Ibn ‘Asakir, Dimashq, ed. al-Munajjid, I, pp. 461–2; al-Tabari, Ta’rikh, ed. de Goeje et al., I, p. 2922; tr. Humphreys, XV, pp. 125–6; al-Maqrizi, al-Mawa‘iz, I, p. 75.

194

the earlier empire

an important role in uniting and mobilising the tribes, it was nevertheless waves of tribal forces, motivated primarily by traditional tribal ambitions and goals, that broke over Syria, Iraq and Egypt from the 630s on. It is unlikely that either Syria or Iraq could have withstood the advance of forces of this kind, given the state of their defences after the end of the last Persian war in 628, only six years before the first Arab advance. The Arab armies were not simply marshalled in Medina and then sent forth with the caliph’s instructions; providing food, fodder and water for an army of thousands of men and animals would have been extremely difficult. The norm was rather for small contingents to expand as other groups gradually joined them on the march; the sources make clear that commanders were expected to engage in such recruiting along the way, to ensure that the newcomers were armed and equipped, and to ‘keep each tribe distinct from the others and in its proper place’.112 In this way a small force could soon swell to thousands as warriors joined its ranks in expectation of adventure, fighting and plunder. The situation was made more difficult by the fact that confronting the Arabs on this scale posed entirely new military problems. Both imperial powers were accustomed to dealing with Arabs as bands of raiders, and had planned their frontier defences accordingly. Watch-towers and forts, many of them abandoned for centuries in any case, were inadequate to deter the forces that now swept past them, and whereas the old Roman system had anticipated incursions by single uncoordinated bands, it was now confronted by penetration at many points simultaneously. It was probably also difficult to determine exactly where the enemy was at any given time, for when battle was not imminent an Arab army tended to fragment into bands of warriors roaming the countryside. Finally, and as the above example shows, Arab strategy was often highly reactive and thus difficult to counter or predict. Incursions into Iraq, for example, seem to have begun when drought in Arabia obliged the tribe of Rabi‘a, of the Banu Shayban, to migrate into Iraqi territory, where the Sasanian authorities permitted them to graze their herds on the promise of good behaviour. But the presence of these tribal elements eventually led to friction, which the Rabi‘a quite naturally interpreted as unwarranted reneging on an agreed arrangement. When they called on their kinsmen elsewhere for support, the crisis quickly escalated into full-scale conflict between Arab and Persian forces.113 It is difficult to guess whether either of the great powers would have been able to stem the military momentum that was building in Arabia, even had they correctly gauged the threat it posed. With Kinda, the Ghassanids 112 113

Ibn ‘Asakir, Dimashq, ed. al-Munajjid et al., I, p. 446. Ibn A‘tham al-Kufi, al-Futuh, I, pp. 88–9.

2c. the arabs to the time of the prophet

195

and the Lakhmids all in a state of either collapse or disarray, the growing strategic power of Islam was able to develop in what otherwise amounted to a political void; the real source of the danger confronting the empires was effectively beyond their reach from the beginning. Byzantium and Persia could fight armies that violated their frontiers, but could not stop the process that was generating these armies in the first place. Initial victories over the Arabs at Mu’tah in Syria in 632 and the battle of the Bridge in Iraq in 634 thus proved no deterrent, as in earlier times would have been the case.114 What overwhelmed the Byzantines and Sasanians was thus the ability of the message and charismatic personality of Muhammad to mobilise the tribal might of Arabia at a level of unity never experienced among the Arabs either before or since. Unprepared for defence on the scale required to counter this new threat and unable to marshal tribal allies of their own to strike at their foe in his own heartlands, both were forced to fight deep within their own territories and suffered defeats that simply encouraged further incursions on a larger scale. Greek and Persian field armies were crushed in one disastrous battle after another, leaving cities to endure sieges without hope of relief and encouraging resistance everywhere else to evaporate in short order.115 114 115

Donner (1981), pp. 105–11, 190–202; Kaegi (1992), pp. 71–4, 79–83. Donner (1981), pp. 119–220; Kaegi (1992), pp. 88–180.

CHAPTER 3

WESTERN APPROACHES (500–600) john moorhe ad

the continuing unit y of the post -roman world Throughout the political history of western Europe, there have been few periods of such dramatic change as the fifth century. In 400 the borders of the Roman empire in the west, by then distinct from the eastern empire which was governed from Constantinople, stood reasonably firm. They encompassed all of Europe south of the Antonine wall in Britain and the Rhine and Danube rivers on the continent, extending eastwards of the Danube’s confluence with the Drava; they also included a band of territory along the African coast, stretching two-thirds of the way from the Straits of Gibraltar to the Nile. But within a hundred years this mighty entity had ceased to exist. North Africa had come under the power of groups known as Vandals and Alans; Spain of Visigoths and Suevi; and Gaul of Visigoths, Franks and Burgundians. The Romans had withdrawn from Britain early in the century, leaving it exposed to attacks from the Irish, Picts and Anglo-Saxons, while in Italy the last emperor, Romulus Augustulus, was deposed in 476 by a general, Odovacer. The supplanter of Romulus was himself deposed and murdered in 493 by Theoderic the Ostrogoth (493–526), who established a powerful kingdom based on Italy. While the empire had weathered the storms of the fifth century largely unscathed in the east, in the west it had simply ceased to exist. Western Europe, one might be excused for thinking, had moved decisively into a post-Roman period, and the middle ages had begun. However dramatic these events may have been, they did not constitute a definitive parting of the ways between the west and what we may now call the Byzantine east. Long-distance trade continued throughout the Mediterranean and beyond, as research on African pots found across a wide area is increasingly making clear.1 Consuls were being appointed for the west in the year 500 and when, a few decades later, the western consulship lapsed, some in the west still dated documents with reference to the eastern consuls who continued to be appointed. The Mediterranean 1

See Loseby (2005), pp. 621–3. See also Ward-Perkins (2005), pp. 87–106.

196

3. western approaches ( 500–600)

197

Antonine wall

Rhine

Danu

FRANCIA Lo i r

be

PANNONIA

e

Drava

Poitiers Rhone

Milan Pavia

Viviers

Verona Po

Faenza

Ravenna

DA

LM

Sirmium ILLYRIA AT

Toulouse Rome Capua Benevento Cumae Nocera Naples

CORSICA

SARDINIA BALEARIC ISLANDS

Seville Asidona Straits of Gibraltar

Pillars of Hercules

TYRRHENIAN S EA SICILY

Cartagena

Malaga

Septem

IA

Thamugadi

Syracuse

Carthage Decimum

(Timgad)

Caput Vada

Cartagena Regained by Byzantium in the mid-sixth century 0 0

250

500 250

750

1000 km

500 miles

Map 10 Lands of the empire in the west in the sixth century

was traversed by members of the intelligentsia and diplomats, such as a legate of Theoderic who made twenty-five trips from Italy to Spain, Gaul, Africa and Constantinople. The west was awash with doctors from the east, among them Anthimius, who lived in Italy and wrote a fascinating book on diet for a Frankish king in which he recommended the use of such foods as leavened bread, beer and mead made with plenty of honey. Another eastern doctor was Alexander of Tralles, brother of the well-known architect Anthemius. Alexander practised medicine in Rome and his Therapeutica

198

the earlier empire

was translated into Latin in the sixth century.2 On the other hand Priscian, who was probably an African, was in Constantinople when he wrote what were to become standard works on Latin grammar;3 we know that Africans in Constantinople were renowned for their Latin accent, but reviled for their poor Greek. Latin manuscripts were copied in Constantinople and Greek ones in Ravenna, the Gothic capital in Italy. Furthermore, despite the political changes in the west, the new rulers there were keen to represent themselves as in some way subservient to the Roman emperors who still ruled in Constantinople. Theoderic the Ostrogoth wrote to Emperor Anastasius that ‘our kingdom is an imitation of yours . . . a copy of the only empire’ and Sigismund the Burgundian informed him that, while he gave the appearance of ruling his people, he believed himself to be merely the soldier of the emperor.4 In these and many other respects, the post-Roman west remained firmly part of the Roman world. the successor states in the west Nevertheless, there had been changes, and seen from Constantinople the political situation of the west in 500 cannot have given cause for joy. Developments in the west had not passed unnoticed by the emperors of the east, amidst the other internal and external problems besetting them in the fifth century. The last decade of the western empire had seen the despatch of new emperors and armies from the east, and Romulus Augustulus’ deposition in 476 was recorded by sixth-century Byzantine authors in terms which suggest they saw it as marking a major change: according to the Chronicle of Marcellinus, Rome was founded 709 years before Octavian Augustus held power, and he died 522 years before it perished in 476.5 Constantinople became a centre for refugees who fled the new kingdoms in the west. African catholics were prominent among these, including a widely reported group who miraculously found themselves able to speak after King Huneric had ordered that their tongues be cut out. There were also people from Italy who were said, early in the sixth century, to have received a warm welcome at the court of Anastasius (491–518), and in one of his works Priscian expressed the hope that Rome and Constantinople would both come to be under the emperor.6 Indeed, emperors who had traditionally had pretensions to rule over the whole known world could not have looked with complaisance on the loss of the western provinces, which constituted the greater part of the territory over which their predecessors had ruled. 2 4

3 Priscian, Grammatici Latini, ed. Keil. Alexander of Tralles, Therapeutica, ed. Puschmann. Theoderic, in Cassiodorus, Variae, I.1.3, ed. Fridh and Halporn, p. 9; Sigismund, in Avitus of Vienne, Epistulae 93, ed. Peiper, pp. 100–1; tr. Shanzer and Wood, pp. 146–7. 5 Marcellinus, Chronicle, ed. Mommsen, p. 91; tr. Croke, p. 27. 6 Priscian, De laude Anastasii imperatoris, lines 242–7, 265, ed. and tr. Coyne, pp. 48–9, 60–1.

3. western approaches ( 500–600)

199

When the Byzantines looked westwards they saw a world dominated by the Mediterranean, and by the year 500 almost all of its coastline formerly within the western empire was under the control of three barbarian kingdoms. Vandals had occupied the bulk of Roman Africa where they proved stern rulers, whose expropriation of the landowning class and persecution of catholics made them unpopular. Using their powerful navy, the Vandals sacked Rome in 455 and withstood major Byzantine attacks in 460 and 468. They faced two other barbarian kingdoms on the opposite shores of the Mediterranean. The Visigoths, originally settled as Roman foederati around Toulouse, had gradually gained control of most of Gaul south of the Loire and begun moving into Spain, while Italy and some adjacent lands were under the control of Ostrogoths.7 They had made their way there in accordance with an agreement concluded with Emperor Zeno (474–91). Although in 497 Zeno’s successor, Anastasius, returned to Italy the palace ornaments, which Odovacer had sent to Constantinople after deposing Romulus Augustulus, this degree of recognition does not imply that the Byzantines were happy to accept the Ostrogothic state. The Vandals, Visigoths and Ostrogoths had far more in common than possessing adjacent kingdoms around the Mediterranean. They were all Arian Christians, adherents of a heresy which denied that the Father and the Son were of one substance as taught by the council of Nicaea (325); this marked them off from both the Byzantines and the great mass of the people among whom they settled. The Byzantines regarded them as speaking a single language and looking the same; together with the Gepids, they were viewed from Constantinople as nations distinguishable in name only.8 They were connected by a system of marriage alliances: one of Theoderic’s daughters had married the Visigothic king Alaric and his sister married the Vandal king Thrasamund, a web of relationships which may have been anti-Byzantine in purpose. Of these three states, that of the Ostrogoths was by far the most dangerous. To the east it included Dalmatia, which gave it a border with the empire hundreds of kilometres long: even if the Ostrogoth ruler had no expansionist designs in the east, he was well placed to influence developments there in turbulent times. So it was that a Byzantine rebel had already sought the aid of the Italian-based Odovacer in 486, a circumstance which may have helped prompt the emperor’s despatch of Theoderic and his Ostrogoths to Italy to discipline Odovacer shortly afterwards; and when, towards the end of Anastasius’ reign, the magister militum Vitalian rebelled against the emperor, he was believed to have sought Theoderic’s 7 On the Vandals, see Courtois (1955); on the Goths, Wolfram (1988); Heather (1991); on the kingdom of Toulouse, Barbero and Loring (2005), pp. 167–71, 174. 8 Pr W, III.2.1–6, ed. and tr. Dewing, II, pp. 8–11.

200

the earlier empire

assistance. Some decades earlier, before moving into Italy, Theoderic had intervened when a rebellion threatened to unseat Zeno, for which the grateful emperor rewarded him with a consulship. An early sixth-century Italian author, apparently referring to these events, spoke of Theoderic as having bestowed the diadem on Zeno and compelled his love, with the implication that he was superior to the emperor.9 It was a perspective unlikely to have been shared in Constantinople. If this were not enough, in 504 one of Theoderic’s generals gained control of Sirmium, a city in Pannonia formerly part of the eastern empire. The Ostrogoths not only kept the city; they went on to advance further into imperial territory. Following a decisive defeat of the Visigoths at the hands of the Franks in 507, Theoderic ruled part of their kingdom as well as that of the Ostrogoths. Constantinople had reason to look with fear on the mighty Ostrogothic kingdom, in particular, from among the states that had emerged around the western Mediterranean. These, however, were not the only successor states to the empire in the west. To the north were territories that had come under the control of other peoples, notably Franks and Burgundians.10 Like the Goths, they had found homes within the borders of the old empire, and they had been integrated into the system of alliances set up by Theoderic; he himself had married the sister of Clovis, king of the Franks (c. 481–c. 511), and one of his daughters had married Sigismund, heir to the Burgundian throne. But around the end of the fifth century Clovis had adopted catholicism, and whatever his motives may have been in taking this step, it is clear that he saw himself as accepting the religion of the emperor. Catholic influence was also strong at the Burgundian court, where Sigismund was converted. More importantly, from a Constantinopolitan perspective the impact of the Frankish and Burgundian intruders on the Roman world would have seemed less than that of the Goths and Vandals: their capacity to harm imperial interests was slight, and with judicious encouragement they could be made to serve imperial policy. According to a strange story told in a seventh-century text, the Frankish king Childeric (c. 463–82) went to Constantinople, where he sought the emperor’s agreement that he should go to Gaul as the emperor’s servant.11 Hence it is not surprising that when conflict broke out in 507 between the Franks under Clovis, who enjoyed Burgundian support, and the Visigoths and Ostrogoths, Anastasius 9 Ennodius, Panegyricus, ch. 14, ed. Vogel, pp. 211–12; ed. and German tr. Rohr, pp. 246–9. On the interpretation, see MacCormack (1981), p. 230. 10 The Byzantines distinguished the Franks and Burgundians from the Goths by calling them ‘Germans’, a shorthand way of indicating that they had come from the lands east of the Rhine, which the Romans had failed to conquer. 11 Fredegar, Chronicle, III.11, ed. Krusch, pp. 95–7. The story gains in plausibility if one takes the name of the emperor which is supplied, Maurice, to have been a slip for Marcian.

3. western approaches ( 500–600)

201

intervened on behalf of the Franks. He dispatched a fleet which ravaged part of the coast of Italy and prevented Theoderic from intervening in Gaul as early as he would have wished; he also made Clovis an honorary consul. It is clear that Constantinople viewed the west in a differentiated way. The Mediterranean lands were occupied by powers that threatened Byzantine interests, but the empire could sometimes act to destabilise its enemies. Theoderic’s last years were clouded by accusations that a group of Roman senators was treasonably corresponding with the emperor, and by his possible over-reaction to reports that Arians were being persecuted in the east. These two issues were recurrent in the history of the Gothic and Vandal states. The Vandal king Huneric had been concerned at the possibility that catholic clergy were sending letters overseas – presumably to the empire – about the succession to the throne; and on one occasion Theoderic acted to stop correspondence from Burgundy reaching the emperor. The Vandals also felt that religious persecution was a tool that could be employed for reasons of diplomacy. The position of the emperor vis-`a-vis catholics in the west had been strengthened by the healing in 519 of the Acacian schism, which had divided the churches of Rome and Constantinople since 484.12 The last years of Theoderic therefore manifested some of the tensions implicit in the relationship between Constantinople and the successor states to the empire around the western Mediterranean. To the north, on the other hand, were powers from whom good could be expected. It was a basic distinction, and its application became clear during Justinian’s military ventures. the vand al war On 19 May 530, the Vandal king Hilderic was deposed by another member of the royal family, Gelimer. Hilderic had enjoyed close relations with Justinian, who was therefore presented with an excellent opportunity to declare war on the Vandals. The deposition of the emperor’s ally was, however, merely a pretext for intervention. According to the African writer Victor of Tunnuna, Justinian’s decision to invade was prompted by the vision of a martyred African bishop, while a passage in the Codex Iustinianus of 534 – which may well have been written by the emperor himself – is eloquent as to the persecution of catholics by the Vandals. It describes their sufferings in language reminiscent of the account written by the African Victor of Vita in the 480s. We have no reason to doubt that Justinian’s invasion, like so many of his activities early in his reign, was motivated by 12

See above, pp. 106, 114–15.

202

the earlier empire

religion rather than by any ideology of imperial renewal.13 We are told that the plan to invade Africa was opposed by his advisers. But the imperial will was not to be trifled with, especially when a bishop reported a vision in which success was promised. In 532 a peace was concluded with Persia, enabling resources to be directed towards the west. Justinian prepared a force which put to sea at about the summer solstice in 533 under the command of Belisarius, a general who had recently risen to prominence in campaigning against the Persians and in putting down a rebellion in Constantinople. The religious nature of the enterprise was highlighted as the patriarch prayed over Belisarius’ ship and placed on one of the vessels a soldier who had recently been baptised. We can follow the Vandal war in some detail, through the eyewitness account of Belisarius’ legal assistant, Procopius. The arrival of the Byzantine forces in Africa occurred in excellent circumstances: Gelimer, unaware of their approach, had sent part of his forces to Sardinia. The invaders landed unopposed south of Carthage at Caput Vada, whence they proceeded towards the capital. They kept close to the shore for some distance before they turned inland and marched to Decimum, some fifteen kilometres outside Carthage. Here Gelimer met them, but after a short encounter he fled, and two days later, on 15 September 533, the Roman army marched into Carthage. Belisarius dined on food that had been prepared for Gelimer, while his soldiers, behaving with remarkable restraint, are said to have bought food in the market. Gelimer summoned forces from Sardinia, but at the battle of Tricamarium, thirty kilometres outside Carthage, the Vandal army was again turned to flight, and Gelimer took up residence among the Berbers on a mountain where he consoled himself by composing sad verses before surrendering. Having quickly gained control of Sardinia, Corsica, the Balearic Islands and Septem, a fort adjacent to the Straits of Gibraltar, Belisarius returned to Constantinople with booty which included the treasures of the Jews that Titus had taken from Jerusalem to Rome in the first century and which the Vandals in turn had taken to Africa in 455. The victorious general paraded through the streets of Constantinople in triumph, and both he and Gelimer performed proskyn¯esis before Justinian. The defeated king was provided with estates in Galatia in Asia Minor, and Belisarius went on to hold a consulship in 535; the largesse he distributed included spoils won on this campaign. Justinian saw to the making of gold plates that depicted the history of his triumphs and legislated for the return of property the Vandals had taken from its rightful owners. In a matter of months 13 On the martyr’s visions, see Victor of Tunnuna, Chronicle, ed. Mommsen, p. 198; ed. Cardelle de Hartmann, ch. 118, p. 38; ed. and Italian tr. Placanica, pp. 38–9; on the persecution of catholics, see Justinian, Corpus iuris civilis, II, Codex Iustinianus, 1.27.1, ed. Krueger et al., p. 77. See also above p. 109.

3. western approaches ( 500–600)

203

the kingdom of the Vandals that had seemed so strong had collapsed, and Africa found itself governed by a praetorian prefect appointed by the emperor. We have no reason to doubt that its inhabitants approved of these developments. Nevertheless, there was still fighting to be done. The nomadic Berbers had been pressing increasingly on the Vandal kingdom, and they were to pose a major problem to Byzantine Africa, for their practice of lightly armed and mobile combat made them difficult opponents. A series of fortifications was quickly erected to deal with them, of which the impressive ruins at Thamugadi still stand, with walls averaging 2.5 metres in thickness and rising to over fifteen metres in height (see figs. 7a and 7b). Archaeological and literary evidence both indicate that, contrary to Justinian’s expectation, the Byzantines never succeeded in occupying all the territory held in Roman times, but the number and extent of the defences they erected makes it clear they planned to stay in Africa. There were also internal troubles, for many of Belisarius’ soldiers married Vandal women, only to see the property they hoped to gain through their wives threatened by Justinian’s legislation for the return of property held by Vandals. They mutinied in 535, and more seriously in 544, after the magister militum and praetorian prefect Solomon had been killed fighting the Berbers. But the ringleader of the rebels was murdered in 546 and towards the end of that year a new general, the energetic John Troglitas, arrived. An expedition led by him in the spring of 548 was crowned with success, and Africa knew peace. the gothic war: early successes Justinian can only have been delighted at Belisarius’ triumph in 533, and his thoughts naturally turned to a more ambitious project. Imperial legislation of April 535 referred to the recovery of Africa and the imposition of servitude on the Vandals, but added that the emperor now hoped to receive from God things greater than these.14 As it happened, it was a propitious time to intervene in Italy. Following the death of Theoderic in 526, his successors had found it hard to step into his shoes, and both his daughter, Amalasuntha, and the man who came to be her rival, Theoderic’s nephew Theodahad, entered into negotiations with the emperor. In the spring of 535 Amalasuntha was murdered, so providing a casus belli.15 The reason Justinian gave for intervening in Italy was different from that provided for the war in Africa; whereas the Vandals had been attacked for their outrageous treatment of the catholic provincials, the Ostrogoths were assaulted because of the weakness of their claim to hold Italy. They had done well, it was 14 15

Justinian, Corpus iuris civilis, III, Novellae, 8.10.2, ed. Krueger et al., p. 74. See Moorhead (2005), pp. 148–9.

204

the earlier empire

Figure 7a Plan of fort at Timgad (Thamugadi) in North Africa, as an example of careful organisation of defences

3. western approaches ( 500–600)

205

Figure 7b Justinianic fort at Timgad

now asserted, to defeat the tyrant Odovacer, but their proper course should have been then to hand Italy back to the empire, rather than keeping it for themselves. As we have seen, the ending of the line of emperors in the west in 476 had not escaped notice in Constantinople. The initial attack on Italy took place from two directions.16 One army occupied Dalmatia, which thereafter remained under almost unbroken imperial control, while Belisarius, at the head of a small force, easily gained control of Sicily in 535. From there he could launch an attack on the Italian mainland which the resources of the Goths were ill-equipped to deal with, concentrated as they were in the north. Theodahad, by then sole ruler, offered to resign his kingdom, a proposal he subsequently retracted, and early in 536 Pope Agapetus arrived in Constantinople to hold discussions with Justinian on Theodahad’s behalf. But the emperor was in no mood for discussion. A law of 536 refers to the regaining of territory from one ocean to the other, an ambition not hinted at in earlier sources, which indicates that imperial designs had become larger.17 In the same year Belisarius crossed to the Italian mainland. The Goths, discontented at Theodahad’s failure to lead effectively, raised on their shields Witigis, a man of modest family but of proven fighting ability, and Theodahad was 16 The account of Procopius again constitutes a detailed primary source, closely followed by e.g. Bury (1923); Stein (1949–59), II. However, Procopius was probably not in Italy after 540, and as time passed he came to look on the war with less favour: Hannestad (1961). 17 Justinian, Corpus iuris civilis, III, Novellae, 30.11.2, ed. Krueger et al., p. 234.

206

the earlier empire

murdered. The new king left Rome for Ravenna, taking hostages and an oath of loyalty from Pope Silverius, who had succeeded Agapetus, and on 9 or 10 December 536 Belisarius occupied the eternal city. In the following February a large Gothic force arrived and laid siege to it, cutting the aqueducts which supplied the city with water and ravaging Christian burial grounds outside the walls, but to no avail. In March 538 Witigis withdrew. Fighting spread in the north of Italy, and the Byzantines enjoyed the initiative, gaining much territory in 539. The Goths counter-attacked in Liguria and razed the great city of Milan to the ground; we are told that the men were killed and the women handed over to the Burgundians. The Frankish king Theudebert intervened, seeking to benefit no one but himself, and by the end of 539 the Gothic capital, Ravenna, was besieged by imperial forces. In his hour of need Witigis asked Khusro I, the shah of Persia, to break the treaty he had concluded with Justinian in 532 and distract him in the east, a ploy which made the emperor incline towards offering the Goths generous terms.18 But Belisarius was confident, and when the Goths offered to accept him as ‘emperor of the west’, an office obviously prejudicial to Justinian’s position, he feigned consent.19 In May 540 he marched into Ravenna, but refused to honour his agreement with the Goths. Before long he returned to Constantinople, taking with him Witigis and his wife Matasuentha, various Gothic notables and at least part of the Gothic treasure. The reception he received from Justinian was cool, the emperor possibly having been disquieted by the title his general had pretended to be willing to accept. Nevertheless in 540 the mighty state founded by Theoderic had apparently collapsed. The historian Procopius observed that when Belisarius entered Rome in 536, ‘Rome became subject to the Romans again after a space of sixty years’,20 and one gains the impression of a smooth imposition of Byzantine power. In March 537 Pope Silverius, who had owed his appointment to Theodahad and had subsequently sworn loyalty to Witigis, was deposed by Belisarius and replaced by Vigilius, a prot´eg´e of the powerful empress, Theodora. By early 537 Belisarius had appointed one Fidelis praetorian prefect, and by the end of the year a comes sancti patrimonii per Italiam, an official with competence in financial matters, seems to have been functioning in the conquered lands. Fidelis’ tenure of the prefecture would have overlapped with the end of that of Cassiodorus, who had been appointed to the post by the Goths in 533 and whose last letters on behalf of Witigis were written towards the end of 537. By the end of 539 a scribe at Ravenna 18 19

On Perso-Byzantine relations in the sixth century, see above, pp. 104–5, 119–20, 135–6. Pr W, VI.29.17–29, ed. and tr. Dewing, IV, pp. 128–33; but see also Pr W, III.11.18–21, ed. and tr. Dewing, II, pp. 106–7 (misleadingly translated in the Loeb edition). 20 Pr W, V.14.14, ed. and tr. Dewing, III, pp. 146–7.

3. western approaches ( 500–600)

207

employed in a document the formula chi-mu-gamma, in accordance with Byzantine practice.21 As early as 535 there had been signs in Rome of discontent with the Gothic government, and the people of Italy, quickly putting aside positive memories they may have had of the reign of Theoderic, accepted the advent of imperial power. In 540 it must have seemed that the Gothic war, like the Vandal war, had come to a wished-for conclusion. In Constantinople, Justinian had a mosaic placed in the ceiling of the Bronze Gate of the palace, showing Belisarius winning victories for him. In the middle of the composition stood Justinian and Theodora, the kings of the Vandals and Goths approaching them as prisoners, and around them the members of the senate who ‘rejoice and smile as they bestow on the emperor honours equal to those of God, because of the magnitude of his achievements’.22 It was the optimism of a golden moment, such as would never again be possible. the gothic war: the resistance of tot il a As it turned out, the war with the Goths was by no means over. Justinian, perhaps afraid of the threat a mighty general could pose, failed to replace Belisarius, and rivalry and corruption became endemic among the Byzantine commanders left in Italy. They showed little inclination to attend to the Gothic resistance that continued north of the Po, and with the coming to power in 541 of King Totila (or Baduila, as his name was spelt on coins) the Goths gained a leader of outstanding calibre. Totila’s attitude to Justinian was expressed in his coinage, on which the portrait of the current emperor was replaced by that of Anastasius, who had recognised the kingship of Theoderic in 497; if Justinian challenged the Goths on the basis of legitimacy, Totila was prepared to dispute his claim. Before long war was raging again. In the spring of 542 the new Gothic king defeated the imperial army at Faenza and captured its standards, before proceeding to the south and taking Benevento, Cumae and Naples. Belisarius was sent back to Ravenna in 544 to deal with the deteriorating situation, but found himself powerless to stop the Gothic advance. Indeed, his conduct of the war in this period displays an uncharacteristic passivity. This may owe something to a severe outbreak of plague afflicting the empire at the time, with its consequent impact on manpower resources. In December 545 Totila besieged Rome and twelve months later entered it. He immediately visited St Peter’s to pray, an act calculated to suggest continuity with Theoderic, who had himself made devotions at the basilica on his one 21 It probably stood for Christon Maria genna, ‘Mary bore Christ’: Die nichtliterarischen lateinischen Papyri, ed. Tj¨ader, II, p. 60, line 75; p. 259 (commentary). 22 Pr B, I.10.19, ed. and tr. Dewing and Downey, pp. 86–7.

208

the earlier empire

known visit to Rome, and, beyond him, with the emperors whose conduct Theoderic had imitated. But the act was hollow. There were few people left in the city, and Totila made no secret of his animosity towards the senate. In fact, he planned to raze the walls of the city, but Belisarius wrote warning him of the harsh judgement of posterity were he to proceed with this course. Perhaps Belisarius was able to play on the vanity of the Gothic king; in any case, Totila behaved foolishly and abandoned Rome, taking members of the senate as hostages. For forty days the city was home to neither man nor beast, but by April 547 Belisarius had moved in and begun work on restoring its defences. During the spring Totila tried to wrest control of the city from him, but failed. Nevertheless, the Goths were still masters of much of Italy, to the extent that Belisarius tended to travel from one place to another by ship rather than overland, and when Justinian recalled his great general to Constantinople a few years later Belisarius felt much more subdued than he had on his returns in 534 and 540. In 549 an Ostrogothic fleet ravaged the coast of Campania and Rome was again besieged; in the following January it fell. Totila established a mint in the city, held races and, in the words of a contemporary, lived there ‘like a father with his children’.23 With Ravenna still in Byzantine hands, Rome came to hold a political significance to which it had long been unaccustomed. Totila moved to Sicily and ravaged it in 550, whereupon the Franks occupied parts of northern Italy. A full decade after Belisarius had seemed to bring the war to a successful end, the situation in Italy was parlous, and Justinian decided to commit resources on a scale never entrusted to Belisarius. An enormous army was placed under the command of the patricius Germanus. He was an impressive figure, for not only was he a cousin of Justinian but he had married Matasuentha, the granddaughter of Theoderic and former wife of Witigis, a tie which allowed him to anticipate limited resistance from the Goths in Italy. Indeed, the birth of a baby son to the couple allowed the historian Jordanes to be hopeful of a future union of the families of Germanus and Matasuentha.24 But Germanus died while preparations for the expedition were still underway, and in 551 the general Narses was appointed to finish the job. The great army set off overland for Italy in April 552. Franks who had settled in Venetia sought to deny it passage on the grounds that it included a large contingent of Lombards, their traditional enemies. The Goths tried to make the road impassable, but Narses was able to make his way to Ravenna, 23 24

LP, LXI.107, ed. Duchesne, I, p. 298; tr. Davis, I, p. 60. Jordanes, Romana et Getica, ch. 314, ed. Mommsen, p. 138; tr. Mierow, p. 141; Momigliano (1955) provides a rich but inconclusive discussion.

3. western approaches ( 500–600)

209

occupying it on 6 June 552. Totila marched out of Rome, and at the end of June or beginning of July the two forces encountered each other at Busta Gallorum, a site in the Apennines.25 Before the troops of both armies Totila performed a stylish war dance on his charger, but the Goths were heavily outnumbered, and the outcome of the battle was inevitable. The Gothic cavalry could not withstand the enemy archers, and both cavalry and infantry fled, Totila dying of a wound received in flight. Numerous Gothic strongholds surrendered as Narses advanced on Rome, which his enemies were no longer strong enough to defend effectively. The city was easily captured and its keys despatched to Justinian. In their despair the Goths put to death senators they found and 300 children they were holding as hostages, but their cause was now hopeless, and the Franks refused to intervene on their behalf. In October a Gothic force did battle with Narses in the south of Italy at Mons Lactarius, near Nocera, but it was defeated, and Narses gave the surviving Goths permission to return to ‘their own land’. Some continued to resist on a local basis until the capture of Verona in 562 or 563, but by the time Narses was recalled, probably not long after the accession of the emperor Justin II (565–78), Italy seemed stable. The Gothic war had lasted far longer than the Vandal war, but its outcome was the same. A puzzling feature of the Gothic war is the failure of the Visigoths to become involved. For much of the war their king was an Ostrogoth, Theudis (531–48), and at one stage his nephew, Ildibad, was prominent in the resistance in Italy, but we have no reason to believe that help from the Visigoths reached Italy. We do know, however, that around 544 a Visigothic force was defeated at Septem, across the Straits of Gibraltar, which suggests an attempted thrust from Spain into what was by then Byzantine Africa. But in 552 a Byzantine force, purporting to answer an appeal for help from a Visigothic rebel, set out for Spain and succeeded in gaining control of a slice of its south-east coast around Cartagena and Malaga. The area has a mountainous hinterland and looks across the sea to Africa, and the defence of Africa may have been the true reason for Byzantine involvement in Spain.26 In any case, this modest success was the culmination of an extraordinary expansion of Byzantine power in the west. Within a few decades Africa and Italy, together with the large islands of the western Mediterranean, Dalmatia and part of Spain had been reintegrated into the empire, so that the poet Agathias could legitimately claim that a traveller could go as far as the sandy shore of Spain where the Pillars of Hercules lay and still tread imperial territory.27

25 27

26 See also Barbero and Loring (2005), p. 184. For a detailed account, see Roisl (1981). Anthologia graeca, IV.3, lines 83–7, ed. and tr. Paton, I, pp. 120–1.

210

the earlier empire

Figure 8a Justinian, Bishop Maximianus, clergy, officials and soldiers, mosaic from San Vitale, Ravenna; Justinian is, symbolically, centrepiece of the group

con stantinople and the west in t he mid-sixth century We may take the years on either side of 550 as constituting a high water mark of Byzantine influence in the west. Economic links between east and west were strengthened; the export of African pottery to the east, which had declined during the Vandal period, seems to have grown during the early period of Byzantine rule. Byzantine relations with the west were particularly in evidence in Ravenna, the capital of Italy, where Bishop Maximianus obtained from Justinian the title of archbishop and relics of St Andrew, a saint whose cult could be seen as constituting a possible rival to that of St Peter in Rome. It is possible that Maximianus’ splendid ivory throne, now to be seen in the Museo Arcivescovile in Ravenna, was made in Constantinople, and it was he who consecrated the church of San Vitale, with its glowing mosaics of Justinian and Theodora. Justinian failed to visit the west, but no one could doubt that the mosaics of San Vitale, whatever the precise liturgical significance of the scenes they portray, were powerful statements of imperial power in the conquered territories.

3. western approaches ( 500–600)

211

Figure 8b Theodora and her courtiers, mosaic from San Vitale, Ravenna

Strange as it may seem, the clearest sign of the centrality of Byzantium in western affairs in the mid-sixth century is to be seen in Constantinople itself and in the variety of westerners, the influential, the ambitious and the captive, who were there. Liberius, whom Theoderic had successively appointed praetorian prefect of Italy and praetorian prefect of Gaul, had defected while on an embassy to Constantinople shortly before the Gothic war. He later participated in Byzantine campaigns in Italy and Spain, and returned to Italy, where he was buried at Rimini. During the war, and in particular after Totila’s capture of Rome in 546, many Roman aristocrats made their way to the royal city. These included Cassiodorus, formerly prominent in Theoderic’s administration, and the leader of the senate, Cethegus; in 554 Justinian gave senators permission to live in Constantinople. The Roman deacon Vigilius was on hand in Constantinople in 537, well placed to become pope when Silverius fell out of imperial favour; when Vigilius died in 555, his successor Pelagius was likewise there, standing in the wings. From the time of Vigilius, imperial confirmation of the election of a pope was needed before he could be consecrated; this accounts for the long intermissions between pontificates that characterised the following period of papal history. Pope Gregory the Great had served as papal legate in Constantinople (c. 579–c. 586) before being appointed as pope in 590.

212

the earlier empire

His two successors would likewise serve in this position before becoming pope. Clearly, after the conquest of Italy, a stint in Constantinople was a valuable item in the curriculum vitae of prospective popes. Maximianus was appointed to the see of Ravenna while at Constantinople in 546 and he was to travel there again, while in 552 the clergy of the province of Milan asked a legate travelling to Constantinople to see what he could do to secure the return of bishop Datius; he had been absent from his see for fifteen or sixteen years, and in the royal city for much of the time. One of Gregory the Great’s acquaintances while he was in Constantinople, the Milanese deacon Constantius, was appointed bishop of his city in 593, while another, the Spaniard Leander, was to become bishop of Seville. In 551 Reparatus of Carthage and other African bishops were summoned to Constantinople; in the following year Justinian exiled Reparatus and replaced him, against the will of the clergy and people of Carthage, with Primosus, his former legate in Constantinople. Members of various Germanic royal families, such as the Ostrogoth Amalasuntha, were also on hand. An eye could be kept on their activities in the City, and they could be called into action as imperial needs required. No less striking is the centrality of Constantinople in the intellectual life of the west. A large volume of literature in Latin was produced there during, and immediately after, the reign of Justinian. It was in Constantinople that the Illyrian, Marcellinus, and the African, Victor of Tunnuna, wrote their chronicles; and although the chronicle of the Spanish Goth, John of Biclaro, was produced in Spain, he wrote it after spending some years in the City. It was in Constantinople that the Goth, Jordanes, wrote his histories of the Romans and the Goths. Cassiodorus worked on his Expositio psalmorum in the City, and it was there that the African, Junillus, wrote his introduction to the study of the Bible, while another African, Corippus, witnessed the accession of Justin II, which he described in a panegyric; and it was from Constantinople that various African theologians came to operate. Somewhat later, the future pope Gregory delivered there the talks which formed the basis of his massive Moralia in Job. Scholars have sometimes doubted Gregory’s assertion that he did not know Greek, on the basis that it would have been difficult for the representative of the pope to have functioned in Constantinople without knowing the language. However, given the flourishing and influential community of Latin-speakers there, Gregory may not have found a command of Greek necessary. the three chapters But at this very time of the centrality of Constantinople in western affairs, events were under way which threatened its position and, as often happened in late antiquity, tensions were expressed in disputes over religion.

3. western approaches ( 500–600)

213

Imperial policy had long sought to bring together adherents of the council of Chalcedon (451), who recognised the ‘unity of Christ’s person in two natures’, and their monophysite opponents, and Justinian made an important attempt to bring about reconciliation between the disputing parties.28 He asked the five patriarchs of the church to anathematise the person and works of Bishop Theodore of Mopsuestia, some of the writings of Bishop Theodoret of Cyrrhus, and a letter attributed to Bishop Ibas of Edessa. These three theologians, all long dead, were held to show Nestorian tendencies, and Justinian believed that their condemnation would be a painless way of conciliating the monophysites, who held an opinion contrary to that of the Nestorians. But the council of Chalcedon had accepted the orthodoxy of Theodoret, and the letter of Ibas had been read out there, so an attack on these thinkers could be construed as an attack on the council. Pope Vigilius refused to accept Justinian’s proposal, whereupon, to the astonishment of the populace of Rome, he was arrested in a church in 545 and conveyed to Constantinople. Years of intrigue followed, in which Vigilius was alternately vacillating and resolute. Finally, in 553, the council of Constantinople condemned the Three Chapters, as they came to be called, and Vigilius accepted its decision. In 554 he set out to return to Rome, but died at Syracuse in June 555, a broken man. As it turned out, Justinian’s efforts did nothing to reconcile the monophysites and the adherents of Chalcedon, but there was an immediate hostile reaction in the west, where it was felt he had gone against the position adopted by the council. So intense were feelings in Italy that it proved difficult to find bishops prepared to consecrate Vigilius’ successor, Pelagius, and a schism broke out in northern Italy (see above, p. 118). There was considerable disquiet in Gaul, and throughout the Visigothic period the Spanish church failed to accept the council of Constantinople. Opposition was, however, strongest in Africa where an episcopate which had seen off the persecuting Arian Vandals was in no mood to be dictated to by a catholic emperor, and the African church flung itself into the controversy with the learning and vigour which had characterised it for centuries. As early as 550 a synod excommunicated Vigilius, and a series of authors wrote attacking Justinian’s position; it was an African chronicler who observed that the council of Constantinople was followed by an earthquake in that city!29 Small wonder that a bishop from northern Gaul, Nicetius of Trier, wrote a strongly worded but theologically incoherent letter to the emperor, reporting that all Italy and the entirety of Africa, Spain and Gaul wept 28 For a more detailed discussion of Christological disputes in sixth-century Byzantium, see above pp. 116–19. 29 Victor of Tunnuna, Chronicle, ed. Mommsen, p. 203; ed. Cardelle de Hartmann, ch. 147, p. 49; ed. and Italian tr. Placanica, pp. 52–3.

214

the earlier empire

over him: ‘O sweet Justinian of ours, who has so deceived you, who has persuaded you to proceed in such a way?’30 western antagonism to t he empire Early Christian history is full of controversies on issues apparently so abstruse that modern scholars have often felt they were really about subjects far removed from the matters being overtly debated, and the controversy over the Three Chapters in the west may have been one where the real issue was unstated. It is possible to interpret the strong stance the west took against Justinian’s line as constituting a response to the impact of his wars of conquest. Doubtless the heads of churches in Africa and Italy sincerely welcomed the coming of Justinian’s armies, but while governed by non-Roman Arians, they had come to enjoy de facto independence from imperial oversight, which they would not surrender willingly. It is no coincidence that one of the most famous assertions of ecclesiastical power ever made vis-`a-vis the emperor was enunciated by Pope Gelasius (492–6) during the period of Ostrogothic power in Italy. The wars created a situation in which an emperor, for the first time in a long while, was able to attempt to impose his will directly on western churches, and some of the opposition to Justinian’s policies may simply have been a reaction against the new reality. But it may also be that opposition to the Three Chapters was a vent for hostility towards, or disillusionment with, the outcome of the wars in the west. If we accept this, we will not be surprised to find Cassiodorus, the best-known collaborator with the Goths among the Romans, writing towards the middle of the century in terms which suggest sympathy for the theologians whose condemnation Justinian was seeking. Nor are other indications of western coolness towards Byzantium lacking in the period after the conquests. The indigenous inhabitants of Africa and Italy initially welcomed the Byzantine armies. In Italy the Gothic government was worried about the loyalty of the populace even before the war began, and the detailed narrative of Procopius makes it clear that its fears were justified. Yet early in the war a Gothic spokesman told the people of Rome that the only Greeks who had visited Rome were actors, mimes or thieving soldiers, suggesting there was already some resentment towards the Byzantines, which the Goths sought to exploit. We are told that during the pontificate of Pope John III (561–74) the inhabitants of the city maliciously told the emperor that ‘it would be better . . . to serve the Goths than the Greeks’.31 The use of 30 Epistulae Austrasicae, no. 7, ed. Gundlach, p. 417. There is a reminiscence here of St Paul (Gal. 3:1). The answer to Nicetius’ questions is ‘the Devil’. 31 Gothic spokesman: Pr W, V.18.40–1, ed. and tr. Dewing, III, pp. 182–3; message to the emperor: LP, LXIII.10, ed. Duchesne, I, p. 305; tr. Davis, I, p. 62.

3. western approaches ( 500–600)

215

the term ‘Greeks’ is interesting, for in Procopius it is a hostile word placed in the mouths of non-Romans. Perhaps the Romans had come to accept, or at least pretend to accept, the barbarians’ assessment of the easterners. The dire state of the Italian economy after the long war, and the corrupt and grasping nature of the Byzantine administration imposed in both Africa and Italy, made imperial government unpopular. Further, Italy’s integration into the empire did not imply reversion to the position of independence from the east which it had enjoyed before the advent of barbarian power, nor were its Roman inhabitants able to enjoy the positions of influence they had held under the Goths; Italy was now a minor part of an empire governed by a far-away autokrat¯or who never troubled to visit the west. Power in Africa and Italy passed to Greek-speaking incomers, and we have evidence for cults of eastern saints, which they presumably brought with them. Needless to say there were loyalists and careerists who supported the Byzantine regime, for example the African poet Corippus, whose epic Iohannis was partly an attempt to justify the imperial cause to his fellow Africans;32 but these represented minority opinion. If this were not enough, opposition to Justinian’s wars even developed in the east. This can be traced through the works of Procopius, which move from a sunny optimism in describing the Vandal war to the sombre tone which increasingly intrudes in the Gothic war and the animosity towards the emperor displayed in the Secret history; but one can also deduce from other sources a feeling that resources had been committed in the west to little profit. However impressive their outcome in bringing Africa and Italy back into the empire, Justinian’s wars had in some ways the paradoxical result of driving east and west further apart. byz antine military d iff iculties in t he west Throughout Justinian’s reign, that part of the empire south of the Danube had been troubled by incursions, in particular those of the Turkic-speaking people known as Bulgars and groups of Slavs whom contemporaries called Antes and Sclaveni. The government dealt with the threat as best it could by building forts and paying subsidies, but following Justinian’s death in 565, the situation deteriorated rapidly. His successor Justin II adopted a policy of withholding subsidies, and in particular he refused a demand for tribute made by Avars, who had recently made their way into the Danube area. The Avars soon showed their mettle. In 567 they joined forces with the Lombards living in Pannonia to crush the Gepids, a victory that signalled the end of the Germanic peoples along the middle Danube. In the following year the Lombards left Pannonia for Italy, whereupon the Avars occupied 32

Cameron, Averil (1985).

216

the earlier empire

the lands they had vacated, the plain of modern Hungary. From there they launched attacks deep into imperial territory, and the renewal of war with Persia in 572 made the Byzantine response to these developments the less effective. In 581 Slavs invaded the Balkans, and before long it became clear that they were moving in to stay. These events had a major impact on the west. The attention of the authorities was now diverted from the newly won provinces, and direct land access to Italy was rendered difficult. Moreover, it may well have been the rise of the Avars that impelled the Lombards to launch in 568 their invasion of Italy where they quickly gained control of the Po valley and areas of central and southern Italy. The Byzantine administration, under the successor of Narses, the praetorian prefect Longinus, proved embarrassingly ill-equipped to cope, and a force eventually sent from the east under Justin’s son-in-law Baduarius was defeated. In 577 or 578 the Roman patricius Pamphronius, who had gone to Constantinople seeking help, was sent away with the 3,000 pounds of gold he had brought with him and told to use the money to bribe some Lombards to defect or, failing that, to secure the intervention of the Franks. In 579 a second embassy was fobbed off with a small force and, we are told, an attempt was made to bribe some of the Lombard leaders. Perhaps we see here the reflection of a change in imperial policy, for while Justin had behaved in a miserly fashion, his successor Tiberius I (578–82) was inclined to throw money at his problems. However, neither strategy succeeded, and it was all too clear that the situation in Italy was desperate. It was time for Constantinople to play the Frankish card again. For the greater part of the sixth century the Franks had steadily been growing more powerful. Their defeat of the Visigoths in 507 was followed by expansion from northern into southern Gaul, while the weakening of the Burgundians and Ostrogoths in the 520s and 530s saw further gains.33 In the early stages of the Gothic war they were in the happy position of being able to accept the payments that both sides made seeking their assistance, but when King Theudebert marched into Italy in 539, he was acting solely in his own interests. Theudebert issued gold coins displaying his own portrait rather than that of the emperor and bearing legends generally associated with emperors rather than kings. He responded to an embassy from Justinian in grandiloquent terms, advising him that the territory under his power extended through the Danube and the boundary of Pannonia as far as the ocean shores.34 Towards the end of Theudebert’s life his forces occupied Venetia and some other areas of Italy, and it was rumoured that he planned to march on the City: such was the fear he inspired in 33 34

See Barbero and Loring (2005), pp. 173–4; Van Dam (2005), p. 200. Epistulae Austrasicae, no. 20, ed. Gundlach, pp. 438–9.

3. western approaches ( 500–600)

217

Constantinople. The settlement of the Lombards in Pannonia by Justinian in about 546 may have represented an attempt to counter the Franks. Following the death of Theudebert in 547, Justinian sent an embassy to his heir Theudebald proposing an offensive alliance against the Goths, but he was turned down and Frankish intervention in Italy continued to be a problem throughout the Gothic war. The advent of the Lombards, however, placed the Franks once more on the far side of an enemy of the Byzantines and they could again be looked upon as potential allies. But the attempts made to gain their help occurred against a highly complex political and military background. It is difficult to reconstruct the web of alliances and animosities that lay behind relations between Constantinople and the disparate parts of the west towards the end of the sixth century. In 579 Hermenigild, the elder son of the Visigothic king Leovigild, revolted against his father. After the rebellion’s suppression, Hermenigild’s wife (the Frankish princess Ingund) and son Athanagild fled to the Byzantines; Athanagild was taken to Constantinople, and his Frankish relatives were unable to secure his return to the west, despite their efforts. A few years later one Gundovald, who claimed to be the son of a Frankish king, arrived in Marseilles. He had been living in Constantinople, but had been lured back to Francia by a party of aristocrats. The emperor Maurice (582–602) gave Gundovald financial backing, and one of his supporters in Marseilles was later accused of wishing to bring the kingdom of the Franks under the emperor’s sway. This was almost certainly an exaggeration, and Gundovald’s rebellion came to naught, but again we have evidence of imperial fishing in troubled western waters.35 In 584 the Frankish king Childebert, the uncle of Athanagild, having at some time received 50,000 solidi from Maurice, sent forces to Italy, but the results were not up to imperial expectations and Maurice asked for his money back. Other expeditions followed, but little was achieved. Finally, in 590 a large Frankish expedition advanced into Italy and made its way beyond Verona, but failed to make contact with the imperial army. This was the last occasion when Constantinople used the Franks in its Italian policy. The fiasco of 590 may be taken as symbolising a relationship which rarely worked to the benefit of the empire. While it may often be true that the neighbours of one’s enemy are one’s friends, Byzantine attempts to profit from the Franks had persistently failed. By the last years of the century the Byzantines were in difficulties throughout the west. Most of Italy was under Lombard control, and severe losses had also been sustained in Africa, although the latter can only dimly be perceived. In 595 Berbers caused alarm to the people of Carthage itself, until the 35 On Gundovald see Gregory of Tours, Libri, VI.24, VII.10–14, VII.26–38, ed. Krusch and Levison, pp. 291–2, 332–6, 345–62; tr. Thorpe, pp. 352–3, 394–8, 407–23.

218

the earlier empire

exarch tricked them into defeat. A geographical work written by George of Cyprus early in the seventh century indicates that the imperial possessions in Africa were considerably smaller than those which the Vandals had controlled, themselves smaller than those which had belonged to the Roman empire.36 The establishment of exarchs in Ravenna and Carthage indicates a society that was being forced to become more military in orientation, and while the Byzantine possessions in Spain are not well documented, it is clear that they tended to diminish rather than grow. east and west: continuing links and growing d ivisions Paradoxically, despite the waning of Byzantine power in the west, the latter continued to be vitally interested in the east. A ready market remained for imported luxury items; goods of Byzantine provenance were included in the early seventh-century ship burial at Sutton Hoo in East Anglia and Radegund, the founder of a convent at Poitiers, petitioned Justin II and his wife Sophia for a portion of the True Cross, which she duly received in 569 (see above, p. 123). At the end of the century, the letters of Pope Gregory the Great reveal a man who saw the empire as central to his world and who had a penchant for wine imported from Egypt, surely one of the few Italians in history of whom this could be said. Byzantine legislation was followed with attention. The Frank Chilperic I did not merely rejoice in the possession of gold medallions that Tiberius I sent him: an edict he issued shows an apparent dependence on a novella of the same emperor.37 Eastern liturgical practice was imitated; on the recommendation of the newly converted Visigothic king Reccared, the third council of Toledo prescribed in 589 that the creed was to be sung before the Lord’s Prayer and the taking of communion ‘according to the practice of the eastern churches’, apparently in imitation of Justin I’s requiring, at the beginning of his reign, that the creed be sung before the Lord’s Prayer. This is one of a number of indications of the increasingly Byzantine form of the public life of Spain towards the end of the sixth century. The chronicle of Marius of Avenches, written in Burgundy, is dated according to consulships and indictional years, until its termination in 581. Inscriptions in the Rhˆone Valley were still being dated according to consulships or indictional years in the early seventh century, and coins were being minted in the name of the emperor at Marseilles and Viviers as late as the reign of Heraclius (610–41). Whatever may be the merits of thinking in terms of ‘an obscure law of cultural hydraulics’, whereby streams of influence were occasionally 36 37

George of Cyprus, Descriptio orbis Romani, pp. x-xi, 33–4. Stein (1920).

3. western approaches ( 500–600)

219

released from the east to water the lower reaches of the west,38 there can be no doubt that the west remained open to Byzantine influence, nor that western authors such as Gregory of Tours and Venantius Fortunatus sought to keep abreast of eastern material in a way that few easterners reciprocated. Emperors, moreover, gave indications that they still regarded the west as important. The marriages which the emperor Tiberius arranged for his daughters are strong evidence of this, for whereas one of them married Maurice, the successful general who was to succeed Tiberius, another married Germanus, the son of the patricius whom Justinian had nominated to finish the war against the Goths in 550, and of his Gothic wife Matasuentha. Tiberius made each of his sons-in-law caesar and, given the strong western associations of Germanus, it is tempting to see the emperor as having thought of a divisio imperii into east and west, something that never seems to have crossed Justinian’s mind. If this was Tiberius’ plan, nothing came of it; but his successor, Maurice, drew up a will appointing his elder son Theodosius as lord of Constantinople with power in the east, and his younger son Tiberius as emperor of old Rome with power in Italy and the islands of the Tyrrhenian Sea. Again, nothing came of this plan, but it was from Carthage that Heraclius, the son of an exarch, launched his successful rebellion against Phocas in 610. It was later believed that at a difficult point in his reign the emperor Heraclius planned to flee to Africa, only being restrained by an oath the patriarch forced him to take. In the mid-seventh century Maximus the Confessor, a complex figure who in various ways links east and west, was accused of having had a vision in which he saw angels in heaven on both the east and the west; those on the west exclaimed ‘Gregory Augustus, may you conquer!’, and their voice was louder than the voices of those on the east.39 Surely, it may appear, relations between Byzantium and the west remained strong. But although the west retained a capacity to absorb Byzantine influences, and emperors after Justinian continued to think in terms of controlling the west, in other ways the sixth century saw the two parts of the former empire move further apart. Justinian’s wars had over-extended the empire, entailing a major weakening of its position on the northern and eastern frontiers, and as warfare continued against the Slavs, Avars and Persians there were few resources to spare for the west, where the territory controlled by Constantinople shrank to scattered coastal fringes. By the end of the century there was little trade between Carthage and Constantinople. East 38 39

See the memorable characterisation of this view in Brown, P. (1976), p. 5. Mansi, XI, cols. 3–4. The Gregory referred to was an exarch of Carthage who had rebelled against Emperor Constans II. On Maximus, see below pp. 231–2.

220

the earlier empire

and west were drifting apart linguistically: there are no counterparts to a Boethius in the west or a Priscian in the east towards the end of the century. In his correspondence as pope, Gregory the Great complained of the quality of translators out of Latin in Constantinople and Greek in Rome: both sets translated word for word without regard for the sense of what they were translating.40 Byzantine historians after Procopius rapidly came to display a lack of knowledge of, or interest in, western affairs. Evagrius Scholasticus, writing towards the end of the sixth century, argued in favour of Christianity by comparing the fates of emperors before and after Constantine, a line of argument that could only be sustained by ignoring the later western emperors.41 The sources available to Theophanes, when he wrote his Chronicle in the early ninth century, allowed him to note the accession of almost every pope from the late third century to Benedict I in 575, but not of subsequent ones. Meanwhile Paul the Deacon, writing in the late eighth century, seems to have regarded Maurice as the first Greek among the emperors.42 One has the feeling that towards the end of the sixth century the west simply became less relevant to easterners. Meanwhile, the west was going its own way. The discontent manifested in Africa and Italy over the condemnation of the Three Chapters may plausibly be seen as reflecting unhappiness at the situation following the wars waged by Justinian. Increasingly, the Italians came to see their interests as not necessarily identical with those of the empire. In Spain, Justinian’s activities left a nasty taste in peoples’ mouths: the learned Isidore of Seville, writing in the early seventh century, denied not only ecumenical status to the council of 553, but also a place among Roman lawgivers to Justinian and patriarchal rank to the see of Constantinople. In Africa, the government’s inability to deal with the Berbers prepared the way for the loss of the province to the Arabs in the following century. It is hard to avoid the conclusion that in the sixth century Byzantium and the west moved significantly apart; one cannot but see the emperor Justinian as being largely to blame. 40 41

Gregory I, Letters, VII.27, ed. Ewald and Hartmann, I, p. 474; tr. Martyn, II, pp. 482–3. Evagrius Scholasticus, Ecclesiastical history, III.41, ed. Bidez and Parmentier, pp. 141–4; tr. Whitby, pp. 189–92. 42 ‘Primus ex Grecorum genere in imperio confirmatus est’: Paul the Deacon, Historia Langobardorum, III.15, ed. Bethmann and Waitz, p. 100; tr. Foulke, p. 113.

CHAPTER 4

BYZANTIUM TRANSFORMING (600–700) an drew lou t h

introduction Most centuries can be said to have been, in one way or another, a watershed for Byzantium, but the case for the seventh century is particularly strong. At the beginning of the century, the Byzantine empire formed part of a political configuration that had been familiar for centuries: it was a world centred on the Mediterranean and bounded to the east by the Persian empire, in which most of the regions surrounding mare nostrum formed a single political entity – the Roman (or Byzantine) empire. It was a world whose basic economic unit was still the city and its hinterland; although it had lost much of its political significance, the city retained the social, economic and cultural high ground. By the beginning of the seventh century, this traditional configuration was already being eroded: much of Italy was under Lombard rule, Gaul was in Frankish hands and the coastal regions of Spain, the final acquisition of Justinian’s reconquest, were soon to fall to the Visigoths. By the end of the century this traditional configuration was gone altogether, to be replaced by another which would be dominant for centuries and still marks the region today. The boundary that separated the Mediterranean world from the Persian empire was swept away: after the Arab conquest of the eastern provinces in the 630s and 640s, that boundary – the Tigris–Euphrates valley – became one of the arteries of a new empire, with its capital first in Damascus (661–750) and then in Baghdad (from 750). By the mid-eighth century this empire stretched from Spain in the west to the valleys of the Oxus and the Indus in the east, far larger than Justinian’s Byzantine empire or even the Roman empire had ever been, and hugely richer than any of its neighbours. The new empire caused Europe, East Asia and North Africa to be reconfigured around it, forcing the Byzantine empire – and the emergent Frankish kingdoms – into virtual satellite status. This radical upheaval, together with the persistent aggression of the Arabs against the remaining Byzantine lands and the incursion of Slavs and peoples hailing from the central Eurasian steppe into the Balkans, accelerated the transition of the cities of the eastern Mediterranean world that was already well under 221

222

4. byz a nt iu m t ra ns forming ( 60 0 –70 0 )

the earlier empire

223

CRIMEA

L A

S E A

Danube

DALMATIA

Pliska

ITALY Rome

Constantinople Thessalonica

Naples

Sea of Marmara

Kyzikos

SARDINIA

BALEARIC ISLANDS

IC

B L A C K

Ravenna

CORSICA

Z

GAUL

A

Cherson

ASIA

Tig ris

MINOR

( A N AT O L I A ) CAPPADOCIA S MT US TA U R

Smyrna

Septem SICILY Carthage

Syracuse

(fell 698)

Antioch

Constantia

RHODES

Under Muslim control by 700

(Salamis)

CYPRUS

Damascus

CRETE

Yarmu k

Jerusalem

Gaza KINGDOM OF

VENETIA

THE LOMBARDS URIA IG

JUDAEAN DESERT

Alexandria

LIBYA

EXARCHATE OF RAVENNA

SINAI

L

PE TH Ravenna NT E Rimini AP

D

A

O

LM

A

EGYPT TI

A

PER

U

LIS

A GI

DUCHY OF SPOLETO

CORSICA

Rome

DU Y CH

ABR

Limit of Byzantine Territory Territory under Byzantine authority

IA

Under Lombard control M

Subject to Slav immigration and Avar attack Areas under Muslim control or occupation Byzantine territory disputed with Muslims

BR

Flaminian Way Byzantine frontiers Lombard frontiers

UT

TIU

EXARCHATE OF CARTHAGE

DUCHY OF NAPLES

CAL

O NT VE

SARDINIA

NE BE

Benevento

OF

DUCHY OF ROME

SICILY

Armenian territories subject to Muslim raids 0

Syracuse

Euphr ate

Kos

250

500 250

Map 11 Byzantium transforming: the empire towards the end of the seventh century Cambridge Histories Online © Cambridge University Press, 2008

750 500 miles

1000 km

s

224

the earlier empire

way. By the end of the century the cities had lost much of their social and cultural significance, and survived as fortified enclaves, if often as market centres, too.1 The only place approximating to the traditional city was Constantinople, and that largely because of the presence of the imperial court; but even Constantinople barely survived, and did so in much reduced circumstances.2 This dramatic transition caused something of a crisis of confidence and even identity for Byzantium. At least twice the emperor entertained the notion of deserting Constantinople and re-establishing the capital of the empire closer to its traditional centre in Rome: in 618 Heraclius (610– 41) thought of moving to Carthage, and in the 660s Constans II (641– 68) settled in Sicily. In both cases we can see how the traditional idea of a Mediterranean empire still haunted the imagination of the Byzantine rulers. In fact, despite the dramatic and permanent changes witnessed by the seventh century, Byzantine reactions can be seen as attempts to preserve what was perceived as traditional. But as always with the Byzantines, one must be careful not to be deceived by their rhetoric. This rhetoric – and, as we shall see, administrative changes that were more than rhetorical – spoke in terms of centralisation, an increasing focus on the figure of the emperor and the court, and a growing influence of the patriarch and clergy of the Great Church of St Sophia in religious matters. In reality, however, events and persons on the periphery were often more important than what was going on at the centre. The transition that started in the seventh century was not completed in that century: not until the late eighth and ninth centuries, when Arab pressure on the Byzantine empire eased after the capital of the caliphate moved eastwards from Damascus to Baghdad, did Byzantium finally complete the transition begun in the seventh. What emerged was an empire and culture focused on emperor and capital; but much of what the centre now stood for was, in fact, worked out not in Constantinople itself, but at the periphery. The history of the Byzantine empire in the seventh century is difficult to reconstruct. Traditional sources are sparse and mostly late.3 We can draw on Theophylact Simocatta’s History and the Paschal chronicle, both of which were probably written at the court of Patriarch Sergius around 630 during the euphoria caused by Heraclius’ triumph over the Persians. The celebrations of Heraclius’ Persian victories by George of Pisidia also belong to this period but history writing in Byzantium stops thereafter until the end of the eighth century. For the political history of the seventh century our 1 The pace of this change in the cities and its interpretation is the subject of an ongoing debate. See Foss (1975), Foss (1977a), emphasising the impact of the Persian invasion in the first quarter of the seventh century; Haldon (1997a) pp. 92–124, 459–61; below, pp. 261, 264, 483–5. 2 For Constantinople, see Mango (1990). 3 On the problems of the literary source materials, see BEINE, I.

4. byz antium transf orming (600–700)

225

principal sources are thus two later works: Nikephoros I, patriarch of Constantinople’s Brief history, composed in the late eighth century and intended as a continuation of Theophylact Simocatta; and the early ninth-century Chronicle ascribed to Theophanes the Confessor. To some extent the dearth of writing from the period 630–790 may be seen to be a consequence of the collapse of much of traditional Mediterranean society. The demise of the ancient city meant the collapse of the educational system’s traditional base: there were fewer and fewer people to write for.4 There was also less to write about: details of the fall of the Byzantine eastern provinces to the Arabs and subsequent defeats and losses would not be welcome material for Byzantine writers, and are either omitted by Nikephoros and Theophanes, or drawn from Syriac or Arab sources. Like these Byzantine historians, we can supplement our sparse resources with oriental historical material. There is an anonymous history of Heraclius ascribed to the Armenian bishop Sebeos and dated to the latter half of the seventh century (see above, n. 1, p. 157). There is also a world chronicle, written in Egypt at the end of the century by Bishop John of Nikiu; however, this only survives in mutilated form in an Ethiopic translation. There are in addition several contemporary and later Syriac chronicles: besides anonymous works, there are those compiled by Elias bar Shinaya, the eleventh-century metropolitan of Nisibis, and Michael the Syrian, the twelfth-century Jacobite patriarch of Antioch, both using earlier sources. Legal sources are also sparse for this period, but the Farmer’s law (Nomos georgikos) probably belongs to the seventh or eighth century, as may the Rhodian sea-law (Nomos Rodion nautikos). Traditionally, therefore, the seventh and eighth centuries have been regarded as the Byzantine ‘dark ages’, though historians have begun to recognise that it is only in respect of traditional historical literary material that one can speak of a paucity of sources for the period. For in fact it was an immensely fruitful period for Byzantine theology, dominated by the figure of Maximus the Confessor, perhaps the greatest theologian of the orthodox east and certainly the greatest Byzantine theologian.5 To make full use of these ‘untraditional’ sources would, however, involve writing a different kind of history, beginning not from the institutional and political, but rather working outwards from the deeply-considered worldview to be found in such writings.6 But one should note that there is a notable lacuna in the theological sources themselves. They are all from the periphery: Maximus writing mostly from North Africa, Anastasius of Sinai and John Klimakos (‘of the Ladder’) from Sinai. Elsewhere, Cyprus 4 5

See Whitby, Michael (1992). The contemporary accounts of the legal process he underwent, with some other documents, are edited in Maximus the Confessor, Scripta saeculi, ed. and tr. Allen and Neil. 6 For a notable attempt by a historian to use these theological sources, see Haldon (1992a).

226

the earlier empire

and Palestine were homes to a good deal of writing, polemical and hagiographical for the most part. We know almost nothing of theology in Constantinople between the middle of the sixth century, such as came from the circle of Justinian, Leontius of Byzantium and Leontius of Jerusalem, and the ninth-century revival of learning – that of the iconodule theologians Nikephoros I, patriarch of Constantinople, Theodore the Stoudite, Patriarch Photios and others. The only exceptions are the Constantinopolitan opponent of iconoclasm, Patriarch Germanos I, and some traces of the theology of the iconoclasts preserved by their opponents. Virtually all the theology that survives from this period of transition belongs to the periphery. This chapter will firstly give an outline of the political history of the period, and will follow this with some account of the transition that the seventh century witnessed. To do otherwise would be nearly impossible, as the elements of the transition – the transformation of the city, the administrative and the religious changes – are not easily datable, and consequently would find no natural place in the narrative history. even ts: persians d ef eated, arabs triumphant, churchmen at odds The century began with Maurice (582–602) on the imperial throne, urging his army to resist the incursion of Slavs who were seeking to cross the Danube from the north bank. Growing discontent culminated in mutiny when Maurice ordered the army to continue their campaign against the Slavs into the winter months, when bare trees would provide less cover for the marauders. Led by Phocas, a relatively minor officer, the army marched on Constantinople and deposed Maurice. Phocas was proclaimed emperor, but was never very secure and faced a number of revolts. More seriously the Persian shah Khusro II (590–628) used Maurice’s murder as a pretext to declare war on the empire, to avenge his former protector (see above, pp. 127, 128). With the invasion of Syria, there began a war that would last until 626–7. In 610 Phocas was deposed by Heraclius, son of the exarch of Carthage, who, according to Theophanes, seized the throne at the invitation of the senate in Constantinople. Heraclius’ ships displayed reliquaries and icons of the Mother of God on their masts: a sign of the continuing authentication of political authority by supernatural means seen in the later decades of the sixth century. Phocas was swiftly overthrown and executed, and Heraclius acclaimed as emperor and crowned by the patriarch in St Stephen’s chapel in the palace. On the same day he married his betrothed, Eudocia, whom he crowned augusta. The situation Heraclius faced was grim. The Persians were now advancing into Asia Minor, taking Caesarea in Cappadocia in 611, and to the north

4. byz antium transf orming (600–700)

227

from across the Danube the Avars were once again a serious menace: in 615 both enemies would make a joint assault on Constantinople. Attempts were made to negotiate a peace treaty with the Persians – immediately, according to the eastern sources; according to the Greek sources, in 615, once the Persian forces had advanced as far as Chalcedon. Anyway the peace efforts were repudiated, as the Persians were convinced that the Byzantine empire lay at their mercy. The war took on the character of a holy war between a Christian army, using icons of Christ and the Virgin as banners, and the predominantly Zoroastrian army of the Persians. Besides advancing into Asia Minor, the Persians invaded Palestine, taking Jerusalem in early May 614, and then Egypt and Libya. The fall of Jerusalem, by now regarded by Christians as their Holy City, was a catastrophe for Byzantium as a Christian empire, and for the emperor as God’s vicegerent on earth. Still worse was the seizure of the relic of the True Cross, which was taken back to the Persian capital, Ctesiphon, along with Zacharias, patriarch of Jerusalem, and those Christian notables who survived the sack of the city; tens of thousands are said to have been killed.7 It was not until Heraclius had managed to negotiate a truce with the Avars that he was able to make a serious attempt to defeat the Persians. From 622 onwards he conducted a series of campaigns against them. In 626, while Heraclius was on campaign, the Persians joined forces with the Avars to besiege Constantinople. Heraclius himself did not return, but sent a contingent of the field army to reinforce the City’s defenders, who were under the leadership of the two regents, Patriarch Sergius and the magister officiorum, Bonus. Constantinople was besieged for ten days by a huge army of various peoples under the command of the khagan of the Avars, while the Persian army under Shahrvaraz held the Asian shore of the Bosporus. The siege failed when the fleet of Slav boats was destroyed by the Byzantine fleet in the Golden Horn, just across from the Church of the Virgin at Blachernae. The success of the Constantinopolitans’ defence of their city was ascribed to the Virgin Mother of God, and it is likely that the famous troparion ‘To you, champion commander’ was composed by Patriarch Sergius to celebrate her victory. Heraclius pressed his attack into the Sasanians’ heartland. The Persians were demoralised by their troops’ failure under Shahrvaraz to secure the City; they were also smarting at the destruction by the emperor’s brother Theodore of another contingent destined for Constantinople. Heraclius’ successes provoked a palace revolt in which Khusro was murdered, and the Persians sued for peace. All the territory they had taken was restored to the Byzantine empire, and the Tigris–Euphrates valley became the frontier once again. Heraclius recovered 7 For the most thorough examination of the various sources for the taking of Jerusalem by the Persians, see Flusin (1992); see also Kaegi (2003a), pp. 78–80.

228

the earlier empire

the True Cross, and celebrated his triumph by taking the relic on a tour of the restored Byzantine territories, before returning it to Jerusalem on 21 March 630.8 It would seem to be at this stage that Heraclius began to face the religious problems that had plagued the Byzantine empire since the council of Chalcedon in 451.9 The schism between those who supported Chalcedon and those who repudiated it, whom their enemies called monophysites, had become institutionalised with a separate monophysite episcopal hierarchy since the consecration of Jacob Baradaeus in 542. The monophysites had their greatest support in the eastern provinces, especially Syria and Egypt; many Christians in Armenia also declined to acknowledge the council of Chalcedon (see also below, pp. 333, 335). After conquering the eastern provinces, Khusro had sought to strengthen his hold over his new subjects by exploiting the Christians’ schisms. At a meeting held in Ctesiphon, Khusro met with leaders of the monophysites, the Armenians and also the Nestorians, the main Christian group established in Persia. These last had rejected the condemnation of Nestorius at the third ecumenical council of Ephesus in 431 and fled to Persia to escape persecution in the Byzantine empire. It was agreed that the Nestorians should retain their position within the traditional Sasanian territories, but that the Persian authorities would support the monophysites in Armenia and those former Byzantine provinces where the monophysites were in a majority, that is, Syria and Egypt. The monophysites welcomed this agreement, their patriarch of Antioch, Athanasius ‘the Camel Driver’, rejoicing at the ‘passing of the Chalcedonian night’. If Heraclius was to be secure in his regained eastern provinces, he needed to gain the support of the monophysites. The policy he pursued was proposed by his patriarch Sergius, who had foreseen this problem and had already begun negotiations with monophysites: Sergius was himself Syrian, possibly with a monophysite background. The proposal was to seek union on the basis of the doctrine of monenergism, i.e. that Christ, while he had two natures, as Chalcedon had affirmed, possessed only a single divine-human activity. This policy achieved some success in Armenia, but the Syrian monophysites (Jacobites) were not amenable and required an explicit repudiation of Chalcedon. Monenergism’s greatest success was in Egypt, where Cyrus of Phasi, appointed patriarch and augustal prefect in 631, reached an agreement with the main monophysite group, the Theodosians.10 On 3 June 633 a solemn eucharist celebrated the union with the Theodosians, on the basis of a carefully phrased pact of union in nine 8 9 10

For the date and the literature cited, see Flusin (1992), II, pp. 293–309. For the earlier religious problems, see Meyendorff (1989), pp. 333–73; above, pp. 99–129. Named after Theodosius, the sixth-century patriarch of Alexandria.

4. byz antium transf orming (600–700)

229

chapters; this placed monenergism in the context of the Cyrilline Chalcedonianism that had been espoused by Justinian and endorsed at the fifth ecumenical council in 553.11 But it was not only some of the monophysites who refused to accept monenergism. As Cyrus was about to celebrate his triumph of ecumenism, also present in Alexandria was the learned and highly respected abbot, Sophronius. To him, the nine chapters amounted to monophysitism. He protested to Cyrus, to no avail, and took his protest to Patriarch Sergius in Constantinople. Sergius was sufficiently alarmed by Sophronius’ protest to issue a ruling on the matter (the Psephos) in which he forbade any mention of either one or two activities in Christ. But that scarcely satisfied Sophronius, who took his complaint to Pope Honorius I in Rome. He seems to have had no success with the pope either, and from Rome he made his way to Palestine, where he was elected patriarch of Jerusalem in 634. In his synodical letter Sophronius exposed the heresy of monenergism, though without explicitly breaching the terms of the Psephos. Before Sophronius’ arrival in Constantinople, Sergius had already communicated the success of the doctrine of monenergism in Alexandria to Honorius, who in his reply used the phrase that was to lead to the refinement of monenergism into the doctrine of monothelitism. That phrase was ‘one will’. Monothelitism, the doctrine that Christ had only one divine will, was proclaimed as imperial orthodoxy in the Ekthesis issued by Heraclius in 638, although this was doubtless composed by Sergius. However, by 638 the immediate purpose of this religious compromise was being overtaken by events, for Heraclius’ triumph over the Persians proved a pyrrhic victory. Even while it was being celebrated, Palestine and Syria began to experience attacks from Arab tribes that within barely more than a decade would lead to the loss of the Byzantine empire’s eastern provinces – this time for ever – and the complete collapse of the Sasanian empire. In 633 there were Muslim attacks on garrisons in Gaza, and the Arab armies soon moved further north, although there is considerable confusion in the sources about the sequence of events thereafter.12 Heraclius mustered an army and sought to defeat the Arabs. The decisive battle took place at the river Yarmuk in 636, when the much larger Byzantine force was routed. Heraclius abandoned the eastern provinces in despair. The year before, Damascus had already fallen to the Muslims – or more probably had been surrendered – and in 638 Patriarch Sophronius surrendered Jerusalem to Caliph ‘Umar bin al-Khattab. Alexandria was taken in 642, and though the 11 12

See above, pp. 117–18. For the standard account, see Donner (1981); Kaegi (1992). For the difficulties inherent in using the Arab sources, see Leder (1992); Conrad (1992). For a revisionist account of the Arab conquests, see Cook and Crone (1977); Crone (1980); Crone (1987). For a lucid account of the whole controversy over the value of early Islamic sources, see Humphries (1991), pp. 69–103; see also below, pp. 365, 367.

230

the earlier empire

Byzantines recaptured it, in 645 it finally fell. By that time Mesopotamia had already fallen, and with it the Sasanian empire. The speed with which the eastern provinces of the Byzantine empire succumbed to the Arabs remains to be explained by historians. However attractive at first sight, the idea that these provinces, with their attachment to monophysitism, were already culturally lost to the empire does not seem to be borne out by the evidence we have: on the contrary, there is much evidence for the continuing power of Hellenism in the eastern provinces well into the seventh century – evidence suggesting that Hellenic culture was more potent there than in the empire’s capital itself.13 When Heraclius died in 641, his death precipitated a dynastic struggle. He was succeeded by two of his sons: Constantine, by his first marriage to Eudocia; and Heraclius, known as Heracleonas, by his second wife Martina, who was also his niece. Martina herself was given a special role to play as augusta. Heraclius’ marriage to his niece after the death of Eudocia had met with opposition at the time, and there was also opposition to the association of Martina as empress with the two emperors. Constantine’s death – the result of poisoning according to a rumour reported by Theophanes – only increased the opposition to Martina and Heracleonas; there were demands that the imperial dignity should be shared with Constantine’s son, also called Constantine, but usually known as Constans. As troops from the Anatolian armies appeared at Chalcedon in support of these demands, Heracleonas seems to have acceded to them. Nevertheless, Heracleonas and his mother were deposed and exiled, together with Martina’s other two sons, and Constans II became sole emperor. Constans inherited the continuing collapse of the eastern provinces to the Arabs: Egypt was slipping away and Muslim raids into Armenia began in 642–3. In 647 the future caliph Mu‘awiya (661–80) led a raiding party into Anatolia and besieged Caesarea, and from there they penetrated further still into Anatolia. The Arabs made no attempt to settle, but huge amounts of booty were taken back to Damascus. Mu‘awiya also realised the need for the Muslims to develop sea power, and in 649 he led a naval expedition against Cyprus, in which Constantia was taken. In 654 Rhodes was laid waste, Kos taken and Crete pillaged. The following year, in an attempt to remove the threat from the sea, the Byzantine fleet under the command of Constans himself engaged with the Arab fleet, but was defeated and Constans barely escaped with his life. The death of Caliph ‘Uthman in 656 precipitated a civil war (fitna) amongst the Arabs: one faction was led by Mu‘awiya, proclaimed caliph in Syria, the other by ‘Ali, son-in-law of the prophet Muhammad. The civil war ended with the death of ‘Ali and the establishment of the Umayyad 13

See Mango (1989); Cameron, Averil (1991a).

4. byz antium transf orming (600–700)

231

dynasty under Mu‘awiya in 661/2 – events provoking the schism in Islam between Sunni and Shiite that still endures. However, those years of civil war provided valuable respite for the Byzantines. Constans was able to turn his attention to the Balkans, where the power of the Avars had waned, and in 658 he led an expedition into the regions settled by the Slavs (Sklaviniai). There he met with considerable success, according to Theophanes, and was able to use the Slavs he captured to repopulate areas in Anatolia that had been devastated or depopulated. This policy of repopulating Anatolian regions by Slavs was to be continued by his successors, Constantine IV (668–85) and Justinian II (685–695/705–711). Constans also inherited his grandfather’s religious policy. By the early 640s, opposition to monothelitism had grown. Behind this opposition was the monk Maximus, known to later ages as ‘the Confessor’; he had been a close associate of Sophronius, who had originally stirred up opposition to monenergism. Maximus found support in Palestine and Cyprus, but more importantly in Italy and North Africa, where he had been in exile since the late 620s. These were areas which, in the sixth century, had protested against Justinian’s condemnation of the Three Chapters as compromising the decisions of Chalcedon.14 In North Africa a number of synods condemned monothelitism, and Maximus pressed home the attack in a series of skilfully argued tracts and letters. In 645 the former patriarch Pyrrhus arrived in North Africa; as a supporter of Empress Martina he had shared her fall. In July that year a disputation between the monothelite Pyrrhus and Maximus was held in Carthage, before the exarch Gregory, in which Pyrrhus admitted defeat and embraced orthodoxy.15 It was perhaps the strength of feeling against monothelitism that led Gregory to allow himself to be declared emperor in opposition to Constans in 646–7, but his rebellion was short-lived; he died the following year defending his province against Arab raiders. Meanwhile, Pyrrhus had made his way to Rome to declare his new-found orthodoxy to the pope, followed closely by Maximus. In 648, in a vain attempt to prevent further controversy, the famous Typos was issued in the name of the emperor by Patriarch Paul, forbidding discussion of the number of activities or wills in Christ. In Rome, Maximus prepared for a synod, together with other Greek monks who had fled west in the face of the Arabs or the heresy of the empire. This was finally held in 649 in the Lateran Palace in Rome, under the newly elected Pope Martin I (649–55): both the Ekthesis and the Typos were condemned, together with the patriarchs Sergius, Pyrrhus and Paul. The extent to which this synod was of Greek inspiration has become clear 14 15

See above, pp. 117–18. The record of the dispute is preserved as Opusculum 28 in Maximus the Confessor, Opera, PG 91, cols. 287–354.

232

the earlier empire

from recent research, which has shown that the Greek Acta of the synod are the original, the Latin version being a translation.16 Such open defiance of the imperial will could not be ignored. Olympius, exarch of Ravenna, was ordered to arrest Martin and compel the bishops gathered in Rome to accept the Typos. When he arrived in Rome, Olympius discovered that, despite his best efforts, Pope Martin’s popularity made it hazardous to try and arrest him. In defiance of the imperial will he made his peace with Martin and departed for Sicily to deal with Arab raiders. There, like Gregory the exarch of North Africa, he may have been proclaimed emperor. But he died in 652. In the following year a new exarch arrived with troops and succeeded in arresting the pope. Martin was brought to Constantinople and tried for treason, with Olympius’ rebellion being cited as evidence. Although condemned to death, Martin’s sentence was commuted to exile and, already ailing, he was sent to Cherson in the Crimea, where he died in 655. Martin felt abandoned by those who should have supported him; his successor had been elected more than a year before his death. By that time, Maximus had already been arrested, likewise tried for treason and sent into exile in Thrace, where attempts were made to break his opposition to monothelitism. When that failed, he was brought back to Constantinople for trial. He was condemned as a heretic, mutilated and exiled to Lazica, where he soon died on 13 August 662. By the time Maximus died in exile, the emperor himself was in selfimposed exile from Constantinople. Around 662 Constans II and his court moved to Syracuse in Sicily. This attempt to abandon the beleaguered Constantinople and re-establish the court closer to the centre of the truncated empire recalls earlier plans by Heraclius, and shows that there was no sense that the Byzantine empire was now confined to the eastern Mediterranean. From his base in Sicily, Constans clearly intended to liberate Italy from the Lombards; before arriving at Syracuse, he had led a campaign in Italy. This had met with some success, though he failed to take Benevento and soon retired to Naples, from where he made a ceremonial visit to Rome. However, his residence in Sicily was extremely unpopular, imposing as it did an unwelcome financial burden on the island. There was also fierce opposition in Constantinople to the loss of the court, and in 668 Constans was assassinated by a chamberlain. Constans II was succeeded by his son Constantine IV. It was during Constantine’s reign that the Umayyad caliph Mu‘awiya made a serious attempt to complete the Arab expansion begun in the 630s, aiming to take Constantinople and with it destroy the only serious opposition to Muslim rule in the Mediterranean. After his victory over ‘Ali in the fitna, Mu‘awiya renewed his offensive against the Byzantine empire. By 670 the islands of 16

Riedinger (1982); see also below, p. 402, n. 29.

4. byz antium transf orming (600–700)

233

Cyprus, Rhodes and Kos, and the town of Kyzikos on the southern coast of the Sea of Marmara, had all been occupied by Arab naval forces. In 672 Smyrna was taken, and in 674 the main attack on Constantinople began. A large Muslim fleet blockaded the city, and for the next four years the same fleet was to blockade Constantinople, retiring in the winter to shelter off Kyzikos. Each year the defences of Constantinople held firm, and in the final naval battle, the Byzantines secured a major victory with the help of Greek fire. First mentioned in the sources on this occasion, Greek fire was a highly inflammable, crude oil-based liquid that was shot out at enemy ships, setting them ablaze.17 At the same time as this naval victory, the Byzantine army was able to surprise and defeat an Arab army contingent in Anatolia. Mu‘awiya was forced to break off his attack on Constantinople and sue for peace. This major victory for the Byzantines proved to be a turning point: the Arab threat to Constantinople receded for the time being and Byzantium’s prestige in the Balkans and the west was enhanced. Embassies from the khagan of the Avars, now restricted mainly to the Hungarian plain, and from the Balkan Slavs arrived in Constantinople, bringing gifts and acknowledging Byzantine supremacy. However, the situation in the Balkans was about to change. The Slavs based there had never formed any coherent political entity, though their presence confined imperial authority to Thessalonica and other coastal settlements. The Bulgars, a Turkic-speaking group whose homeland was to the north of the Sea of Azov, had long been a power among the nomadic peoples of the Eurasian steppes. The Byzantines had maintained friendly relations with them, and had supported them against the Avars. But with the arrival of another people – the Khazars – the Bulgars’ khanate began to split up, and one group led by Asparuch arrived at the Danube delta around 680, intending to settle south of the river in traditionally imperial territory. The Byzantines saw no threat in the Bulgars, but were unwilling to allow them south of the Danube. A Byzantine fleet arrived at the river mouth and troops moved up from Thrace, intending to expel the Bulgars. The Bulgars avoided open battle but, as the Byzantine forces withdrew, took them by surprise and defeated them. Constantine IV concluded a treaty with Asparuch, granting the Bulgars the territory they already held. As a result of this presence, several Slav tribes hitherto loyal to Byzantium recognised the overlordship of the Bulgars and became their tributaries, and a Bulgaro-Slav political structure started to develop, with its capital at Pliska. This independent, periodically hostile presence so close to the City, in principle able to control the route from the Danube delta to Constantinople, would prove a long-standing threat to the stability of the empire.

17

Haldon and Byrne (1977); Haldon (2006a).

234

the earlier empire

Figure 9a The imperial fleet burns the ships of Thomas the Slav with discharges of Greek fire (illustration from the Madrid Skylitzes, an illuminated manuscript copied in mid-twelfth-century Sicily)

Figure 9b Twenty-first-century Greek fire: experiment by John Haldon and colleagues to re-create Byzantium’s most feared weapon

The enforcement of monothelitism as imperial policy, though it secured papal acquiescence in the years immediately following Martin’s arrest and exile, was bound to prove ultimately unacceptable to the west, which saw the council of Chalcedon as endorsing the Latin Christology of Pope Leo I (440–61). By 680 Constantine had come to the conclusion that

4. byz antium transf orming (600–700)

235

religious unity with the west was more important than the fragile possibility of union with the monophysites – now mostly lost to the Umayyad caliphate – offered by monothelitism. He proposed to Pope Agatho (678– 81) the calling of an ecumenical council to condemn monothelitism. Agatho enthusiastically concurred, and held synods in Italy and England to prepare for the coming ecumenical council. Armed with these synodical condemnations of monothelitism, the papal legates arrived in Constantinople. The sixth ecumenical council met in Constantinople from 3 November 680 until 16 September 681. Monenergism and monothelitism were condemned, and the patriarchs Sergius, Cyrus, Pyrrhus, Paul and Peter anathematised, together with Pope Honorius. There was no word, however, of the defenders of the orthodoxy vindicated by the council, Martin and Maximus, who had suffered at the hands of Constans; nor were the emperors Constans or Heraclius mentioned. Constantine IV himself was hailed, at the final session, as a ‘new Marcian’ and a ‘new Justinian’. The latter part of Constantine’s reign saw the Byzantine empire regain a certain stability. In 684–5 he led a successful military expedition into Cilicia, forcing Caliph ‘Abd al-Malik to sue for peace and pay tribute to the Byzantines (see below, pp. 344, 381–2). Religious reconciliation with Rome led to peace with the Lombards in Italy, brokered by the pope. In North Africa, the Byzantines were able to halt the advance of the Arabs through alliances with Berber tribes, though this only bought time until the Berbers themselves converted to Islam. Constantine IV died in 685 and was succeeded by his son, Justinian II. It is worth noting that both Constantine IV and Constans II had deposed their brothers in the course of their reigns – in Constantine’s case, despite open opposition from senate and army – in order to secure the succession of their eldest sons. Justinian sought to build on the relative stability achieved by his father, leading an expedition into the Balkans and reaching Thessalonica. He continued the policy of both his father and his grandfather of transporting Slavs into Anatolia. He also transported some of the population from Cyprus to Kyzikos, depopulated during the siege of Constantinople, and ferried Mardaites from northern Syria and Lebanon to the Peloponnese and elsewhere. Whether or not Justinian was responsible for the breach of the truce with ‘Abd al-Malik in 692, he suffered military disaster when his Slav troops deserted. As a result, several Armenian princes once again acknowledged Muslim suzerainty. In 692 Justinian called a council, known as the quinisext or fifth-sixth council (see below, pp. 244–7). In so doing, he followed both his father’s example and that of his namesake, declaring his credentials as emperor and guardian of orthodoxy. This was also manifest in his coinage: the image of the emperor was displaced from the obverse of the coin to the reverse, and replaced with an image of Christ, the source of his authority as emperor.

236

the earlier empire

Figure 10 Nomisma of Justinian II, showing bust of Christ on the obverse for first time on gold coins (left); the emperor is shown on the reverse (right)

In 695, Justinian was overthrown in a palace coup and replaced by Leontius (695–8), the recently appointed strat¯egos of the theme of Hellas. Justinian had his nose slit and was exiled to Cherson, where his grandfather had earlier exiled Pope Martin. Leontius’ reign lasted three years, during which he witnessed the end of Byzantine rule in North Africa. That defeat, and the consequent loss of Carthage, provoked another rebellion in which Leontius was deposed in favour of Apsimar, the droungarios of the Kibyrrhaiotai fleet, who changed his name to the more imperial-sounding Tiberius. Tiberius II reigned from 698 to 705, during which time Asia Minor was subjected to continual Arab raiding. He was replaced by Justinian II, who returned with the support of the Bulgar khan Tervel, slipping into the City through one of its aqueducts. Justinian’s final six years were ones of terror and vengeance, brought to an end by a military coup; thereupon three military leaders succeeded one another for short and inglorious reigns, until the accession in 717 of Leo III, the emperor who subsequently introduced iconoclasm.18 ad ministrative change At the beginning of the seventh century the administration of the empire, both civil and military, was essentially what had emerged from the reforms of Diocletian and Constantine in the late third and early fourth centuries. By the end of the eighth century quite different forms of administration were in place. Although we have a fairly clear picture of early seventhcentury Byzantine administration, for the late eighth century the picture is less clear; and because the evidence is both sparse and open to diverse interpretations, the nature and pace of administrative change in this period is still a matter of debate. However, in general terms the change can be described as follows: at the beginning of the seventh century the empire 18 The emperors were: Philippikos (711–13), Anastasios II (713–15) and Theodosios III (715–17). On Leo III’s religious policy, see below, pp. 279–82.

4. byz antium transf orming (600–700)

237

was divided into provinces ruled by civil governors who, though appointed by the emperor, were responsible to the relevant praetorian prefect (the provinces being grouped into four prefectures), and the army was organised quite separately; at the end of the eighth century the empire was divided into districts called themes (themata), which were governed by a military commander (strat¯egos) who was responsible for both the civil and military administration of the province, and directly responsible to the emperor. Let us now look at the changes involved in more detail.19 Civil administration In the civil administration inherited from the reforms of Diocletian and Constantine, alongside the administration of the empire through the prefectures, there were also departments called res privata and sacrae largitiones, administered by counts (comites), who belonged to the imperial court (the comitatus). The comes rei privatae was responsible for all land and property belonging to the state, including the collection of rents and claiming for the state all property that lapsed to it. Originally the comes rei privatae had been concerned with the emperor’s private property, as the name suggests, but the distinction between that and state property had long been elided. The comes sacrarum largitionum controlled the mints, the gold (and probably silver) mines and the state factories in which arms and armour were decorated with precious metals. He was also responsible for paying periodic donatives in gold and silver to the troops, and dealt with the collection or production of clothing and its distribution to the court, the army and the civil service. The praetorian prefects were responsible for the fiscal administration of the prefectures into which the empire was divided. These prefectures consisted of provinces, governed by governors (with various titles), and were themselves grouped into dioceses, governed by vicarii. The praetorian prefects were responsible for the rations, or ration allowances (annonae), which formed the bulk of the emoluments of the army and the civil service, and also for the fodder, or fodder allowances, of officers, troopers and civil servants of equivalent grades. They had to maintain the public post, and were responsible for public works, roads, bridges, post-houses and granaries which did not come under the care of the urban prefects in Rome and Constantinople, the city authorities in the provinces, or the army on the frontiers. In order to do all this, the praetorian prefects had to estimate the annual needs of their prefecture and raise the money through a general levy, or tax, called the indiction. The whole operation of raising this tax and servicing the running of the empire was overseen by the praetorian prefects, 19 What follows is deeply indebted to Haldon’s subtle and powerfully argued account of the administrative changes: Haldon (1997a), pp. 173–253. See also below, pp. 266–7.

238

the earlier empire

who delegated it to their vicarii and governors. Only the praetorian prefect in whose prefecture the emperor and his court were located was attached to the court; once the court was permanently settled in Constantinople, this meant the praetorian prefect for the east (Oriens). Also influential in the comitatus were senior officials of the sacrum cubiculum, the eunuch chamberlains (cubicularii). By the end of the eighth century, the fiscal administration was organised rather differently. The distinction between the public and the ‘sacred’ (i.e. pertaining to the person of the emperor) had gone, and instead of the res privata, the sacrae largitiones and the prefectures, there were a number of departments, or sekr¯eta, of more or less equal status. Besides the sakellarios and the heads of the three great departments – the logothete of the Drome (tou dromou), the general logothete (tou genikou) and the military logothete (tou strati¯otikou)20 – there were several other senior administrators. Among these were two treasurers, the chartoularios of the sakellion, in charge of cash and most charitable institutions, and the chartoularios of the vestiarion, in charge of the mint and the arsenal. Other heads of state establishments included the great curator (megas kourat¯or), in charge of the palaces and imperial estates, and the orphanotrophos, in charge of orphanages. In addition there was an official called the pr¯otasekr¯etis, in charge of records. Directly responsible to the emperor were the principal magistrates, the City prefect (responsible for Constantinople), the quaestor (in charge of the judiciary) and the minister for petitions (who dealt with petitions to the emperor). A rather obvious, and superficial, change is that of language: whereas the older system used Latin titles, the new system used predominantly Greek titles. This reflects the change in the official language of the empire from Latin, traditional language of the Roman empire, to Greek, language of Constantinople and the Hellenistic east; a change dating from the time of Justinian. More deeply, it can be seen that the change involved a reshuffling of tasks, so that they all became subject to a fundamentally civil administration based on the court. The genikon and strati¯otikon derived from the general and military departments of the prefectures (in fact, the prefecture of the East, as we shall see); the sakellion from the sacellum, the personal treasury of the emperor within the sacrum cubiculum; and the vestiarion from the department of the sacrum vestiarium within the sacrae largitiones. The position of the sakellarios perhaps gives a clue to the nature of the changes. In charge of the emperor’s personal treasury, this official’s eventual rise to pre-eminence was a function of his closeness to the emperor and suggests a shift from an essentially public administration, its structure determined by the need to administer a far-flung empire, to an 20

On these officials, see below, p. 273.

4. byz antium transf orming (600–700)

239

administration focused on the court, in which the empire is almost reduced to the extent of an imperial command. The background to this is the dramatic shrinking of the empire in the first half of the seventh century. The loss of the eastern provinces followed by North Africa and, by the end of the eighth century, Italy too, together with the Slavs’ occupation of the Balkans and the emergence of the Bulgar realm south of the Danube, meant that the Byzantine empire had shrunk to the rumps of two prefectures, of the East and Illyricum. Reorganisation of the civil administration took the form of Constantinople incorporating the administrative offices of the empire into a court structure. The growing power of the sakellarios can be traced back to the time of Justinian; by the mid-seventh century, judging from the role he played in the trial of Maximus the Confessor, he was a powerful courtier who took personal charge of matters of supreme importance to the emperor. Logothetes also feature in the sources from the early years of the seventh century, but officials bearing traditional titles, such as praetorian prefect, not to mention civil governors of provinces, continue to appear in the sources well into the eighth century. This would suggest that there was a substantial period of overlap, with the new administration emerging while the old administration still retained some of its functions. However, the full picture only emerges when we consider the changes in the military administration. Military administration The reforms of the Roman army by Diocletian and Constantine separated it from the civil administration, so that governors of provinces no longer commanded a provincial army, although they were still responsible for raising funds to support it. The army was divided into two parts: there were troops protecting the borders, the limitanei, under the command of duces; and there was a field army, the comitatenses, which was mobile and organised in divisions under the command of the magistri militum. In addition there were the palace troops and the imperial bodyguard, whose titles changed throughout the fifth and sixth centuries. By the ninth century a quite different system had emerged, with the army divided into divisions called themes, based in provinces also called themes (themata), each under the command of a strat¯egos. There is no general agreement about how quickly this change took place, or why: was it the result of some planned reorganisation, or simply a fumbling reaction to the problems of the seventh and eighth centuries? There is, however, general dissent from the theory which once commanded much support, and is associated with the name of the great Byzantinist George Ostrogorsky, which saw the thematic army as the result of a deliberate reorganisation of army and empire by Heraclius. The result, supposedly, was a peasant army,

240

the earlier empire

based in the themes, in which land had been allotted to peasant families as small holdings, in return for each family providing and equipping a soldier.21 This somewhat romantic idea of the middle Byzantine empire resting on the popular support of a free peasantry has been generally abandoned. The transition is now thought to have occurred after Heraclius’ reign, and change was probably gradual. Part of the problem is terminology. The word theme originally meant a military unit, and references to themes in sources relating to the seventh century may refer to military units, rather than to the land where they were stationed. But even if it seems that the reference is to territory, we cannot be sure that such a reference is not an anachronism, since our sources date from the ninth century when territorial themes were in place. As with the changes in civil administration already discussed, it is possible – indeed likely – that the two sets of arrangements overlapped; even though there are references to strat¯egoi and themes in the seventh century, there is still mention of provinces (eparchiai), governors, and use of such titles as magister militum well into the eighth century. While it is impossible to provide a detailed timescale, one can reasonably suggest that the themes developed in the following way. After the Byzantine army was defeated by the Arabs, the troops retreated over the Taurus mountains into Anatolia. The years following the defeat saw continual raiding by Muslim forces into Anatolia, leading finally in the 660s and 670s to a concerted attempt by Mu‘awiya to advance across Asia Minor and take Constantinople (see above, pp. 232–3 and below, p. 372). In this prolonged crisis, the Byzantine armies were stationed in the provinces of Asia Minor. They would have been provisioned in the traditional way, by a levy raised by the local governors from the civilian population. The areas that came to be called the themes of the Armeniakoi and the Anatolikoi were the groups of provinces where the armies commanded by the magistri militum per Armeniam and per Orientem took up their stands. The theme of the Thrakesioi covered the provinces in western Anatolia to which the army of the magister militum per Thraciam withdrew after the Arab victories in Syria and Palestine. The theme of the Opsikion was made up of the armies of the magistri militum praesentales, some of which had probably long been established in the area just across the Bosporus from Constantinople. The name derives from the title of the officer (comes Obsequii) who, during the reign of Heraclius, was appointed to command the praesental armies on the emperor’s behalf. The Karabisianoi, the fleet, formed part of the old quaestura exercitus, probably based at Samos (see also below, p. 267). It seems likely that the army corps took up the positions into which they would become embedded sometime 21 Brief account in Ostrogorsky (1968), pp. 95–8. More detail in Ostrogorsky (1958). The romanticism of Ostrogorsky’s vision emerges more clearly in Ostrogorsky (1962). See also below, p. 266.

4. byz antium transf orming (600–700)

241

around the middle of the seventh century. Why and at what stage the civil administration declined, to be replaced by the military government of the strat¯egoi, is much less clear. Presumably the overriding need to supply a standing army, together with the decline of the ancient, city-based economy, meant that the strat¯egoi, backed up by the increasingly powerful officials forming part of the imperial court, gradually assumed the functions of the old governing elite. The elite lost much of its raison d’ˆetre because of the court-centred nature of the civil administration. Legal administration Given the profound changes in civil and military administration beginning in the seventh century, at first sight it is surprising that so little legislative activity seems to have occurred in this century. One has nothing to set beside the major attempts at legislative reform of the fifth and sixth centuries, embodied in the Codex Theodosianus and the Codex Iustinianus. Apart from the Farmer’s law, whose date is disputed and which is anyway a compilation of materials from Justinian’s era and before, the emperors seem to have initiated very little legislation; what remains is primarily ecclesiastical in nature, for example Heraclius’ edict of 632 requiring the compulsory conversion of the Jews, his Ekthesis, and Constans II’s Typos (see above, pp. 229, 231). In contrast, the quinisext council called by Justinian II represents a major recapitulation of canonical legislation, which can be compared with the law-code of Justinian II’s great predecessor (see above, p. 108). The explanation for the lack of legislative activity in the secular sphere is probably to be found in the dual nature of Roman legislation. Legislation was not only a body of rules governing day-to-day behaviour, but more importantly a way of enunciating the world-view and set of values embraced by the Roman – or Byzantine – empire. As John Haldon has put it: Seen from this perspective, the legal ‘system’ became less a practical instrument for intervening in the world of men in order to modify relationships or individual behaviour, but more a set of theories which represented a desired (if recognisably not always attainable) state of affairs. Emperors needed to issue no new legislation, therefore, but rather to establish (or to re-establish) the conditions within which the traditional system would once again conform to actual practice.22

religion and the church It is generally recognised that from the later sixth century onwards there was an increasing desire to have direct access to the power of the holy. 22

Haldon (1997a), p. 259.

242

the earlier empire

Again, this cannot be demonstrated beyond peradventure, since the means of access – cults of saints and their relics, and perhaps even the veneration of icons – were already well established by the sixth century. Traditionally, imperial authority had been justified by the divinely protected status of the emperor, expressed through an imperial cult. The Christianisation of the imperial cult tended rather to enhance its authority than to diminish it, since the representative of the only God was hardly reduced in status in comparison with a divine emperor holding a relatively lowly position in the divine pantheon.23 It seems to be demonstrable that this Christian imperial authority and that of the hierarchy of the Christian church, which was closely bound up with it, were reinforced by holy men and holy images claiming immediate access to supernatural power. It seems, too, that even traditional imperial authority was increasingly expressed through images that spoke of a more immediate sacred authority. This becomes evident at the beginning of the seventh century from the use of icons of Christian saints as military banners, especially of the Mother of God; from the way in which Christian armies are seen as fighting for the Virgin, with her protection and even her assistance; and from the role claimed for the Virgin as protector of the city of Constantinople. A sacralisation of authority is also manifest in the increasing significance attached to coronation by the patriarch in the making of an emperor; this was always conducted in a church from the beginning of the seventh century, and in the Great Church of St Sophia from 641. The institutional church, indeed, may well have felt itself threatened by the proliferation of the holy in the seventh century: the church in the Byzantine east certainly failed to establish the kind of control over the holiness present in saints, their images and their relics, that the popes and bishops had won in the western church.24 But if there is little evidence of tension between the proliferation of the holy and the church hierarchy in the Byzantine east in the seventh century,25 there is certainly evidence of tension between the centre and the periphery in geographical terms. Despite the wealth of theological literature that survives from the seventh century, we know little about theology at the capital, for the simple reason that by the ninth century no one in Constantinople wanted to be reminded of it. Theology in Constantinople was subservient to the emperor, and to the politically inspired doctrines of monenergism, monothelitism and, in the next century, iconoclasm. Resistance to all of these – a resistance that was finally recognised as ‘orthodoxy’ – came from the periphery, and in the long term especially from the monks of Palestine, who had long been known for their commitment to Chalcedonian orthodoxy. This fact had curious long-term consequences for orthodox 23 25

24 See Brown, P. (1976). See Dagron (2003). In the eighth century, the ready support the iconoclast emperors seem to have found among the higher clergy may possibly be evidence of a reaction on the part of the church hierarchy.

4. byz antium transf orming (600–700)

243

Byzantium, and is worth pursuing briefly here. Resistance to monenergism began with Sophronius, who had been a monk in Palestine and later became patriarch of Jerusalem; resistance to monothelitism was led by Sophronius’ disciple Maximus, whose impact on the orthodox in Palestine was such that they were called Maximians by the monothelites in Syria and Palestine.26 In the second half of the seventh century dyothelite (‘orthodox’) Christians in Palestine found themselves in a new situation. Previously they had been adherents of an imperial orthodoxy that had been backed up, in the last resort, by force. Now they found themselves in a situation where their religious position was opposed by other Christian groups – monophysite, monothelite and even Nestorian – and by non-Christians like Jews, Samaritans, Manichees and, eventually, by Muslims. They had both to defend what they believed in and to work out exactly what their faith amounted to. In order to do this, they had to pay attention to matters of logic and definition, for the only way to defend and commend their position was by convincing others; they could no longer appeal to the secular arm. One element in this refining of the presentation and understanding of the Christianity of the ecumenical councils was dialogue with – or polemic against – the Jews. After a long period when there was scarcely any dialogue with Jews, or even simple refutation of Judaism, the second half of the seventh century witnessed an extraordinary burgeoning of such works. Most come from the provinces seized by the Arabs: Syria, Palestine, the Sinai peninsula and Cyprus. It is clear from some of these works that Jews themselves took the initiative, forcing Christians to produce fresh defences of doctrines such as the Trinity and practices such as veneration of saints, relics and icons.27 Alongside such doctrinal clarification there was also celebration of the doctrines of Christianity in liturgical poetry, which came to form the backbone of monastic worship and again stemmed principally from Palestine. This eventually became the worship of the orthodox – that is Byzantine – church, and of those churches which learnt their Christianity from Byzantium. The crucial century for this definition, defence and celebration of orthodoxy was the period from 650 to 750. It is epitomised in the works of John Damascene, an Umayyad civil servant turned Palestinian monk, who thought of himself as a Byzantine Christian. Its first test was the iconoclasm of Byzantine emperors, beyond whose political reach these Christians lived.28

26 The view that Maximus was himself a Palestinian, propounded in the Syriac Life (Maximus the Confessor, Syriac Life), seems to be losing credibility among scholars. 27 See D´ eroche (1991); Cameron, Averil (1996a). 28 For these developments and John Damascene’s part in them, see Louth (1996b). See also below, p. 283.

244

the earlier empire

As we have seen, this form of Christianity was called Maximianism by its enemies, but it owed more to Maximus than simply its attachment to dyothelite Chalcedonianism, as declared at the Lateran synod of 649 and vindicated at the sixth ecumenical council of 680–1 (see above, pp. 231, 235). For Maximus’ genius as a theologian was to draw together the several strands of Greek theological reflection into an imposing synthesis. One strand in this synthesis was the dogmatic theology of the great patriarchs of Alexandria, Athanasius and Cyril, which formed the basis for the dogmas endorsed by the ecumenical councils from the fourth to the sixth centuries. Another strand was the Christian Hellenism of the fourthcentury Cappadocian fathers, Basil of Caesarea, Gregory Nazianzen and Gregory of Nyssa. A further strand was constituted by the ascetic wisdom of the fourth-century Egyptian desert fathers; and of their successors in the Judaean desert to the east of Jerusalem, in the coastal desert of Gaza and the barren mountains of the Sinai peninsula. These three strands Maximus wove together, the final tapestry being shot through with the Neoplatonic metaphysics of Pseudo-Dionysius the Areopagite, believed to be in reality an early sixth-century Syrian monk (see above, pp. 111–12). It was this theological vision of Maximus which inspired the more soberly expressed, even dry doctrinal synthesis that we find in John Damascene. Maximus’ vision, in which humankind, the cosmos and the scriptures themselves were all interrelated, was reflected in the domed interior of the Byzantine church. In that space, as Maximus explained in his reflections on the divine liturgy called the Mystagogia, the liturgical ceremonies involving the clergy and the people celebrated the whole unfolding of the Christian mystery, from creation to Christ’s second coming, in a way that probed the depths of the human heart and illuminated the mysteries of the cosmos.29 But to turn from what may seem giddy heights – albeit expressed in such gesture, movement, melody and colour as to impress the simplest of Byzantine Christians – we see a more detailed picture of the life of the Byzantine church in the seventh century emerging from the 102 canons of the quinisext council, called by Justinian II in 692.30 Like his predecessor and namesake, Justinian II wished to mark his reign and manifest his exercise of imperial power by calling an ecumenical council. Hitherto, all councils regarded as ecumenical had been called to deal with some pressing doctrinal issue, but with the monenergist/monothelite controversy now settled, there was no doctrinal issue to provide occasion for an ecumenical council. However, the previous two ecumenical councils, the second and third of Constantinople, had issued only doctrinal canons, whereas all the 29 30

For an introduction to the theology of Maximus, see Louth (1996a). DGA, ed. and French tr. Joannou, I.1, pp. 98–241; tr. in Nedungatt and Featherstone (eds.) (1995), pp. 41–185.

4. byz antium transf orming (600–700)

245

earlier ones had dealt with both doctrinal and disciplinary issues. Thus the council Justinian eventually called, which issued only disciplinary canons, was regarded as finishing off the work of the previous two councils (the fifth and the sixth ecumenical councils) and was therefore called the quinisext council. It is also known as the Trullan council (in Trullo) from the domed chamber (troullos) in the palace where proceedings took place. The 102 canons issued by the council cover many aspects of the life of Christians, both their religious duties and their behaviour in secular life. The first two canons affirm and define the existing tradition, of which the rest of the canons constitute a kind of completion: canon 1 affirms the unchanging faith defined at the previous six ecumenical councils; and canon 2 confirms the body of disciplinary canons already accepted by the church.31 The rest of the canons complete this body of canonical material, and the whole body of legislation constituted by this council can be compared in some ways to Justinian’s code, in that it is intended as the final statement of an ideal of Christian life, expressed through much quite detailed legislation. It remains the foundation of the canon law of the orthodox church. In this context it is worth drawing attention to the last canon, which affirms that the administration of penalties in accordance with the canons must take account of the quality of the sin and the disposition of the sinner, for the ultimate purpose of canon law is to heal, not simply to punish. This canon reaffirms a principle already expressed in earlier canons,32 usually called the principle of ‘economy’ (oikonomia). It is not unlike the way in which in seventh-century secular law used Justinian’s code as an ideal, trying to fit the ideal to concrete issues rather than promulgating fresh legislation (see above, p. 241). One guiding principle of the canons of the quinisext council was to define the practices of the Byzantine church in conscious opposition to the developing customs of the Latin west. For instance, canon 55 forbids fasting on Saturdays and Sundays, except for Holy Saturday, and is explicitly directed against the practice of fasting on Saturdays during Lent found in the city of Rome. More important are the canons that allowed for a married pastoral clergy. Although restricted to priests and deacons – since on appointment to the rank of bishop, a married man had to separate from his wife, who took the veil (canons 12 and 48) – this too is in conscious opposition to the Roman canons; it would be some centuries, however, before a celibate priesthood was strictly enforced in the western church. A similar independence of Rome is manifest in canon 36. This prescribed the order of the patriarchates and, following the canons of the first ecumenical council of Constantinople (canon 3) and the ecumenical council 31 32

For a succinct account of the development of Christian canon law, see Louth (2004). Canons of St Basil, no. 95, DGA, ed. and French tr. Joannou, II, pp. 193–8; see also no. 3, DGA, II, pp. 100–1; Canons of St Gregory of Nyssa, no. 1, DGA, II, pp. 203–9.

246

the earlier empire

of Chalcedon (canon 28, which had been repudiated by Rome), ranked Constantinople second after Rome, with equal privileges. Although the papal legates accepted the canons, Pope Sergius I (687–701) refused to sign them and Justinian’s furious attempt to enforce papal consent only exposed the limits of his power in Italy. Sergius’ introduction of the singing of the Agnus Dei into the mass at Rome is perhaps to be seen as a snub to the council (see canon 82, discussed below).33 Although Pope John VII (705–7) seems to have accepted the canons of the council in 705, when Justinian was restored to the imperial throne, this represented no lasting endorsement of them by the western church. Other canons regulated the life of the local church, still understood as essentially an urban church ruled by a bishop although, as we have seen, the reality of the city was fading fast. Urban churches were grouped into provinces, under the leadership of a metropolitan bishop, and these provinces were to convene once a year (canon 8). Bishops were to live in their sees, and must return to them as soon as possible if they fled during ‘barbarian’ raids (canon 18). This anxiety that the bishop should stick to his city was partly to ensure his continuing pastoral care, but also his control of the church’s financial interests; the local churches were frequently considerable landowners with their estates being administered by the bishop. The requirement that bishops reside in their own sees was taken seriously, as is evident from the more abundant later evidence, especially from the Komnenian period, when the empire was even more focused on Constantinople and provincial sees were regarded as exile by their bishops.34 There are also canons against selling the sacraments and purchasing church office (what the west later called simony: canons 22–3). Legislation concerning monasticism, like much earlier legislation, attempted to confine monks to their monasteries and control the power of holy men (canons 40–9). Legislation concerning the laity forbade various entertainments, such as playing dice (canon 50); watching mimes, animal fights or dancing on stage (canon 51); the observance of civic ceremonies such as the Calends, Vota or Brumalia, which had pagan associations, as well as female dancing in public, dancing associated with pagan rites, cross-dressing, the use of comic, satyr or tragic masks, and the invocation of Dionysus during the pressing of grapes for wine (canon 62). All of this the church regarded as ‘paganising’, though such practices should probably not be thought of as the survival of paganism outright, but rather the continuance of traditional forms of worship involving the laity.35 Canons also forbade the confusion of traditional liturgies with the Christian sacraments – for example canon 57 forbidding the offering of milk and 33 34

LP, LXXXVI.14, ed. Duchesne, I. p. 376; tr. Davis, I, p. 89. 35 See Haldon (1997a), pp. 327–37. See Angold (1995), pp. 139–262.

4. byz antium transf orming (600–700)

247

honey on Christian altars – and others regulated the institution of marriage and the circumstances of divorce (canons 53, 54, 72, 87, 92 and 93). Several canons dealt with relations between Christians and Jews. Canon 11 forbade eating unleavened bread with Jews, making friends with them, consulting Jewish doctors or mixing with Jews in the baths; canon 33 forbade the ‘Jewish’ practice of ordaining only those of priestly descent. Both these canons illustrate the way in which Jews were permitted to exist, but separately from the orthodox society of the empire. In fact, the seventh century had seen the beginning of a more radical policy towards the Jews: forced baptism on pain of death. Maximus the Confessor expressly objected to such a policy introduced by Heraclius in 632,36 and the policy was introduced again in the eighth and tenth centuries, by Leo III (717–41) and Romanos I Lekapenos (920–44) respectively. But the more normal Byzantine attitude to the Jews, to be preserved as a standing witness to the truth of Christianity with limited civil rights, is that envisaged by the canons of the Trullan council.37 Two canons bear witness to the place of religious art in the Byzantine world. Canon 100 forbids pictures that excite immoral pleasure, and emphasises how easily the bodily senses move the soul. Canon 82 is concerned with religious paintings and forbids the depiction of Christ as a lamb, a popular form of religious art that picked up the words of John the Baptist about Jesus as the ‘lamb of God who takes away the sin of the world’ (John 1: 36). However, the canon argues, such symbolism has been fulfilled since God has come in human form; now the reality of the Incarnation is to be expressed by depicting the Incarnate Word as a man. Such concern for the content of religious images, expressed in theological terms, prefigures the controversies of the next two centuries caused by iconoclasm. The comparatively settled picture of Christian life in the Byzantine empire presented by the canons of the quinisext council is not, however, the whole story. The second half of the seventh century saw the production of apocalyptic texts, composed in Syriac. One of these, soon translated into Greek and subsequently into Latin, was ascribed to the early fourthcentury bishop Methodius (of Olympus, according to the Syriac original; of Patara, according to the Greek translation).38 The Apocalypse of PseudoMethodius responds to the loss of the eastern provinces to the Arabs – termed Ishmaelites or ‘wild ass of the desert’ – by recounting the history of the Middle East since biblical times. It predicts the final overthrow of the Ishmaelites at Jerusalem by the king of the ‘Greeks’ (so the Syriac; ‘Romans’ in the Greek version), whose victory will usher in the end of the world.39 36 38 39

37 See above, p. 116. See Devreese (1937). This confusion as to his see is also found in manuscripts of authentic writings by Methodios. See Alexander (1985); Syrian chronicles, tr. Palmer et al., pp. 222–50.

248

the earlier empire

The emergence of such apocalyptic hopes and fears at the end of the seventh century contrasts sharply with the spirit of the early sixth-century Chronicle of John Malalas, written partly to demonstrate that the world had survived the transition from the sixth to the seventh millennium from the creation (i.e., c. ad 500) without disaster. The end of the seventh century saw the Byzantine empire still in a process of transition and redefinition: the Arab threat to Constantinople would continue well into the eighth century, and iconoclasm is probably to be seen as a further stage in the empire’s search for its identity and ways of expressing this in the aftermath of the crisis of the seventh century.40 But there were scarcely any signs of incipient iconoclasm at the end of that century. The quinisext council invested a clearly articulated theological significance in religious art, and the process observed since the end of the sixth century of authenticating political authority by imagery invoking the supernatural was taken a stage further at the end of the seventh century: Christ’s image appeared on the obverse of imperial coinage, the imperial image being consigned to the reverse (see fig. 10 above, p. 236). But the structures of the society that would eventually emerge from this period of crisis can already be seen, albeit in inchoate form; so too can some of its limitations, when compared with Justinian’s vision of the Roman empire which it claimed to embody. Already there is a sense in the legislation of the quinisext council that the customs of those Christians who looked to Constantinople were different from those who looked to Rome: a gap that would widen as Rome moved from the Byzantine emperor’s sphere of influence to that of the Franks. The Mediterranean Sea was no longer to unite the territories that bordered it, but would come to separate the several societies which claimed the heritage of that lost unity. 40 Not all scholars accept that the seventh century should be regarded as a crisis for the Byzantine empire: see Treadgold (1997), pp. xvi, 287–413; Treadgold (1990). On iconoclasm, see below, pp. 278–84.

PA RT I I

T H E M I D D L E E M P I R E c. 7 0 0 – 1 2 0 4

CHAPTER 5

STATE OF EMERGENCY (700–850) mar ie - fran ce au z e´ py

an impenetrably d ark age? The so-called Byzantine iconoclast period is a ‘dark age’ whose obscurity is only randomly illuminated by the few remaining sources, and even these are difficult to interpret. Apart from in Italy, no archives have been preserved. The contemporary sources comprise two chronicles – that of Theophanes the Confessor, covering the period up to 813, and the Breviarium of Patriarch Nikephoros, which stops at 769 – and an account of Leo V’s reign whose author is known as the ‘Scriptor incertus’. The only near-contemporary chronicle for the reigns of Michael II and Theophilos is that of George the Monk, probably completed in 846 and reworked in 871–2,1 with Theophanes Continuatus being the most important of the later chroniclers to cover this period. Other sources include the Acts of the second council of Nicaea (787); these contain several extracts from the ruling of the iconoclast council of Hieria (754), which they set out to refute. Further sources include a legal code called the Ecloga (741); the Farmer’s law (or Nomos georgikos) – though this is not dated with precision; the Taktikon Uspensky (842–3); the correspondence of the monk Theodore the Stoudite, and of Bishop Ignatios of Nicaea (known as Ignatios the Deacon) from the first half of the ninth century; numerous saints’ Lives; and the polemical anti-iconoclast literature.2 We can add to these sources others of Arab, Syriac, Armenian and Greek origin from the caliphate, as well as several inscriptions and numerous seals. This slightly simplified overview of the internal, written sources for the 150 years of iconoclasm reveals both the paucity of material, and how inadequate it is for understanding the profound transformation of the empire in this period – a transformation demonstrated by the fact that, from the 850s onwards, nothing was ‘as before’, even if it is difficult to date the reforms whose effects historians observe. This lack of source-material forces us either to project forwards, based on the situation in the seventh century, or backwards, from the state of affairs in the second half of the 1 2

Afinogenov (2004). Introduction to the sources: PMBZ, Prolegomena; Brubaker and Haldon (2001).

251

252

the midd le empire Danub e

Pliska

Beroia M a r i ts a mo Stry

Va

Mesembria Anchialos Bersinikia Develtos Skopelos

Adrianople

r

n

da

r

Markellai

Probaton

Bulgarophygon Thessaloniki Mount Athos

Cephalonia

Arkadiopolis Selymbria Heraclea Constantinople

Abydos

Patras Corinth Argos Aegina Sparta

Cape Malea

0 0

100

200 100

300 km 200 miles

Thira

MEDITERRANEAN

SEA

Map 12a The empire in the eighth and ninth centuries: key towns in the Balkans

ninth, and the history of eighth-century Byzantium tends to be highly hypothetical, an overlapping of past and future, rendering the period itself virtually non-existent. As a result, while recognising that the period was one of profound institutional restructuring, historians are hesitant about gauging the continuity or discontinuity of these institutions from the late Roman to the medieval empire.3 3 Oikonomides (1996a); Oikonomides (2002); Haldon (1999a), pp. 107–11; Brandes (2002a), pp. 480–98; Haldon (2003a), pp. 727–8.

5. state of emergency (700–850)

B L A C K

253

S E A Sinope

Amastris Hieron Hieria

Nikomedeia Nicaea Malagina Prousa Ankyra

Kyzikos

Nakoleia Pergamon Akroinon Sardis Ephesos

Koloneia

Sebasteia

Amorion

Ikonion

Chonai

Charsianon

Trebizond Dazimon

Caesarea Rodandos S Koron AIN NT Tyana OU M Lulon S U Cilician Gates R Ulukisla

TA

U

nder Maea

Amaseia Euchaita

Side

Kibyrrha

Seleukia

M E D I T E R R A N E A N

0 0

100

200 100

300 200

400

S E A

500 km 300 miles

Map 12b The empire in the eighth and ninth centuries: key towns in Asia Minor

This period of history is further obscured by the shadow of religion, to the extent that it takes its name from imperial religious doctrine – iconoclasm – rather than from the Isaurian or Amorian dynasties. Thus our sources are not only sparse, they are also biased; with the exception of juridical or administrative documents, they were written by the iconoclasts’ enemies, the iconodules, and they are all of clerical or monastic origin. They paint a picture of a period in which religious questions obscure everything else, and the Isaurian emperors Leo III (717–41) and Constantine V (741–75), who initiated and championed iconoclasm, are subject to virtual damnatio

254

the midd le empire

memoriae.4 No pro-Isaurian texts remain, apart from prescriptive ones, and our information about them is both minimal and hostile. All this conspires to make the eighth century not only an empty period, but also off-putting; nothing positive could possibly have happened then. Although Theodora’s defence of her husband Theophilos helped mitigate the damnatio memoriae imposed on the Amorian emperors Michael II (820–9) and Theophilos (829–42) because of their iconoclasm, nonetheless the sources – all written under their successors, the Macedonians – are highly critical of them simply for being predecessors of Basil I. This negative presentation has had a lasting effect on historical writing about the period, especially the eighth century; for example, archaeological finds from these dark centuries were, until recently, dated either to the seventh or the ninth century, exaggerating the dearth of sources for the intervening years. Most important of all, however, has been the presentation of the period’s historical narrative in terms of a history of the church: essentially the heretical emperors’ persecution of the church for venerating icons, and the valiant defence of the institution by monks and patriarchs. Modern historiography has tended not to question this presentation or the periodisation it imposes. ‘Iconoclasm’ (730–843) has been broken down into two periods, the first (730–87) mainly covering the Isaurian dynasty, and the second (815–43) the reigns of Leo V (813–20) and his Amorian successors, separated by an iconodulic interval which began with the second council of Nicaea in 787. Although this division should be questioned for imposing a religious frame of reference on a period characterised principally by a struggle for survival against enemies who threatened the empire’s very existence, it is nevertheless consistent with the course of events. Against the background of long-term structural reforms of domestic policy, one can indeed distinguish three different epochs: the first, under the great Isaurians Leo III and Constantine V, one of violence when the empire was saved from destruction; the second, under Irene and her successors, a time when the earlier period’s gains dwindled away and war loomed large; and finally, in the early ninth century, an era of returning prosperity, when the spectre of war receded. At the same time the empire’s geopolitical situation changed completely. Whereas at the end of the seventh century the empire could still, albeit with difficulty, lay claim to being universal, by the mid-ninth century it had become a Balkano-Mediterranean state. Even if throughout this period, those whom we call the Byzantines continued to call themselves Romans, others began calling them Greeks.5 4 5

Rochow (1994), pp. 123–71. Opus Caroli, IV.28, ed. Freeman and Meyvaert, p. 557; Paul the Deacon, Historia Langobardorum, IV.46, V.10–11, VI.57, ed. Bethmann and Waitz, pp. 135, 149–50, 185; tr. Foulke, pp. 200, 223–5, 303.

5. state of emergency (700–850)

255

earthquake, pl ague and continuous war fare The eighth century was not only a time of obscurity, but also one of adversity, when man and nature conspired to bring the empire almost to the point of extinction. A series of natural catastrophes afflicted Constantinople and its hinterland in the middle of the century. An earthquake brought down the walls of the City on 26 October 740. The Justinianic plague (see above, p. 123; below, pp. 478–9) returned one last time; starting in Mesopotamia and travelling through Sicily and the Peloponnese, the epidemic reached the capital in 747, emptying the City of its inhabitants.6 The winter of 763–4 was so harsh that the Black Sea and the Sea of Marmara froze, with huge icebergs on the Bosporus threatening the City’s sea walls during the thaw. In 766, a drought affected Constantinople’s water supply.7 As if nature’s depredations were not enough, man brought incessant warfare to the empire. At the beginning of the eighth century the Arabs were waging a war of annihilation against the empire. Enormous Arab land and sea forces encircled Constantinople in 717, attempting unsuccessfully to seize it, and this operation was repeated at Nicaea in 727.8 There followed almost ritualised warfare in Asia Minor, with annual raids by the caliph’s armies; although they seldom succeeded in capturing Byzantine cities, they ravaged the countryside and carried off the population and livestock. The Arabs could also mount both large-scale invasions – Harun al-Rashid’s expedition in 782 reached as far as Chrysopolis, opposite Constantinople9 – and sea raids, such as those launched on the Sicilian coast by ships from Ifriqya. The Isaurians managed to save the empire by raising an army which could go to the aid of besieged cities, but which was also capable of defeating enemy armies in open country, as it did at the battle of Akroinon in 740 (see below, p. 386).10 During the upheavals caused by the transition of power from the Umayyads to the Abbasids, the Byzantines took the offensive. They raided beyond the Taurus in 751 to Melitene and in either 754 or 755 to Theodosioupolis, transplanting these cities’ populations.11 The offensive also took place by sea. The fleet of the Kibyrrhaiotai destroyed an Arab fleet 6 Nikeph., ch. 67, ed. and tr. Mango, pp. 138–41; Theoph., ed. de Boor, I, pp. 422–4; tr. Mango and Scott, pp. 584–6; Stathakopoulos (2004), pp. 382–5. 7 Theoph., ed. de Boor, I, pp. 434, 440; tr. Mango and Scott, pp. 600–1, 607–8. 8 Mango (2005). 9 Theoph., ed. de Boor, I, p. 456; tr. Mango and Scott, pp. 628–9; Brooks (1900), pp. 737–9. 10 Nikeph., ch. 61, ed. and tr. Mango, pp. 130–1; Theoph., ed. de Boor, I, pp. 405–6, 411; tr. Mango and Scott, pp. 559–62, 570–2. 11 Nikeph., chs. 70, 73, ed. and tr. Mango, pp. 142–7; Theoph., ed. de Boor, I, pp. 427, 429; tr. Mango and Scott, pp. 589–91, 592–4; Lilie (1976), pp. 164–5.

256

the midd le empire

off Cyprus in 748;12 and the organisation of a Sicilian fleet during the 750s put paid to half a century of incessant Arab raiding on Sicily.13 Even after the empire had been saved from annihilation and a lasting border between the two empires had been drawn at the Taurus, warfare continued: but now it was waged by new enemies. On the Arab side, the annual raids in Asia Minor continued to ravage Cappadocia and were particularly dangerous when the empire was weak, or the caliphate in a strong position. Such was the case under Empress Irene (797–802) when, in 798, a detachment got as far as Malagina and stole a saddle from the imperial stables. And during the revolt of Thomas the Slav in 823, Thomas’ supporters allowed Arab raiders to reach Bithynia.14 Finally, there were two such emergencies under Theophilos: when Caliph al-Ma‘mun (813–33) led successful raids in person from 830 to 832, capturing Lulon and Tyana; and again in 838, when Caliph al-Mu’tasim (833–42) responded to the emperor’s attack on Sozopetra in Syria in the previous year, defeating him roundly at Dazimon and capturing Amorion, capital of the theme of the Anatolikoi and cradle of the ruling dynasty. In the 150 years of conflict, only two major truces were concluded – in 782 and 806; both imposed humiliating terms on the Byzantines, who had to pay tribute, and altogether they suspended the fighting for fewer than five years. The chain of signal-towers functioning by the ninth century, possibly even by the eighth, shows the permanent nature of the conflict with the caliphate. The towers ran from the northern entrance of the Cilician Gates, on the border at Lulon (near Ulukis¸la), to the Pharos of the imperial palace in Constantinople, and signals alerted the emperor to Arab incursions within the hour.15 But in the ninth century it was attacks by sea, independent of the caliphate, which did most harm. Euphemios, turmarch of Sicily, summoned the Aghlabids of Kairouan to help him in his rebellion against Michael II, and in 827 they landed and besieged Syracuse (see below, p. 462). Thus began the conquest of Sicily which was to drag on for the rest of the century. It was probably in this same year that Crete was attacked and soon taken by exiles from Cordoba; they had captured and subsequently been driven out of Alexandria by the caliph’s army.16 Crete and Sicily served as bases for raids on the islands and the Aegean littoral, now under constant threat; so, too, were the coasts of southern Italy, and Bari was captured around 842. 12 Nikeph., ch. 68, ed. and tr. Mango, pp. 140–1; Theoph., ed. de Boor, I, p. 424; tr. Mango and Scott, pp. 586–7; Lilie (1976), p. 164, n. 9. 13 McCormick (2001), pp. 865–72. 14 Theo. Stud., no. 475, ed. Fatouros, II, p. 683. 15 Pattenden (1983); Zuckerman (1994); SD, ed. and French tr. Auz´ epy, p. 238, n. 282; Constantine Porphyrogenitus, Three treatises, ed. and tr. Haldon, pp. 132–5 (text), 254–5 (commentary). 16 Tsougarakis (1998), pp. 30–41; Malamut (1988), I, pp. 72–6.

5. state of emergency (700–850)

257

On the northern border a new front was opened in 754 against the Bulgars in Thrace. Constantine V had made Thrace more secure through repeated campaigning in the last fifteen years of his reign, particularly with his victory at Anchialos in 763. On Constantine’s death, Thrace enjoyed a state of peace, with its network of kastra where garrisons were stationed17 and its renovated roads.18 This is illustrated by Empress Irene’s progress in 784 when she went as far as Beroia and Anchialos, which she ‘ordered to be rebuilt’.19 Relations with the Bulgars were in fact so close that in 776–7 Khan Telerig sought refuge in Constantinople, where he was baptised in the presence of Leo IV (775–80) and married one of Empress Irene’s relations.20 But under Irene and Constantine VI (780–97) the northern border became very dangerous once again. The Bulgars crushed the armies sent against them – as at Markellai in 792 – and their power increased still further under Khan Krum (c. 803–14). The efforts of Nikephoros I (802–11) to get the better of them ended in disaster in 811. The Bulgars annihilated the imperial army after it had seized their capital Pliska; Nikephoros was killed and his skull was used by Krum as a drinking goblet.21 This disaster was compounded two years later by the defeat at Bersinikia near Adrianople; the resulting fall of the Byzantine towns and fortresses of Beroia, Probaton, Anchialos and Mesembria allowed the Bulgars to devastate Thrace and Macedonia. The state of emergency which arose after Bersinikia led to the seizure of the throne by Leo ‘the Armenian’, strat¯egos of the theme of the Anatolikoi; he organised the defence of Constantinople, while the people assembled at Constantine V’s tomb, crying: ‘Arise and help the state which is perishing!’22 The death of Krum and the victory of Leo V in 816 changed the situation once again; a treaty was signed that same year and brought peace for more than three-quarters of a century, allowing for the reconstruction of the region.23 One portion of the empire which saw little warfare in this period was mainland Greece and the Peloponnese, following massive Slavonic immigration there in the seventh and eighth centuries; occasional Byzantine military expeditions were sufficient to ensure continued overlordship. Constantine V subjugated the Sklaviniai of Macedonia around 759;24 and in 17 18 19 20 21

Theoph., ed. de Boor, I, p. 447; tr. Mango and Scott, pp. 617–18. ˇ cenko (1972). Mango and Sevˇ Theoph., ed. de Boor, I, p. 457; tr. Mango and Scott, pp. 630–1. Theoph., ed. de Boor, I, p. 451; tr. Mango and Scott, p. 622. Theoph., ed. de Boor, I, pp. 489–91; tr. Mango and Scott, pp. 671–4; Chronicle of 811, ed. and French tr. Dujˇcev, pp. 210–17; tr. Stephenson (2006), pp. 87–90. On the settlements at Pliska, see Henning (2005), pp. 42–3 and nn. 11–14 on p. 50. 22 Scriptor incertus, De Leone Armenio, ed. Bekker, pp. 338–40; Theoph., ed. de Boor, I, pp. 496, 498–503; tr. Mango and Scott, pp. 679–80, 682–6. 23 Treadgold (1984). 24 Theoph., ed. de Boor, I, p. 430; tr. Mango and Scott, pp. 594–6; Ditten (1993), pp. 234–5.

258

the midd le empire

783 Staurakios the eunuch, logothete of the Drome under Irene, led an expedition against the Sklaviniai in Macedonia (or Thessaly25 ), Greece and the Peloponnese, returning victorious with booty and captives. Although previously considered of great importance, this expedition has recently been re-evaluated so far as the Peloponnese is concerned, with archaeological and sigillographic records showing the claims of the so-called Chronicle of Monemvasia to be exaggerated. According to the Chronicle, the Peloponnese had been abandoned to the Avaro-Slavs for 218 years – from the sixth year of the reign of Maurice (582–602) to the fourth year of Nikephoros I’s reign – with the exception of the eastern part, from Corinth to Cape Malea, where the emperor sent a strat¯egos.26 In reality, the Slavs were never completely beyond imperial control in the eighth century. In Thessaly, the seals of Slav archontes27 attest imperial recognition, and even in the less politically organised Peloponnese, the Slavs came into contact with the Greek population, which was more numerous in the eastern part of the peninsula. This contact occurred not only at the level of military administration, as shown by seals of strat¯egoi and droungarioi found around Argos, but also of church administration, as shown by seals and the presence of bishops at the second council of Nicaea.28 Two revolts by the Slavs in the Peloponnese had to be put down by military force in the course of the ninth century. First came the rebellion of the Slavs of Patras in the reign of Nikephoros I, recounted by Constantine VII Porphyrogenitus (945–59); this eventually led to the appropriation of both them and their property by the archbishopric of Patras, which was then asserting its independence from Corinth. Secondly, in 842 the Melingoi and Erizites further to the south rose in revolt and, once suppressed, were subjected to tribute.29 War was not only caused by enemy attacks on the empire; imperial succession could also lead to bloody civil war. On the death of Leo III on 18 June 741, his son Constantine V had to put down a revolt led by his brother-in-law Artabasdos; Constantine was supported by the fleet of the Kibyrrhaiotai and the armies of the Anatolikoi and the Thrakesioi, while the Opsikion, Thrace and the Armeniakoi backed Artabasdos. In the summer of 742 Artabasdos was proclaimed emperor in Constantinople, which was then besieged and, on 2 November 744, captured by Constantine.30 25 26

Oikonomides (1999–2000), p. 62. Lemerle (1963), pp. 10, 16–17; Lampropoulou et al. (2001), pp. 206–8, 220; Turlej (2001), pp. 125–58. 27 Seibt (1999); Seibt (2003a). 28 Avramea (1997), pp. 86–104; Avramea (2001). 29 DAI, chs. 49, 50, pp. 228–45; Turlej (2001), pp. 93–111. 30 Nikeph., chs. 64–6, ed. and tr. Mango, pp. 132–7; Theoph., ed. de Boor, I, pp. 413–15, 419– 21; tr. Mango and Scott, pp. 572–6, 580–4; Speck (1981); PMBZ, #632; Rochow (1994), pp. 21–9; Stathakopoulos (2004), pp. 377–8. For the most recent discussion of the debate over the chronology of the siege of Constantinople, see Nichanian, ‘Aristocratie et pouvoir imp´erial’ (PhD thesis, 2004), pp. 519–44.

5. state of emergency (700–850)

259

Some eighty years later, Thomas the Slav, turmarch of the foideratoi in the Anatolikoi, rebelled against Michael II on Christmas Eve 820; Michael was Leo V’s successor and was generally considered to have been responsible for his murder.31 The rebellion lasted three years and nearly succeeded, for Michael could only count on Constantinople and its fleet and the themes of Opsikion and the Armeniakoi, whereas Thomas controlled everything else, including the tax revenues, and had been crowned emperor at Antioch. Thomas laid siege to Constantinople in 821–2, but was foiled by a Bulgar attack and Michael’s destruction of his fleet using Greek fire. In the end he took refuge in Thrace, where he was captured by Michael at Arkadiopolis in October 823.32 The battle fought against the army of the Armeniakoi in 793 by Constantine VI at the head of all the other armies might also be considered a struggle for the succession. The Armeniakoi did not want Irene to be co-ruler with her son Constantine, who had been reconciled with her after a spell of sole rule (790–2). They wanted to remain loyal to him alone and to their strat¯egos Alexios Musele(m), who had been blinded for alleged complicity in a plot in favour of the caesar Nikephoros (son of Constantine V and uncle of the emperor) after the debacle at Markellai at the hands of the Bulgars in 792. Although this third civil war was apparently confined to the army, the other two, which both lasted over two years and involved sieges of the capital, also affected the civilian population. Thus, war dominated the eighth and early ninth centuries. Under the Isaurians it was a war of survival which overshadowed society as a whole and still threatened the empire’s very existence at the beginning of the ninth century, as can be seen from the Constantinopolitans’ reaction in 813 at the tomb of Constantine V. After 830, however, war affected mainly the armed forces in operations on the border with the caliphate, on the islands and in Italy; and even in these far-flung areas, its impact varied. Thrace, a land of reconquest and colonisation from 750, was ravaged by the enemy during the years 811 to 813, but regained its peace and prosperity after 816. Asia Minor suffered heavy but uneven losses. As the caliphs abandoned hopes of conquering the empire, a border gradually emerged between the Aegean and the Black Sea, along a line from Seleukeia to Trebizond; this became a sort of no-man’s-land, frequently changing hands, and studded with nineteen fortresses, according to Ibn Khurradadhbih.33 Regions close to the border were particularly exposed, such as Cappadocia and the Pamphylian and Lycian coasts; so too were the Anatolikoi, Armeniakoi and Opsikion, which bore the brunt of the Arab attacks. In contrast, the inhabitants of the Thrakesioi and the Black Sea coast were largely spared.34 The Kibyrrhaiotai 31 See Afinogenov (2001) for the argument that the revolt began under Leo V, as against Lemerle (1965). 32 Lemerle (1965). 33 Haldon and Kennedy (1980), pp. 85–6. 34 Lilie (1976), pp. 169–78 and map, p. 186; Haldon (1997a), pp. 106–7.

260

the midd le empire

successfully protected the Aegean islands in the eighth century, but after the Arab conquest of Crete in the 820s they became front-line targets. At the same time Sicily, after fifty years of peace following an earlier fifty years of raids, also came under attack. Finally, Constantinople itself was besieged four times, twice by foreign armies, in 717–18 and 813. It is difficult to determine the precise impact of both plague and continuous warfare on population figures; however, this was already being felt in the seventh century, and the population level seems to have reached its nadir in the eighth,35 although the situation was not uniform. Greece and the islands appear to have been more densely populated than the rest of the empire in the eighth century, but this changed after a century of Arab raiding. In Constantinople, the low point was the plague of 747–8; according to Patriarch Nikephoros, plague emptied the City of its inhabitants, and although it is difficult to give precise figures for the surviving population, estimates vary from 40,000 to 70,000.36 Constantinople was a case apart; while greatly affected by the plague, it also profited from war in demographic terms, thanks to the influx of refugees into the capital – a topic rarely studied.37 d epopul ation and ruralisat ion The overall demographic decline had two important consequences: shortage of manpower became a principal factor in imperial policy, and this in turn transformed the landscape of the empire. Human spoils, the captives who followed in the train of victorious armies, were a constant feature of the wars against the Arabs, Bulgars and Slavs. Prisoner exchanges and the refusal to hand over fugitives are more often mentioned in Arabic sources than in Byzantine ones, but show that manpower had become a precious commodity; the withholding of captives led to the disastrous campaign against the Bulgars at Bersinikia in 813.38 The emperors conducted a veritable settlement policy. Constantine V and Leo IV settled prisoners taken on the Arab border – from Germanikeia, Melitene and Theodosioupolis – in newly-constructed kastra in Thrace.39 Constantine repopulated Constantinople in 756 with natives of Greece and the islands,40 and in the 760s he installed migrant Slavs on the Asiatic coast of the Black Sea near 35 37 38

36 Mango (1990), pp. 51–62; Magdalino (1996a), p. 18. Laiou (2002b), p. 50. Theoph., ed. de Boor, I, p. 397; tr. Mango and Scott, p. 546. al-Mas‘udi, al-Tanbih, ed. de Goeje, pp. 188–92; tr. Carra de Vaux, pp. 255–8; Theoph., ed. de Boor, I, pp. 498–9; tr. Mango and Scott, p. 682–3; Skyl., ed. Thurn, p. 12; French tr. Flusin and Cheynet, pp. 12–13. 39 Nikeph., chs. 70, 73, ed. and tr. Mango, pp. 142–5; Theoph., ed. de Boor, I, pp. 427, 429, 451; tr. Mango and Scott, pp. 589–94, 621–3; Lilie (1976), pp. 164–5. 40 Nikeph., chs. 67–8, ed. and tr. Mango, pp. 138–41; Theoph., ed. de Boor, I, pp. 422–4, 429; tr. Mango and Scott, pp. 584–5, 592–4.

5. state of emergency (700–850)

261

the Bosporus.41 But it was Nikephoros I who pursued a settlement policy most vigorously. According to Theophanes’ Chronicle, in 807 he began by moving to Thrace people in Asia Minor found to be without fixed homes, and in 809–10 he settled impoverished soldiers from the themes of Asia in the Sklaviniai of Greece and Macedonia. The so-called Chronicle of Monemvasia corroborates Theophanes’ information for the region of Sparta.42 Later, according to the Life of Athanasia of Aegina, manpower shortages forced Theophilos to issue an edict requiring Roman widows to marry barbarian immigrants.43 Likewise, Theophilos adopted the policy of settling defectors from the caliphate in Asia Minor: initially the Banu Habib of Nisibis were installed as freebooters on the border, and the famous Persian unit commanded by Theophobos was transferred to the regions of Sinope and Amastris, later to be dispersed throughout the themes (see below, p. 393).44 The empire required men, both to join the army and to pay the taxes which provided for its upkeep: this could explain both the forced conversion under Leo III of the Jews and Montanists,45 and also the persecution of monks by Constantine V between 767 and 770, forcing them to return to lay status and marry.46 The scarcity of men also transformed the landscape and economy of the empire between the end of the sixth century and the seventh. Certain regions, such as Mount Athos, were abandoned and would not be repopulated until the ninth century.47 The empire was no longer a network of cities, but rather a rural state supervised from Constantinople, the metropolis which survived behind its walls. The fate of the cities of antiquity has been much studied48 and there is agreement that the standard model – that of cities being abandoned for fortresses or refuges built on higher ground – needs refining. To begin with, a large number of cities were abandoned outright – including Anemurion49 and Tyana – even if some, such as Pergamon, revived in the tenth century.50 However, despite frequent examples of refuges built beside an ancient city – in Calabria one 41 Nikeph., ch. 75, ed. and tr. Mango, pp. 148–9; Theoph., ed. de Boor, I, p. 432; tr. Mango and Scott, pp. 597–9. 42 Theoph., ed. de Boor, I, pp. 482, 486; tr. Mango and Scott, pp. 661–3, 666–8; Chronicle of Monemvasia, ed. Lemerle, p. 10. 43 Life of Athanasia of Aegina, ed. Halkin, ch. 2, p. 181; tr. Sherry, pp. 143–4. 44 Ibn Hawqal, Surat, ed. Kramers, I, pp. 211–13; French tr. Kramers and Wiet, I, pp. 205–6; French tr. in Vasil., pp. 419–21; al-Tabari, Ta’rikh, ed. de Goeje et al., III, p. 1235; tr. Bosworth, XXXIII, p. 85; French tr. in Vasiliev (1935), I, p. 294; Cheynet (1998a). 45 Through baptism, they became eligible for enrolment in the army: Theoph., ed. de Boor, I, p. 397; tr. Mango and Scott, p. 546; Rochow (1991), p. 104; Haldon (1999a), pp. 260–2. 46 Nikeph., chs. 80, 83, ed. and tr. Mango, pp. 152–3, 156–9; Theoph., ed. de Boor, I, pp. 437–9; tr. Mango and Scott, pp. 604–5; SD, ed. and French tr. Auz´epy, pp. 36–7 (introduction). 47 Lefort (1991), pp. 67–8. 48 Kirsten (1958); Brandes (1989); Foss (1996a); Brandes and Haldon (2000). See also below, pp. 470–2, 482–5. 49 Russell (2002). 50 Rheidt (2002), p. 624.

EXARCHATE OF RAVENNA

Cherson

Rimini

B L A C K DUCHY Rome OF ROME

THRACE Constantinople

DUCHY OF SARDINIA DUCHY OF NAPLES

ARMENIAKOI Euchaita

Nicaea OPSIKION CALABRIA

THRAKESIOI

SICILY

Samos

HELLAS

AN AT I KO LI O

Amorion

Chonai KIBYRRHAIOTAI Kibyrrha Euphra te s

Flaminian Way 0 0

100

200 100

300 200

400

500 300

600 km 400 miles

S E A

Crete

M E D I T E R R A N E A N

S E A

Map 13 The empire under militarised rule: army units and embryonic themes, earlier eighth century

CHERSON

B L A C K

S E A

Y

AEGEAN S

THRAKESIOI

PELOPONNESE

M

200 100

I

300 200

400

500 300

600 km 400 miles

I KO

ANATOLIKOI

S

M E D I T E R R A N E A N

Map 14 Administrative organisation: themes in the later ninth century

S E A

CHALDIA KOLONEIA

SEBASTEIA

CHARSIANON

KIBYRRHAIOTAI

SA

IA

100

O AT

ON

P

M TI

N IA GO

C

LA

AL

L CI

OPSIKION O

LA

L HE LIS PO KO NI

NG

THESSALONIKI

LA

AR M EN IA

OB

Thessaloniki

O

B

PH PA

CA PP AD O

AR

IA

CEPH

SI

TH RA CE STRYMON MACEDONIA

DYRRACHION

I RIO LLA KE OU

D

DALMATIA

IA

264

the midd le empire

could cite the case of Locri, abandoned for Gerace (Hagia Kyriake) before 78751 and examples from Asia Minor are legion52 – the inhabitants often remained inside the ruined city, even when they had no means of rebuilding it: they withdrew to a defensive position, fortifying only a small part of the city with materials from the ruins. Such was the case at Ankyra,53 Amorion,54 Side55 and Sardis.56 Ephesos combines the two patterns: one small part of the ancient city surrounding the port was reused and fortified, and a fortress was built on higher ground nearby, around the cathedral of St John (now Selc¸uk).57 Furthermore, fortifying a reduced space was in no way incompatible with small groups living in other districts inside the perimeter of the ancient city, as at Amorion, or with the presence of a kastron on higher ground further off, as is well illustrated by the Miracles of St Theodore at Euchaita under Constantine V.58 Finally, there were the cities created on virgin sites or on the sites of ancient fortresses, strongholds on high ground which contained the newly constituted civil, ecclesiastical and military administration within their walls. Such were the Thracian kastra of Probaton and Bulgarophygon constructed under Constantine V,59 ranking high on the list of bishoprics, or Santa Severina (Nikopolis) in Calabria.60 There were many variants, but the essential pattern was that cities shrank to a quarter of their previous size; all of them – whether old or new – were now fortified and their main function had changed. The city had become above all a local branch of the state; both from a military point of view – as garrisons or refuges for the surrounding rural population – and from an ecclestiastical perspective – as the residence of the bishop. However, it should be noted that the economic function of the city as a place for markets and fairs did not disappear. In the context of the demographic and economic depression shown by numismatic records, the written sources sometimes give paradoxical information; one example is the remission by Constantine VI of fees amounting to 100 pounds of gold (7,200 nomismata) for the fair of St John at Ephesos in 795 – an enormous sum which continues to puzzle historians.61 One explanation might be the industry of local peasants, whose villages (ch¯oria) had become the basic unit of the fiscal system, so vividly pictured in the Farmer’s law.62 Indeed, the Arabic geographers describe the

51 52 54 57 60 61 62

Prigent (2002), p. 938; Noy´e (1998), p. 234. 53 Foss (1977a). Brandes (1989), pp. 111–20; Haldon (1997a), p. 108. 55 Foss (1996b), pp. 24–46. 56 Foss and Scott (2002). Lightfoot (1998). 58 Zuckerman (1988), pp. 198–9. 59 Kountoura-Galake (1997). Foss (1979a), pp. 103–15. Prigent (2002), pp. 939–46. Theoph., ed. de Boor, I, p. 469; tr. Mango and Scott, pp. 644–5. Farmer’s law, ed. and Russian tr. Medvedev et al.; ed. and tr. Ashburner; Lemerle (1979), pp. 35–51; Brandes and Haldon (2000), pp. 148–9. See also, pp. 487, 488–9.

5. state of emergency (700–850)

265

empire as having no cities and being made up of prosperous districts with fortresses and villages, often in caves or underground.63 the armed f orces The emperors in the eighth century confronted an empire that had been ruralised and depopulated, and teetered on the brink of ruin. The situation was particularly dire when Leo III took power in 717, as Constantinople was besieged by land and sea by enormous Arab forces. Leo followed a succession of emperors – six in the twenty-two years following the first deposition of Justinian II in 695 – which marked a crisis for the empire.64 The failure of the siege in 717 had considerable repercussions; this was the first major setback in the Arab conquest which had begun in the 630s and had continued unrelentingly ever since (see below, pp. 370–7; above, p. 221). However, it did not guarantee the security of an empire still very much under threat. It was because of this threat that the Isaurian emperors and their successors, with the exception of Irene, introduced reforms designed to strengthen the empire against its enemies. It is often difficult to unscramble the Isaurians’ reforms from the attempts of Heraclius and his successors to cope with the Arab invasions and the loss of the eastern provinces; only in the ninth century do we become better informed about the state of play. But there is now a tendency to reassess the role played by the Isaurians, long minimised because of their discredited religious policies. The threat which hung over the empire explains why absolute priority was given to the army. Organising the army and maintaining it well required that the state apparatus be placed under direct authority of the emperor; victory depended on the purity of faith of the Christian people. The empire’s defences were fragmented into small units stationed in fortresses, so that walls became as important as men (see above, fig. 2 on p. 57). Repair of the walls went on incessantly, the unrivalled champion in this being Michael III (842–67), at the very end of the period. In Constantinople after the earthquake of 740 Leo III paid for the restoration of the walls, previously the inhabitants’ responsibility, out of the imperial treasury, raising the City taxes by 8 31 per cent (a tax of one miliar¯esion per nomisma, called the dikeraton). Inscriptions on the land walls near the Sea of Marmara record this work.65 Theophilos, for his part, repaired the walls near the Blachernae district and large sections of the sea walls.66 At Nicaea 63 Hudud al-‘alam, tr. Minorsky, ed. Bosworth, pp. 156–7; Ibn Hawqal, Surat, ed. Kramers, I, p. 200; French tr. Kramers and Wiet, I, pp. 194–5; Haldon (1997a), p. 112. 64 Haldon (1997a), pp. 74–82. 65 Nikeph., ch. 63, ed. and tr. Mango, pp. 130–3; Theoph., ed. de Boor, I, p. 412; tr. Mango and Scott, p. 572; Meyer-Plath and Schneider (1938–43), II, nos. 7, 12, 13, 16, 24, 29a, 32, pp. 124, 127, 130–2. 66 Meyer-Plath and Schneider (1938–43), II, no. 61, p. 141; van Millingen (1899), pp. 183–4.

266

the midd le empire

the walls were restored by Leo III and Michael III.67 Leo III (or Leo IV) sent a spatharios to repair the walls of Rodandos on the Arab border, and Leo IV sent the strat¯or Isaac, better known as Theophanes the Confessor, to repair those of Kyzikos.68 Following the pattern at Constantinople, these repairs were paid for by the local inhabitants, though carried out under the orders of an imperial official. The human element of the empire’s defence is more complicated than that of the physical walls, and many points remain unclear. At the beginning of the eighth century the army was based in the provinces and made up of different corps redeployed throughout Asia Minor after 636, following their withdrawal from the east. The troops of the magister militum per Orientem were stationed in central Asia Minor, their name Hellenised to Anatolikoi (from the Latin Orientales) and magister militum being translated to strat¯egos. The troops of the magister militum per Armeniam were deployed in northern Asia Minor and took the Greek name of Armeniakoi; those of the magister militum per Thracias, which had been sent as reinforcements to the east, were pulled back to the west coast of Asia Minor and took the name of Thrakesioi. Finally, that part of Asia Minor closest to Constantinople had remained the quarters of the imperial guard, the Opsikion (see above, p. 240). These army units became known as themes (themata), a generic name whose etymology is disputed. By the beginning of the ninth century, theme had come to mean the territory on which a corps was stationed, with each unit’s strat¯egos based in the theme’s capital: Amorion in the Anatolikoi, Euchaita in the Armeniakoi, Chonai in the Thrakesioi and Nicaea in the Opsikion. Similarly, at the end of the seventh century new army corps were installed in Sicily, Hellas and Thrace, each commanded by a strat¯egos. This currently accepted model for the emergence of the themes has replaced that suggested by George Ostrogorsky.69 A further development saw the subdivision of the existing army corps, or themes, for tactical reasons; at the beginning of the ninth century the theme of Cappadocia was created in the Anatolikoi, with its capital at Koron; and in the Armeniakoi, the themes of Paphlagonia and Chaldia were established. But themes could also be subdivided for political reasons, as was the case with the Opsikion. All too often involved in plots, it was broken up into the Boukellarioi (768) and Optimatoi (c. 775) and these were simultaneously demoted from combat to rearguard units. The Opsikion theme proper retained only the western part of its former territory (Phrygia, the Hellespont and western Bithynia). Moreover, several new themes were created: in Crete, probably in the eighth century, and 67 68 69

Schneider and Karnapp (1938), nos. 29, 36, pp. 49, 51–2. Gr´egoire (1908); Methodius, Life of Theophanes, IX.15, ed. Latyshev, pp. 10–11. Haldon (1997a), pp. 208–20; Haldon (1999a), pp. 71–4; Ostrogorsky (1968), pp. 95–8. See also above, pp. 239–40.

5. state of emergency (700–850)

267

in Macedonia, the Peloponnese and Cephalonia in the ninth century. As regards the fleet, the Kibyrrhaiotai (from the city of Kibyrrha in Caria) had reinforced the Karabisianoi at the end of the seventh century and then replaced them to ensure protection of the southern coasts of Asia Minor, which were also guarded by the fleet of the Aegean Sea in the eighth century.70 Access to the western Mediterranean was controlled by the fleet of the Helladikoi, which revolted in 727, and a fleet of Sicily also appeared in the 750s.71 In organisational terms, themes were subdivided into tourmai, droungoi and banda. The tourma, at the head of which was a turmarch, and the bandon, under the orders of a count (kom¯es), were assigned to a territory, whereas the droungos was a tactical unit, under the orders of a droungarios.72 The numbers of men in a unit varied by region and over time, and figures suggested for individual themes are no more than rough estimates.73 According to Theophanes, Constantine V mobilised the entire Byzantine army against the Bulgars in October 773 (or 774); this totalled some 80,000 men, drawn both from the provincial – thematic – units and from the elite regiments stationed in the capital (tagmata). This figure is considered reasonable by some scholars, but far too high by others,74 and should be compared with information given at the end of the ninth century by Leo VI in his Tactica: according to Leo, the cavalry themes had 4,000 men each, 2,000 per tourma.75 From the mid-eighth century on, the tagmata stationed at Constantinople consisted of the Schools (scholai) and excubitors (exkoubitoi), old guard units which over the centuries had become largely ceremonial. Constantine V made them operational again and it was these tagmata who surrounded the emperor on the battlefield and whose arms were provided by the state.76 The tagmata – who prefigured the professional army of the tenth and eleventh centuries – were reinforced by Constantine V’s successors, though the Schools were demoted for a time under Irene, punishment for their fervent iconoclasm in hindering the meeting of the iconodule council in 786; they were replaced by another tagma, the Arithmos, formed out of banda from various themes.77

70 71

For a summary: Haldon (1999a), pp. 86–7 and table 3.1. McCormick (2001), pp. 865–972; Theoph., ed. de Boor, I, p. 405; tr. Mango and Scott, pp. 559–60; Avramea (1997), pp. 101–4. 72 Haldon (1999a), pp. 110–14. 73 Treadgold (1995), pp. 64–75; Haldon (1999a), pp. 113, 314, n. 68. 74 Theoph., ed. de Boor, I, p. 447; tr. Mango and Scott, p. 617; Treadgold (1995), p. 64; Haldon (1999a), p. 102. 75 Haldon (1999a), p. 110. 76 First mentioned in the sources by Theoph., ed. de Boor, I, p. 442; tr. Mango and Scott, p. 610. 77 Theoph., ed. de Boor, I, p. 462; tr. Mango and Scott, pp. 635–7; Haldon (1984), pp. 239–42.

268

the midd le empire

Unlike the troops of the tagmata who were full-time, those of the provinces were only mobilised for campaigns during the summer months and sometimes served outside their theme; natives of the Asia Minor themes might even be employed in Europe.78 However, some thematic soldiers were permanently on duty if they guarded a fortress, since fortresses had standing garrisons. The conditions under which soldiers were recruited is a question upon which much has been written; however, it remains an open question, involving as it does the entire system of taxation, an understanding of which, in turn, depends on one’s interpretation of the empire’s monetary circulation. There is agreement on some points: that thematic soldiers were based throughout the whole territory of a theme; that they were responsible for their own maintenance, since they had to present themselves for service with their equipment;79 and that the administration provided for their needs on campaign. Various suggestions have been put forward as to how soldiers paid for their equipment and how their service was remunerated. One is that the tenth-century system – that military service was inextricably linked with the holding of inalienable, exempted land – had, in fact, been in place since the seventh-century withdrawal of troops from the east: in effect, troops whom the state could no longer pay in cash were paid in land. Based on the numismatic records available before c. 198080 and on the disappearance of imperial estates between the sixth and tenth centuries,81 this interpretation is intellectually tempting and has been favoured by those who consider military service as a fiscal obligation, bound up with a plot of exempted land.82 However, as has often been noted,83 such an interpretation jars with the fact that those few contemporary sources which mention soldiers contain no evidence of compulsory military service in connection with land; these sources include the Ecloga, the Chronicles of Theophanes and Nikephoros, the Lives of Philaretos and Euthymios the Younger, and the letters of Theodore the Stoudite. Chapter Sixteen of the Ecloga, often cited but yet to be examined in detail, gives an idea of the position of soldiers under the Isaurians. When soldiers (strati¯otai) were enrolled (strateuomenoi), their name and place of origin were inscribed on the theme’s military roll, as well as on that of the central office of the strati¯otikon in Constantinople. Enrolment entailed military responsibilities: going to war, when called up, fully equipped with horse, harness and arms. It also entailed benefits: soldiers received a regular 78 Theoph., ed. de Boor, I, p. 462, 470–1, 491; tr. Mango and Scott, pp. 635–7, 645–8, 673–4; Life of Philaretos the Merciful, ed. Ryd´en, pp. 72–5. 79 Life of Philaretos the Merciful, ed. Ryd´ en, pp. 72–5; Lemerle (1979), pp. 59–60. 80 Hendy (1985). 81 Treadgold (1983a), pp. 628–31; Treadgold (1995), pp. 171–6. 82 Oikonomides (1988a); Oikonomides (1996a), pp. 37–40. 83 Lemerle (1979), pp. 59–64 ; Haldon (1993), pp. 20–9.

5. state of emergency (700–850)

269

wage, which was paid whether or not they were on campaign, as well as extra wages during combat, booty, and bonuses from the emperor or strat¯egos. The regular wage (roga) was an annuity due to those who held the office of soldier, the so-called strateia. The household (oikos) in which an enrolled man lived, and which he could leave for another, was also probably exempt from supplementary taxes and impositions, although there is no mention of this in the Ecloga. Furthermore, the term strateia was not confined to soldiers. Under the Isaurians, every individual inscribed on the administrative roll by virtue of his office – in effect, everyone in imperial office, including ecclesiastical office – was deemed to hold a strateia, and thus entitled to receive a roga ‘from the hand of the emperor’, as well as rations in kind. Strateiai could either be bought or granted out by the emperor.84 The ways in which military strateiai changed hands are not altogether clear. Elderly soldiers might be taken off the rolls and thus forfeit their strateia; but on the other hand, soldiers’ widows had to furnish a fully equipped soldier – or the equivalent sum – if they wanted to keep the strateia, i.e. to continue receiving the roga and, perhaps, tax exemptions for the household. This latter practice was abolished under Irene, but it must subsequently have been re-established, as we find it in use around 840 in the Life of Euthymios the Younger.85 Irene’s measure forced her successor Nikephoros to make up for the soldiers who had thus been lost to the army, giving rise to the second in the list of this emperor’s ‘vexations’ decried by Theophanes. Nikephoros decreed that the poor from the villages should be enrolled as soldiers (strateuesthai); they were to be equipped at the expense of their fellow villagers who, in addition, were to pay 18 21 nomismata for each poor man so enrolled. The latter sum has generally been understood to be the price paid for the equipment.86 But it could also be understood as the price of the soldier’s strateia. This would imply that the enrolled man’s roga was paid not directly to him – though he retained any extraordinary earnings – but to those who had jointly bought the strateia for him (the syndotai) and who, according to article 18 of the Farmer’s law, paid his taxes and had the use of his land. tax ation and the provinces The debate over the arrangements for recruitment of the army is connected with the debate over whether the tax system worked in cash or in kind; and 84 Ecloga, XVI.3–4, ed. and German tr. Burgmann, pp. 222–5; tr. Freshfield, pp. 102–3; Nikeph., ch. 80, ed. and tr. Mango, pp. 152–3. 85 Nikephoros, Antirrhetici, III.62, cols. 491–2; French tr. Mondzain-Baudinet, pp. 268–9; Theo. Stud., no. 7, lines 61–3, ed. Fatouros, I, p. 26; Life of Euthymios the Younger, ed. Petit, p. 18. 86 Theoph., ed. de Boor, I, p. 486; tr. Mango and Scott, pp. 666–8; Lemerle (1979), pp. 62–3; Haldon (1993), pp. 25–6; Oikonomides (1996a), p. 39.

270

the midd le empire

this, in turn, depends on how one believes money circulated. It is unanimously agreed that circulation dwindled between 650 and 850. However, recent studies have drawn attention to the vitality of Sicilian gold coinage, as well as to the role played by the miliar¯esion – the silver coin worth 121 of the nomisma, introduced by Leo III in 721. The latter innovation helps explain the great importance of the fortress of Lulon, located on the border with the caliphate at the heart of a mining region. The circulation of copper coins has also been reassessed; this had previously been assumed to be non-existent, as account had not been taken of the many anonymous folleis preserved among archaeological finds in Turkish museums, notably at Ankyra.87 Availability of money was vital for the army, in order to pay the rogai of the officers and men. The enemy was well aware of this and the ‘Wells Fargo’ wagon of the rogai, accompanied by the strat¯egos, was a key target for attack. In 809 the Bulgars seized the rogai of the army on the Strymon, totalling 1,100 pounds of gold, or 79,200 nomismata. Two years later, the Arabs captured the rogai of the theme of the Armeniakoi, which amounted to 1,300 pounds, or 93,600 nomismata – over three times the annual tribute of 30,000 nomismata paid to the Arabs by Nikephoros I in 806.88 It is generally thought that the roga was paid every four years, though no contemporary text declares as much.89 The state also needed to issue allowances in kind to maintain the army on campaign. The system described in the military Treatise compiled under Constantine VII Porphyrogenitus, but based on documents of the Isaurian period,90 was probably at least partially in place from the eighth century. Under this system, the pr¯otonotarios of the theme – the highest civilian official in the thematic administration, first mentioned in the ninth century – would supply each military staging-post (apl¯ekton) with barley for the horses and other necessities; these outgoings would be recorded in and deducted from the theme’s account in the eidikon, the central office of the tax administration.91 We can see this supply system at work in 782 under Irene. According to al-Tabari, she supplied Harun al-Rashid ‘with guides and markets’ for his journey back from Bithynia to the caliphate, after he had negotiated a peace treaty in which this featured as one of the clauses.92

87 Morrisson (1998); Morrisson (2001); Pitarakis (1998), p. 170; Lightfoot (2002). See also below, pp. 469, 470, 483–4. 88 Theoph., ed. de Boor, I, pp. 482, 484, 489; tr. Mango and Scott, pp. 661–5, 671–2; see also TC, p. 11. 89 Haldon (1999a), p. 124; Treadgold (1988), p. 352. 90 Constantine Porphyrogenitus, Three treatises, ed. and tr. Haldon, pp. 96–7. 91 Constantine Porphyrogenitus, Three treatises, ed. and tr. Haldon, pp. 116–17. 92 al-Tabari, Ta’rikh, ed. de Goeje et al., III, p. 504; tr. Kennedy, XXIX, p. 221; cited by Brooks (1900), p. 736.

5. state of emergency (700–850)

271

The sources do not really allow us to decide between the two interpretations of the tax system currently on offer. Those who argue that taxes were paid in cash do so mainly by reference to the tax on Constantinopolitans for repairs to the City’s walls; they maintain that the hearth tax, the kapnikon, as well as the property tax, described by the Farmer’s law in a village context, were paid in cash from the time of Constantine V.93 This would indeed appear probable for the kapnikon, which Nikephoros I extended to the inhabitants of church lands – those belonging to bishoprics, monasteries and pious foundations – and we even know the rate at which the kapnikon was paid under Michael II: two miliar¯esia per household.94 But it is harder to say the same of the property tax. Two contemporary letters support the idea of a property tax called the syn¯on¯e, paid in kind: both were written between 820 and 843 by Bishop Ignatios of Nicaea to the pr¯otonotarios Nicholas, complaining on behalf of the men of his church. Although these men were theoretically exempted from the syn¯on¯e, as they were from forced labour and other impositions – and despite having already sent the grain of the syn¯on¯e to the public granaries that year – the syn¯on¯e was still being demanded of them, together with an additional six modioi per male inhabitant. Opinions as to whether the property tax was paid in cash or kind depend on whether one interprets the word syn¯on¯e as a property tax or as a requisition; it was originally the Greek translation of the Latin coemptio, the compulsory, fixed-price sale to the state of goods needed for the army.95 One might also take into account the case, under Theodora and Michael III, of the poor who were imprisoned by the dioik¯et¯es of Prousa for non-payment of taxes; the abbot of Agauroi gave them 100 nomismata, which had been earmarked to pay the taxes of his own monastery.96 It is not clear, however, which tax was involved here. This debate is compounded by another concerning the kommerkiarioi, whose numerous seals have been found for the period between 650 and 730, dated by indiction and stamped with the emperor’s effigy. Their legends also mention warehouses (apoth¯ekai) and the names of several provinces. There is general agreement that the kommerkiarioi reported to the central office of finances in Constantinople, the genikon logothesion. Those who believe taxes were paid in kind argue that the kommerkiarioi were responsible for depositing the tax proceeds needed for the year’s campaigns in the apoth¯ekai; and there is undeniably a certain correspondence between the dates and provinces mentioned on the seals and military expeditions.97 On the other 93 94

Oikonomides (1996a), pp. 29–39; Oikonomides (2002), pp. 980–1. Theoph., ed. de Boor, I, pp. 486–7; tr. Mango and Scott, pp. 666–8; Life of George of Mytilene, ed. Phoutoules, ch. 5, p. 35; TC, p. 54. 95 Ignatios the Deacon, Correspondence, ed. and tr. Mango, nos. 7, 8, pp. 38–45; Kazhdan (1992); Oikonomides (1996a), pp. 70–2 as against Haldon (1994a); Kaplan (2001). 96 Life of Eustratios of the monastery of Agauroi (Abgar), ch. 15, ed. Papadopoulos-Kerameus, p. 378. 97 Brandes (2002a), pp. 281–426.

272

the midd le empire

hand, those who argue that taxes were paid in cash see the kommerkiarioi as private entrepreneurs who controlled the silk trade for a fixed period in the provinces where they managed the apoth¯ekai for the state.98 After 730 the seals of the kommerkiarioi disappear for about a century and are replaced by impersonal seals ‘of the imperial kommerkia’ from one city or province or another. This implies reform under Leo III, the exact terms of which are unknown. However, this might explain the appearance in our sources of the kommerkion, or indirect tax on transactions. The kommerkion is mentioned under Constantine VI in connection with the fair at Ephesos, and under Irene and Nikephoros I in connection with the customs offices on the Bosporus (Hieron) and the Dardanelles (Abydos).99 At the beginning of the ninth century the kommerkiarioi reappear in the Balkans and even in the Danish port of Hedeby, where seals of the kommerkiarios Theodore have been found.100 This survey of the army and the tax system needed to maintain it, to ensure the empire’s survival, can only be fragmentary and provisional, and our knowledge is constantly being expanded with the publication of new sources, such as seals. However, the most important fact remains the Isaurians’ mobilisation of all the empire’s resources for the army, and the resultant militarisation of society, which also has the effect of obscuring civilian life. We know little about the civil administration of the provinces before a thematic civil administration under the authority of the pr¯otonotarios was installed at the beginning of the ninth century. All we know is that it was carried out by chartoularioi and eparchs, who had a role in the tax system, and by the dioiketai ‘of the provinces’ who were the collectors of taxes.101 Another consequence of this militarisation was the creation of a military aristocracy, whose titles were reward for senior command. The Isaurians surrounded themselves with men who often had their origins outside the empire: for example under Leo III the patrikios Beser, or Artabasdos, strat¯egos of the Anatolikoi and later brother-in-law of the emperor; and under Leo IV, the five strat¯egoi appointed after the expedition of 778, four of whom were Armenian. Leo V, himself of Armenian origin, was the son of a patrikios named Bardas – perhaps the strat¯egos of the Armeniakoi in 771. Leo married the daughter of the patrikios Arsavir, also an Armenian and probably the nephew of Bardanes Tourkos (i.e. Khazar), the strat¯egos of the Anatolikoi who revolted against Nikephoros I in 803. It is also under the Isaurians that 98 99

Oikonomides (1986a); Oikonomides (2002), pp. 983–8. Theoph., ed. de Boor, I, pp. 469–70, 475, 487; tr. Mango and Scott, pp. 644–7, 653–4, 668; Brandes (2002a), pp. 583–9. 100 Brandes (2002a), pp. 562–64; McCormick (2001), p. 227; Cheynet (2003), p. 51. 101 Oikonomides (2002), p. 989; Brandes (2002a), pp. 195–225; Brandes and Haldon (2000), pp. 169– 71.

5. state of emergency (700–850)

273

family names first make an appearance – most often as sobriquets applied to the iconoclasts – and aristocratic families were constituted. Several of the latter, such as the Kamoulianoi and the Melissenoi, who appeared under Constantine V, were to have a long history. A good part of the aristocracy of the ninth century owed its standing to the brilliant military career of an ancestor in the previous century.102 cen tral ad ministration and imperial ideology The state of emergency in the empire also led to a tightening up of the administration and a change in the emperor’s role. The departments established in the seventh century remained in operation – the logothesia, each under the direction of a logothete – and we know about their organisation in detail from the Taktikon Uspensky. The general (genikon) logothesion was a sort of finance ministry, collecting taxes and distributing money. The strati¯otikon logothesion was the department managing the army. The logothesion of the Drome (tou dromou) managed the roads, intelligence and diplomacy. Other departments or functions grew in importance during this period, and there has been considerable debate as to their continuity from Roman institutions. The sakellarios, originally a eunuch who kept the emperor’s purse, became a key figure who, by the end of Theophilos’ reign, was the chief organiser of expenditure and had more authority than the general logothete himself. Likewise, the office of the eidikon, a treasury whose functions are unclear, makes its appearance in the ninth century.103 The tightening up of the administration around the emperor enabled him to govern more directly, especially since the offices of the logothetes were located in the Great Palace. This immediacy of power is also a feature of the military sphere: from the time of Heraclius, emperors had led their armies into battle in person. This was particularly true under the Isaurians and Leo V, who went on campaign nearly every year. This tradition of the warrior emperor makes Irene’s reign even more anomalous: as a woman, she could not lead the army. In diplomatic relations with newer, neighbouring states, the emperors continued a policy of impressing their subjects with the empire’s superiority and prestige. Theophilos adorned the reception hall of the Magnaura with a throne surrounded by automata of roaring lions and chirping birds in a plane tree, which Liudprand of Cremona described in the tenth century.104 John the Grammarian’s embassy to Baghdad on behalf of Theophilos was 102 103 104

Cheynet (2000), pp. 288–302; Kountoura-Galake (2004); PMBZ, #4244. Brandes (2002a), pp. 106–72, 427–79. Lemerle (1986), p. 178 and n. 27; Liudprand, Antapodosis, VI.5, ed. Chiesa, p. 147; tr. Wright, pp. 207–8.

274

the midd le empire

celebrated for its richness and splendour. The organ which Pepin the Short received as a gift from Constantine V also contributed to the empire’s renown amongst the Franks (see below, p. 414). Imperial building projects had the same goal and here, too, Theophilos was a master. He remodelled the Great Palace where, it is thought, Constantine V had built the church of the Mother of God of the Pharos in the previous century;105 of the many buildings added by Theophilos, the best-known are the Triconch of the Sigma and the Sigma itself. Across the Bosporus Theophilos constructed the Palace of Bryas, which has yet to be identified with certainty, and he adorned St Sophia with the bronze doors that are still in place.106 This restructuring of the emperor’s image went hand in hand with the reinforcement of dynastic rule. From Heraclius on, rulers crowned their eldest sons as co-emperors, although the form of coronation sometimes varied. In 776 Leo IV added to the ceremony an oath of loyalty to both emperors, which civilian and military officials, as well as notables, had to swear; this vow not to accept any emperor other than Leo’s newly crowned son Constantine was signed by all and deposited in St Sophia.107 The gold coinage gives an excellent example of the insistence on dynastic rule under the iconoclast emperors, for on their coins both the Isaurians and Theophilos showed images not only of their descendants, sometimes including their daughters, but also of their ancestors. So Constantine VI’s father, grandfather and great-grandfather are all squeezed in on the reverse of his nomisma.108 In this as in other areas, Irene is an exception; she was the only sovereign in the history of the empire to put her bust on both sides of the nomisma. The Isaurians appear to have done most to boost the dynastic aspect of the imperial office. Indeed, it was Constantine V who created the legitimising concept of porphyrogenitus for his son Leo; being born-in-the-Porphyra – the chamber in the imperial palace covered with red marble – would become a prerequisite for the Macedonian emperors.109 At the Easter ceremonies in 769, Constantine V made official the hierarchy of court titles given to members of the imperial family: his sons were given the titles of caesar and nobelissimos, as recorded in the Book of ceremonies.110 Indeed the emperor’s reception for the poor – given on the eighth day after Christmas in the Hall of the Nineteen Couches – should be dated to Constantine’s period 105 Theoph., ed. de Boor, I, p. 444; tr. Mango and Scott, pp. 612–14; Jenkins and Mango (1956), pp. 134–5; Magdalino (2004), pp. 20–3. 106 Ricci (1998); Treadgold (1988), p. 323; Mango (1967). 107 Theoph., ed. de Boor, I, pp. 449–50; tr. Mango and Scott, pp. 619–21. 108 DOC, III.1, pp. 292, 406–10; Morrisson, Catalogue des monnaies byzantines, II, pp. 466, 514–15. 109 Dagron (1994), pp. 112–13. 110 DC, I.43, 44, ed. Reiske, pp. 217–29; Diehl (1905), pp. 269–302; Mango and Sevˇ ˇ cenko (1972), p. 390.

5. state of emergency (700–850)

275

Figure 11a Gold coin of Constantine VI showing his ancestors squeezed onto the reverse of the nomisma (below)

of rule, since a token of a ‘pauper of the Nineteen Couches’ dating from his reign has been found.111 oaths and the empress irene Besides insisting on dynastic rule, the iconoclast emperors made other innovations in the imperial office: they were keen on justice, and their personal relations with their subjects were marked by the intensive use of oaths. In fact, it would appear that by use of silentia they somehow sought approval of their decision-making from the elite, and even from the people. So far as justice is concerned, later chroniclers have given Leo V and Theophilos the reputation of emperors close to their people, eager to right the wrongs committed by their officials; and the Isaurians, who were accused by their opponents of being litigious and fond of trials, demonstrated in both the prologue and the text of the Ecloga just how much they considered 111

Bendall and Nesbitt (1990).

276

the midd le empire

Figure 11b Gold coin of Irene, portraying her on both sides of the coin

good justice to be a vindication of their imperial office.112 Its clear style, easily comprehensible and focusing on the practical essentials of civic life, ensured that this Isaurian compendium would long remain in use. Another innovation of the Ecloga was to institute salaries for judges, so that their decisions no longer depended on gifts offered by those on trial.113 The widespread use of oaths as a political instrument also appears to be an iconoclast novelty, and Constantine V was probably the instigator. According to Agapius of Membij he used oaths to reinforce the hereditary transmission of imperial power, thereby binding his subjects to his son Leo. According to Theophanes, Constantine also invoked the protection of his other sons,114 and as we have seen, Leo IV made taking an oath of loyalty to his son into a ceremony involving the army and the entire body politic. The swearing of loyalty oaths was subsequently taken to extremes by Constantine VI and Irene; they demanded that oaths be sworn to both of them, but on several occasions ordered further oaths to the exclusion 112 TC, pp. 30–1, 87–8, 93–4; Nikephoros, Antirrhetici, III.62, cols. 487–8; French tr. MondzainBaudinet, p. 267; Simon (1994), pp. 12–16. 113 Ecloga, preface, lines 96–109, ed. and German tr. Burgmann, pp. 166–7; tr. Freshfield, p. 70. 114 Agapius of Membij, al-‘Unwan, ed. and French tr. Vasiliev, PO 8.3, p. 544; Theoph., ed. de Boor, I, p. 450; tr. Mango and Scott, p. 621; Svoronos (1951), pp. 109, 119–21.

5. state of emergency (700–850)

277

of one or the other, thus provoking the revolt of the Armeniakoi in 793. Constantine V put oath-taking to even broader use. In 766 he commanded that his subjects swear an oath not to bow down before religious images, and this was one of the points which made it difficult for the bishops to renounce iconoclasm twenty years later: they feared – rightly – the accusation of perjury, a crime for which the Ecloga prescribes the cutting out of the tongue.115 It appears that the swearing of oaths was practised widely in this period, so much so that Theodore the Stoudite and the patriarch Nikephoros complained, and Irene sought to limit it.116 Another innovation under Leo III was the use of the silention; under Justinian I this had been a restricted council, but under Leo it became a type of special assembly in the Great Palace, convened by the emperor when he wanted to announce a particularly solemn decision. According to the Life of Stephen the Younger, a silention could even involve assembling all the people, when it was convened in the Hippodrome.117 The populist tendencies of the iconoclast emperors, remarked upon by Nikephoros, are also evident in their use of mockery as a political weapon: under Constantine V the conspirators of 766, as well as the monks and Patriarch Constantine II, were all subjected to public ridicule in the Hippodrome; likewise, under Michael II, Thomas the Slav was mocked by the army before his execution.118 The characteristics of the iconoclast emperors – warlike, just and close to the people – bring a certain coherence to their reigns in the eighth and ninth centuries, in contrast to that of Irene. Perhaps because she was a woman and her position as basileus was in the strict sense exceptional, she took exactly the opposite course from her predecessors in every sphere. By blinding her son she made the dynastic transmission of power impossible; she abolished iconoclasm; she got rid of the elite corps of soldiers; she sued for peace when the battle was nearly won; she chose a patriarch from among the laity; and she set up an intermediary power between emperor and administration, entrusting the latter to her close advisers, the eunuchs of the Bedchamber. Thus Staurakios, logothete of the Drome in 782, remained in office until his disgrace in 800 and played a dominant role in Byzantine politics for twenty years.

115 Theoph., ed. de Boor, I, p. 437; tr. Mango and Scott, pp. 604–5; Nikeph., ch. 81, ed. and tr. Mango, pp. 154–5; Mansi, XIII, col. 61; Ignatios the Deacon, Life of Tarasios, ch. 14, ed. and tr. Efthymiadis, pp. 86, 177; Nikephoros, Refutatio, ch. 23, ed. Featherstone, pp. 49–50; Ecloga, XVII.2, ed. and German tr. Burgmann, pp. 226–7; tr. Freshfield, p. 106. 116 Nikephoros, Antirrhetici, III.62, cols. 487–8; French tr. Mondzain-Baudinet, p. 267; Theo. Stud., no. 7, lines 37–41, ed. Fatouros, I, p. 25; Burgmann (1981). 117 Christophilopoulou (1951); SD, ch. 40, ed. and French tr. Auz´ epy, pp. 139–41, 235–7. 118 Nikephoros, Antirrhetici, III.64, cols. 493–4; French tr. Mondzain-Baudinet, pp. 270–1; Theoph., ed. de Boor, I, pp. 437–8; tr. Mango and Scott, pp. 604–6; Rouan (1981), pp. 425–8; TC, p. 69.

278

the midd le empire

The imperial office as reinforced by Constantine V was proof against this singular reign. It withstood both the diplomatic failures which led to the establishment of a rival empire in the west, and the military failures in the face of the caliphate’s renewed offensive in Asia Minor; but in the end it was laxity in financial matters that led to Irene’s downfall. The revolt against her came not from the army but from the offices of finance; Irene was deposed and replaced by the minister of the treasury, the general logothete Nikephoros, who had to reimpose the taxes which the empress in her benevolence had abolished. culture, purif ication and the drive against idol atry The religious policy of the period has deliberately been left until last, to prevent it eclipsing all other aspects, as so often happens. In fact, the religious direction taken is hardly surprising, inspired as it was by the misfortune of the times and the need to save the empire. However, it has been completely distorted by the violence of the anti-iconoclast polemics, transforming a well-thought-out and sound decision into the deranged impulse of ignorant men inspired by Satan. Iconodule propaganda is now being subjected to closer scrutiny. One example is the allegation that the iconoclasts were lacking in education (amathia). Because the chroniclers Theophanes and Nikephoros maintain that classical culture was dead at the beginning of the eighth century – for which Theophanes holds Leo III directly responsible119 – and because polemical works accuse the iconoclasts of ignorance, it is generally considered that Byzantine society under the iconoclast emperors lost access to classical culture. This accusation is not altogether unfounded; with the disappearance or drastic contraction of towns, and the ruralisation of Asia Minor and the Balkans, there must have been an equivalent reduction in the number of books. Even in Constantinople, books became rare. A work such as the Parastaseis, in which self-styled philosophers show their non-comprehension of the masterpieces of classical art surrounding them in the capital, proves that there was no great flourishing of classical culture in the eighth century.120 However, texts written in refined Greek started to emerge in the second half of the ninth century, presupposing continuity of literary studies during the eighth and ninth centuries, and there are several indications of such continuity. The saints’ Lives of the period attest a functioning elementary and secondary educational system,121 and there were some deeply learned men of letters of the beginning of the ninth century, including George Choiroboskos, deacon of St Sophia, 119 Theoph., ed. de Boor, I, p. 405; tr. Mango and Scott, pp. 559–60; Nikeph., ch. 52, ed. and tr. Mango, pp. 120–1. 120 Mango (1975a); Parastaseis syntomoi chronikai, ed. Preger; tr. Cameron and Herrin. 121 Lemerle (1986), pp. 81–2, 108–15.

5. state of emergency (700–850)

279

John the Grammarian, patriarch of Constantinople (837–43) and Leo the Mathematician. The Isaurians’ reputation for ignorance has been reinforced in a sort of vicious circle by the fact that, at first sight, no manuscript of a scholarly nature has survived from this period. The famous uncial manuscript of Ptolemy’s ‘Handy Tables’ with its miniature of the zodiac was attributed to the ninth century until it was demonstrated that the astronomical calculations it contains could only have been done during the reign of Constantine V.122 The accusation of amathia brought against the iconoclasts must be considered in its ideological context. In characterising the iconoclasts as ignorant, the iconodules certainly wanted to accuse them of grammatical ignorance; but above all they wanted to accuse them of ignorance of divine truth and thus blindness to the reality of the Incarnation, which had made representation of Christ possible. It is unlikely that the appearance of minuscule – a highly important development for Greek letters – could have occurred in a world devoid of culture, and its replacement of uncial script might almost be compared to the invention of the printing press. Minuscule’s use of ligatures, accents and punctuation made it far quicker to write, easier to read and more economical in its use of paper and ink. Codicologists now date the use of minuscule in manuscripts to sometime around the 780s, in two distinct geographical areas: Constantinople and Palestine.123 The earliest extant minuscule manuscript with a precise date is the ‘Uspensky Gospels’, dated to 835 by a notice of the scribe Nicholas, a monk of the Constantinopolitan monastery of Stoudios.124 Exactly where minuscule writing originated remains unknown, but it is curious that the imperial palace has never been considered. The religious policy of the iconoclast emperors Leo III, Constantine V and Leo IV in the eighth century, and of Leo V, Michael II and Theophilos in the ninth, should be understood as just one aspect of their struggle to ensure the empire’s survival. Various explanations have been offered for Leo III’s sudden ban on venerating icons in 730. The apparent aniconism of the eastern part of the empire is one suggestion. This was Leo’s birthplace and seat of Bishops Constantine of Nakoleia and Thomas of Claudiopolis; according to Patriarch Germanos’ letters read out at the second council of Nicaea, these two bishops had forbidden veneration of icons in their sees even before 730. Another suggestion is the supposed influence of Islam and Judaism; early in the 720s Caliph Yazid II (720–24) issued an edict banning Christian images, and the iconoclasts invoked the Old Testament ban on pictorial representation (Exodus 20:4). But the argument of eastern 122 Ptolemy, ‘Handy Tables’, Biblioteca Apostolica Vaticana, MS. gr. 1291 (see fig. 12); Lemerle (1986), pp. 83–4; Wright (1985). See fig. 13. 123 Fonkiˇ c (2000), pp. 181–2. 124 ‘Uspensky Gospels’, St Petersburg, Rossiiskaia Natsionalnaia Biblioteka, RNB gr. 219 (see fig. 13).

280

the midd le empire

Figure 12 Zodiac from Ptolemy’s ‘Handy Tables’ – a hint of intellectual life continuing in the eighth century

aniconism does not stand up to scrutiny, and Jewish and Muslim influences were only indirect.125 The chroniclers do offer one plausible explanation for Leo III’s decision: that iconoclasm was triggered off by Leo’s false interpretation of the terrifying volcanic eruption at Thira in 726 as a manifestation of the wrath of God. He thought, wrongly they say, that God’s wrath had been caused by the idolatry of the empire’s subjects in venerating religious

125

Auz´epy (2004), pp. 135–42.

5. state of emergency (700–850)

281

Figure 13 Manuscript of the Uspensky Gospels, dated to 835 – the earliest dated example of the new ‘speed-writing’ in minuscule, easier to read as well as faster to write than the previous form, uncial

images.126 According to Theophanes, Leo launched a propaganda campaign in 725, and in the following year, the icon of Christ said to have been situated above the Great Palace’s bronze gate (Chalke) was reportedly destroyed.127 This culminated on 7 January 730 with Leo’s declaration against icons, during a silention in the Hall of the Nineteen Couches. Patriarch Germanos refused to subscribe and was forced to resign, and opponents of the decisions taken at the silention, clerics and laymen alike, were persecuted.128 This explanation given by the chronicles is plausible enough. An emperor who considered himself divinely commanded to ‘tend the most Christian people’ in a manner pleasing to God, thereby obtaining military victories and domestic peace,129 would understandably be concerned by such violent and repeated manifestations of God’s wrath against his people. In the 126 Nikeph., chs. 59–60, ed. and tr. Mango, pp. 128–9; Theoph., ed. de Boor, I, pp. 404–5; tr. Mango and Scott, pp. 558–60. 127 Theoph., ed. de Boor, I, p. 405; tr. Mango and Scott, pp. 559–60; Auz´ epy (1990); Auz´epy (2004), pp. 133–4. 128 Nikeph., chs. 60, 62, ed. and tr. Mango, pp. 128–31; Theoph., ed. de Boor, I, pp. 404–5, 407–9; tr. Mango and Scott, pp. 558–60, 562–7. 129 Ecloga, preface, lines 21–31, ed. and German tr. Burgmann, pp. 160–1; tr. Freshfield, pp. 66–7.

282

the midd le empire

Bible, the reason why God abandons his people to the point where they are vanquished and led off to captivity in Babylon – a prospect anything but theoretical in Byzantium of the 720s – is none other than idolatry. Iconoclasm, in banning the veneration of religious images through such gestures and ritual as bowing down (proskyn¯esis), kissing (aspasmos), and the burning of lights and incense, is in fact nothing but a rejection of idolatry.130 In all probability, Leo III aimed to please the Almighty by choosing iconoclasm; by banning idolatry, he might induce God to stop granting victories to the enemy and avert the threat of defeat and captivity from his people. And this choice might be considered a good one, seeing that God’s wrath did in fact subside. The danger from the Arabs, and then from the Bulgars, abated and Isaurian propaganda was, according to Patriarch Nikephoros, able to ascribe these successes, as well as the longevity of the emperors, to iconoclasm.131 Furthermore, it was precisely the renunciation of iconoclasm that coincided with defeat and then, under Nikephoros and Michael I, military disaster; and its reinstatement by Leo V in 815 was followed a year later by victory over the Bulgars. Thus, iconoclasm could well be considered as the religious component of the overall strategy for the empire’s survival. It is difficult to assess the cult of images at the beginning of the eighth century and to determine whether the Isaurians’ charge of idolatry was justified. Scholarly opinion is divided, particularly as to when this cult arose.132 At the council of Hieria in 754, the Isaurian bishops maintained that worship of images became widespread after the sixth council in 680.133 This is possible, though it has not been proved. What we know of devotional practices would suggest that Leo III’s diagnosis was not exaggerated. At the second council of Nicaea in 787, the bishops cited a passage from the Miracles of Cosmas and Damian as an authority favouring icons and their cults: this tells of a woman who was cured by drinking the scrapings from a fresco representing the saints.134 And, in a letter to Louis the Pious (814–40) seeking to justify his attachment to iconoclasm, Michael II listed the practices he considered inadmissible: fragments of icons mixed in the eucharist, icons which served as altar tables, or as sponsors in baptism or monastic tonsure.135 The popularity of icons and their cult might also be explained by the misfortunes of the times. Icons were a refuge when daily life was disrupted and institutions no longer worked; they permitted a direct relationship 130 131 132 133

Mansi, XIII, cols. 221C, 277D, 353C. Nikephoros, Antirrhetici, III.70, cols. 503–4; French tr. Mondzain-Baudinet, p. 278. Brubaker (1998) as against Delierneux (2001); Auz´epy (2004), pp. 152–6. Mansi, XIII, cols. 217A, 221C, 225D; Adversus Constantinum Caballinum, ch. 13, col. 329. See also p. 235. 134 Mansi, XIII, col. 68A–D; van den Ven (1955–7), no. 57, p. 356. 135 Michael II and Theophilos, Letter to Louis I, pp. 478–9; McCormick (1994a), pp. 145–9.

5. state of emergency (700–850)

283

with an intercessor, an individual devotion which did not need any church or clergy.136 This would explain why Leo III’s decision in 730 met with resistance – perhaps amplified by iconodule propaganda at the beginning of the ninth century, but still historically certain. Leo III overrode Patriarch Germanos’ opposition, replacing him with Anastasios, his synkellos, but then came up against the patriarchs of Rome and Jerusalem. John of Jerusalem, with the support of John Damascene, virulently opposed the imperial policy of iconoclasm and sent round a synodal decision hostile to the emperor’s proposition. This synodal decision, with later modifications, was transformed into a pamphlet entitled Adversus Constantinum Caballinum.137 For his part, Pope Gregory II (715–31) rejected the synodal decision of Anastasios and wrote Leo letters of reproach. The only version we have of these is one revised and corrected by later polemicists, who put into Leo’s mouth the famous saying: ‘I am emperor and priest.’138 Gregory’s successor, Gregory III (731–41), condemned iconoclasm in November 731 at a provincial council whose acts have been lost.139 There appears to have been no further debate on iconoclasm after this, even though Artabasdos authorised the cult of images while he held Constantinople.140 Constantine V finally decided to legitimise iconoclasm throughout the church by summoning a council; although this purported to be the seventh ecumenical council, it is generally known as the iconoclast council of Hieria. Judging by the quotations from his writings by Patriarch Nikephoros, collected under the title Peuseis, Constantine was himself a good theologian. He prepared the ground for the assembly with a public awareness campaign similar to that of his father before the silention of the Nineteen Couches. Constantine timed the council to coincide with the vacancy in the see of Constantinople following the death of Anastasios, assembling 338 bishops in the palace of Hieria on the eastern shore of the Bosporus from February to August 754.141 This council must have had a disciplinary aspect, for it issued numerous (non-extant) canons, but its main decision was to ban the making and venerating of religious images upon pain of punishment. The argument hinged on the portrayal of Christ: this was rejected on Christological grounds, since Christ (being God and man) could not be delimited in a material figure without falling into the error of Nestorianism or into the confusion of monophysitism.142 Unlike the decisions of the silention of 136 138

137 Auz´ Brown, P. (1973); Haldon (1997a), pp. 356–62. epy (1995a). LP, XCI.24, ed. Duchesne, I, p. 409; tr. Davis, II, p. 16; Gouillard (1968), pp. 253–76; PseudoGregory II, Letters, ed. and French tr. Gouillard, no. 2, pp. 298–9; Dagron (2003), pp. 158–66. 139 LP, XCII.2–3, ed. Duchesne, I, pp. 415–16; tr. Davis, II, pp. 19–20; McCormick (2001), pp. 867–8. 140 Theoph., ed. de Boor, I, p. 415; tr. Mango and Scott, pp. 575–6; Nikeph., ch. 64, ed. and tr. Mango, pp. 132–5; Gero (1977), pp. 15–19 as against Speck (1981), pp. 127–32. 141 Nikeph., ch. 72, ed. and tr. Mango, pp. 142–3; Theoph., ed. de Boor, I, pp. 427–8; tr. Mango and Scott, pp. 589–92; MS, XI.24, ed. and French tr. Chabot, II, p. 520. 142 Council of Hieria (754), Horos, Mansi, XIII, cols. 241, 244, 245, 252, 256.

284

the midd le empire

the Nineteen Couches, the council of Hieria appears at first to have been accepted without demur. Iconoclasm became the orthodoxy of the empire, if not the oikoumen¯e. The papacy waited fifteen years before condemning the imperial policy of iconoclasm and the council of Hieria at the Lateran council of 769. It was an internal crisis in the empire that brought Constantine V to violent action in the summer of 766, when a plot against him was thwarted. In fact the trouble began in November 765 with the execution of the monk Stephen the Younger, a friend of the conspirators. Stephen’s execution was followed by the persecution of officials and a compulsory oath for all imperial subjects, abjuring veneration of saints’ images. The climax came in August 766, when ceremonies of ridicule were organised in the Hippodrome – on 21 August against the monks, and four days later against the conspirators. Finally, there was a purge of those in positions of command. New strat¯egoi, faithful to the emperor and his doctrines, were named to the themes of the Anatolikoi and Thrakesioi. Patriarch Constantine, suspected of complicity in the plot, was replaced by Niketas on 16 November 766, before undergoing cruel and humiliating punishment almost a year later in October 767. The strat¯egos of the Thrakesioi, Michael Lachanodrakon, conducted a virulent campaign against both icons and monks in his theme, burning the former and forcing the latter to discard their monastic habits. Betrayal by the men he had trusted led the emperor to harden his policy. The persecution he launched against the monks was peculiar, aimed not at making them iconoclasts – which they had been since 754, like all imperial subjects – but at making them renounce their monastic state and take up clerical, civil or military positions. However, not all were affected, for Leo IV later named monks to the highest episcopal sees.143 rev italising the church of constantinople The Isaurians’ religious policy was not limited exclusively to iconoclasm, although this was its major feature. It also had international ramifications, defending and extending the rights of the church of Constantinople. Thus, relations with the pope were always connected with imperial policy in Italy; Rome was subject to the empire, at least until 751, when the exarchate of Ravenna was seized by the Lombards. Indeed, according to Theophanes, Leo III’s iconoclasm was the reason why Italy seceded.144 But the Roman Liber pontificalis reports that Gregory II was initially opposed to the emperor’s attempt to bring the taxes of the province of Rome into line with 143 Nikeph., chs. 79–84, ed. and tr. Mango, pp. 150–61; Theoph., ed. de Boor, I, pp. 436–43, 449, 453; tr. Mango and Scott, pp. 603–12, 619–21, 625–6; SD, ed. and French tr. Auz´epy, pp. 21–42 (introduction). 144 Theoph., ed. de Boor, I, p. 413; tr. Mango and Scott, pp. 572–3.

5. state of emergency (700–850)

285

those of the other provinces, particularly those payable by the churches; it was only later – at some indeterminate date after 725 – that Gregory refused the emperor’s demand that he should accept iconoclasm. The reaction of Leo III to papal opposition, which stiffened after the silention of the Nineteen Couches and Germanos’ resignation, was highly effective. The church of Rome was deprived of the patrimony of St Peter both in Sicily, where imperial strat¯egoi had been sent since the end of the seventh century, and in Calabria, which was dependent on the strat¯egos of Sicily.145 Thus after 732–3 the income from certain estates traditionally allocated to the church of Rome for, amongst other things, the lighting of places of worship and maintenance of the poor, reverted to the central tax administration – a hefty annual sum of some 350 pounds of gold, or 25,200 nomismata.146 This measure perhaps explains the relatively high number of issues from the mint at Syracuse, in a period of general monetary restriction. The mint never struck silver coins, only copper and gold, and in fractions of the nomisma that had disappeared in the rest of the empire; although devalued after 820, this coinage circulated throughout Europe as far as the Crimea.147 In 743, Constantine V partially compensated Gregory III’s successor, Zacharias (741–52), for the loss incurred by the reallocation of these revenues by granting him the estates of Ninfa and Norma to the south of Rome. Lying between the hills of Volsci and the sea, they were not far from Terracina and Gaeta which were later claimed by the duchy of Naples, when it was an ally of the Byzantines.148 Finally, Leo III decided to harmonise political and ecclesiastical structures – probably in tandem with the measures of 732–3, and in any case before 754 – by placing such regions as were under the direct or indirect authority of the eastern empire beneath the patriarch of Constantinople’s jurisdiction. The bishops of Illyricum, Crete, Sicily, Calabria and of the duchy of Naples, formerly subordinate to Rome, found themselves under the authority of Constantinople – although in the case of Naples, this lasted only until 769.149 The transfer of these regions to Constantinople was probably accompanied by the confiscation of possessions from the church of Rome, such as happened in Istria in the 770s.150

145 146

LP, XCI.16, 17, ed. Duchesne, I, pp. 403, 404–5; tr. Davis, II, pp. 10, 11–12. Theoph., ed. de Boor, I, p. 410; tr. Mango and Scott, pp. 567–70; Hadrian I, Letter, Mansi, XII, col. 1073; Noy´e (1998), p. 233. 147 Morrisson (1998); Morrisson (2001). 148 LP, XCIII.20, ed. Duchesne, I, pp. 433, 438 and n. 45; tr. Davis, II, pp. 20, 46 and n. 78; Saxer (2001), p. 531; CC, no. 64, ed. Gundlach, p. 591; partial tr. King, pp. 289–90; McCormick (2001), p. 878. See also p. 443. 149 Hadrian I, Letter, Mansi, XII, col. 1073; Anastos (1957). 150 Saxer (2001), p. 531; CC, no. 63, ed. Gundlach, p. 590; partial tr. King, p. 289; McCormick (2001), p. 878.

286

the midd le empire

The popes never accepted this reorganisation at patriarchal level, which was coupled with an adjustment of ecclesiastical structures to match the military situation. This entailed a similar reorganisation at episcopal level, disregarded until recently because of scholarly doubts about the sourcevalue of the list of bishoprics known as the Notitia of the iconoclasts (Notitiae episcopatuum). Recent study, however, has shown that on several points this text gives an accurate picture of the church in the eighth century.151 In Calabria and Thrace, the kastra built to accommodate the military and administrative authorities received the status of both city and bishopric from the emperor; the imperial right to bestow such status had been acknowledged at the council of Chalcedon (canon 17) and incorporated into canon 38 of the council in Trullo. Calabrian examples include Gerace, sometime before 787,152 and the creation of the archbishopric of Santa Severina, probably after 736.153 In Thrace, kastra such as Bulgarophygon, Skopelos and Develtos, which had been built or renovated by Constantine V and settled with captives taken on the Arab border, were made bishoprics, as their bishops’ presence at the second council of Nicaea shows.154 In Greece the bishoprics of Epirus Primus encircled the Peloponnese from Cephalonia to Aegina. They were sometimes located on smaller islands such as Orobe, which also served as relays for the fleet, as the numerous seals found on them attest.155 The route to Italy was thus guarded by a military as well as an ecclesiastical network. This use of the church provides the background to Constantine V’s policy towards the monasteries. His persecution cannot be described as bloody, seeing that it caused only two deaths, but he lifted previous exemptions from both individuals and property, reimposing liability to contribute to the state: monasteries and monastic lands as well as episcopal estates were sold, confiscated for military purposes or reallocated to the armed forces.156 The Isaurians saw the church as a reputable institution for which they were responsible, and the patriarchs as enforcers of the imperial will in ecclesiastical matters. Their expectations of subordination could sometimes be harsh and humiliating, as when Anastasios was paraded round the Hippodrome on an ass after the defeat of Artabasdos and then restored to the patriarchal throne; but the gift of the extension of jurisdiction westwards was ample compensation. The later iconodule patriarchs never questioned this gift to their institution made by an iconoclast emperor whom they had anathematised. In the Acts of the second council of Nicaea, Patriarch Tarasios (784–806) omitted translating into Greek the passages of 151 152 154 155

Kountoura-Galake (1996b), pp. 121–43; Prigent (2002). 153 Prigent (2002), pp. 939–46. Noy´e (1998), p. 234; Prigent (2002), p. 938. Kountoura-Galake (1997); Darrouz`es (1975), p. 54; Lamberz (2004), pp. 74–5. NE, no. 3, p. 245 (text); Darrouz`es (1975), p. 37–8; Lamberz (2004), pp. 62–3; Avramea (1997), p. 99.

5. state of emergency (700–850)

287

Pope Hadrian I’s letter demanding the restitution of the patrimony of St Peter.157 Bishops formed the most important rank of the ecclesiastical hierarchy, residing in their sees and taking responsibility for charitable works, amongst other things. Judging by iconoclast-linked hagiographical sources, the bishop was deemed a model of holiness even worthier of emulation than the monk. However, it is possible that the episcopal office was the equivalent of a strateia, which could be purchased and which carried with it rights to the bishopric’s revenues. In the Ecloga clerics are sometimes described as strateuomenoi, and even Tarasios admitted after the second council of Nicaea that most of the bishops present had bought their office, leaving them open to charges of simony by the monks.158 Iconoclasm provided the cement for the edifice, with duly anointed clergy delivering the people from idolatry through celebration of the eucharist. Constantine V sought to spread the doctrine adopted at Hieria throughout Christendom (see above, pp. 283–4) and conducted a vigorous diplomatic campaign aimed at Pippin the Short (751–68), sending numerous embassies and, it would seem, several eastern patriarchs. This campaign, whose success Pope Paul I briefly feared, finally ended in failure in the wake of the crisis of 766. The council of Gentilly (Easter 767) ratified the Carolingian rejection of iconoclasm, making possible its subsequent condemnation by the papacy at the Lateran council in 769; meanwhile, at least some of the eastern patriarchs sent a written condemnation of iconoclasm to Rome, which arrived in August 767.159 The church of Constantinople was thereupon cut off from the other churches, and this probably explains Irene’s desire to put an end to the situation. from the second council of nicaea (787) to t he s y n o d i ko n o f o rt h o d ox y (843) The switch in imperial religious policy was the work of Irene and of Tarasios, the patriarch whom she had chosen after the death of his predecessor, Paul IV. Formerly head of the imperial chancellery (pr¯otasekr¯etis), on Christmas Day 784 Tarasios was elevated from layman to the patriarchal throne – a step frowned upon by Pope Hadrian I, as the Lateran council had banned this type of episcopal election.160 Irene and Tarasios sought to give their new 156 158

157 Lamberz (2001). SD, ed. and French tr. Auz´epy, pp. 36–7 (introduction). Life of George of Amastris, chs. 16, 23–4, pp. 26–7, 36–7; Auz´epy (1992), pp. 60–2; Mansi, XIII, col. 474E; Auz´epy (1988), pp. 18–19. 159 McCormick (1994a), pp. 113–32; Auz´ epy (1999), pp. 215–28. 160 Ignatios the Deacon, Life of Tarasios, chs. 8–17, ed. and tr. Efthymiadis, pp. 78–91, 174–8 and p. 14 (introduction); Theoph., ed. de Boor, I, pp. 458–61; tr. Mango and Scott, pp. 631–5; LP, XCVI.20, ed. Duchesne, I, p. 476; tr. Davis, II, p. 99.

288

the midd le empire

policy a solemnity comparable to that of the council of Hieria by calling an ecumenical council, and they gained the support of the pope, who sent two legates. However, this abrupt change in position did not go unopposed. The council first assembled on 1 August 786 in the church of the Holy Apostles in Constantinople, but was dispersed by soldiers of the tagmata – the schools and excubitors – who had the backing of several of the bishops taking part.161 This did not discourage the empress. That autumn she tricked the soldiers of the tagmata into dispersing, by sending them to Asia Minor in response to an alleged Arab attack. She then had them disarmed, and reconstituted the City units with soldiers taken from the thematic army corps.162 Tarasios, for his part, disarmed episcopal opposition; his first, novel, strategy was to invite the monks – possibly a group opposed to the bishops – to participate in the council, while allowing the bishops to keep their posts only on condition that the most notorious iconoclasts among them made a public admission of error. He also passed off two eastern monks as official envoys from the patriarchates of Antioch and Alexandria, so as to justify the council’s claim to ecumenical status.163 Finally, it was not in Constantinople but at Nicaea – a place full of symbolism for Christendom – that the 365 bishops and 132 monks assembled from 11 September to 1 October 787, and declared the making and venerating of religious images an article of faith.164 Christological arguments were abandoned in favour of other, less substantial ones, or even for unproven affirmations: the antiquity of icons and their cult; the impossibility of idolatry – necessarily pagan – after the coming of Christ; insistence on the incarnation of Christ, which rendered acceptable representation of that which had been seen; dismissal of the charge of idolatry by virtue of the name inscribed on the icon. The council’s ruling that icons should receive proskyn¯esis, after thirty years’ prohibition as idolatry, was not accepted as easily as iconodule propaganda would have us believe. The Libri Carolini mention civil war, and many of our early ninth-century sources – which are admittedly iconodule – deplore the number of people, clergymen included, who remained convinced iconoclasts, even if they stopped short of considering Constantine V a saint.165 During the reign of Michael I (811–13), Theophanes’ Chronicle mentions the heretics known as Athinganoi in Phrygia – probably the successors of the 161 Mansi, XII, cols. 990B–991D, XIII, col. 459C; Ignatios the Deacon, Life of Tarasios, ch. 26, ed. and tr. Efthymiadis, pp. 100–1, 183; Theoph., ed. de Boor, I, pp. 461–2; tr. Mango and Scott, pp. 634–7. 162 Theoph., ed. de Boor, I, p. 462; tr. Mango and Scott, pp. 635–7; Haldon (1984), pp. 239–42. 163 Auz´ epy (1988); Auz´epy (1999) pp. 211–28. 164 Darrouz` es (1975); Lamberz (2004); Janin (1975), pp. 427–41; Herrin (1987), pp. 417–23. 165 Opus Caroli, II.24, III (praefatio), ed. Freeman and Meyvaert, pp. 282, 329; SD, ch. 38, ed. and French tr. Auz´epy, pp. 137–8, 232–4; Theoph., ed. de Boor, I, pp. 488–9, 496–7; tr. Mango and Scott, pp. 668–72, 679–82; Nikephoros, Antirrhetici, III.69, cols. 501–4; French tr. Mondzain-Baudinet, pp. 277–8.

5. state of emergency (700–850)

289

Montanists whose forced conversion under Leo III had failed – as well as the Paulicians. The latter had been active in the region of the Pontus since the end of the seventh century and under Theophilos and especially under Michael III, they became a military threat to the empire on the borders with the caliphate.166 They were persecuted at the instigation of Patriarch Nikephoros, who pressed for capital punishment for them as well as for iconoclast abbots. Religious policy changed again in 815 when Leo V (813–20), raised to the throne to save the empire, reinstated iconoclasm. Much like Leo III before him, he was unable to convince his patriarch, who was sent into exile. The change in policy was formalised by a provincial council assembled after Easter by the new patriarch, Theodotos Kassiteras; Theodotos was of the Melissenoi family and, like his predecessor, was a layman promoted overnight to the patriarchal throne. Hieria was re-established and the second council of Nicaea overturned. The cult of icons was forbidden, but this new iconoclasm was less intransigent: there was no longer any question of idols or idolatry, and images which were not actively venerated, for example those suspended high up, were permitted. New emphasis was laid on the argument that, since man is made in the image of God, any additional material image is superfluous.167 In its preliminary stages between Christmas 814 and Easter 815, this new position initially met with staunch opposition from the patriarch, Nikephoros; from the bishops who had supported Tarasios or who had been trained by him, such as Euthymios of Sardis, Michael of Synada and Theophylact of Nikomedeia; and from many monks who had received their instruction during the iconodule interval. Foremost among the rebellious clergymen was Theodore, head of the great Constantinopolitan monastery of Stoudios. Theodore had already distinguished himself by his intransigence both to imperial power and to the patriarch, notably when Tarasios acceded to Constantine VI’s second marriage. Theodore went into exile for a second time in 815, and was followed by other bishops and abbots who rejected the return to iconoclasm, notably Theophanes the Confessor, Makarios of Pelekete and Niketas of Medikion from Bithynia. This generation of anti-iconoclasts is well known, for it was celebrated in numerous saints’ Lives.168 Nevertheless, they were kept on the sidelines and iconoclasm remained the religious law of the empire under Michael II – who continued Leo V’s policies, despite arranging his murder – and Theophilos. Michael II recalled those who had been exiled, but did not restore them to their positions, and the persecution under Theophilos, 166 Theoph., ed. de Boor, I, p. 495; tr. Mango and Scott, pp. 678–9; Gouillard (1965), pp. 307–12; Lemerle (1973), pp. 75–90. 167 Alexander (1953); Alexander (1958b). 168 Alexander (1958a), pp. 136–55; Byzantine defenders of images, ed. Talbot.

290

the midd le empire

Figure 14 The emperor who resumed the drive against ‘idols’ and his fellow iconoclasts were execrated by iconodules after the restoration of icon-veneration in 843 and throughout the later history of Byzantium and orthodoxy: this miniature from the Theodore Psalter of 1066 from the Stoudios Monastery in Constantinople shows Theodore the Stoudite in discussion with Leo V and Patriarch Nikephoros, beside iconoclasts defacing the image of Christ

denounced in the hagiographic texts, seems to have been motivated more by political than by religious considerations.169 Theophilos’ death when his son was barely two years old brought about a final change in religious policy, less well documented than the earlier ones. Yet again the patriarch was got rid of: John the Grammarian (837–42) was replaced by Methodios, an iconodule monk with so much influence that Theophilos had preferred to keep him close at hand in the palace. But this time the re-establishment of images was definitive; it was not only the result of a decision by an assembly of church authorities, but also took liturgical form on 11 March 843, the first Sunday of Lent. This Sunday became a celebration of the re-establishment of images and was soon named the Feast of orthodoxy, when the Synodikon of orthodoxy was read out. This document, celebrating the triumph over the iconoclasts’ heresy, has become the symbol of the orthodox faith, receiving successive additions over the centuries concerning other heresies.170 From the 850s iconoclasm belonged to the past in so far as Patriarch Methodios had purged the clergy, but the repeated ‘U-turns’ in religious policy over more than a century had lasting consequences. With the victory of iconodulism the patriarchate was able to affirm the church’s autonomy from imperial power, which was accused of encroaching on its domain by imposing religious norms, as encapsulated in the celebrated phrase put in Leo III’s mouth: ‘I am emperor and priest.’ Moves towards an independent sphere for the church had begun in Rome in the seventh century with the monothelite crisis. The pope then affirmed his autonomy in the eighth century by associating himself with the Carolingians and creating his own temporal base. In the ninth century it was the patriarch of Constantinople who, through opposing the imperial policy of iconoclasm with the help of unprecedentedly vehement texts, excluded the emperor from questions of dogma and assumed the position of head of the church. This new 169 170

Dagron (1993), pp. 143–7; Auz´epy (2003), pp. 432–9. Gouillard, ‘Synodikon de l’Orthodoxie’, pp. 120–33, 161–8.

5. state of emergency (700–850)

291

equilibrium between church and imperial power – albeit unequal because the emperors still named the patriarchs – was expressed in the Eisag¯og¯e; probably drawn up by Photios, this described the patriarch as the ‘image of Christ’, whereas the emperor is only the ‘legitimate authority’.171 By the end of the iconoclast period the empire had been transformed. Henceforth it would be characterised by religious peace; by a new orthodoxy now fixed in liturgical and iconic repetition; by a provincial administration structured by war around the theme, which had become the sole civil and military administrative unit; and by a central administration regrouped around the emperor but divided into large departments. The shock of the invasions had passed and, owing in part to the survival of its capital, the empire metamorphosed into a great Balkano-Anatolian power, administered coherently and financed through an effective tax system. This transformation made possible the initiatives and achievements of the period that followed. 171 Auz´ epy (1999), pp. 300–11; Afinogenov (1994); Afinogenov (1996); Dagron (1993), pp. 202–7; Dagron (2003), pp. 223–35.

CHAPTER 6

AFTER ICONOCLASM (850–886) s hau n tou ghe r

introduction Two emperors dominate the generation or so following iconoclasm, Michael III the Amorian (842–67) and Basil I the Macedonian (867–86).1 The story of this pair is intimately intertwined, although it climaxed with the assassination of the former at the instigation of the latter on the night of 23 September 867 in Michael’s bedroom in the palace of St Mamas. Thus began the long ascendancy of the Macedonian dynasty, which witnessed the peak of Byzantium’s power. A clear understanding of the reigns of Michael and Basil is, however, fraught with difficulty given the nature of our main narrative sources. These are both late – dating to the mid-tenth century – and polarised.2 The Macedonians were naturally keen to justify the ousting of Michael III, so he is depicted in Theophanes Continuatus and Genesios as unworthy of imperial power and deserving of his fate.3 The Macedonians were also concerned to present Basil in the best possible light, as God-favoured and preordained to rule.4 The most famous expression of this is the Life of Basil (which forms book five of Theophanes Continuatus’ chronicle), written in the reign of his grandson Constantine VII Porphyrogenitus (945–59), though we also have Leo VI’s Funeral oration for Basil I (dated to 888) and poems and artefacts from Basil’s reign.5 However, a hostile view of Basil is provided by the chronicle of Symeon the Logothete, which also treats Michael more ambiguously.6 Despite these sources’ 1 There is still no monograph on Michael III, but see PMBZ, #4991. For Basil I we have Vogt (1908); see also PMBZ, #832. 2 For the narrative sources in general see Markopoulos (2003) with bibliography; for Michael’s reign: Karlin-Hayter (1971). 3 For the Chronographia of Theophanes Continuatus see Nickles (1937); Jenkins (1948); Jenkins (1954). On Genesios, see the introduction to Gen., tr. and comm. Kaldellis. 4 See Agapitos (1989). 5 On the authorship of the Life of Basil see Sevˇ ˇ cenko, I. (1992a). For the funeral oration: Leo VI, Funeral oration for Basil I, ed. and French tr. Vogt and Hausherr; Adontz (1933). For other sources and artefacts see e.g. Markopoulos (1992); Magdalino (1987); Brubaker (1999a), pp. 147–200. 6 On the chronicle of Symeon the Logothete: Bury (1912), pp. 455–9; Jenkins (1965); Karlin-Hayter (1991a); Wahlgren (2001). See now Symeon [Magister] the Logothete, Chronicle, ed. Wahlgren.

292

6. after iconocl asm (850–886)

293

polarity and emphasis on court politics it is clear that there was continuity in the goals of the two regimes. The security of the east was paramount, although the west was still of concern. The government also had to cope with the Arab naval menace and the potential Bulgar threat from the north. New opportunities were seized when they arose, among Moravians, Armenians and the Rus. Such were the achievements of an era that it is usually characterised as a decisive turning-point, if not a belle ´epoque. But it is clear that they followed upon an already advancing recovery, as much a cultural as a political revival. One should also recognise that this was not a period of unbroken success. court politics 8 42 –867 The elimination of Michael III was the culmination of a power struggle that had been going on for most of his reign. When Michael succeeded his father in 842 he was only a child, probably two years old.7 This necessitated the establishment of a regency headed by his mother, Empress Theodora (842–56).8 The dominant force, however, appears to have been the eunuch Theoktistos, who was logothete of the Drome and epi tou kanikleiou, and a key agent of the Amorian dynasty.9 Theoktistos appears to have had no love for the empress’ brother Bardas, who found himself excluded from political power. When Michael approached adulthood, but was still constrained by the wishes of Theodora and Theoktistos – who arranged his marriage to Eudocia Dekapolitissa despite his supposed attachment to Eudocia Ingerina – it seems that Bardas seized the moment and persuaded his nephew to consent to a plot to remove Theoktistos. In 855 the eunuch was murdered in the palace and Theodora soon found herself formally barred from government. She still appears to have hankered after position and influence, for she was implicated in a plot against Bardas. He, however, was firmly in the ascendant, and is credited with running the empire until his death in 866. Bardas’ government is often glowingly praised, as is his fostering of intellectual life by the establishment of the school of the Magnaura.10 His importance is reflected by his developing career as well as the careers of those close to him. He served as domestic of the Schools and rose through a series of titles, eventually attaining the honour of caesar. His brother Petronas replaced him as domestic of the Schools. Photios, a relative of the Amorian house11 and a close ally of Bardas, became patriarch of Constantinople in 858 on the deposition of Ignatios, who had opposed Bardas. 7 9 10

8 On Theodora: Garland (1999), pp. 95–108; Herrin (2001), pp. 185–239. Mango (1967). For a positive assessment of Theoktistos: Gr´egoire (1966), pp. 105–8. Bury (1912), pp. 161, 439; Jenkins (1966), pp. 160, 164; Ostrogorsky (1968), pp. 223–4. On the school: Lemerle (1986), pp. 183–5. 11 Mango (1977), esp. pp. 9–12.

294

the midd le empire

What the ultimate ambitions of Bardas were is a moot point; his career was cut short in 866 when he was murdered while preparing to embark on an expedition to Crete. Afterwards the reason given for his death was that he had been plotting to overthrow Michael, but it is possible that Bardas had simply fallen victim to the ambitions of others, not least the emperor’s favourite, Basil the Macedonian. The origins of Basil the Macedonian are obscure and the story of his rise to prominence and power is spiced with colourful episodes, still the stuff of analysis and debate.12 Basil’s sobriquet, ‘the Macedonian’, is thought to refer to his provenance from the theme of Macedonia, not Macedonia itself, although the Life of Basil asserts that his ancestors were originally settled in Macedonia. The Macedonian claim recalled the famed figures of Philip and Alexander the Great.13 Basil seems also to have had Armenian blood in his veins,14 and the dynasty was to claim descent from the Arsacids.15 Our narrative sources relate that the infant Basil was among the citizens of Adrianople seized and transported across the Danube by the Bulgar khan Krum (c. 803–14), and Symeon the Logothete specifies that he was born in the reign of Michael I (811–13).16 While some scholars accept this date of birth, estimating that Basil was fifty-five years old when he became emperor,17 others are less sure.18 Certainly the narrative sources also depict Basil as still a young man when he made his way to Constantinople to make his fortune in the mid-850s. The stories about Basil’s developing career, dependent on the favour of a variety of patrons, suggest a lowly background, undermining the Life of Basil ’s assertion of his not undistinguished ancestry. His peasant origins are not contested by modern historians, and are in fact supported by the Davidic imagery embraced by Basil.19 Having been taken in by Nicholas, keeper of the church of St Diomedes – in whose porch Basil had slept on his first night in Constantinople – Basil soon moved on to the service of Theophilitzes; he ended up working for Michael III himself, having cemented a social relationship with the wealthy Peloponnesian 12 13

Moravcsik (1961). Schminck (2000), pp. 67–8, argues that this is why Basil was fancied to be a Macedonian, rejecting the view that Adrianople was in the theme of Macedonia. The Life of Basil claims that Basil’s mother was descended from both Alexander the Great and Constantine the Great: Life of Basil, pp. 215–16. On the Macedonian link with Constantine, see Markopoulos (1994). 14 Adontz (1933–4). Ostrogorsky (1968), p. 232, n. 2, and Schminck (2000), pp. 64–7, are sceptical, but most seem happy with Basil’s Armenian origin, e.g. Treadgold (2001), p. 133. 15 Markopoulos (1994), p. 163. According to the Life of Ignatios (cols. 565, 568) and Pseudo-Symeon (Chronicle, p. 689) Photios invented a genealogy for Basil which made him a descendant of the Armenian king Tiridates. 16 GMC, p. 817. On the Bulgarian episode see Kislinger (1981). 17 Treadgold (2001), p. 133. 18 For example Brooks (1911); Adontz (1933–4). 19 See Markopoulos (1992), pp. 227–8; Markopoulos (1994), pp. 161–2; Brubaker (1999a), pp. 183–93; Dagron (2003), pp. 199–200.

6. after iconocl asm (850–886)

295

Figure 15 Illustration from the Madrid Skylitzes: Michael III racing near the church of St Mamas

widow Danelis along the way.20 Genesios relates that Michael first met Basil after hearing of his participation in a wrestling match and summoning the wrestlers to come before him; but our other sources assert that the encounter occurred when Basil managed to break in one of the emperor’s horses. This episode has suspicious overtones of Alexander and Bucephalus,21 but it does tie in with Basil’s subsequent career. He was enrolled among the imperial grooms, and after the failure of Theodora’s plot against Bardas he was made head groom (pr¯otostrat¯or); the previous incumbent of the post had been executed as a conspirator. Perhaps Basil’s equine skills endeared him to Michael, since the emperor’s passions included hunting and chariot-racing; his enthusiasm for equestrian pastimes is also conveyed by his construction of luxury stables. Our narrative sources make clear that there was an intense bond between Michael and Basil.22 It has been suggested that there was in fact a sexual relationship between the two men, although some scholars have no truck with this hypothesis.23 The appointment of Basil as parakoim¯omenos (after the fall of his predecessor Damian) does suggest an unusual state of 20 After becoming emperor, Basil rewarded Nicholas, his family and the institution. On Theophilitzes: Tougher (1999), p. 154. For Danelis: Tougher (1997b), pp. 27–8, 129–30, with bibliography. 21 Gen., IV.26, ed. Lesm¨ uller-Werner and Thurn, p. 78; tr. Kaldellis, p. 98 makes explicit the echo, and also recalls Bellerophon and Pegasus. 22 For example, the strong language used to describe their relationship: e.g. GMC, pp. 825, 832; Gen., IV.27–8, ed. Lesm¨uller-Werner and Thurn, pp. 78–80; tr. Kaldellis, pp. 98–9. 23 For discussion: Tougher (1999). For acceptance of Michael’s homosexuality, see Schminck (2000), pp. 61–4; for rejection, Treadgold (1997), p. 943, n. 11.

296

the midd le empire

affairs, seeing that the position was normally assigned to eunuchs.24 The name of the post itself, though generally understood as ‘chief eunuch’ or ‘grand chamberlain’, literally means ‘sleeping beside’ and thus indicates close physical proximity. The question of the relationship between Michael and Basil is further complicated by the fact that Basil married Eudocia Ingerina. Symeon the Logothete provides further information about this union: Michael arranged for Basil to marry Eudocia, having separated him from his first wife Maria. However, Eudocia was to remain as the emperor’s mistress, while Basil was to have Michael’s sister Thekla in recompense. Some historians have accepted these details, as well as Symeon the Logothete’s report that the future Leo VI (886–912) was Michael’s son.25 The account and its interpretation can, however, be questioned.26 The hostility of the Logothete should not be overlooked, nor should details be cherry-picked.27 It is perfectly possible that Leo was a son of Basil. Whatever the truth about the relationship between Michael, Basil and Eudocia, it is clear that the two men were still close in 866 when Bardas was assassinated. Following their return to Constantinople after aborting the Cretan expedition, Basil was adopted by Michael, given the title of magistros and then quickly crowned as co-emperor. Since Basil had Michael murdered in his bedchamber just over a year later, their relations obviously deteriorated. Perhaps Basil had always set his sights on sole power, or perhaps Michael and Basil simply lost trust in one another. The sources present alternatives according to their biases, and we must judge for ourselves. Clearly Michael suffers from receiving negative treatment, and some have sought to defend his reputation, even to the extent of claiming him as a great emperor.28 Others have recognised the hostility of the Macedonians but have felt that to reinterpret Michael as great is going too far, and that the sources’ calumny contains a grain of truth.29 It is acknowledged, too, that whatever one thinks of Basil’s motives his reign was generally successful, even if the Life of Basil exaggerates his greatness. Certainly Basil built on the achievements of the age of Michael III, and it is salutary to turn from court conflicts to the continuities of foreign policy. 24 25 26 27

Tougher (1997a), pp. 171–2. The classic interpretation is Mango (1973a). Treadgold (1997), p. 453 accepts Mango’s case. E.g. Kislinger (1983), pp. 128–32; Karlin-Hayter (1991b); Tougher (1997b), pp. 43–5. For instance Symeon the Logothete declares that Constantine, too, was a son of Michael. This is argued away by asserting that Constantine was a son of Basil and Maria (e.g. Adontz (1933), p. 509); but there is no Byzantine testimony to this effect. It is possible that Constantine was a son of Basil and Eudocia. 28 On Michael’s image: Kislinger (1987); Liubarsky (1987). Henri Gr´ egoire was a notable champion of Michael, e.g. Gr´egoire (1966), pp. 105–15; see also Karlin-Hayter (1989). 29 E.g. Bury (1912), p. 162; Jenkins (1966), pp. 156–7; Ostrogorsky (1968), p. 223. Treadgold (1997), pp. 450–5 accepts the negative image of Michael.

6. after iconocl asm (850–886)

297

ex ternal af fairs 8 50–886 The main concern of Byzantium was the Muslim threat.30 The focus fell naturally on the eastern frontier, beyond which lay the Abbasid caliphate. By the mid-ninth century, however, the caliphate was no longer launching full-scale invasions, and the raids into Byzantine territory were largely headed by the amirs of Tarsus and Melitene. The reign of Michael III was marked by a series of successes against the Arabs on land: there were Byzantine victories in 855 and 859, the latter led by the emperor himself. Inscriptions at Nicaea and Ankyra from this period recording their fortification by Michael are suggestive of a concerted effort.31 However, 863 is the famous date, often seen as a turning-point in the Byzantine–Arab conflict on the eastern frontier; thereafter the Byzantines were able to go on the offensive, eventually triumphing in the tenth century.32 In this year Michael’s uncle Petronas defeated the army of the amir of Melitene, who was killed in the engagement.33 During Basil’s reign, however, the Byzantines were preoccupied with crushing the Paulicians. The Paulicians were a religious group of Armenian origin deemed heretical by orthodox Byzantines, and they formed distinctive communities in the eastern borderlands.34 Following the restoration of icons under Theodora they were severely persecuted, but found sanctuary on the upper Euphrates, and Tephrike became their power centre. The Paulicians joined the Byzantines’ enemies on the eastern frontier, assisting the raids of the amir of Melitene. Their leader Karbeas died in 863, but his nephew and successor Chrysocheir appears to have been even more formidable, penetrating into Asia Minor. It was the domestic of the Schools and relative of Basil, Christopher, who managed to defeat the Paulician leader in 872, though Tephrike was only taken in 878. Basil’s efforts against Arab targets had more limited success, and his reign witnessed defeats such as the failed attack on Tarsus in 883. It seems that the reorganisation of the eastern frontier in the second half of the ninth and early tenth centuries was as important as military victories for increasing Byzantium’s strength.35 The Byzantines did not just face land war in the east; the Arabs were also a potent naval threat. The struggle for security at sea had intensified after Muslims originally from Spain had seized Crete, a vital strategic location, in the 820s (see above, p. 256). The Byzantines tried to rectify this situation. In the first year of the regency Theoktistos led an expedition to Crete, and Michael and Bardas were preparing to sail there when the caesar was 30 For Byzantium and the Arabs, see below, ch. 9; Vasiliev (1935–68). See also the translation of al-Tabari, Ta’rikh, general ed. Yarshater. 31 Gr´ 32 Whittow (1996a), p. 311. egoire (1966), p. 110. See also above, p. 265. 33 Huxley (1975). 34 See Lemerle (1973); Ludwig (1998); above, p. 289. 35 Whittow (1996a), p. 315.

298

the midd le empire

murdered in 866. The reoccupation of Crete was clearly a consistent goal, but one only achieved in 961. The Byzantines are, however, credited with a successful assault on Damietta in Egypt in 853.36 A naval response was also called for in the case of Sicily, as the Arabs extended their control of the island and threatened southern Italy: their castle-by-castle advance culminated in the fall of Syracuse in 878.37 Despite this event Basil I launched a concerted effort to maintain Byzantine power in the west.38 When the Arabs threatened Ragusa in 867, the emperor responded emphatically, no doubt as concerned to stem the expansion of the Arabs as to tackle the specific problem of southern Italy and Sicily. To address the latter situation in 868 Basil entered into alliance with the Frankish emperor Louis II (855–75), who was campaigning against the Arabs in southern Italy on his own account, with the Byzantines supplying naval assistance. This arrangement was cemented with the engagement of Basil’s eldest son Constantine to Louis’ daughter. However, the alliance foundered, and Louis’ ambitions faltered and then died with him in 875.39 Despite this setback Basil maintained his aspirations. Otranto was occupied in 873, and three years later Bari was regained, as was Taranto in 880. In the closing years of Basil’s reign the general Nikephoros Phokas (grandfather of the future emperor of the same name) was active in southern Italy, and increased Byzantine control of Apulia and Calabria.40 It appears that the successes of the early Macedonians were assisted by the revival of the imperial fleet and the creation of new naval themes.41 Basil was well served by admirals such as Niketas Ooryphas and Nasar (anticipating Himerios under Leo VI), who were active throughout the Mediterranean; one success was the temporary occupation of Cyprus. Thus although Sicily slipped inexorably from Byzantine control, the empire did provide some response to the Arab naval threat. Yet this remained intractable, persisting into the reign of Leo VI (see below, pp. 499–500). A strong presence was, however, re-established in southern Italy, and was soon enhanced. Byzantine ambitions there remained live down to the twelfth century. Besides the Muslims, the Byzantines’ other major bugbear had been the Bulgars on the northern frontier, with their centre near the lower Danube at Pliska.42 As recently as 811 Nikephoros I (802–11) had been killed on campaign against them. Khan Krum subsequently ventured against Constantinople, only to die in 814 (see above, p. 257). Following his death there was an extended phase of peace between Byzantium and the Bulgars. 36 38 40 41 42

37 See below, p. 462. Gr´egoire (1966), pp. 106–7. 39 See below, p. 419. See Gay (1904), pp. 79–145; Kreutz (1991), pp. 41–7, 62–3. On Nikephoros Phokas: Gr´egoire (1953); Tougher (1997b), pp. 204–7; below, pp. 504, 560. Ahrweiler (1966), pp. 96–9. See also above, p. 286. Fine (1983); Whittow (1996a), pp. 262–85.

6. after iconocl asm (850–886)

299

For the mid-ninth century the key issue was religion. Under Khan Boris (c. 852–89) Christianity was spreading in the Balkans, and Boris contemplated conversion. He sought missionaries from the Franks, but Byzantium was probably uneasy at Frankish interference so close to Constantinople. The exact course of events is controversial, but whether or not Boris was threatened by a Byzantine invasion, Boris turned to Byzantium for a Christian mission.43 In the mid-860s Boris was baptised, taking the Christian name of his godfather, Michael III himself. Thus it looked as if Byzantine cultural influence in Bulgaria was assured. However, in 866 Boris turned to the papacy, seeking advice and an archbishop from Pope Nicholas I (858–67), who was then happy to score points against Constantinople.44 Papal missionaries replaced Byzantine ones. But Boris found his plans for the Bulgarian archbishopric thwarted, and in 870 cannily returned to the Byzantine fold; he procured an archbishop by skilful manoeuvring at the time of the 869–70 church council in Constantinople. Byzantine cultural influence was secured, although this also fuelled the political ambitions of Bulgaria, which were to burst forth under Boris’ even cannier son, Symeon (893–927). For the reigns of Michael III and Basil I, though, the relationship between Byzantium and Bulgaria was remarkably peaceful, and this probably freed up military energy for release elsewhere. For Byzantium the traditional concerns in the sphere of foreign affairs were the Arabs and Bulgaria, but new problems arose. The most dramatic came from the north.45 In 860 a Rus fleet suddenly appeared before Constantinople, having sailed across the Black Sea, though probably not from Kiev, which was yet to develop as a political centre. The raiders subjected the suburbs around the imperial city to plunder. Michael III was away on campaign, but he hurried back when informed of the assault. However, the fleet soon departed, perhaps simply through having amassed enough booty rather than being driven away by an act of God. While it seems that the Rus were already known to the Byzantines, the events of 860 made a deep impact. Byzantium responded to the Rus’ subsequent request for a mission, although this mission does not seem to have lasted long (see below, p. 320). The relationship remained mixed, with further Rus raids in the tenth century but also trading treaties and Rus serving with the imperial forces. Diplomatic and cultural contacts intensified, leading ultimately to the conversion of Prince Vladimir of Kiev and his people c. 988 (see below, pp. 325–6). 43 For traditional acceptance of a Byzantine invasion see e.g. Whittow (1996a), p. 282. For the view that Boris freely turned to the Byzantines: Shepard (1995a), pp. 239–40; Zuckerman (2000a), pp. 118–200. 44 On the conflict between Rome and Constantinople under Photios: Dvornik (1948); Dagron (1993), pp. 169–83; Simeonova (1998a). See also below, pp. 318–19. 45 On the Rus: Franklin and Shepard (1996); Whittow (1996a), pp. 241–62. For the raid of 860 see also Vasiliev (1946).

300

the midd le empire

Soon after the Rus raid, another avenue for Byzantine cultural influence opened up, when Prince Rastislav of Moravia (846–70) requested churchmen.46 It is likely that Rastislav, sandwiched between Franks and Bulgars, turned to Byzantium in the hope of securing a political counterweight. The Byzantines embraced the opportunity, despatching in 863 the famous brothers Constantine and Methodios. They hailed from Thessaloniki, and Constantine had especially strong bonds with the court and intellectual circles of Constantinople.47 To pursue their mission in Moravia they sought to spread the word in the language of the Slavs, and to this end developed a Slavic alphabet, the first of its kind, and translated many religious texts into the literary language they coined. Their mission dissolved after the death of Rastislav and disengagement of other Slav princely patrons, who came under Frankish pressure. But their disciples had an impact in the newly Christianised Bulgaria of Boris, where they ended up as refugees after being expelled from Moravia in 885. Installed at Ohrid and Pliska, they were entrusted with the creation of a Slavic clergy and expounding Christianity in comprehensible Slavic, lessening the need for Byzantineborn clergy; but Greek remained the language of court ceremonial and, probably, the liturgy.48 Thus the outcome of the mission to Moravia had unintended consequences, not necessarily advantageous to Byzantium in so far as they nurtured the aspirations of Symeon, Boris’ son. A final development lay to the east.49 Armenia had fallen under Arab overlordship from the end of the seventh century, and the leading Armenian families (the Bagratuni and the Artsruni) had assisted in the Arab sack of Amorion in 838. But with the decentralisation of the Abbasid caliphate there came the opportunity for greater independence, and this was exploited by Ashot I Bagratuni (‘the Great’), prince of princes, who in 884 was crowned king of Armenia. Under Michael III and Basil I political relations with Armenia were fostered, Basil recognising Ashot as prince of princes (arch¯on t¯on archont¯on).50 These friendly relations persisted into the tenth century, and assisted in Byzantium’s expansion eastwards. court politics 8 67–886 Compared with the shenanigans of the court politics of Michael III’s reign, court politics in the reign of Basil I seem relatively tranquil. Turbulence is concentrated at the extremes of his period of rule. Following the assassination of Bardas and the elevation of Basil, one of the main conspirators, 46 For the mission to Moravia and its aftermath, see below, pp. 316–18; Dvornik (1970); Tachiaos (2001). 47 Constantine had already been on a mission to the Khazars: see below, p. 315. 48 See Goldberg (2006), pp. 270–1, 280–1; Hannick (1993), pp. 930–4. 49 For Armenia, see below, ch. 8; Whittow (1996a), pp. 195–220; Grousset (1947). 50 On ecclesiastical relations between Armenia and Byzantium under Photios, see below, pp. 351–2.

6. after iconocl asm (850–886)

301

Bardas’ son-in-law Symbatios, felt sidelined and came out in revolt. This was apparently aimed at Basil, not Michael, but was anyway suppressed, and Symbatios and his accomplice Peganes found themselves mutilated and forced to beg in the streets of Constantinople. With the assassination of Michael, Basil clearly had to justify himself, and judging by the amount of propaganda produced under the new regime much energy was devoted to this task. Following the murder, one of Basil’s first main steps was to depose Patriarch Photios (858–67, 877–86) and reinstall Ignatios. These acts of Basil have been viewed in the light of ecclesiastical politics, namely the tensions between Rome and Constantinople that had set in during the patriarchate of Photios and the supposed internal division between ‘extremists’ and ‘moderates’. The existence of these opposing ‘politico-religious’ groups was formulated by Francis Dvornik, who saw them as competing for control over church and state in Byzantium.51 The extremists were identified as traditional and conservative Christians, mainly monks and their supporters. The moderates were considered those more in touch with the realities of earthly life, and more willing to compromise in the matter of Christian ideals. Amongst their number Dvornik included the secular clergy and intellectuals. Basil’s ousting of Photios and favouring of Ignatios was read by Dvornik as signifying that the emperor threw in his lot with the extremists in a bid to secure support for his regime. Even if one accepts the existence of these two factions in Byzantine society – and surely such a formulaic reading of history is open to question – it is perfectly possible that Basil was simply motivated by the desire to eliminate a political rival: Photios had close connections with the Amorian dynasty and could have headed opposition to the usurper.52 However, Photios later returned to favour, taught Basil’s children, and became patriarch again upon Ignatios’ death in 877.53 Ironically, it seems that the rehabilitated Photios had a part to play in contributing to, and even shaping, the ideology of the new regime.54 Basil was cast as the legitimate God-favoured restorer of the Roman empire. The appeals to Armenian ancestry and Davidic imagery appear to have had some input from Photios, even while he was still in exile.55 The dynasty’s ideological concerns, including devotion to the prophet Elijah and the archangel Gabriel, are reflected in the illustrated copy of the homilies of Gregory Nazianzen presented to Basil c. 880, the commissioning of which has been attributed to Photios (see fig. 16).56 Photios’ fingerprints have also 51 53

52 Tougher (1997b), pp. 76–8. Dvornik (1948), pp. 9–18. On Photios’ fall, exile and rehabilitation see e.g. Dvornik (1948), pp. 136–7, 161–73; Dagron (1993), pp. 176, 180–1; Simeonova (1998a), pp. 247–8, 280–91. 54 Tougher (1997b), pp. 32, 70–1. 55 See Ciccolella (1998), esp. pp. 325–8; Markopoulos (1992), pp. 226–9; Magdalino (1987), p. 58. 56 See Brubaker (1999a), esp. 147–200; Dagron (2003), pp. 193–9.

302

the midd le empire

been detected on Basil’s legislative work, which was to culminate in the issuing under Leo VI of the Basilika, a revised Greek version of the Justinianic corpus.57 The model of Justinian, the ‘great restorer’ of the sixth century, seems to be evident in other spheres of Basil’s activity, too. Basil’s interest in the empire’s western approaches is notable, and the example of Justinian’s reconquest may have influenced him. Basil also has the reputation of being a great builder, particularly in the Life of Basil.58 While it is possible that this text overplays the emperor’s architectural achievements, as did Procopius those of Justinian in The Buildings,59 it is clear at least that Basil was responsible for a new complex in Constantinople, encompassing a polo ground, gardens, a courtyard, and the Nea Ekklesia (‘new church’), which celebrated the dynasty.60 The difficult court politics towards the end of the reign revolve around Basil’s heir Leo, but can perhaps be opened up to reveal larger issues. Basil’s intended successor was his eldest son Constantine, but he died from a fever in 879. As next oldest, and already a co-emperor, Leo became heir apparent. The future of the dynasty looked assured, as Leo was duly married to Theophano Martinakia, a relative of both the Amorians and Eudocia Ingerina.61 However, relations between Basil and Leo were strained.62 A common explanation for this is that Basil disliked Leo because he was Michael III’s son, but perhaps the answer lies more in a clash of personalities and wishes. Leo was not content with Theophano. The Life of Euthymios has Leo vividly recall how Basil threw him to the floor and beat him when Theophano told him that Leo was having an affair with Zoe Zaoutzaina.63 Relations deteriorated to the point that Leo was suspected of plotting against Basil, and was shut up in the palace apartment of the Pearl, a confinement that is thought to have lasted for three years, from 883 to 886.64 It is evident that Leo had formed his own group of supporters, such as Andrew the domestic of the Schools and Stephen magistros, but it is not certain that they had hatched a plot. Indeed the narrative sources depict Leo as victim of the machinations of Photios’ circle. It looks as if towards the end of Basil’s reign the Amorians were preparing to stage a comeback. It has been suggested that this move was also inspired by discontent with Basil’s western initiatives, which seem to have incurred some internal opposition.65 That a major plot against Basil headed by John Kourkouas was exposed in March 886 adds to the air of political crisis. Maybe from realisation that Leo had been innocent, or perhaps simply because difficult circumstances required a show of dynastic unity, Basil restored Leo on 20 July 886, the feast day of Elijah, one of the patrons of the Macedonian 57 59 61 63 65

58 Life of Basil, pp. 321–41. See e.g. F¨ogen (1998), esp. pp. 11–12. 60 See Magdalino (1987). See e.g. Ousterhout (1998), esp. p. 129. See also above, p. 111. 62 Tougher (1997b), pp. 42–67. Tougher (1997b), pp. 134–40. 64 Jenkins (1965), pp. 101–2. Life of Euthymios, ed. and tr. Karlin-Hayter, p. 41. Vlysidou (1991).

6. after iconocl asm (850–886)

303

Figure 16 Basil I being crowned by the Angel Gabriel, who flanks him together with the Prophet Elijah

dynasty.66 This event was followed by Basil’s death just over a month later, and Leo was thrust to sole power. The question of his complicity in the demise of the dynasty’s founder remains open.

66

See for instance Magdalino (1987); Magdalino (1988a); Dagron (2003), pp. 192–8.

304

the midd le empire conclusion

It is clear then that the period c. 850–c. 886 was one of great moment in the recovery and advance of the Byzantine empire. Whether this was due to individuals such as Theoktistos, Michael, Bardas and Basil, or to favourable conditions such as an already progressing revival and altered international circumstances, is a matter for debate. Probably there is an element of both. However, the failures of the period and the persistent problems, such as the Arabs’ occupation of Sicily and naval threat, should not be overlooked, nor should the uncertainties of the evidence. But whatever overall verdict is reached on this phase of the empire’s life, it surely qualifies as one of the most intriguing periods of Byzantine history.

CHAPTER 7

RELIGIOUS MISSIONS s e rgey a. ivan ov

introduction Although Christianity would seem by its very nature to be a missionary religion, both the sense of what ‘mission’ means and the specific motivations of missionaries have varied as each generation reads afresh the Gospels’ injunctions. Early Christians were keen to stress the ‘international’ character of their religion and the primordial equality of all peoples, yet a different conceptual system was embedded in the very language in which the early Christian apologists wrote. St Paul already uses the term barbarian, with its implicit contrast between ‘us’ and ‘them’. Early Christians also appropriated the discourse of the Roman world, which was similarly permeated with the spirit of empire. If the empire was ‘the world’, then those beyond the imperial borders were automatically assigned to an ‘other’ world, not inhabited by real people. Primitive Christianity opposed this kind of logic. St Christopher, for example, was – according to his Life – ‘from the race of dog-heads, from the land of cannibals’;1 but this did not prevent him becoming a Christian martyr. Does this imply that natural savagery could be eradicated? An answer can be found in another legend – the ‘Tale of St Christomeus’ – one of the apocryphal stories of the wanderings of the apostles Andrew and Bartholomew. The legend tells how a certain cannibal was visited by an angel, who breathed grace into him and ordered him to assist the apostles. When the inhabitants of ‘the city of the Parthians’ incited wild beasts against the preachers in the circus, Christomeus asked God to give him back his former nature: ‘and God heeded his prayer and returned his heart and mind to their former savagery’. This monster then tore the beasts to pieces, whereupon many of the pagans died of fright. Only after this did Andrew come up to Christomeus and say: ‘“the Holy Spirit commands that your natural savagery should leave you” . . . and in that moment his good nature returned’.2 The legend is clearly designed to 1 2

Acta sanctae Marinae et sancti Christophori, ed. Usener, p. 57. ‘Tale of St Christomeus’, Cod. Hier. Sab. 373, fols. 117–29, Jerusalem; Cod. Brescian A III 3, fols. 142–5, Brescia. I am grateful to A. I. Vinogradov for allowing me access to this text (BHG 2056), which he is preparing for publication.

305

306

the midd le empire Novgorod

L. Ilmen

in

n

Polotsk

a

Rostov We ste r

D

LITHUANIANDOMINATED RUS (C14)

v

RUS Kiev

MORAVIA

Volg a

Chernigov Pereiaslavl Dn iep er

T isza

HUNGARY

Szombor

KHAZARIA

P

ASU SM OU NT AI N

I A N

A

Constantinople

BALKANS

Terek

I A CAU C ZIC

Bitola

AZ

S E A

ALANIA

Nizhnii Arkhyz LA

B L A C K

L. Ohrid

KH

S

AB

MOESIA

Itil (Astel)

A

CRIMEA

ube Dan

Sarai

C

ban Ku

S

S ALBANIA

Nicaea

E A

ASIA MINOR Patras

(ANATOLIA) CAPPADOCIA

Athens

Ephesos

Tokale Kilise (Göreme)

is gr Ti

E

up hr ate s

L I B YA EGYPT

le

Ni

Aphrodito

Kerch

Philae

(Bosporus)

CRIMEA

Phoullai(?)

SUDAN MAKURRIA Dongola

Doros(?) Cherson

Faras

ALODIA

0 0

250

500 250

750 500 miles

1000 km

ETHIOPIA

(ABYSSINIA)

Map 15 Byzantine religious missions

Caffa Sougdaia Bishoprics

NUBIA

AKALE GUZAY Axum

GOTHIA

H I M YA R

Sanaa ( Y E M E N )

7. religious missions

307

glorify Christomeus and its superficial message is that even a cannibal can become a Christian. Yet the deeper message – which perhaps reveals itself despite the author’s best intentions – is precisely the opposite: that there is always a beast sleeping within any barbarian. By taking the first step, by assimilating the discourse of barbarism, early Christians were also well on the way to assimilating a Roman conceptualisation of barbarians.3 In Christian apologetics one increasingly finds the idea that Christianity was useful to the empire because it could help in moderating barbarian savagery; not, one might think, a matter of prime concern for persecuted Christians. This notion is already fully formed in the writings of Origen. It could have prompted missionary undertakings, but in fact did not. From a Christian viewpoint, conversion was something so fundamental that it could not depend on the paltry efforts or specific initiative of mere humans. Oddly, not even the apostles in their apocryphal wanderings were portrayed as missionaries in the proper sense of the word. Among the agents of the initial Christianisation of the barbarians we find merchants, mercenaries, hostages and political exiles: that is, missionaries without a mission as such. If priests travelled to barbarian lands, it was only in order to minister to Romans in foreign captivity.4 The Syrian monks probably constituted the only group of deliberate propagandists for the faith. state-sponsored missions in the age of just inian During the sixth century Christian space was very significantly expanded, thanks above all to centralised missionary policies.5 Emperors began to receive state visits from barbarian rulers, showering them with gifts and baptism. In 522 Justin I (518–27) baptised Tzathus, king of Lazica, gave him a Byzantine bride and declared him his own son. In 527, Justinian (527–65) baptised Grod, prince of the Bosporan Huns, and Grep, ruler of a Germanic people, the Heruli, Justinian was also active beyond the empire’s borders, and his missionary initiatives extended in several directions. Thus, in Abkhazia many new churches were constructed at a fair distance from the sea. These churches were clearly intended for the barbarians; they contain baptisteries suitable for adult baptism. The expensive building materials and the high quality of the construction-work suggest that the empire was footing the bill. 3 4

For more details see Ivanov (2002). Seminal articles on Byzantine missionary activities in general include: Beck (1967); Hannick ˇ cenko, I. (1988–9); Shepard (2002a). The only monograph is Ivanov (2003). (1978); Sevˇ 5 For a detailed account of the main missionary undertakings of the sixth century see Engelhardt (1974).

308

the midd le empire

Justinian’s aims were purely political, as is clear from the account of the baptism of the Abkhazians by Procopius of Caesarea. The empire began to intervene in the internal affairs of this barbarian tribe so as to counter the influence of Sasanian Persia. This political pressure had a mildly civilising tinge. From an attempt to persuade the barbarians to renounce their ‘savage’ rituals it was but a short step to full-scale Christianisation. This in turn led to the overthrow of the authorities associated with the pagan religion and from there it was another small step to attempted colonisation. Such a policy could also have unforeseen consequences: the repudiation of Christianity because of its association with imperial expansionism.6 The Caucasian Tzani also became the targets of a state mission.7 The principal agent of the dual policy here – combining threats and Christian proselytism, church building and deforestation – was the Byzantine commander Sittas. Where Byzantium had no direct political interest, it likewise had no active interest in missions. The sincerity of the conversion was of no concern. According to Procopius, Justinian: persuaded all [the Heruli] to become Christians. Thus, having exchanged their way of life for one more mild, they resolved in all things to adopt Christian customs and on the basis of a treaty of alliance to cooperate with the Romaioi (Romans). Yet among themselves they are fickle, and adept at doing harm to their neighbours. And they engage in indecent intercourse even with donkeys. They are the most disgusting of all peoples.8

For the Byzantines, barbarian ‘mildness’ and ‘Christianity’ meant only one thing: forbearance from attacking the empire. We should not imagine, however, that every mission in this period was accomplished by armed force or with narrowly political aims. Missionary activity in Abyssinia was different. Unfortunately, Greek authors say not a word about it, and we shall encounter such silences again, many times. Yet the local Ethiopian sources are far from reticent. They tell us that a group of monks from Byzantium settled in the region of modern Akale Guzay. These ‘righteous men from Baraknakh’ were murdered by locals during a pagan uprising, and thereby became the first Abyssinian martyrs. Another group of seven or nine ‘Roman saints’ arrived in Axum and yet another missionary was Michael Aragawi, whose Ethiopian Life reveals a few details of his preaching.9 Although the chronological indicators in the ‘Roman 6 PR W, VIII.3.18–19, 21; VIII.9.10–12; I.12.3; VIII.2.17; II.29.15, ed. and tr. Dewing, V, pp. 80–1, 132–5; I, pp. 96–7; V, pp. 68–9; I, pp. 532–3. See also Evagrius Scholasticus, Ecclesiastical history, IV.22, ed. Bidez and Parmentier, p. 170; tr. Whitby, p. 221. 7 PR W, I.15.24–5, ed. and tr. Dewing, I, pp. 136–7; PR B, III.6.6–8; III.6.11–12, ed. and tr. Dewing and Downey, pp. 206–7, 208–9. 8 PR W, VI.14.33–6, ed. and tr. Dewing, III, pp. 410–13. 9 Life of Michael Aragawi, ed. van den Oudenrijn, p. 19.

7. religious missions

309

saints’ file are contradictory, scholars usually date them to the late fifth and early sixth centuries. Around the end of the 530s the southern Arabian state of Yemen broke free of Ethiopian patronage and established close links with Byzantium. A sign of the strengthening of Yemen’s religious contacts with the Byzantines was the construction of a large church in Sanaa, built in red, yellow and black marble and adorned with mosaics in the Constantinopolitan manner.10 It is not clear exactly when or by whom the Byzantines were asked to send a teacher of Christianity for the re-Christianised country; it may have been the occupying Ethiopian authorities or, more logically, the local inhabitants themselves. We have only one text, and a dubious one at that: the Life of Gregentius, bishop of the Homerites (Himyarites). In the story of Byzantine missions, Gregentius is as significant as he is mysterious. No reliable data about him has survived. His extant Life is late and most likely fictitious.11 Appended to it is a text known as the Laws of the Himyarites which, even if it is not an authentic piece of legislation, remains an example of Byzantine missionary thinking, albeit abstract and from a later period. The striking feature of the Laws is that their rules for the newly converted Arabs are much stricter than the rules in force in the Christian empire itself (see also above, pp. 186–7). The Laws turn practically every civil offence into a criminal one, and virtually all private law becomes public law. The Romans themselves would never have dreamed of abiding by such ferocious requirements.12 Overall, the Laws of the Himyarites represent a totalitarian missionary utopia. As for the Arab tribes immediately bordering Byzantium to the east, conversions of pagan bedouin to orthodox Christianity were rare. Here, as so often, the empire was more preoccupied with averting heresy than with making Christian converts.13 Travelling up the Nile, Justinian’s emissaries reached a multi-confessional sanctuary on the island of Philae, at the outer limits of the imperial possessions. The temple was converted into a church of St Stephen. The first extant inscription left by a native is dated as early as 537: ‘I, Theodosius the Nubian’.14 The history of the mission to Sudan is much better known than any other, because it involved the rivalry between Justinian and his wife Theodora, patrons of Chalcedonism and monophysitism respectively. The main source – virtually our only source – is John of Ephesus, himself a monophysite. In John’s account, the idea of a mission to Sudan was conceived in the circle of the monophysite patriarch of Alexandria, who lived in exile in Constantinople under the patronage of Theodora. Theodora 10 11 13

al-Tabari, Ta’rikh, ed. de Goeje et al., I, pp. 934–6; tr. Bosworth, V, pp. 217–21. 12 Papathanassiou (1996). Life of Gregentios, p. 107; Berger (2001), pp. 57–61. Evagrius Scholasticus, Ecclesiastical history, VI.22, ed. Bidez and Parmentier, p. 238; tr. Whitby, p. 314. 14 Fontes historiae Nubiorum, ed. Eide et al., III, no. 325, p. 1181.

310

the midd le empire

turned to her husband for support, but he had his own plans to dispatch a Chalcedonian embassy to Sudan from Egypt. John’s subsequent narrative unfolds like a thriller. The imperial couple sent two missions, racing each other, but Theodora’s cunning ensured that her own mission, headed by Julian, arrived first: [Julian] handed over the empress’ letters . . . And [the Nubians] also received magnificent gifts, many baptismal garments and all in abundance. And they immediately . . . believed in the Christian God . . . Then he taught them . . . and also intimated to them the following: ‘Be forewarned that among Christians there are disputes concerning the faith . . . for this reason the empress has sent us to you.’15

Julian then explained to the Nubians how they should respond to the emperor’s mission. On arrival, Justinian’s envoy immediately handed over the emperor’s letter and gifts to the Nubians, and then his missionaries ‘began to teach them as they had been ordered, and they said, “Our Roman emperor has instructed us to propose that, if you become Christians, you should join the church and those who adhere to it, and not those who have been cast out.”’16 However, according to John of Ephesus, the barbarians firmly rebuffed him. The intrigue here revolved less around conversion than around the rivalry between monophysitism and Chalcedonism. Yet after the expulsion of Justinian’s embassy, Julian stayed in Sudan for two more years, showing great zeal and instructing the barbarians in Christianity daily: from nine in the morning until four in the afternoon he would conduct his lessons naked, sitting up to his neck in water in a cave, because of the unbearable heat: ‘Yet he endured this, and instructed and baptised the king, his magnates, and many people with them.’17 The initial baptism of Sudan took place between 537 and 539, whereupon Julian returned to Byzantium. In 565 Theodosius, Patriarch of Alexandria, had his prot´eg´e Longinus ordained as the new bishop of Nubia. Longinus was immediately arrested by Justinian and imprisoned for three years; but eventually he managed to escape to Sudan, where he spent some six years. According to John of Ephesus, Longinus ‘taught, enlightened and instructed them anew, and he built a church there, and appointed clergy, and taught them the entire order of the services and all the rules of Christianity.’18 It would appear that Longinus’ major achievement was the training of local clergy. This enabled the new religion to put down roots in Sudan, where it survived for many centuries.19 15 17 18 19

16 JE, IV.7, ed. and Latin tr. Brooks, II, pp. 138–9. JE, IV.7, ed. and Latin tr. Brooks, II, p. 138. JE, IV.7, 49, ed. and Latin tr. Brooks, II, pp. 139, 175–6. JE, IV.8, ed. and Latin tr. Brooks, II, p. 140. Excavations at Faras have revealed the tomb of Bishop John who died in 606, aged eighty-two: presumably he was one of Longinus’ trainees.

7. religious missions

311

The fashion for the new religion spread further still, and Longinus was invited to a tribe further south, the Alodians. It is curious, however, that John of Ephesus says nothing about any mission to the Makurrah, though their land lay between Sudan and Alodia. Only the Latin chronicle of John of Biclaro mentions the conversion of the Makurrah, which he dates to the year 569;20 we can surmise that they were converted by the Chalcedonian patriarch of Alexandria, with the aim of annoying his monophysite rivals. In the ruins of Dongola, capital of Makurria, the remains of several ‘Byzantine’ churches have been identified. We do not know how long the Makurrah remained Chalcedonian. At some time in the late sixth or early seventh century they joined up with Sudan and accepted monophysitism. No Greek source contains even a single word about this rich and dramatic story of the Byzantine mission to the middle Nile: again we come up against silence. Justinian’s successors could be as ambitious as he was. According to John of Biclaro the Garamantes, Berbers living in the Libyan desert, were baptised under Justin II (565–78),21 while Maurice (582–602) is associated with an attempt to Christianise Byzantium’s great eastern rival, Persia. Christianity had been known in Persia from a very early period. After Christianity became the Roman empire’s state religion, any deterioration in the relations between the two superpowers would lead to persecution of Persian Christians. As divisions within Christianity deepened, the Persian authorities began to encourage Nestorianism, and this gradually expanded to become the second religion of Iran (see above, p. 144). The Persian ruler Hormizd IV (579–90) was notably tolerant of all Christians in Persia, including Chalcedonians, and this gave rise to a legend about the Persians’ own conversion.22 This legend, preserved only in Latin tradition, probably reflects hopes generated in the empire by developments in Persia. In 590 the shah was deposed and his son, Khusro II (590, 591–628), fled to Byzantium. The prince regained the throne with the aid of troops provided by Maurice. According to the Shahnama (Book of kings), in this new spirit of friendship the emperor sent Khusro ‘a cross ornamented with jewels’ and garments embroidered with crosses.23 During this period, the Chalcedonians were in favour. Here too, we learn of Byzantine activities from all kinds of sources, but with one conspicuous exception – the Byzantines themselves. Why does Theophylact Simocatta, who recounts 20 John of Biclaro, Chronicon, ed. Mommsen, p. 212; ed. Cardelle de Hartmann and Collins, p. 61; tr. Baxter Wolf, p. 60. 21 John of Biclaro, Chronicon, ed. Mommsen, p. 212; ed. Cardelle de Hartmann and Collins, p. 61; tr. Baxter Wolf, p. 60. Christian features in the language of the Garamantes’ descendants, the Tuareg, suggest that their forebears had contact with Latin-speaking missionaries. 22 Fredegar, Chronicle, IV.9, ed. Krusch, pp. 125–6; tr. Wallace-Hadrill, pp. 7–9. 23 Firdausi, Shahnama, French tr. Mohl, VII, pp. 145–6; tr. Warner and Warner, VIII, pp. 307–8.

312

the midd le empire

Maurice’s dealings with the Persians in great detail, not say a word about his Christianising activities? However, these achievements were short-lived: in 602, as soon as Maurice was murdered, Khusro launched an attack on Byzantium, and ‘from the Euphrates to the east, the memory of the Council of Chalcedon was obliterated utterly’.24 The sixth century was an age of grandiose missionary undertakings, but there were also smaller-scale ones. The ‘Legend of seven bishops of Cherson’, for example, reflects local hagiographical tradition. One version of this legend, which probably originated in the sixth century, includes a certain Ephraim among the Christian missionaries in Cherson. According to this variant, Ephraim had been sent to convert ‘the land of the Tauroscythians which borders on Cherson’. It is noteworthy that in later centuries this mythical Ephraim was recast as having converted a number of barbaric tribes: Turks, Huns and Hungarians.25 The sixth century was also the age of the parting of ways between Chalcedonian and ‘heretical’ Christianity (see above, pp. 116–18, 212–13). The subsequent large-scale missions of the Nestorian and monophysite churches, involving conversions in Central Asia and China, had nothing to do with Byzantium. Henceforth only a ‘heretic’ could allow himself an elevated, ‘pan-Christian’ attitude towards missions. One such champion of unalloyed apostolic evangelism was the sixth-century Alexandrian traveller Cosmas Indicopleustes. In his Christian topography, Cosmas presents a kind of bird’s eye view of world-wide evangelisation.26 This sense of universality was all but lost by the imperial church. the lull in mission work A policy of ‘state Christianisation’ persisted into the seventh century. Heraclius converted a ‘Hunnic’ leader in 619 and Caucasian Albania in 628.27 Not until the ninth century do we hear of any further centralised initiatives on the part of the imperial authorities to convert distant barbarian tribes.28 Yet this very decline created substantial scope for local and personal initiatives of a kind which had perhaps existed before, but which the large-scale state ventures had overshadowed.

24 26

25 Ivanov (2003), pp. 82–3, 192. Pigulevskaia (1946), p. 262. Cosmas Indicopleustes, Christian topography, III.65–6, ed. and French tr. Wolska-Conus, I, pp. 502–7; tr. McCrindle, pp. 117–21. 27 There is also an unreliable later tradition that Heraclius converted the Croats: DAI, ch. 31, pp. 148–9; see also Konstantin Bagrianorodnyi, Ob upravlenii imperiei, ed. and tr. Litavrin et al., p. 376 (commentary). 28 Note, however, the baptism of the Bulgar khan Telerig in 777: Theoph., ed. de Boor, I, p. 451; tr. Mango and Scott, p. 622.

7. religious missions

313

Apart from cursory allusions to certain exiles who carried out pastoral work in the Crimea and Khazaria,29 we have the Life of Stephen, bishop of the Crimean city of Sougdaia. The Greek version of this Life is very brief, although slightly fuller versions exist in Slavonic and Armenian. Stephen was born in Cappadocia and received his bishopric in Sougdaia. The establishment of an episcopal see at a place like Sougdaia, on the edge of the barbarian world, is noteworthy in itself, since the town had only been founded in the later seventh century. Sougdaia was home to pagans (Khazars and Circassians) as well as Christians, and Stephen was an active preacher: ‘When the pagans heard that he worked wondrous miracles, they believed in the Lord, and a countless multitude was baptised. And he appointed many presbyters and deacons for them.’30 The missionary had established good relations with the local Khazar commander (the tarkhan), who ‘while exercising his power would regularly come to St Stephen and would listen and do as [Stephen] told him. And the saint instructed him much on the path to salvation.’31 Stephen died in office at the end of the eighth century. In the second quarter of the seventh century, a new and fearsome enemy appeared on Byzantium’s eastern borders: the Islamicised Arabs. Preaching Christianity to them directly was difficult in the extreme, except in the case of prisoners-of-war. Yet already by the eighth century we find instances of Arabs voluntarily converting to orthodoxy, as in the Life of Stephen Sabaites.32 What we have here, for the most part, is the apostasy and subsequent reconversion of Christians who had accepted Islam while in Arab captivity. One of the preachers to such people was Romanos the Neomartyr, executed in 778.33 According to the legendary Life of Theodore of Edessa, who most likely lived between 776 and 856, this bishop baptised none other than Mu‘awiya, ‘caliph of Baghdad’, at the caliph’s own request! Another missionary among the Arabs was Elias the Younger, a Sicilian who was shipped off to North Africa in the mid-ninth century.34 All these missionaries preached at their own risk. They would never have counted on support from Byzantium. The system of bishoprics in the Balkans virtually collapsed with the incursions of the Slavs. The Byzantine reconquest began in Greece in the eighth century and was followed by the Christianisation of the Slav groupings that had settled there. The Chronicle of Monemvasia relates how Emperor Nikephoros I ‘concerned himself with rebuilding churches and with 29 30 31 32 33 34

Menologion of Basil II, col. 181. Life of Stephen of Surozh, ch. 15, ed. Vasilievsky, p. 86 (longer Life). Life of Stephen of Surozh, ch. 30, ed. Vasilievsky, p. 95 (longer Life). Life of Stephen Sabaites, pp. 544–6. Life of Romanos the Neomartyr, ed. and Latin tr. Peeters, pp. 422–3. Life of Elias the Younger, ed. and Italian tr. Rossi Taibbi, pp. 24–6, 32–4.

314

the midd le empire

turning the barbarians themselves into Christians’.35 The emperor’s methods are interesting: ‘He installed [the fugitive inhabitants of Patras] where they had been before, together with their own pastor . . . and he offered Patras the status of a metropolitanate . . . Therefore the barbarians, too, were instructed in the faith and baptised, and they joined the Christian faith.’36 The chief prerequisite for the baptism of the barbarians on reconquered imperial lands was thus the organisation of a network of bishoprics. t he mid -ninth-century upswing A notable upswing in the empire’s missionary activities may be observed around the turn of the ninth century. A growing interest in converting the barbarians is traceable in the Life of Andrew the Apostle, written by the monk Epiphanios. The author made his own journey in the footsteps of the apostle, and his narrative combines hagiographic commonplaces with first-hand observations: ‘And from there he went to Bosporus . . . where we ourselves found Bishop Kolymbadios, who knew ten languages.’37 The emphasis which Epiphanios places on the bishop’s polyglot skills could reflect his own interest in the apostolic succession from St Andrew. Thus the ‘first-called’ apostle is regarded not just as a miracle-worker (his main characteristic in the apocryphal ‘Wanderings’) but as a practising missionary. Epiphanios’ text is also important as the first, albeit timid, attempt to describe a missionary as he ‘really’ was: ‘Seeing that the apostles were unselfish, exhausted, pallid, without even sandals on their feet and dressed only in tunics, and that despite this divinely inspired words issued forth from them – seeing this, people did not wish to part from them.’ The role attributed to Patriarch Photios (858–67, 877–86) in ninthcentury Byzantine missionary activity is usually exaggerated;38 we lack firm evidence as to any deliberate plans he might have had to convert the barbarians. However, several of the missionary undertakings of the period were initiated by Emperor Michael III (842–67). If we believe Niketas Byzantinos,39 Michael was associated with some kind of coordinated religious work among the Muslim Arabs.40 The ninth century also saw the beginning of missionary progress on the empire’s northern periphery. This is indicated by the celebrated episcopal notice outlining the ‘see of Gothia’: ‘The metropolitanate of Doros: [bishops] of the Chotziroi, Astel, Chwales, the 35 Svod drevneishikh pis’mennykh izvestii o slavianakh, ed. Gindin et al., II, p. 330; Kislinger (2001), p. 202. See also above, p. 258. 36 Svod drevneishikh pis’mennykh izvestii o slavianakh, ed. Gindin et al., II, p. 330; Kislinger (2001), pp. 202–3. However, see the doubts voiced by Turlej (2001), pp. 109–11. 37 Vinogradov, ‘Grecheskie zhitiia apostola Andreiia’ (PhD thesis, 2001), pp. 152–3; Greek traditions, ed. Vinogradov, p. 145. See also Mango (2002a). 38 Hurbaniˇ 39 Niketas Byzantinos, Confutatio, col. 672. c (2005). 40 See the Life of Peter of Argos, ed. Kyriakopoulos, pp. 244–5.

7. religious missions

315

Onogurs, Reteg, the Huns, Tamatarcha’.41 At the end of the notice, unconnected with the previous text, we find additional information: ‘the eparchy of Gothia: [the bishop of] the Chotziroi near Phoullai and near Charasion, which is called Black Water. [The bishop of] Astel: Astel is the name of a river in Khazaria, and there is a fortress there.’42 The metropolitanate of Doros, as described in the notice, encompassed an enormous territory including part of the Crimea, the northern Caucasus and the north-eastern Caspian region – that is, all the territory of the Khazar khaganate. Even if this list of bishoprics is in fact merely a rough draft, it is extremely revealing; its compilation implies expansive missionary ambitions on the part of the church of Constantinople. Hopes of converting the khaganate apparently rose in Byzantium at the start of the ninth century, in the course of a multifaceted diplomatic offensive to the north.43 Around 860, as part of the same initiative, Michael III sent Constantine the Philosopher (the future St Cyril) on a mission to Khazaria. Constantine saw it as an evangelising opportunity, if we may believe his Life. He said to the emperor: ‘If you command, lord, on such a mission I shall gladly go on foot and unshod, lacking all the Lord forbade His disciples to bring.’ The emperor answered, saying: ‘Well spoken, were you to do this [on your own]! But bear in mind the imperial power and honour, and go honourably and with imperial help.’44

This discussion neatly encapsulates two views of missionary activity: Constantine’s remark alludes to Christ’s instructions to his disciples (Matthew 10:9–10), whereas the emperor counters to the effect that a missionary from Byzantium is at the same time an ambassador, and so the Gospel’s insistence on simplicity does not apply to him. Here mission manifestly merges with diplomacy. In the event, the results of Constantine-Cyril’s work among the Khazars were not particularly impressive: ‘about two hundred of these people were baptised, having cast off heathen abominations and lawless marriages’.45 Soon afterwards the Khazar khaganate adopted Judaism as its state religion. The Life of Constantine-Cyril also relates how the saint took time out from his Khazarian diplomacy to mount, at his own initiative, a missionary raid on the people of Phoullai in the Crimea, felling their sacred oak tree.46 41 42 43 44

NE, no. 3, pp. 241–2 (text). See also above, p. 286. NE, no. 3, p. 245 (text). ‘Astel’ is identifiable as the Khazar capital, Itil. Shepard (1998), pp. 19–20. Life of Constantine-Cyril, ch. 8, ed. Angelov and Kodov, p. 95; ed. Floria, pp. 148–9; tr. Kantor, p. 43. 45 Life of Constantine-Cyril, ch. 11, ed. Angelov and Kodov, p. 102; ed. Floria, pp. 163–4; tr. Kantor, p. 61. 46 Life of Constantine-Cyril, ch. 12, ed. Angelov and Kodov, p. 103; ed. Floria, pp. 165–6; tr. Kantor, pp. 63–5.

316

the midd le empire

Later Michael III ordered the brothers Constantine and Methodios to create a Slavonic alphabet for the translation of the Scriptures (see below, fig. 19). It was Michael who sent Constantine and Methodios to Moravia and who initiated the baptism of both the Bulgars and the Rus. Michael’s role in these missions has been somewhat overshadowed by Basil I’s subsequent successes, as Basil appropriated his predecessor’s initiatives for himself. However, Basil the Macedonian (867–86) seems to have been the first Byzantine emperor seriously to consider himself on a par with the apostles in missionary matters.47 It was during Basil’s reign that the feast of Pentecost acquired missionary connotations. In mosaics in the church of the Holy Apostles, which Basil extended and decorated, the apostles were clearly represented as missionaries. Similar representations can be found in miniatures48 and frescoes49 of the period. It is interesting, for example, that in the frescoes of the Tokale Kilise church in Cappadocia, the ‘peoples’, ‘tribes’ and ‘tongues’ who turn to the apostles are virtually obscured by huge depictions of emperors in Byzantine imperial ceremonial dress (fig. 17). In this sense the emperors are indeed ‘equal to the apostles’.50 The sources consistently stress the role of the emperor in the conversions of the Bulgars, Rus and the north-western Balkans.51 Byzantium’s religious embassy to Moravia in 863 and the activities of Constantine and Methodios laid the foundations of Slav written culture. Yet in no way does this justify the oft-made claim that the Moravian mission was the high-point of Byzantine missionary activity. Although the brothers are frequently labelled ‘apostles of the Slavs’, Moravia had in fact received Christianity without Byzantine involvement. True, Prince Rastislav’s letter to the emperor mentions that the Moravians had been visited by ‘many preachers . . . from the Greeks’,52 which might be taken to imply that there were Byzantine missionaries in Moravia before Cyril and Methodios. However, this phrase’s context undermines such an interpretation: the Byzantines are contrasted with all previous missionaries to Moravia, including ‘Greeks’; the implication is that these particular Greeks were not reckoned ‘Byzantines’. The Cyrillo-Methodian embassy itself should be viewed more as a unique event than as an integral part of an overall missionary strategy. The brothers from Thessaloniki did not undertake it as churchmen; when they did acquire ecclesiastical office, they observed the Latin rather than the Greek rite; and, yet again, the most striking feature of contemporary Greek sources is their total silence about the mission. Left to their own devices, lacking imperial assistance, the orthodox teachers also came into conflict with the German clergy – ‘the Franks’ – who 47 49 51 52

48 Brubaker (1999a), pp. 238–45. TC, p. 341. See also above, p. 301. 50 Jolivet-L´ Epstein (1986), p. 77, fig. 99. evy (2001), pp. 259, 261, plates 31, 33. See DAI, ch. 29, pp. 124–7. Life of Methodios, ch. 5, ed. Angelov and Kodov, p. 188; ed. Floria, p. 186; tr. Kantor, p. 111.

7. religious missions

317

Figure 17 Fresco from Tokale Kilise church in Cappadocia, showing emperors between apostles and ‘the nations’

were supported by the neighbouring east Frankish realm. The work of Methodios and his followers in Moravia can be pieced together from the Lives of Methodios and Clement of Ohrid, and also from the legal code known as the Court law for the people, which the Byzantines helped compile. These sources make clear that, although the brothers lacked political support for their activities in Moravia, from the start they made the same demands on the barbarians as they would have made on subjects of the empire. This concerned, above all, the laws of marriage: polygamy was forbidden, as was marriage to any relative, to godparents, and so forth.

318

the midd le empire

The Byzantine missionaries were admirably consistent: they made no distinction between the elite and the masses, between neophytes and Byzantines. Such an attitude was bound to make the Slav elite wary of orthodox churchmen. This was one of the reasons for the ultimate failure of the CyrilloMethodian mission. After Methodios’ death some of his followers were expelled from Moravia, others were sold into slavery. The empire, for its part, showed no interest in the fate of the enterprise. Constantinople made no attempt to absorb Moravia into its own sphere of ecclesiastical jurisdiction; it did not quarrel with Rome about the introduction of the Latin rite into Moravia; nor did it intervene to defend its own envoys from harassment by the Franks. In the late seventh century, the Bulgars had seized much of the Byzantine province of Moesia south of the lower Danube. The conquests of Khan Krum (c. 803–14) greatly extended their dominions to the south, bringing a sizeable number of Greek-speaking Christians under Bulgar sway (see above, p. 257). The influence – including religious influence – of these Greek-speaking Christians on the incoming Bulgars can be traced in sources from the early ninth century. As usual, missionary activity was initiated by captives as well as by local Christians. The Life of Blasios of Amorion53 and the tale of Prince Enrabotas illustrate this.54 The conversion of the Bulgars took place in the mid-ninth century in several stages and in complex competition with the church of Rome. Theophanes Continuatus claims that Boris of Bulgaria (c. 852–89) was coaxed towards Christianity both by his sister, who had spent some time in the empire as a hostage, and also by a Byzantine captive, a monk named Theodore Koupharas; however, he also alleges that Boris’ final decision to convert was made after a severe drought in 864 or 865.55 All the sources on the conversion of Bulgaria56 tend to stress the wonder of divine intervention, the role of famine, the emperor’s diplomatic skills or the persuasiveness of Boris’ entourage, but nowhere do we find a word about Byzantine missionaries. Indeed, the Greek sources make plain that the Bulgars would never have accepted Christianity were it not for exceptional circumstances. Photios himself calls the conversion of Bulgaria ‘improbable’,57 which supports the impression that it was not a pre-planned action. The first attempt to establish Greek Christianity in Bulgaria was a failure. The Greeks were obviously unprepared for the methodical persistence of missionary work. The extent to which the two sides spoke, as it were, different languages can be seen from the long letter sent by Patriarch 53 55 57

54 Theophylact of Ohrid, Martyrium , cols. 193–7. Life of Blasios of Amorion, pp. 660–1. 56 See Speck (2000), pp. 342–61. See also above, p. 299. TC, pp. 162–3. Phot., no. 2, ed. Laourdas and Westerink, I, p. 51.

7. religious missions

319

Photios to the newly baptised Prince Boris.58 Photios’ arrogant tone, wholly unsuitable for a missionary epistle, reflects the general attitude of the Greek clergy in Bulgaria. Boris found the behaviour of the empire’s minions so irritating that as early as 866 he rejected their ministrations and turned instead to the Roman church. Vacillating between Constantinople and Rome, playing off one Christian centre against the other, Boris sent an extensive set of questions to Pope Nicholas I (858–67) in Rome. Boris’ letter has not survived, but we do have the pope’s answers. This document is in striking contrast to Photios’ epistle. The papal letter is respectful and specific. Through it, by inference, we can see which of the Byzantine demands the newly baptised barbarians had found most irksome. The Greeks fussed about Bulgarian marriage ritual;59 they forbade visits to the baths on Wednesdays and Fridays;60 they required worshippers to stand in church with their arms crossed over their chests; those without their belts fastened were banned from receiving communion,61 and so on. In some cases, Nicholas indicates that he understands the principles laid down by the Greeks, but that he disagrees with their rigorist approach which could scare neophytes away from Christianity altogether. He proposed distinguishing the essential from the secondary. Such flexibility was alien to the Byzantines of the ninth century. After an elaborate contest in ecclesiastical politics, Bulgaria returned once more to the fold of the Constantinopolitan church. Theophanes Continuatus writes that: through the emperor’s continual admonition, through formal receptions and still more through magnanimous generosity and gifts, [Basil I] made them accept an archbishop and agree to their land being filled with bishops. And through them, and also through the pious monks whom the emperor summoned from the mountains and from the caves in the earth and sent thither, this people . . . allowed itself to be caught in Christ’s net.62

Such meticulousness in carrying out a programme of conversion is due, above all, to the Byzantines’ strong sense that the Bulgarian land was originally theirs and must inevitably be returned to them in time. The Bulgarians were well aware how their country was viewed by its mighty neighbour, and they understandably regarded Byzantine Christianity as a potential threat. That is why in the 880s Boris was happy to receive 58 59

Photios, Letter to Khan Boris, ed. Laourdas and Westerink; tr. White and Berrigan. Nicholas I, Responsa, ch. 3, ed. Perels, pp. 569–70; German tr. Heiser, pp. 403–5. See also above, p. 299. 60 Nicholas I, Responsa, ch. 6, ed. Perels, p. 572; German tr. Heiser, pp. 409–10. 61 Nicholas I, Responsa, chs. 54, 55, ed. Perels, p. 587; German tr. Heiser, pp. 450–2. 62 TC, p. 342.

320

the midd le empire

Methodios’ followers after they were expelled from Moravia. The problem was that the Greek clergy did not know the Slavonic language. The training of local clergy reduced the Bulgarian church’s dependence on Byzantium. In 860 a people called Rus mounted an attack on Constantinople (see above, p. 299). And ‘soon’, according to Theophanes Continuatus ‘an embassy came from them to the ruling city, asking that they be brought into communion through divine baptism; and thus it came to pass.’63 In his circular to the eastern patriarchs, Photios depicts the Rus as under Byzantium’s spiritual authority; despite their previous reputation for savagery, the Rus were now ‘subjects and friends’,64 and had received a Byzantine bishop. A century later, a different version was concocted by Constantine VII Porphyrogenitus: through generous distribution of gold, silver and silk garments [Basil] also inclined towards compliance the invincible and godless people of the Rhos [Rus]. He concluded peace treaties with them and persuaded them to join in the salvation of baptism and to accept an archbishop ordained by Patriarch Ignatios; and the archbishop appeared in their country and the people loved him.65

Then we read of how the bishop was asked by the Rus to cast the Gospel into the fire, but the book would not burn. One mission is more likely than two; the embassy travelled to the Rus under Michael III, but Michael’s achievements were later attributed to Basil I. Whether this short-lived conversion occurred in 863 or 867, this is the earliest surviving Greek account of a religious mission dispatched to distant barbarians in the name of the central authorities in Constantinople. The mission brought no perceptible long-term results; in the tenth century, when Byzantine sources again begin to speak of the contemporary Rus, there is not the slightest hint of any ‘baptism’. m issions to the al ans, hungarians and rus Deliberate Christianisation of Alania – a barbarian power stretching from the Kuban to the Terek – began during the second patriarchate of Nicholas I Mystikos (901–7, 912–25), more precisely between 914 and 918. We possess an invaluable source for this mission in the form of the patriarch’s letters. The attempt to convert the northern Caucasus had been instigated by the Abkhazian principality rather than directly by Byzantium, but Nicholas was personally responsible for several significant initiatives. In the first place, he sent missionaries to Alania drawn from his own closest associates (whereas Constantine and Methodios had held no ecclesiastical office); secondly, he did not send and then ignore them, but kept continual watch 63

TC, p. 196.

64

Phot., no. 2, ed. Laourdas and Westerink, I, p. 50.

65

TC, p. 342.

7. religious missions

321

over their activities (again in contrast with Constantinople’s indifference to the brothers from Thessaloniki); finally, Nicholas set in motion a process which soon led to the inclusion of the see of Alania within the Constantinopolitan patriarchate. In the context of Byzantine traditionalism, this was revolutionary. Dioceses had been founded before, even in foreign realms – Bulgaria, for example – but always within the historical boundaries of the Roman empire. The lands to the north of the Caucasus were completely ‘other’, and their entry into the patriarchal ambit, followed by the vast lands of Rus, opened a new page in ecclesiastical history. The missionaries to Alania give us the first intimations of just how difficult it was to convert barbarians. Peter, archbishop of Alania, complained to Nicholas that his ‘sorrows are many and great is the affliction of [his] evils’.66 He added that Nicholas, who had never been in exile, could not hope to understand his torments. In reply, the patriarch objected, ‘. . . your wisdom was not being sent out for your comfort . . . but to labours and toils and distresses’, before advising him to ‘consider the blessed heralds of the Gospel, in whose number you have been found worthy to be enrolled . . . and cease to lament and to be dismayed because human affairs do not run as we would have them!’ Nicholas then consoled him, declaring that ‘. . . your portion of honour [is] equal to that of the apostles’ own’.67 Peter and another envoy, Euthymios, are the first Byzantine missionaries, in the proper sense, whose names survive in a Greek source. For the first time we read of the conversion of pagans not as an act of divine providence but as hard and often thankless work. And we read of missionaries as real people: self-sacrificing, perhaps, but also prone to despair. The appearance of such figures in Byzantine writings is an important sign that the culture of the Romaioi was developing a rather more realistic view of barbarians. This development is particularly evident in the advice which Nicholas gives to his missionaries. In a letter to Peter, the patriarch formulates his position on marriage among the Alans: As for what you write of matters respecting marriage which are opposed to the church order, and of other habits which give a more pagan character to those practising them, your wisdom is aware that so sudden a conversion of pagan life into the strictness of the Gospel is not easily achievable. You should therefore continually apply your doctrine and salutary exhortation in a paternal and generous spirit . . . and where you find them recalcitrant, bear it with long-suffering, especially if the disobedient belong to the upper class of this nation and are not governed but governors. Towards their subjects you may perhaps be able to behave rather more austerely and despotically . . . but towards the powerful ones, who are quite capable of counteracting the salvation of the whole nation, you must reflect whether, if we behave too harshly to them, we may not unawares exasperate them the more, and thus turn everything upside down.68 66

NM, no. 135, pp. 436–7.

67

NM, no. 135, pp. 438–41.

68

NM, no. 52, pp. 284–7.

322

the midd le empire

Thus the Byzantines softened their previously inflexible stance on polygamy, especially among the nobility. The failure of the CyrilloMethodian mission in Moravia had been to a large extent caused by missionary rigour on precisely this issue (see above, pp. 317–18). Apparently the Greeks had learned useful lessons from their Latin rivals in Bulgaria and Moravia. And yet the fruits of the mission of Euthymios and Peter were not longlasting; al-Mas‘udi relates that after 932 the Alans ‘turned away from their new beliefs and expelled the bishops and the priests who had been sent by the emperor of Rum’.69 We do not know the circumstances in which the Byzantine church reappeared in Alania, although recent research has shown that this could have been as early as the 960s,70 and we shall return to the Christianisation of Alania below. After half a century of Byzantino-Hungarian military clashes and political contacts, around 948 envoys of Fajsz, prince of the Magyars, arrived in Constantinople. A few years later there came Bulcsu, who was baptised by Constantine VII Porphyrogenitus, and his example was then followed by Gyula. According to John Skylitzes, the latter ‘took with him a monk by the name of Hierotheos, renowned for his piety. Theophylact [patriarch of Constantinople] consecrated him bishop of Tourkia [Hungary]. Once there, Hierotheos converted many people from their barbarian errors to Christianity.’71 The metropolitanate of Tourkia appears in none of the official lists, but it probably existed until at least the mid-eleventh century.72 Archaeological evidence suggests that the Byzantine mission was especially active in the region of Szombor.73 Finds of Byzantine reliquary crosses in Hungary are distributed along the course of the Danube and the Tisza. They number about forty, with fifteen dating from between the mid-tenth and the mideleventh century (fig. 18). The ruling family from G´eza I (972–97) onwards accepted the Latin rite, although Greek clerics remained in the southern Hungarian lands well into the twelfth and even thirteenth centuries. Around the turn of the tenth century Niketas the Paphlagonian compiled a cycle of panegyrics in honour of the apostles. He depicts Andrew,74 Bartholomew75 and Matthew as thoroughgoing missionaries; Matthew is even said to have preached to the ‘Ethiopians . . . in their own language’.76 Unlike his predecessors writing on similar themes, Niketas shows awareness of a clash of cultures, though he decides not to describe how his protagonists overcame it. Thomas the Apostle, for example: 69 70 71 72 75

al-Mas‘udi, Muruj al-dhahab, ch. 479, ed. Pellat, I, pp. 228–9; rev. French tr. Pellat, I, p. 173. Beletsky and Vinogradov (2005). Skyl., ed. Thurn, p. 239; French tr. Flusin and Cheynet, pp. 201–2. 73 Gy¨ 74 NP, cols. 64–5, 68. Oikonomides (1971), pp. 527–31. orffy (1976), pp. 175–8. 76 NP, cols. 280–1. NP, col. 208.

7. religious missions

323

(a)

(b) Figure 18 (a) Reliquary pectoral cross found in Hungary, with Christ crucified on obverse and the Virgin with her arms raised in prayer on the reverse (b) Reliquary pectoral cross found in Orosh´aza, Hungary; the poorly rendered figure is of the Virgin or an unidentified saint

. . . arrived among these people who are revolting in appearance but even more repulsive in their disposition. What was it like for him to associate and converse with them on questions of piety! He complained quietly about the burdensomeness of associating with these peoples, and suddenly the solution to all his difficulties appeared.77

As far as the eulogist is concerned, relief comes in the form of intervention by Christ, so he does not delve into the specific techniques of apostolic missionary practice.78 Another indication of the Byzantines’ growing interest in missionary activity can be seen in the Life of the apostle Thomas, contained in the late tenth-century collection of reworked saints’ Lives of Symeon Metaphrastes. All the early versions of Thomas’ acts derived from Gnostic accounts which – contrary to official Christianity – emphasise the harmfulness of 77

NP, col. 136.

78

NP, col. 140.

324

the midd le empire

marriage and wittily describe miracles and transformations. In these earlier versions, the problems of mission as such merit just one phrase: Thomas complains that, as a Jew, he cannot preach to the Indians. Nowhere do these versions explain how Thomas managed to solve his problems. Yet when Symeon Metaphrastes embarked on writing a commentary on Thomas’ acts, missionary problems become a central theme: Thomas was sent to India, which was utterly barbaric . . . That which is rooted over the course of a prolonged period turns into habit, which is stronger than any arguments of reason. Arriving in such a country, the apostle did not behave arrogantly and provocatively, did not start talking grandly and boastfully, and refrained from many things which might have made him seem haughty, insufferable and smug . . . With dirty hair, a pallid face, dry and thin . . . dressed in a dirty threadbare cloak, he prepared himself . . . for meek and humble behaviour. He did not immediately start criticising them, did not reproach them with anything, and decided not to resort to such devices as severity. For he knew: what has become fixed in our souls through long habit cannot easily be eliminated, but is more likely to be changed by persuasion than force. Therefore he resorted more to gentleness, to kind manners and pleasant words . . . He came before them not with arrogance and superciliousness, and not with grandiloquence, but with deeds and signs . . . The Indian people were inducted into the mysteries and the seed of the Word was implanted in their souls. Thomas’ preaching [was disseminated] to such an extent that it reached the king himself, though it did not enter deeply into his consciousness.79

In Metaphrastes’ work Thomas is transformed from the showy magician of earlier tradition into a modest, industrious missionary. Interestingly, Metaphrastes’ ideal preacher conducts his propaganda ‘from below’. The text of the ‘Commentary’ is an implied polemic with those who deal with barbarians without bothering to conceal their contempt. Such an attitude in no way abrogated the imperial conception of barbarians as targets for conquest. In real life these two types of discourse – the imperial and the missionary – coexisted. In Constantine VII Porphyrogenitus’ Book of ceremonies we find the following paean to God: ‘He has enlightened the peoples . . . [and] glorifies imperial benefactors with victories and subjects barbarians to their right hand.’80 Elsewhere in the same work Constantine introduces chants that are to be performed at Pentecost: ‘God, who tamed the godlessness of the nations with tongue-like manifestations of flame, promises through you, brave lords, to conquer and annihilate pagan godlessness. The emperor So-and-So, the joy and reviver of the Romaioi, will force the alien tongues to become of one tongue in

79 80

Symeon Metaphrastes, Life of Thomas the Apostle, pp. 156–67. DC, I.9, ed. Reiske, I, p. 59; ed. and French tr. Vogt, I, p. 54.

7. religious missions

325

faith.’81 The inner contradictions of these passages are self-evident to us, but not to the Byzantines. After the first ‘baptism’ of the Rus in the 860s, there is a long gap in the sources. The only references to Byzantium’s attempts at Christianisation are to be found in the Rus primary chronicle where, after the conclusion of the peace treaty of 911, the emperor provides the Rus envoys with an escort ‘so as to show them the beauty of the churches . . . and to instruct them in their [i.e. the Byzantines’] faith’.82 That there were Christians in the Kievan elite is shown by the fact that a priest called Gregory went to Constantinople as part of Princess Olga’s entourage. The first extensive evidence, however, is provided by the stories of the baptism of Olga in 954–583 or 957,84 as told by the Rus chronicler and by Skylitzes. Despite the detailed account in the Rus chronicle,85 we cannot be certain when or even where Olga was baptised. There is no doubt, however, about the fact that the princess accepted Byzantine orthodoxy; she took the baptismal name of Helena in honour of the empress, who thereby became her godmother. Yet her relations with her godparents soon deteriorated to such an extent that in 959 she requested bishops from the Saxons. The conversion of the princess did not in itself lead to the Christianisation of Rus. This process was instigated in 988 by her grandson Vladimir. Unfortunately not a word about the conversion of Rus can be found in the works of contemporary Byzantine authors; the details have to be extrapolated from Rus, Arabic and western sources and, once again, we encounter the extraordinary silence of the Greek sources. What, for all this, do we know about the Byzantines’ involvement in the conversion? The Rus primary chronicle states that, following envoys from the Muslims, the ‘Germans’ and the Jews, ‘the Greeks sent a philosopher to Vladimir’.86 Scholars have speculated as to who this anonymous ‘philosopher’ might have been. Most likely he is merely the chronicler’s generalised representation of a Byzantine missionary, and the term ‘philosopher’ harks back to Constantine-Cyril the Philosopher. The chronicle puts a long speech into this philosopher’s mouth.87 The speech, supposedly delivered in Vladimir’s presence, is overburdened with names and details that were hardly central to the teaching of Christianity. It is ponderous in the extreme, and hardly likely to have attracted and held the attention of a curious pagan. We cannot treat it as a standard missionary text, routinely regurgitated by Greek missionaries for the conversion of barbarians. 81 82 83 84

DC, I.9, ed. Reiske, I, p. 59; ed. and French tr. Vogt, I, pp. 54–5. PVL, p. 20; RPC, pp. 68–9. Litavrin (1999), pp. 435–6. See also the arguments for 946 in Zuckerman (2000b). Nazarenko (2001), pp. 285–6. For detailed surveys of other hypotheses on the time and place of Olga’s baptism, see Nazarenko (2001), pp. 219–310; Featherstone (2003); Tinnefeld (2005b). 85 PVL, p. 29; RPC, p. 82. 86 PVL, p. 40; RPC, p. 97. 87 PVL, pp. 40–8; RPC, pp. 97–109.

326

the midd le empire

Although the episode with the ‘philosopher’ has clearly been inserted into the chronicle from some other work unconnected with Vladimir, it is not pure fantasy. Aspects of the philosopher’s conduct remind us of other missionaries. For example, a painting of the Last Judgement is shown by the philosopher to Vladimir,88 just as a painting of the Last Judgement had figured in the conversion of Boris of Bulgaria by the Byzantine missionary Methodios; and the reliance on citations from the Old Testament is reminiscent of Constantine-Cyril’s speech to the people of Phoullai. Be that as it may, in the chronicle’s account none of these ploys impressed Vladimir. He refused to be baptised, and said: ‘I will wait a little more.’ Next, according to the chronicle, the prince sent his own embassies to various countries in order to ‘test the faiths’. In Constantinople the emperor and patriarch did everything possible to impress the envoys with the pomp of the service in St Sophia, and ‘they were astonished’.89 Yet despite the envoys’ warm reception in Constantinople, and despite their very positive reactions, Vladimir attacked and captured the Byzantine city of Cherson. We need not enter into the scholarly debates and attempt to explain this extraordinary turn of events. Vladimir’s baptism, according to the Rus primary chronicle, was merely a corollary to the negotiations about the return of Cherson, a precondition for receiving the emperor’s sister Anna as his bride;90 the negotiations concluded, Anna travels to Cherson not with a metropolitan for Rus, nor even with a staff of missionaries, but merely with the clergy of her personal entourage. It is left to Vladimir to say: ‘let those who have come with your sister baptise me’; so ‘the bishop of Cherson together with the emperor’s sister’s priests instructed Vladimir in the faith and baptised him.’91 After his baptism Vladimir ‘took his imperial bride, and Anastasios [the Chersonite who had betrayed the town to the Rus] and priests from Cherson . . . and he also took ecclesiastical vessels and icons.’92 This suggests that providing liturgical vessels had not been reckoned a missionary responsibility of the princess’ entourage, nor had anyone had the foresight to bring vessels from Constantinople in anticipation of the baptism of Rus; instead they were simply trophies plundered by Vladimir in Cherson. Although the Rus metropolitanate most probably existed from around 990, nothing is heard about it until 1039. Later Russian chronicles attempted to fill this lacuna with tales of local conversions,93 and even with the story of a certain ‘philosopher’ named Mark the Macedonian, who was allegedly sent by Vladimir on a

88 89 90 93

ˇ cenko, I. (1988–9), pp. 25–6. PVL, p. 48; RPC, p. 110. See also Sevˇ PVL, p. 49; RPC, p. 111. 91 PVL, p. 50; RPC, p. 113. 92 PVL, p. 52; RPC, p. 116. PVL, p. 50; RPC, p. 112. Nikon chronicle, PSRL 9, pp. 63–4; tr. Zenkovsky and Zenkovsky, I, pp. 110–11.

7. religious missions

327

mission – in the event unsuccessful – to the Volga Bulgars in 990.94 But this is highly dubious information from a late source. We learn something of the activity of senior Byzantine clergy in Rus from the series of questions that were put to Metropolitan John II. On the one hand, the metropolitan’s general approach is plain enough in his injunction to ‘adhere to strictness rather than to the custom of the land’.95 He rejects anything that is ‘far from present-day piety and the becoming way of life of the Romaioi’.96 Yet in two of his responses John shows a degree of tolerance: firstly he urges that sorcerers and magicians should not be punished with mutilation;97 and secondly, he allows priests to wear animal skins under their robes ‘because of the terrible cold and frost’.98 But such concessions to local conditions are rare. Besides the metropolitanate at Kiev, in the eleventh century there were bishoprics in perhaps as many as eight other towns, including Chernigov, Pereiaslavl’, Polotsk and Novgorod, and in the twelfth century in at least three more. The prelates in all these towns were Greek-born, but the only detailed information comes from the Life of an eleventh-century bishop, Leontios of Rostov (though the text was composed in the twelfth century, some time after the death of its hero). According to his Life, Leontios had been preceded by Theodore and Ilarion, Byzantines who, ‘unable to endure the abuse and persecution, fled home to the [land of the] Greeks’.99 Initially Leontios, too, had little success; driven out by the pagans, he moved to the edge of the town and built himself a hut. Children began to visit him and he gave them instruction, and then adults, too, would come. Eventually Leontios was invited back to the citadel, where he set about cautiously instilling Christianity, with encouragement and gentleness. Leontios’ success was again brief: he died in a pagan uprising. Although its ‘facts’ are probably fictitious, the Life of Leontios reflects contemporary Byzantine missionary practice; or at any rate it reflects the impressions of such practice that were formed in Rus. One serious problem for Byzantine churchmen was their ignorance of the local language. Metropolitan Nikephoros addresses the Kievans thus: ‘I have not been granted the gift of tongues, like the divine Paul, so as to carry out my tasks in that language [i.e. Slavonic], and therefore I stand amongst you voiceless and am much silent.’100 94 Nikon chronicle, PSRL 9, pp. 58–9; tr. Zenkovsky and Zenkovsky, I, p. 109; see also Kniga stepennaia, I, pp. 111–13. 95 John II, Canonical responses, ed. Beneshevich, p. 109. The text survives in Greek and in Slavonic versions, though the two do not always coincide. Here we draw on both versions. 96 John II, Canonical responses, ed. Beneshevich, p. 110. 97 John II, Canonical responses, ed. Beneshevich, pp. 110–11. 98 John II, Canonical responses, ed. Beneshevich, p. 114. 99 Life of Leontios of Rostov, ed. Titov, p. 4. 100 Text in Nikephoros, Sermon, ed. Nazarenko, p. 569; Nikephoros, Works, ed. Polianskii, p. 186.

328

the midd le empire l ater byz antine mission-work

From the eleventh century onwards101 Byzantium’s only remaining pagan neighbours were the nomadic peoples of Asia Minor and the Black Sea steppes. Missions sent to them typically achieved swift successes which could just as easily be reversed. In 1048 the Pecheneg leader Tyrach was converted.102 Several years earlier Kegen, leader of a Pecheneg splintergroup, had ‘received holy baptism, himself and those with him. And a certain pious monk named Euphemios was sent, who set up a sacred font beside the Danube and provided holy baptism for all.’103 This conversion provoked a certain amount of controversy in Byzantium. John Mauropous viewed it with great enthusiasm,104 his friend Michael Psellos was quite sceptical,105 while Michael Attaleiates was downright hostile: ‘there is no point in trying to bleach the Ethiopian.’106 Alexios I Komnenos (1081–1118) was praised for his missionary endeavours: For the emperor . . . was fond of teaching our doctrines and was a real missionary by choice and in his manner of speech; he wanted to bring into the fold of our church not only the Scythian nomads, but also the whole of Persia, as well as the barbarians who inhabit Libya and Egypt and follow the rites of Muhammad.107

Anna Komnena returns to this theme elsewhere: ‘I for my part would call him “the thirteenth apostle”.’ In contrast with the church fathers, who had reckoned the world already baptised or about to be baptised, Alexios took a realistic view both of the extent of the unbaptised world and of the complexity of the task before him. We should note, nevertheless, that there is no firm evidence that Alexios ever dispatched any religious missions beyond the old limits of the empire. Theophylact, archbishop of Bulgaria two generations after its conquest by Byzantium, spent the first half of his life – until 1092 – at the Constantinopolitan court, and the second part in provincial Slav Ohrid. He composed the extended Life of Clement of Ohrid, his remote Slav predecessor in his see. Theophylact’s missionary principle, as it emerges from the Life, may be formulated thus: when helping barbarians adapt to Christianity, one should take them as they are, and one should simplify Christianity to make 101 102 103 104 105 106 107

For more detail on Byzantine missions of the second millennium, see Ivanov (2007). Skyl., ed. Thurn, p. 459; French tr. Flusin and Cheynet, p. 380. Skyl., ed. Thurn, p. 456–7; French tr. Flusin and Cheynet, pp. 378–9. John Mauropous, Quae . . . supersunt, no. 182, ed. de Lagarde, pp. 143–6. Michael Psellos, Orationes panegyricae, ed. Dennis, p. 63. Attal., ed. Bekker, pp. 30–1; ed. and Spanish tr. Per´ez Mart´ın, p. 25. Al., VI.13.4, ed. Reinsch and Kambylis, I, p. 199; ed. and French tr. Leib, II, p. 81; tr. Sewter, pp. 211–12. See also XIV.8.8, ed. Reinsch and Kambylis, I, p. 457; ed. and French tr. Leib, III, p. 181; tr. Sewter, p. 466.

7. religious missions

329

it accessible to their understanding. He praises Cyril and Methodios for creating ‘an alphabet which matched the coarseness of the Slavonic tongue’ (see fig. 19).108 He likewise appreciates the flexibility of his own hero, Clement, in dealing with the barbarians: ‘Knowing the coarseness of the people and their extraordinary obtuseness in mastering Scripture, [Clement] . . . devised the following scheme: for every festival he composed sermons that were simple, clear, containing nothing deep or subtle, the type of sermon that could not escape the comprehension of even the dullest of Bulgarians.’109 Theophylact explains that in Bulgaria only ‘wild’ trees had grown, bearing no ‘cultured’ fruit; but Clement ‘ennobled the wild plants through grafts, in order (as I think) thus to nurture human souls.’110 What we have here is not so much a narrative of Clement’s specific missionary activity, but more of a parable: Byzantine culture cannot be forced on the barbarians; it must be carefully grafted onto their own culture.111 With few exceptions, the emphasis shifts in the twelfth century from the conversion of barbarians to their subjugation. In the abundant panegyrical literature, emperors’ victories wholly overshadow their missionary achievements. One of the period’s few known preachers to the barbarians was Nicholas Hagiotheodorites, metropolitan of Athens, who died in 1175.112 Moreover, according to Euthymios Tornikes, in this period Byzantium reestablished bishoprics in the cities that had been captured by the Seljuqs, but not, he stresses, in new places.113 The work of Byzantine missionaries in Alania continued, although we have no direct sources on the subject. The seat of the metropolitanate of Alania is thought to have been a town in the vicinity of modern Nizhnii Arkhyz, but we do not even know the town’s name. Active church-building continued in the northern Caucasus, although the architecture of the extant churches is more reminiscent of Abkhazia than of Byzantium. A few dozen Greek inscriptions attest the presence of Greeks. Apparently there was an attempt to adapt the Greek alphabet so as to render local languages. Vestiges of Byzantine orthodoxy, albeit sometimes in heavily distorted form, have been detected in the pagan beliefs of the modern inhabitants of the northern Caucasus, especially the Ossetians. There exists a unique written document, a report by Theodore, metropolitan of Alania. In 1225 Theodore sent to the Nicaean patriarch, Germanos II, a report on his journey to the Caucasus. The report’s general conclusion is that Christianity in Alania has withered: ‘Alas, on apostolic foundations there was built a house of straw and cane, 108 109 110 111 112 113

Theophylact of Ohrid, Life of Clement, II.7, ed. Milev, p. 80; ed. Iliev, p. 82. Theophylact of Ohrid, Life of Clement, XXII.66, ed. Milev, p. 132; ed. Iliev, p. 101. Theophylact of Ohrid, Life of Clement, XXXIII.68, ed. Milev, p. 134; ed. Iliev, p. 102. Floria et al. (2000), pp. 202–3, 208. Euthymios Tornikes, Syngraphai, ed. Papadopoulos-Kerameus, pp. 159–60. Euthymios Tornikes, Syngraphai, ed. Papadopoulos-Kerameus, p. 183.

330

the midd le empire

Image removed for rights reasons

Figure 19 Early example of the Glagolitic script invented by Constantine-Cyril to cater for the Slavonic language

and it has fallen victim to fire’;114 ‘the Alans are Christians only in name.’115 Theodore complains at the lack of proper missionary experience, though at the same time he is proud of his own modest successes in this area.116 114 116

115 Theodore of Alania, Alanikos, col. 409. Theodore of Alania, Alanikos, col. 400. Theodore of Alania, Alanikos, cols. 405, 409.

7. religious missions

331

Figure 20 Early example of Cyrillic script: an early eleventh-century inscription in Bitola, commissioned by John Vladislav (see below, p. 529); although now bearing the name of Constantine-Cyril, this form of writing was a pragmatic adaptation of the Greek alphabet to accommodate the Slavonic language’s distinctive sounds and gradually replaced the brand-new, accurate, but intricate script devised by Constantine-Cyril himself

State missionary activity limped on after the restoration of Greek power in Constantinople, but a weakened empire and the strengthening of her Islamic neighbours forced the emperors to show extreme caution.117 However, centres of orthodoxy in close contact with the barbarians, for example in the Crimea, were active in conversion-work. Late Byzantine baptisteries suitable for the baptism of adults have been found at several Crimean sites, suggesting possible missionary activity on the part of the local cave monasteries. In the late Byzantine period some new ecclesiastical provinces were created on barbarian territory. Among them was a bishopric instituted at Sarai, capital of the Golden Horde. In 1276 Bishop Theognostos of Sarai sent the patriarch of Constantinople, John Bekkos (1275–82), a list of questions arising from his pastoral work. Many questions reveal the missionary character of his concerns.118 The patriarch’s answers display considerable tolerance. In its final period the empire was eventually able to work out an integral and flexible ideology of mission. Realism characterises the missionary activity of the patriarchate in general. Thus in September 1365 a new bishopric of 117 118

Duc., XX.4, ed. Grecu, p. 135; tr. Magoulias, p. 112. Theognostos of Sarai, Questions, col. 136 (Slavonic version); appendix 1, col. 10 (Greek text).

332

the midd le empire

Achochia is mentioned in a patriarchal document.119 The bishopric was perhaps created for the migrant Abkhazian population. Around 1317, archbishoprics were founded in Lithuania and the Caucasus,120 although an attempt to convert the Lithuanian prince Olgerd ended in failure.121 As an example of this more practical approach to mission one might point to Gregory Palamas: while in Muslim captivity in 1354, Gregory conducted religious disputations, and in Nicaea he preached Christianity in the streets, on his own initiative.122 And yet, even on the eve of its downfall, Byzantium could not fully shed its cultural snobbery or arrogance. The very term barbarian refers to a political discourse dominated by Roman imperial rhetoric in which Christianity does not fit comfortably. Missionary ideas are also the losers at the level of folkloric discourse. Thus among the Byzantines there was a widely held belief that the northern tribes of Gog and Magog had been locked behind iron gates by Alexander the Great. No writer from among the Romaioi ever took the trouble to consider whether Gog and Magog might be baptised. For the Greeks, the cultural stereotype was stronger than the religious principle: speaking in terms of Gog and Magog they could indulge themselves, lumping together all barbarians as a seething, subhuman mass. In the Byzantine mind the concept of universal Christianity was linked to the idea of world empire, which the Byzantines never entirely abandoned. This aspect of their outlook could easily be dubbed expansionism. Dimitri Obolensky proposed that Byzantium maintained an enormous, highly complex and diffuse system of international ties, which he called the ‘Byzantine commonwealth’. However, if we look closely at the fabric of the relations between the Romaioi and the world around them, we see that there was as much isolationism as there was expansion. In Byzantine missionary activity we find a paradoxical yet characteristic instance of isolationism, in the form of barbarians being converted by a stylite.123 The image of a static, lone missionary contradicts the basic concept of activism that the idea of proselytising normally implies. Yet in this image we can see the distillation of a specifically Byzantine perception of mission. 119 121 122 123

120 NE, no. 17, pp. 399–400 (text), 182 (commentary). MM, I, p. 477. NG, XXXVI.40–1, ed. Schopen and Bekker, III, pp. 520–1. Philippidis-Braat, ‘La Captivit´e de Palamas’, p. 161. Life of Symeon Stylite the Younger, ed. van den Ven, I, p. 112.

CHAPTER 8

ARMENIAN NEIGHBOURS (600–1045) t. w. gre e n woo d

introduction Anyone wishing to unravel the history of the relationship between Byzantium and Armenia from late antiquity into the eleventh century has to confront a series of historical and historiographical challenges. The most immediate, and intractable, of these is one of definition: what does ‘Armenia’ mean? Although Armenia is used to express a territorial entity in contemporary texts, both Armenian and non-Armenian in origin, its precise meaning varies according to the date and the context in which it is used. Far from finding a single, stable definition of Armenia, one discovers multiple ‘Armenias’.1 Thus a seventh-century Armenian geographical compilation depicts ‘Great Armenia’ as comprising not only regions currently recognised as Armenian but also those with historic associations.2 Successive provinces of Armenia were imposed and superimposed by external powers, each with a particular scope. The kingdom of Armenia, re-established in 884, bore little relation to its Arsacid precursor and increasingly represented only the Bagratuni kingdom centred on Ani, excluding rival kingdoms in Vaspurakan, Siwnik‘ and elsewhere. Given the absence of stable territorial boundaries and in the light of significant Arab settlement in certain districts from the end of the eighth century, there have been attempts to construct Armenian identity in terms of a blend of confessional, linguistic and cultural features. Once again the evidence supports a plural and inclusive definition. Instead of a community of believers, united around a single confession and recognising the spiritual authority of a single leader, the Armenian church embodied a spectrum of doctrinal interpretations, revolving largely, but not exclusively, around the acceptance or rejection of the council of Chalcedon.3 This interpretation is at odds with the conventional outline of Armenian church history supplied by the majority of the Armenian sources, which advertise a pronounced anti-Chalcedonian, monophysite character after 600. Yet the faint 1 2 3

Hewsen (2001) offers a comprehensive sequence of maps. Anania of Shirak, Geography, ed. Soukry, pp. 29–35; tr. Hewsen, pp. 59–70. Garso¨ıan (1999a) to 700; thereafter Mah´e (1993).

333

300

200

C

K

S

E

A

C

A

IA

A

ZI

P

L

LA

S

B

400 km

200 miles

A

100

C

100

0 0

N

Phasis

IB

SH

IA

R

A SE

’I JET AR KL

E

I

R

ND R Aruch NA E TA O VA L. Sevan P Aghaman Nakhchawan S Avan Mren BASEAN Vagharshapat ATS Vardanakert Araxe GHE s Dvin K A Theodosioupolis EKE RI E GR V A A B ND Mount Ararat N SIW Artsap k NIK Bagavan Daroynk Nakhchawan FOURTH Khram

A

Sebasteia (Sivas)

DARANAGHI

ARMENIA Melitene

TA RO

L. Van

N Bitlis

V

Sozopetra

A

S

P

U

R

A

K

A

N

A

L

B

N

IA

ax es

K

Sebastopolis

A

Ar

AN ) IJ A AN B ER TAK Z A A RP ( AT

Germanikeia

M

E D

I A

Approximate frontier c. 715–850

Nineveh

Map 16 Armenia 591–850

Frontier 591–630

8. armenian neighbours (600–1045)

335

Figure 21 View from within the walls of Ani, looking south across what was the heart of the flourishing eleventh-century city to the ruined cathedral and beyond to the hills in the Republic of Armenia

impression of a pro-Chalcedonian, and arguably pro-Byzantine, party may still be traced and other schismatic traditions may have survived long after their suppression elsewhere.4 Nor is there good evidence for either linguistic or cultural uniformity. Whilst the written form of the Armenian language may once have possessed such a quality, it seems inherently unlikely that contemporary speech was ever uniform. An eighth-century cleric, Stephen of Siwnik‘, identified seven dialects, all associated with remote, mountainous districts.5 As for cultural uniformity, one has only to think of the selective histories, sponsored by princely houses to their own glory and the denigration of others, the multiple versions of the History of Agathangelos describing the conversion of Armenia or the different traditions surrounding the relics of Gregory the Illuminator, to appreciate that the past was essentially plastic, at the disposal of contemporary writers to develop and rework as they thought fit.6 When one considers the fragmented, isolating topography of the central Caucasus region, the individual districts of varying size, wealth and potential, the harsh continental climate, the dispersed settlement pattern focused upon the village, the frontier status of the region through the period, partitioned between Rome and Persia and then Byzantium and the caliphate, 4 5 6

Garitte (1960); Arutiunova-Fidanjan (1988–9). Stephen of Siwnik‘, Meknut‘iwn, ed. Adontz, p. 187. See respectively TA; van Esbroeck (1971a); van Esbroeck (1971b).

336

the midd le empire

the lack of organic national political institutions, the long-standing doctrinal divisions within the Armenian church, the presence of different dialects and languages, even the potential for different interpretations of the past, one can only conclude that ‘Armenia’ and ‘Armenian identity’ are complex and elusive terms defying concrete definition and characterised by fluidity and plurality. Instead of maintaining the fiction of a united Armenia or a singular Armenian identity, Armenian diversity and incongruity deserve to be highlighted. A second challenge is the uneven treatment in the primary sources of the relationship between Armenia and Byzantium. At times, it receives significant coverage but more often it remains frustratingly obscure, the periods between 730 and 850, and between 925 and 980 being particularly opaque. This may reflect a genuine lack of engagement. But it is also possible that the outline of Armenian history presented by the majority of Armenian sources is intentionally partial. Arguably, Armenian authors anticipated a similar collective historical experience to that of the people of God in the Old Testament and therefore stressed those contexts which replicated the biblical paradigm, including valiant but ultimately unsuccessful resistance against an oppressive and impious empire, exile and return. A neighbouring Christian polity, particularly one which adhered to a rival confession of faith, did not sit comfortably with this model and its influence was therefore downplayed or ignored. Armenian histories are much more than simple vehicles for the preservation of factual information; rather they are complex compositions which need to be handled with care and exploited only after careful textual criticism. Silence on the subject of Byzantium and the imperial church should not be mistaken for lack of contact. Finally, insofar as the literary sources record the development of Byzantium’s relationship with Armenia, they tend to do so in terms of the principal Armenian political and ecclesiastical leaders. As we shall see, Byzantium cultivated multiple ties with several noble houses at the same time. In a society characterised by intense competition between and within princely families, in which those with ambition and ability attracted followers, acquired lands and amassed wealth at the expense of those who did not, it paid to develop links with as many potential clients as possible. Some of this evidence survives only through contemporary Armenian colophons and inscriptions, sources whose historical potential has not been fully exploited. By drawing on these materials, as well as the twin disciplines of numismatics and sigillography, a more complex, nuanced picture of their relationship begins to emerge. political and conf essional flux (591–661) In 590 the fugitive Sasanian king Khusro II (590, 591–628) appealed to Emperor Maurice (582–602) for military assistance against the usurper

8. armenian neighbours (600–1045)

337

Bahram Chobin, offering generous terms, including substantial territorial concessions in Armenia. These were accepted by Maurice, and after the defeat of Bahram in 591 the frontier shifted eastwards.7 The following decade witnessed unprecedented cooperation between the two ‘great powers’ across Armenia. Maurice and Khusro II set out to strip their respective Armenian sectors of soldiers for service in distant conflicts. Two rebellions from the middle of this decade attest the resulting sense of bewilderment among the Armenian elite.8 Only the uprisings in the 770s and the resistance to the forces of Michael IV (1034–41) in 1041 outside Ani reveal a similar desperation. The first of the two rebellions collapsed when threatened by imperial and Persian forces acting in concert. The second ended in bloodshed. An army under the general Heraclius and Hamazasp Mamikonean defeated the rebels, killing the majority and capturing the remainder who were taken back to Theodosioupolis and executed. The only rebel to escape fled to Khusro II but was returned, tortured and killed. The role of Hamazasp Mamikonean challenges the standard picture of Armenian helplessness in the face of implacable imperial oppression. Here is an Armenian noble serving imperial interests inside Armenia. The suspicion must be that there were other Armenian princes prepared to work with the new regime. When war with Persia broke out after Maurice’s assassination in late 602, as Khusro II sought to recover those districts previously ceded, several Armenian princes fought for Byzantium. In 605, the Byzantine forces defending the district of Bagrevand against Khusro were led by the local Armenian lord Theodore Khorkhoruni who entered into negotiations with the Persians only after Byzantine forces had withdrawn.9 Significantly, it took at least five seasons of campaigning for the Persians to expel the Byzantine forces from Armenia (603–7). Moreover, the fighting was not restricted to those western districts which had been under imperial control for generations but was concentrated further east, across the districts recently acquired by Byzantium. Such a holding strategy would have been inconceivable without local support. The decade after 591 also witnessed pressure upon those districts now under imperial control to conform to imperial orthodoxy. Although Catholicos Moses II (574–604) refused to attend a council in Constantinople convened to establish union between the churches and remained in the Persian sector at Dvin, Maurice ordered the council of Chalcedon to be preached in all the churches of the land of Armenia, threatening ‘to unite them in communion through the army’.10 A second catholicos, John 7 8

Whitby, Michael (1988), pp. 297–304. See above, p. 169. Seb., chs. 15–18, ed. Abgaryan, pp. 87–90; tr. and comm. Thomson and Howard-Johnston, I, pp. 32–5. 9 Seb., ch. 32, ed. Abgaryan, pp. 109–10; tr. and comm. Thomson and Howard-Johnston, I, pp. 60–2. 10 Seb., ch. 19, ed. Abgaryan, p. 91; tr. and comm. Thomson and Howard-Johnston, I, p. 37. See also above, pp. 169–70.

338

the midd le empire

of Bagaran, was established at Avan, provocatively situated just across the border. John is usually titled ‘anti-catholicos’ and dismissed as little more than the creature of Maurice with an ephemeral influence upon Armenia. However, there is good evidence for a sizeable body of support for John, at and below diocesan level. After the election of Abraham as catholicos (perhaps in 606, probably in 607), five bishops and nineteen leaders of religious communities, including those linked to the ‘holy cathedral’ and the church of St Hrip‘sime in Vagharshapat, acknowledged their error and returned to the anti-Chalcedonian party.11 Moreover, there were repeated attempts at ecclesiastical reconciliation. In 604, the Byzantine commander in Armenia, Sormen, wrote to the temporary head of the monophysite party, Vrt‘anes, noting that they had met and corresponded on this subject many times. Sormen expressed a hope that they could meet ‘like fellow brothers, joint heirs in baptism and sons in the faith of our father St Gregory’, revealing thereby his own Armenian ancestry.12 This spirit of compromise, which was not reciprocated, seems to find an echo in the remarkable karshuni version of Agathangelos.13 This transposes the key events in the original narrative of the conversion of Armenia to different, contemporary locations. Thus of the seventy-seven virgins who accompanied St Hrip‘sime, forty are assigned to Dvin and thirty-seven to Avan, thereby establishing the equal sanctity of both sees. Gregory the Illuminator baptises in the western district of Ekegheats‘; he meets King Tiridates fifteen kilometres from Theodosioupolis; and he dies in Daranaghi. This radical revision represents a rare witness to the intellectual tradition of the pro-Chalcedonian party in Armenia after 591 and a very subtle development – or rather, subversion – of Armenian tradition. Even the Byzantines’ defeat at Persian hands in Basean, probably in 607, and their subsequent loss of key fortresses, including Theodosioupolis, did not mark the end of operations in Armenia. The following year, a Byzantine counter-attack in the district of Theodosioupolis was repulsed, whilst in 610 the city’s inhabitants were transferred to Ecbatana in Persia, suggesting an ongoing threat. In 613, another Byzantine army marched through these districts. When Heraclius (610–41) launched a significant campaign in 624 against Theodosioupolis and then Dvin, he was advancing through districts which had been incorporated into provincial and episcopal structures for generations. Evidently he was looking to attract additional support. In autumn 624, Heraclius appealed to the princes and leaders of the lands of Albania, Iberia and Armenia by letter, urging them to come and serve 11 12 13

Book of letters, ed. Izmireants‘, pp. 151–2; ed. Pogharean, pp. 298–9; French tr. Garso¨ıan, pp. 514–15. Book of letters, ed. Izmireants‘, p. 90; ed. Pogharean, p. 231. van Esbroeck (1971a); Cowe (1992).

8. armenian neighbours (600–1045)

339

him together with their forces but threatening reprisals and subjugation if they refused.14 It is impossible to gauge the response to his appeal but it seems that many Armenian princes preferred to support Khusro II.15 Only one late source refers explicitly to Armenians being attracted into imperial service before Heraclius’ defeat of the Persian army at the battle of Nineveh on 12 December 627.16 The years between 624 and 628 witnessed a complex series of military manoeuvres and engagements in the Transcaucasus.17 Three primary strategic considerations seem to have guided Heraclius. He courted potential allies across the Transcaucasus and from the steppe world to the north. The decisive impact of Turkic forces in 627 and 628 cannot be exaggerated. Secondly, such a strategy drew Persian armies away from Constantinople and into an environment in which logistical pressures dictated that possession of the larger army was no guarantee of success. Thirdly, whether or not instructed by his father, Heraclius had recognised the potential for striking at the centre of the Sasanian kingdom from the north, using Armenia as a bridgehead.18 Such considerations go a long way towards explaining why Armenia continued to command such attention from successive emperors throughout the seventh century and beyond. When Byzantine forces were expelled in 607, the monophysite party in the Armenian church was already in the ascendant and remained so throughout the reign of Khusro II. The latter began to favour the expanding monophysite confession across his dominions in preference to the Nestorian church of the east. In the aftermath of Heraclius’ triumph and the return of the True Cross to Jerusalem on 21 March 630, the fissures within the Armenian church were reopened. The recently appointed catholicos Ezra (630–41) was invited to attend a church council at Theodosioupolis, probably in early 631, and under threat of the creation of a second catholicos he accepted union. Statements that Ezra was ‘a humble and gentle man’ and that ‘no indecorous word ever passed from his mouth’ reflect a partisan opinion.19 In reality his accommodation with Heraclius is likely to have provoked considerable antagonism, an echo of which may be found in the exile of John of Mayragom, an ardent monophysite whose own catholical ambitions had been thwarted by Ezra’s election.20 An inscription commemorating Ezra, partly in Greek and partly in Armenian cryptograms, has been unearthed at Avan; evidently Ezra wished to associate himself 14 HA, II.10, ed. Arak‘elyan, p. 132; tr. Dowsett, pp. 79–80. The History of the Albanians has been variously, and wrongly, attributed to Moses Daskhurants‘i or Moses Kaghankatuats‘i; the identity of the compiler is unknown. 15 Theoph., ed. de Boor, I, p. 311; tr. Mango and Scott, p. 443. 16 Chronicon ad 1234, ch. 99, ed. Chabot, I, pp. 233–4; Syrian chronicles; tr. Palmer et al., p. 137. 17 Howard-Johnston (1999). 18 Kaegi (2003a), pp. 22–3. 19 Greenwood (2002), pp. 360–3. 20 Yov., XVIII.15–30, ed. Emin, pp. 77–80; tr. Maksoudian, pp. 99–100.

340

the midd le empire

with the church founded there by John of Bagaran and the confessional tradition espoused by him.21 Ezra’s choice of Avan was also dictated by political circumstance, since Dvin still lay in the Persian sector. The deposition of Khusro II did not give Heraclius possession of the whole of Armenia. In 628, Khusro II’s successor, Kavad II, appointed Varaztirots‘ Bagratuni as governor (marzban) of Armenia. Only under the terms of a subsequent treaty in the summer of 630, between Heraclius and the latest claimant to the Sasanian throne, Boran (630–31), were those districts ceded to Maurice returned to Byzantine control. Even then, Persian influence over eastern and southern Armenia persisted. In autumn 637, the leading Armenian prince, Mushegh Mamikonean, responded to a Persian call-to-arms, raising 3,000 troops whilst Gregory, lord of Siwnik‘, contributed 1,000.22 Both fell at the battle of al-Qadisiyya on 6 January 638. With the benefit of hindsight, such loyalty to the Sasanian cause might seem misguided, but the success of the Arab conquest of Persia was still far from assured at that time. The loyalty of Varaztirots‘ Bagratuni and Mushegh Mamikonean to Sasanian Persia may also explain the promotion of ‘new men’ to the office of ‘prince of Armenia’ in the Byzantine sector of Armenia after 630, a title used to denote the principal client. Mzhezh Gnuni and his successors, David Saharuni and Theodore Rshtuni, all came from minor noble houses. Although the narrative sources reveal little beyond this sequence, epigraphic evidence supports the proposition that this decade saw an intense Byzantine campaign to attract a broad spectrum of support. Three inscriptions, recording the foundation of churches at Aghaman (completed 636/7), Bagavan (August 639) and Mren (between 638 and mid-640), all give a regnal year of Heraclius and accord him a laudatory epithet.23 Contemporary regnal formulae and protocols used in imperial documents and legislation repeat this combination. These inscriptions therefore attest an otherwise lost body of correspondence between Byzantium and Armenia. The inscriptions at Aghaman and Mren also confirm that imperial honours were distributed and were prized by their recipients. The founder of the small church at Aghaman chose to define himself as Gregory elustr – i.e. illustris, no more than a middle-ranking imperial title by this time. This reveals a considerable down-reach on the part of the imperial authorities into individual Armenian districts, for Gregory was not the lord of the district in which he sponsored his church. The founder of the church at Mren, David Saharuni, is titled patrikios, kouropalat¯es and sparapet of Armenia and 21 22 23

Greenwood (2004), inscription A.6 and p. 41. Seb., ch. 42, ed. Abgaryan, p. 137; tr. and comm. Thomson and Howard-Johnston, I, pp. 98–9. Greenwood (2004), inscriptions A.4, A.5 and A.7 and pp. 43–7, 62–78.

8. armenian neighbours (600–1045)

341

Syria. His remit encompassed all Armenia and must postdate the death of Mushegh Mamikonean at al-Qadisiyya in 638. The extension of his command beyond the boundaries of Armenia into Syria is unprecedented and suggests that Heraclius was prepared to make remarkable concessions in his efforts to forge an effective opposition to the Arab invasions after the fall of Syria, one in which Armenian military resources had a leading role to play. The contention that Heraclius invested heavily in a network of Armenian clients is supported by the numismatic evidence. Seven different issues of silver hexagrams from the reign of Heraclius and four issues of Constans II (641–68) have been discovered in hoards or during excavations in Armenia, the latest issue being struck between 654 and 659.24 This flow of Byzantine silver into Armenia has traditionally been linked to the presence of Byzantine forces; however, in light of the epigraphic evidence and the elite’s prosperity, reflected in the numerous church foundations, one is tempted to speculate whether this silver was minted for, and paid to, Armenian clients. Armenia had been integrated into the Sasanian silver-based monetary system for centuries and silver coins would have been familiar to Armenians. This strategy proved effective during the following decade. When an Arab raiding party advanced from northern Syria through the Bitlis pass in autumn 640 and sacked Dvin, Theodore Rshtuni ambushed the invaders during their retreat, albeit without much success.25 A second Arab raid, attacking from the south-east through Azerbaijan in summer 643, encountered stiff resistance. One of its divisions, numbering about 3,000, was heavily defeated by Theodore Rshtuni outside the fortress of Artsap‘k‘. The major centre of Nakhchawan in the Araxes valley held out. These operations showed the offensive and defensive potential of Armenia and may have deterred further attacks. Armenia was not insulated from the political turmoil engulfing Constantinople after the death of Heraclius. The failed coup by Valentinus in 645 seems to have prompted widespread changes in the military hierarchy across Armenia. The new commander, Thomas, was anxious not to damage the agreement established with Khorokhazat, leader of continuing Persian resistance against the Arabs in Atrpatakan (Azerbaijan). Thomas visited him and promised that Theodore Rshtuni would be taken to Constantinople.26 This episode illustrates how the interests of two clients did not necessarily coincide. Khorokhazat faced growing recalcitrance from Albania 24 25

Mousheghian et al. (2000a). Seb., chs. 42, 44, ed. Abgaryan, pp. 138–9, 145–7; tr. and comm. Thomson and HowardJohnston, I, pp. 100–1, 109–11. 26 Seb., ch. 44, ed. Abgaryan, pp. 142–3; tr. and comm. Thomson and Howard-Johnston, I, pp. 106–8.

342

the midd le empire

and was looking for assistance in deterring Armenian support for dissident elements. In choosing to back Khorokhazat, Byzantium precipitated a crisis in Armenia. Theodore Rshtuni was soon restored to his command but the relationship was clearly strained. In 652 the governor of Syria (and later caliph) Mu‘awiya (661–80) induced him to switch sides, promising inter alia that Armenian forces would not be employed in Syria and that Arab forces would not be stationed in Armenia unless invited to repel a Byzantine attack.27 In response, Constans II travelled to Armenia to shore up his support and undermine his erstwhile client. He advanced to Theodosioupolis and there received the submission of a disparate group of Armenian princes and their armed forces. Evidently they believed that it was in their long-term interests to return to imperial service. Constans II moved on to Dvin and stayed with Catholicos Nerses III (641–61). He attended a service with his host in the cathedral church of St Gregory, during which the liturgy was celebrated in Greek and the council of Chalcedon was proclaimed. Only one anonymous bishop refused to participate but this tells us little about the ongoing confessional tensions within the Armenian church; presumably anti-Chalcedonians did not attend. Constans II did not remain in Armenia long, being forced to return and defend Constantinople in 654. Thereafter Byzantine fortunes fluctuated, imperial forces being driven out of Armenia twice, but by the first half of 656, Hamazasp Mamikonean was securely installed as kouropalat¯es and prince of Armenia.28 At the same time, honours were distributed to the other princes and treasures to the soldiers, confirming that the benefits of imperial service were not confined to a few but were spread broadly among the elite. Nerses III returned from exile in Tao after ‘the lord of Rshtunik‘ had died and the Arab invasion had come to an end’, indicating an earlier date, perhaps 656, than is generally admitted. Constans II was determined to exploit the unexpected breathing space afforded by the outbreak of civil war or fitna across the caliphate. He sought to establish a broad network of clients across the Transcaucasus. Juansher, prince of Albania, and the princes of Siwnik‘ quickly submitted.29 In autumn 659, the emperor undertook a second progress eastwards lasting several months.30 He ventured into Media, meeting and rewarding loyal clients including Juansher, who requested and received a fragment of the 27 Seb., chs. 48–9, ed. Abgaryan, pp. 164–8; tr. and comm. Thomson and Howard-Johnston, I, pp. 135–42. 28 Seb., chs. 50–2, ed. Abgaryan, pp. 169–77; tr. and comm. Thomson and Howard-Johnston, I, pp. 143–53. 29 Seb., ch. 52, ed. Abgaryan, p. 175; tr. and comm. Thomson and Howard-Johnston, I, p. 153; HA, II.19–21, ed. Arak‘elyan, pp. 180–2; tr. Dowsett, pp. 115–17. 30 HA, II.22, ed. Arak‘elyan, pp. 183–6; tr. Dowsett, pp. 118–20.

8. armenian neighbours (600–1045)

343

Figure 22 The southern fac¸ade of the palatine church of Aruch, built by Mu‘awiya’s principal Armenian client, Gregory Mamikonean, and his wife Heline in 670. A columned palace has been excavated immediately to the south of the church, attested by the capital in the foreground

True Cross. Constans was also seeking to attract others, including Persians who wished to fight on against the Arabs. He was still in Armenia in spring 660, at Vagharshapat, where he rewarded Juansher a second time. A later text suggests that the emperor was present at the inauguration of the impressive church of Zvart‘nots‘.31 Whilst this cannot be proved, his involvement would have done much to bolster the standing of its founder Nerses III and the pro-Chalcedonian party across Armenia. Intriguingly, the terse inscription commemorating Nerses’ role is in Greek rather than Armenian.32 In the event, Constans II’s vision of a chain of clients did not survive beyond the conclusion of the fitna. As the lynchpin of the network, Hamazasp was swiftly removed and replaced by his brother Gregory Mamikonean, previously a hostage of Mu‘awiya. Juansher transferred his allegiance to the ‘king of the south [Mu‘awiya]’, when ‘the emperor of the Romans [Constans] took the dregs of his forces and hastened across sea and land to cross to the . . . distant islands of the west’.33 It seems very likely that the principal Byzantine clients had been displaced or turned by late 661 or early 662. 31 32 33

HA, III.15, ed. Arak‘elyan, p. 317; tr. Dowsett, p. 207. Greenwood (2004), inscription A.18 and p. 41. HA, II.27, ed. Arak‘elyan, p. 193; tr. Dowsett, pp. 124–5.

344

the midd le empire

in depen dence and integration under isl am (661–850) After 661, the limitations of the primary sources make it much harder to trace the interaction between Byzantium and Armenia. The conventional approach has been to treat this dearth of information as evidence for the exclusion of Byzantine influence. Armenian colophons and inscriptions together with isolated textual references collectively support an alternative view, of persistent, wide-ranging Byzantine engagement until 730 but a more limited focus thereafter, concentrated on and operated through those districts bordering imperial territory. The second sustained period of civil war across the caliphate after 680 afforded a fresh opportunity for Byzantine intervention. According to Lewond’s History, Armenia repudiated Arab sovereignty by refusing to pay tribute, probably in 682, but it is impossible to prove Byzantine influence lying behind this decision.34 A later Armenian source records how an Iberian prince, Nerses, massacred the Arab forces in Armenia during the time of Catholicos Israel I (667–77).35 The Arab blockade of Constantinople between 674 and 678 supplies an appropriate historical context for just such a diversionary campaign but a Byzantine connection remains conjectural (see also pp. 233, 372). Constantine IV (668–85) was eager to exploit contemporary disorder across the caliphate. In 685, he invaded Cilicia and threatened northern Syria, compelling the new caliph, ‘Abd al-Malik (685–705) to sue for peace on very generous terms on 7 July 685.36 This campaign may have been coordinated with the devastating Khazar raid into Armenia during which Gregory Mamikonean and Nerses were killed in battle on 18 August 685.37 According to Theophanes the Confessor, Justinian II (685–95, 705–711) ratified the truce with ‘Abd al-Malik soon after his accession although its term was extended to ten years and an additional provision was inserted, requiring the parties to share the tax revenue of Cyprus, Armenia and Iberia.38 A subsequent passage under the same year entry adds that Justinian II despatched a strat¯egos, Leontius, into Armenia. He subjugated Armenia, together with Iberia, Albania, Boukania (probably Vaspurakan) and Media, imposed taxes on those countries and remitted a large sum to Justinian. The changes to the treaty make sense when viewed in the aftermath of this raid. The revenue arrangements may reflect a more fundamental partition, of sovereignty. Gregory Mamikonean’s successor as prince of Armenia was Ashot Bagratuni, titled patrikios. Since he also brought an icon of the 34 35 36 37 38

Lew., ch. 4, ed. Ezean, p. 15; tr. Arzoumanian, p. 54. Yov., XX.18–19, ed. Emin, p. 93; tr. Maksoudian, p. 106. Theoph., ed. de Boor, I, p. 361; tr. Mango and Scott, pp. 503–4. Lew., ch. 4, ed. Ezean, pp. 15–16; tr. Arzoumanian, pp. 54–5. Theoph., ed. de Boor, I, p. 363; tr. Mango and Scott, pp. 506–7.

8. armenian neighbours (600–1045)

345

incarnation of Christ ‘from the west’ for his church at Daroynk‘, forty kilometres south of Mount Ararat, it seems likely that he was a Byzantine client.39 After Ashot’s death – confronting Arab raiders in the Araxes valley in 689 – a number of Armenian princes switched allegiance. This prompted Justinian II to travel to Armenia in person, as his grandfather Constans II had done in similar circumstances. Justinian summoned the princes to him, taking some of their sons hostage, while rewarding others: he raised Nerses Kamsarakan, the lord of Shirak, to the rank of prince of Armenia and the patrikios and exarch Varaz(tr)dat was made prince of Albania.40 He then returned to Constantinople, taking with him Catholicos Sahak III (677–703) and five bishops. Theophanes likewise reports Justinian’s visit to Armenia although he places it too early, in his second year, and wrongly associates it with the Mardaites.41 A remarkable, pro-Chalcedonian account of Armenian ecclesiastical history, which survives only in Greek, records that Sahak and his bishops accepted Chalcedon at a council convened in Constantinople in the fifth year of Justinian II, although on their return to Armenia and under pressure, they reneged.42 This revival in Byzantine fortunes occurred in the context of the second fitna. Even before his final victory over his main rival in 691, Caliph ‘Abd alMalik was turning his attention to Byzantium. Contrary to the traditional view, it seems very probable that it was ‘Abd al-Malik, not Justinian II, who broke the ten-year truce.43 The heavy Byzantine defeat in 692 at Sebastopolis occurred deep inside newly secured Byzantine territory, indicating an Arab offensive (see below, p. 384). Several Armenian clients promptly transferred allegiance but the Byzantine position did not collapse overnight. A colophon confirms that the principal Byzantine client in 689, Nerses Kamsarakan, was still alive in 696 and in contact with Constantinople.44 The region of Fourth Armenia also resisted. Although Muhammad bin Marwan, the governor of al-Jazira, campaigned there in 694/5, evidently it had not been subjugated in 701/2 when Baanes ‘Heptadaimon’ switched sides.45 Perhaps most surprisingly, in 702 Smbat Bagratuni rebelled and defeated an Arab force at Vardanakert, being rewarded with the title kouropalat¯es.46 A 39 40 41 42

Lew., ch. 5, ed. Ezean, p. 16; tr. Arzoumanian, p. 55. ST, ed. Malkhaseants‘, p. 101; French tr. Dulaurier, p. 129. Theoph., ed. de Boor, I, p. 364; tr. Mango and Scott, pp. 507–8. Narratio de rebus Armeniae, chs. 144–5, ed. Garitte, pp. 46–7 (text), pp. 350–6 (commentary); French tr. Mah´e, p. 437. 43 Proposed by James Howard-Johnston in a seminar paper, ‘Byzantium and ‘Abd al-Malik’ (11 March 2003, Oxford). 44 Mat‘evosyan (ed.), Hishatakaranner, no. 28, pp. 21–2; Socrates Scholasticus, Ecclesiastical history, pp. 9–13, 35–40. 45 Theoph., ed. de Boor, I, pp. 368, 372; tr. Mango and Scott, pp. 514, 519. 46 Lew., ch. 10, ed. Ezean, pp. 31–5; tr. Arzoumanian, pp. 64–6.

346

the midd le empire

parallel account of this uprising, but with a Kamsarakan spin, affords useful corroboration.47 The aftermath of this rebellion remains confused. Lewond maintains that Smbat withdrew into Tao and that Catholicos Sahak III negotiated a three-year peace. According to the History of the Albanians, however, military operations continued.48 Dvin fell to a joint Byzantine-Armenian force whilst the Arabs captured a fortress in Sevan only after a three-year blockade. Both sources agree that a Byzantine force then suffered a heavy defeat. Lewond adds that this occurred in Vanand in the first year of Caliph al-Walid I (705–15). The Byzantine troops fled and the Armenian rebels suffered severe reprisals, with 800 men in Nakhchawan and 400 in Khram being imprisoned in churches and then burnt alive. Ominously, the lord of Shirak, Nerses Kamsarakan, was summoned to Syria in 705; his fate is not recorded. Smbat kouropalat¯es escaped into Byzantine territory and was settled in the city of Phasis in Lazica. This sequence of events – a rebellion by Armenian princes, contact with Emperor Tiberius II Apsimar (698–705), the despatch of Byzantine forces, a successful counter-offensive by Muhammad bin Marwan followed by the burning alive of Armenian princes – is corroborated by Theophanes.49 The only significant difference is chronological. Theophanes records this sequence of events under one year, AM 6195 (702/3) but it seems more likely that they were spread across several years (702–5). Aside from the failed attempt at union in the time of Justinian II outlined above, relations between the churches after 661 are almost entirely obscure. In 719, however, Catholicos John III (717–27) stated unequivocally that the six catholicoi after Komitas (between 628 and 705) were all Chalcedonian, exempting only his immediate predecessor Elias (703–17) from criticism.50 As outlined previously, Ezra, Nerses III and Sahak III all engaged in discussions with the imperial church but none of their correspondence or other writings survives. Indeed the only extant letter between 628 and 705 is a draft Armenian ‘Defence’ of the monophysite position, prepared in 649 for despatch to Constans II.51 Arguably, no records or letters associated with these catholicoi survive precisely because of their confessional perspective. An exchange between Patriarch Germanos I (715–30) and Catholicos John III from the 720s does survive, defining and defending their respective positions in great detail.52 Conceivably this correspondence marks the 47 48 49 50 51

Yov., XXI.1–5, ed. Emin, pp. 95–8; tr. Maksoudian, pp. 107–9. HA, III.16, ed. Arak‘elyan, pp. 317–18; tr. Dowsett, pp. 207–8. Theoph., ed. de Boor, I, p. 372; tr. Mango and Scott, pp. 519–20. Book of letters, ed. Izmireants‘, pp. 221–2; ed. Pogharean, pp. 475–6. See now Greenwood (2008). Seb., ch. 46, ed. Abgaryan, pp. 148–61; tr. and comm. Thomson and Howard-Johnston, I, pp. 114– 32; Thomson (1998). 52 Book of letters, ed. Izmireants‘, pp. 358–95; ed. Pogharean, pp. 414–66.

8. armenian neighbours (600–1045)

347

final breach between the churches and was preserved because it articulated the differences. Confessional tensions at the highest level need not have deterred other contacts. Colophons reveal that four patristic works were translated into Armenian in Constantinople between 713 and 717 by David hypatos and Stephen of Siwnik‘.53 After 730, Byzantine influence persisted but on a more limited scale. An inscription on a tombstone located in a crypt at Nakhchawan in Shirak commemorates ‘the blessed lord Artawazd Kamsarakan apo hypat¯on patrikios and prince of Armenia, son of Hrahat patrikios lord of Shirak and Asharunik‘’.54 Artawazd was the grandson of Nerses Kamsarakan mentioned previously. Evidently Byzantine titles continued to be awarded during the eighth century to Armenian princes. Artawazd does not feature in any other source, which is surprising given his rank of ‘prince of Armenia’. His omission is hard to explain unless one views him as a second, rival prince of Armenia and client of Byzantium. When the third fitna erupted, two groups of Armenian princes may once again be discerned. One party, under Ashot Bagratuni, remained loyal to Caliph Marwan II (744–50); the other under Gregory Mamikonean, looked to Constantine V (741–75). Having taken refuge in Tao, ‘they relied upon the forces of the king of the Greeks, who were in the regions of Pontos, because there was a treaty of peace between them, at the command of the emperor Constantine’.55 After blinding Ashot Bagratuni, perhaps in 748, Gregory went to Theodosioupolis and broadcast news of his victory. Evidently Theodosioupolis was under his, or Constantine’s, control and he was attempting to attract further support. His success or otherwise in this initiative is not recorded by Lewond, who simply notes that he died in agony at an unspecified date and was replaced for a short time by his brother.56 Whether Lewond’s hostility stems from a political (anti-Mamikonean) or confessional (anti-Chalcedonian) perspective is unclear. Again this temporary Byzantine revival in Armenia was halted by the resolution of the strife within the caliphate. In 754, Constantine V transferred the population of Theodosioupolis to Thrace. Lewond adds that many from the surrounding districts also left and ‘placed themselves on the side of the pious king’, a rare favourable view of Constantine V.57 This transfer may represent a tactical withdrawal at the end of a series of initiatives in Armenia rather than the original goal. Armenian princes did not risk rebellion against the dominant, controlling power without support, or expressions of support, from a rival power 53 54 55 56 57

Mat‘evosyan (ed.), Hishatakaranner, nos. 31–4, pp. 24–6. Greenwood (2004), inscription A.13 and pp. 75–6. Lew., ch. 26, ed. Ezean, p. 123; tr. Arzoumanian, p. 120. Lew., ch. 26, ed. Ezean, pp. 123–4; tr. Arzoumanian, pp. 120–1. Lew., ch. 29, ed. Ezean, p. 129; tr. Arzoumanian, p. 124.

348

the midd le empire

other than in exceptional circumstances. At first sight, the complicated series of rebellions across Armenia in the 770s fall into that category. At no stage do the narrative sources indicate any Byzantine involvement.58 Two of the rebel leaders, Artawazd and Mushegh Mamikonean, are said to have begun their uprisings by killing local Arab tax-collectors. New administrative arrangements and fiscal burdens at district level may have precipitated their actions. On the other hand, Artawazd moved into Iberia and later reappears as strat¯egos t¯on Anatolik¯on whilst Mushegh’s rebellion apparently took the form of a prolonged, and ultimately unsuccessful, siege of Theodosioupolis. This strategy is hard to fathom unless one accepts that Byzantine support was anticipated. No Byzantine campaign is recorded but it may have been planned; in 777 a large Byzantine army, under Armenian commanders, attacked Germanikeia and devastated the surrounding region.59 For the following five decades, there is very little evidence for Byzantine involvement in Armenia. In 788 as many as 12,000 people under the leadership of Shapuh Amatuni, his son and other Armenian nobles were granted refuge within the empire by ‘the emperor Constantine’. Lewond portrays this as a reaction to hardships inflicted by the caliph and his representatives, specifically the seizure of land.60 It is in the last quarter of the eighth century that several quasi-independent Arabic emirates emerged, ruling districts previously under Armenian control.61 At the same time, members of the Bagratuni princely house exploited their status as preferred Abbasid clients to secure a dominant position. After 775, Byzantine attention was concentrated on potential clients in those districts of Iberia which abutted imperial territory. Ashot Bagratuni, established in neighbouring Klarjet‘i, was appointed kouropalat¯es before 826.62 Byzantine strategy towards Armenia came to operate on and through the remote district of Sper which bordered the theme of Chaldia. The first ninth-century Armenian prince known to have been accorded an imperial title was another Ashot Bagratuni, prince of Sper; he was appointed patrikios and apo hypat¯on by Theophilos (829–42).63 Intriguingly, his appointment is recorded in the context of Byzantine operations against Theodosioupolis, Basean and Vanand, all to the south and east of Sper. Although these operations have been compressed into a single campaign and linked to a major Byzantine offensive against Sozopetra, Melitene and Fourth Armenia undertaken in 837, they could equally comprise separate campaigns spread over a number of years.64 58 59 60 61 62 63 64

Lew., ch. 34, ed. Ezean, pp. 137–52; tr. Arzoumanian, pp. 129–38. Theoph., ed. de Boor, I, p. 451; tr. Mango and Scott, p. 623. Lew., ch. 42, ed. Ezean, pp. 168–9; tr. Arzoumanian, p. 149. Ter-Ghewondyan (1976). Martin-Hisard (2001); Martin-Hisard (2002); Abashidze and Rapp (2004). ST, ed. Malkhaseants‘, p. 144; French tr. Dulaurier, p. 171. Laurent (1980), pp. 249–52.

8. armenian neighbours (600–1045)

349

This targeting of Theodosioupolis and its surrounding districts mirrors the pattern of Byzantine offensives outlined previously, whilst the Khurramite rebellion under Babek afforded a suitable opportunity (see below, p. 390). Caliph al-Mu‘tasim (833–42) responded swiftly to this Byzantine threat. In 838, his forces inflicted a heavy defeat upon Theophilos at Dazimon and captured Amorion. Genesios reports that Armenian forces under the ‘Vasparakanites’ (presumably the leading Artsruni prince) and the prince of princes (probably Bagarat Bagratuni, prince of Taron) participated in these campaigns.65 This represents a rare instance of active service by Armenian forces against Byzantium. It illustrates how closely the leading Armenian princes now identified with caliphal interests and the degree to which Byzantine influence over them had waned. arm enia resurgent, byz antium expectant (850–1045) In 850, Caliph al-Mutawakkil (847–61) sent Abu Sa‘id Muhammad bin Yusuf to Armenia to collect the so-called ‘royal taxes’. Although these were apparently paid, relations between representatives of Abu Sa‘id and the principal Artsruni and Bagratuni princes deteriorated rapidly and all parties took up arms.66 In 852, Bugha al-Kabir embarked on a series of ruthless campaigns to quash Armenian resistance. The principal noble families were targeted and many leading members were either killed or captured and despatched to the Abbasid capital, Samarra. A few, however, escaped. In 853 or 854, Gurgen Artsruni sought refuge with Gregory Bagratuni, prince of Sper.67 Gregory had recently captured an unidentified Byzantine fortress called Aramaneak. When the Byzantine ‘general of the east’ – an Armenian rendering of strat¯egos t¯on Anatolik¯on – attempted to recover Aramaneak, both princes opposed him. He was so impressed by Gurgen’s courage that he informed Michael III (842–67), who invited Gurgen to Constantinople. Gurgen declined but he did persuade Gregory to return the fortress and also fought against Bugha’s troops when they attacked ‘the Greek forces in their fortresses’. This is the first recorded contact between an Artsruni prince and Byzantium for many generations. Significantly it took place in Sper while Armenia was in turmoil. Nor was this the limit of Byzantine ambitions. In 858, after Gurgen had returned to Vaspurakan, he was confronted by Gregory Artsruni at the head 65 Gen., III.13, ed. Lesm¨ uller-Werner and Thurn, p. 47; tr. Kaldellis, pp. 62–3. The prince of princes could have been Bagarat Bagratuni’s brother, Smbat Abu’l ‘Abbas, at this time. 66 TA, ed. Patkanean, pp. 106–212; tr. Thomson, pp. 173–275; Yov., XXV–XXVII, ed. Emin, pp. 113– 35; tr. Maksoudian, pp. 116–26. 67 TA, ed. Patkanean, pp. 194–5; tr. Thomson, pp. 258–9.

A

Z

P

A

IA

S E A

S

H

IA

B L A C K

BK

C

A

N

A SE

A

L

D

I

SPER

K Sebasteia (Sivas)

O L O N E I A DEGIK

Symposion Larissa Tzamandos

Lykandos

hrates Eup

Khtskawnk

Manzikert

RO

Mush Bitlis

Narek

MOKK

L. Sevan

Vagharshapat Dvin A r axes

SI W

Archesh

APAHUNIK

N

Aparank

Perkri L. Van

Artzike Aght amar

U S P V A

R

A

K

NIK

Kapan

A

s xe ra

Her

AZERBAIJAN

tes hra Eup

Tarsus

Ani

Ult is

BASEAN

Oulnoutin TA

Shirakawan

Kars

TA O

K A R Theodosioupolis IN NALIS Q A A N YSID MA E M I R ATE

anias Ars

Melitene

Kur

Ardanuji

A

A

Trebizond

H

N

C

Border in 850 Frontier in 1064 0 0

Map 17 Armenia and imperial expansion 850–1045

100

200 100

300 200 miles

400 km

8. armenian neighbours (600–1045)

351

of Abkhazian and Iberian troops.68 Having failed to attract Gurgen, it seems that Byzantium had switched its attention to a second displaced Artsruni prince and backed his bid to seize Vaspurakan. Although Gregory was unsuccessful, the imperial administration evidently had a strategic vision which extended far beyond those districts adjacent to imperial territory. Therefore when Photios became patriarch of Constantinople in 858 and re-established contact with the Armenian church, he did so in the context of renewed Byzantine engagement across Armenia. The sequence and chronology of the letters exchanged between Photios (858–67, 877–86) and several Armenian correspondents, including Catholicos Zacharias (855–76), remains contentious, as does the authenticity of one of Photios’ letters to Zacharias.69 Collectively the correspondence attests Photios’ determination to heal the long-standing confessional breach. The council of Shirakawan, convened in 862 by Zacharias, represents the first fruits of Photios’ initiative.70 Canons 13 and 14 respectively condemn two groups: firstly, convinced monophysites who masquerade as Chalcedonians, for personal gain; and secondly, those who have apparently accepted Chalcedon, but still cannot help themselves from adopting the traditional Armenian charge – that the council’s ruling on the unity of Christ’s person was, in fact, Nestorian. As Jean-Pierre Mah´e puts it, ‘le cas pr´evu e´tait la conversion de monophysites au dyophysisme et non l’inverse.’71 The aftermath of this council is unknown but just before his deposition in 867, Photios observed in an encyclical letter that ‘today, the covenant of the Armenians worships purely and in orthodox fashion the Christian faith.’72 By the time Photios was reappointed patriarch on 26 October 877, conditions had altered dramatically. His ‘spiritual brother’ Zacharias had died and the prince of princes, Ashot Bagratuni, was now entrenched as the preeminent client of the caliph and wary of Byzantine initiatives. Although Photios made considerable efforts to engage with Ashot, sending conciliatory letters addressed to ‘your most eminent piety’, despatching a relic of the True Cross and even reporting that relics of the three most revered Armenian saints had been found in Constantinople, he was unable to recover lost ground.73 The final letters chart the breakdown in discussions with Ashot and his spiritual advisers. Both sides reverted to their traditional positions, defining and rebutting in meticulous detail the doctrinal errors of the other. Although these letters are not dated, the heavy defeats suffered 68 69 70 71 72 73

TA, ed. Patkanean, pp. 198–9; tr. Thomson, pp. 262–3. See Dorfmann-Lazarev (2004) and Greenwood (2006a) for opposing views. Akinean and Ter-Pawghosean (1968a), cols. 261–6; Maksoudian (1988–9). Mah´e (1993), p. 495. Phot., no. 2, ed. Laourdas and Westerink, I, p. 41, lines 43–4. On the letter, see Akinean and Ter-Pawghosean (1968b), col. 439. On the True Cross, see Mat‘evosyan (ed.), Hishatakaranner, no. 50, pp. 40–3. On the relics, see van Esbroeck (1971b), pp. 401–4; Greenwood (2006b).

352

the midd le empire

by the Byzantine forces at Melitene in 882 and Tarsus in 883 provide a likely terminus ante quem (see above, p. 297). Around 925, Patriarch Nicholas I Mystikos (901–7, 912–925) reflected that Photios had pursued ecclesiastical reconciliation with Armenia without success, implying no correspondence on this subject between the churches in the intervening forty years.74 Frustratingly there is no evidence for contacts with the feuding members of the Artsruni house in Vaspurakan after 858. A little more is known about relations with the extended Bagratuni family. Photios acknowledged Ashot Bagratuni’s concern for his recent travails and joy at his restoration in 877, suggesting contact before he had regained the patriarchate. Moreover Ashot learned about the discovery of the Armenian relics during an embassy from Basil I (867–86) in 878. In spite of these initiatives, it was not Ashot, prince of princes, who was appointed kouropalat¯es but his cousin Ashot, prince of Taron, at an unspecified date before 878.75 In a final letter, Photios described the Taronites who inhabited Fourth Armenia as orthodox.76 It may well be the case that Ashot was rewarded for his orthodoxy. Alternatively the relative proximity of Taron to imperial territory may have influenced the appointment. Either way, Byzantium developed ties simultaneously with several Bagratuni princes. Three decades of ambitious military and ecclesiastical initiatives beyond the eastern frontier, lasting from 854 to 883, were followed by an era of consolidation. Little-known figures, controlling districts much closer to imperial territory, were induced to acknowledge imperial sovereignty. After the accession of Leo VI (886–912), Manuel, lord of Degik, was given a written guarantee of immunity, taken to Constantinople and appointed pr¯otospatharios.77 At the same time, other Armenians were appointed to separate commands along the frontier, usually organised around individual fortresses, and encouraged to expand into adjacent districts. Thus Melias (or Mleh in Armenian) was first appointed turmarch of Euphrateia and Trypia.78 In 908, he captured the kastron of Lykandos and became its kleisouriarch. He then advanced to Tzamandos and constructed a kastron. Later he annexed Symposion. In 915 he was appointed strat¯egos of the newly-created theme of Lykandos. Melias’ lordship thereby gained an administrative and legal identity within the Byzantine state. The network of themes created piecemeal along the eastern frontier reflected the local achievements of men such as Melias. Inevitably there were losers as well as winners. For every Melias, there were figures like Ismael ‘the Armenian’, kleisouriarch of Symposion, who was killed by raiders from Melitene. It would be wrong, however, to assume that this time of consolidation on the frontier coincided with any break in relations with Armenian princes 74 76 77

75 TA, ed. Patkanean, pp. 218–24; tr. Thomson, pp. 282–8. NM, no. 139, pp. 450–1. Phot., no. 284, ed. Laourdas and Westerink, III, p. 94, lines 3194–6. 78 DAI, ch. 50, pp. 238–41. DAI, ch. 50, pp. 238–9.

8. armenian neighbours (600–1045)

353

beyond the frontier. Again, several isolated references indicate continued contact with key Bagratuni princes. After Ashot, prince of princes, had been crowned king on 26 August 884 by Catholicos George II (877–97) using a crown brought from the caliph, Basil I acknowledged him as his ‘beloved son’.79 Leo VI addressed Ashot I’s son Smbat I Bagratuni (‘the Martyr’) (c. 890–913) in the same way after he succeeded his father in about 890, sending him ‘fine weapons and ornaments and clothing embroidered with gold and cups and chalices and golden belts studded with gems’.80 In 892 Smbat captured the city of Dvin and sent its commanders to the emperor in chains, although it seems that this campaign was his own initiative rather than a joint operation.81 When the prince of Taron, Krikorikios (‘little Gregory’), captured his two cousins in battle in the mid-890s, Smbat wrote to Leo VI, interceding for their release.82 Evidently he believed that the emperor could influence the actions of Krikorikios and in this he was proved right. This incident is reported in chapter forty-three of Constantine VII Porphyrogenitus’ De administrando imperio, whose importance has long been recognised.83 It describes how several members of the princely family of Taron across two generations were drawn into the political and cultural orbit of Byzantium; the titles, marriages and properties variously granted to them; and the consequences of such engagement for the very existence of the principality. The chapter ends with the patrikios Tornikios offering to cede his territories to the emperor, Romanos I Lekapenos (920–44). Although Tornikios died before completing this transfer of sovereignty, he left a will – a Byzantine rather than an Armenian custom – devising the same. His cousins complained to Romanos, who agreed to exchange his inheritance for Oulnoutin, a strategically placed kastron in the west of Taron. This chapter reveals much else besides, not least the collection and retention of information gained during diplomatic exchanges; a legal dispute between different members of an Armenian family over title to their property in Constantinople, encouraged if not inspired by the imperial authorities; and complaints to Romanos from three other Armenian princes over payments made to Krikorikios. It is worth remembering, however, that this chapter affords a partial view of diplomatic relations with one particular princely house and the territorial rights conceded to Romanos. The following three chapters trace imperial claims to the Qaysid emirate of Manzikert, to specific districts and kastra around Theodosioupolis and to the kastron of Ardanuji in Klarjet‘i; they do not supply an exhaustive account of relations with every Armenian princely house. 79 80 81 82

Yov., XXIX.13, ed. Emin, p. 140; tr. Maksoudian, pp. 129, 272–3. Yov., XXXI.2, ed. Emin, p. 158; tr. Maksoudian, p. 138. Yov., XXXI.9-13, ed. Emin, pp. 160–1; tr. Maksoudian, pp. 138–9. 83 Shepard (2001). DAI, ch. 43, pp. 188–91.

354

the midd le empire

A better impression of the range of Armenian contacts is supplied by the protocols for imperial correspondence preserved in the Book of ceremonies.84 The list, which has been dated to between 918 and 922, identifies not only the prince of princes of Greater Armenia and the prince of Vaspurakan, ‘who now is honoured as prince of princes’, but also seven other Armenian princes. Yet arguably even this list does not do justice to the range of potential correspondents. It identifies only the leading representative of each princely house, but, as we have seen in respect of Taron above, several members of the same house could be in direct relationship with the emperor. In addition to the activities of Armenian commanders on the frontier, and diplomatic links, Byzantium could also intervene directly using its military forces. A Byzantine force attacked Theodosioupolis as early as 895, whilst in 915 Ashot II Bagratuni (‘the Iron’) (914–c.928), son of King Smbat I ‘the Martyr’, returned from exile in Constantinople at the head of a Byzantine army, intent on re-establishing himself in the districts previously held by his father.85 In the event, neither campaign was followed up but such apparently isolated actions need to be placed in the context of heavy Byzantine defeats in the Balkans, at Bulgarophygon in 896 and Anchialos in 917. Only after peace had been achieved in 927 were Byzantine forces redirected to the east.86 Thereafter key fortresses under Arab control were systematically targeted. Melitene capitulated in 934 and Theodosioupolis in 949, both after years of persistent pressure and blockade. At the same time, every effort was made to ensure that neighbouring Armenian or Iberian princes were not antagonised. Conceivably this strategy was devised after two early reverses. In 922 when a Byzantine army attacked Dvin, it was opposed by the same Ashot II ‘the Iron’ who had benefited from imperial support seven years before.87 Only in exceptional circumstances did an Armenian prince fight against imperial troops. Arguably his own interests had been prejudiced by this advance. Secondly, an attempt was made in 923 to seize control of Ardanuji, located beyond the frontier in Klarjet‘i, by infiltrating troops under the guise of a visiting diplomatic mission.88 Although this kastron had been offered to Romanos I Lekapenos by its prince, the threat by neighbouring Iberian princes to make common cause with local Arabs precipitated a rapid withdrawal. Frustratingly it is at this very moment, with Byzantium poised to utilise all three approaches – administrative, diplomatic and military – that our source-material peters out. There is sufficient evidence, however, to confirm 84 85

DC, II.48, ed. Reiske, I, p. 687; Martin-Hisard (2000). TA, ed. Patkanean, p. 231; tr. Thomson, pp. 294–5; Yov., LVI.1–4, ed. Emin, p. 292–3; tr. Maksoudian, pp. 201–2. 86 Whittow (1996a), pp. 316–17. See also below, p. 509. 87 ST, ed. Malkhaseants‘, p. 170; French tr. Macler, pp. 24–5. 88 DAI, ch. 46, pp. 214–23.

8. armenian neighbours (600–1045)

355

that the eclipse in Bagratuni power – epitomised by Smbat I’s murder in 913 and perpetuated by the long confrontation between Ashot II ‘the Iron’ and Smbat’s nephew, also called Ashot – forced Byzantium to reappraise its position and recognise Gagik Artsruni as the pre-eminent figure.89 Shortly after the death of Catholicos John V in 925, Gagik I Artsruni (908–c.943) wrote to Nicholas I Mystikos, seeking to secure the succession for his preferred candidate through a ceremony in Constantinople. Nicholas’ reply, addressed to Gagik ‘prince of princes’, was uncompromising in its defence of orthodox belief, maintaining that Gagik’s candidate would need to be instructed in sound doctrine and ecclesiastical government.90 At the same time Nicholas noted the ‘confession of friendship’ by which Gagik was ‘attached to our Christ-loving emperor and to our most holy church of God’; his own orthodoxy was not at issue. This relationship had practical implications. According to Ibn al-Athir, in 931 the lord of Vaspurakan, Ibn al-Dayrani (the Arabic version of [Gagik] son of Derenik) proposed and participated in a joint campaign with Byzantine forces against the Qaysid amirs.91 During the Artsruni ascendancy, Byzantium retained ties with other noble houses. The leading Bagratuni after 929, Abas, held the title of magistros, reflecting both the continuing demise of his family’s fortunes and a closer link to Byzantium than many commentators have credited.92 A letter written in about 933 by Theodore Daphnopates to the bishop of Siwnik‘, reprimanding him for teaching monophysite doctrine, reveals the spread of Byzantine interest eastwards.93 Yet it is clear that Byzantium did not enjoy a monopoly of influence across Armenia. Mindful of recent Sajid intervention and devastation, Armenian princes remained wary of Muslim powers to the east and south, however ephemeral these proved to be. Thus when Saif al-Dawla, the future Hamdanid amir of Aleppo, marched north through the Bitlis pass to Lake Van in 940, several Armenian princes responded to his summons and submitted, including one of Gagik’s sons and Ashot, son of Krikorikios, prince of Taron.94 Although the sources contradict one another over the course of his campaign and the identity of the Artsruni client, they confirm that Armenian princes were prepared to recognise the sovereignty of an enemy of Byzantium if they believed this would serve their own interests. Ibn Hawqal offers a second example, listing those Armenian princes who paid tribute to the Sallarid ruler of 89 90

DC, II.48, ed. Reiske, I, p. 687. NM, no. 135, pp. 446–51. A second, unrelated letter from Gagik to the patriarch and the emperor survives: Book of letters, ed. Izmireants‘, pp. 295–301; French tr. Garso¨ıan, pp. 540–9. 91 Ibn al-Athir, al-Kamil, in Vasil., p. 153. 92 DAI, ch. 44, pp. 198–9. 93 Theodore Daphnopates, Correspondance, ed. and French tr. Darrouz` es and Westerink, no. 10, pp. 108–41. Intriguingly the original letter from the bishop was in Armenian. 94 DAI: Comm, p. 169; Whittow (1996a), pp. 319–20.

356

the midd le empire

Figure 23 The southern fac¸ade of the palatine church of Aght‘amar, constructed by the architect Manuel for King Gagik I Artsruni of Vaspurakan between 915 and 921 on an island in Lake Van. The external walls are lined with figural and decorative sculpture, inspired by biblical and Artsruni history; frescoes, now badly damaged, cover the interior

Azerbaijan, Marzuban, in 955 and the considerable amounts due.95 It is unclear whether such sums were actually remitted or whether this liability lapsed after Marzuban’s death in 957, but the principle, however shortlived, seems established. By contrast, there is no evidence that Byzantium imposed any financial burdens upon its Armenian clients. In the event, Saif al-Dawla did not develop a bloc of Armenian support. His victories over Byzantine forces provoked a series of counter-offensives. The successes enjoyed by Nikephoros Phokas after 955 drew Byzantium southwards, into Cilicia and northern Syria, away from active military engagement in Armenia (see below, p. 517). As observed above, campaigns across Armenia had been directed against those emirates and their bases which historically had posed the greatest threat. This strategy concluded with the capture of Theodosioupolis in 949. Although the military focus shifted south, it seems that the nexus of relationships with Armenian princes and clerics continued to be maintained and developed. Admittedly there is very little evidence of Byzantine involvement in Armenia between 935 and 976, but it is during this period that significant confessional tensions emerged within the Armenian church. Catholicos Anania I (943–67) reasserted 95 Ibn Hawqal, Surat, ed. Kramers, II, pp. 354–5; French tr. Kramers and Wiet, II, pp. 347–8; Minorsky (1953), pp. 519–20.

8. armenian neighbours (600–1045)

357

his authority over the dissident see of Siwnik‘ at the council of Kapan in 958, but was succeeded by Vahan I of Siwnik‘ who ‘wished to develop friendship and agreement with Chalcedonians’.96 Vahan I was deposed in 968 by the council of Ani and sought refuge with the king of Vaspurakan, Apusahl Hamazasp (953/8–72). Byzantine influence in these events may be inferred. A colophon records the visit of a priest named Pantaleon to Constantinople in January 966 at the command of Apusahl Hamazasp, ‘king of kings of the house of Armenia’.97 The colophon adds that this occurred in the time of Nikephoros, ‘emperor of the Greeks, valiant and virtuous, victorious in battles against the heathens’. Pantaleon returned safely ‘through the power of the Holy Cross and the prayers of the Holy Apostles and the grace of both our kings, Nikephoros and Hamazasp’. Not only was Apusahl in direct contact with Constantinople; in the eyes of the author, Nikephoros II Phokas (963–9) enjoyed joint sovereignty with the Artsruni king. Nor is this the only evidence of continued Byzantine engagement. Whilst the four chapters devoted to Armenian and Iberian affairs in the De administrando imperio largely recount past episodes rather than present circumstances, their very inclusion is significant. In 966 or 967, after the death of its prince, Ashot, Taron came under Byzantine control. Two years later, Bardas Phokas, nephew of Nikephoros and doux of Chaldia and Koloneia, advanced to Manzikert and destroyed its walls.98 Thus within fifteen years of the compilation of this work, Taron had been incorporated into the empire and the potential threat posed by Manzikert neutralised. In 974, John I Tzimiskes (969–76) travelled to Armenia. According to our only source, the twelfth-century Armenian historian Matthew of Edessa, King Ashot III Bagratuni (‘the Merciful’) (953–77) assembled all the leaders of the countries of the east, including Sennacherim, lord of Vaspurakan, and their forces.99 Having opened lines of communication with Ashot, the emperor advanced to Mush in Taron and camped outside the fortress of Aytsik‘. His forces came under overnight attack, although the circumstances and outcome are obscure. At some point thereafter, Tzimiskes was handed a letter, apparently from Catholicos Vahan I. This detail is hard to interpret, given Vahan’s deposition six years before. The two leaders then made a treaty whereby Ashot III ‘the Merciful’ supplied 10,000 troops in return for notable gifts. Several elements in this account – specifically the leadership role accorded to Ashot, the skirmishes at Aytsik‘ and Vahan’s letter – may reflect a Bagratuni spin or a conflation of different episodes. Scholars have generally interpreted Ashot’s attendance upon the emperor at the head of a large army as a defensive precaution. Yet his conduct also befits a loyal client, 96 97 98 99

ST, ed. Malkhaseants‘, p. 181; French tr. Macler, p. 41. Hovsep’yan (1951), no. 51, cols. 117–20. ST, ed. Malkhaseants‘, p. 183; French tr. Macler, p. 44. ME, I.17, ed. Melik‘-Adamean and Ter-Mik‘ayelean, pp. 22–4; tr. Dostourian, pp. 27–8.

358

the midd le empire

responding to an imperial summons and supplying military assistance at a designated location. Tzimiskes’ subsequent letter to Ashot ‘shahanshah [originally a Persian royal title, ‘king of kings’] of Great Armenia and my spiritual son’, describing his victorious campaign of 975 into Syria and Lebanon, then becomes apposite.100 The degree to which Armenian princes had been drawn into the orbit of Byzantium can be seen through their involvement in the rebellions which erupted against Basil II (976–1025) and Constantine VIII (1025– 8) after 976. Bardas Skleros had the support of Gregory and Bagarat, sons of Ashot, prince of Taron, and Zap‘ranik, prince of Mokk‘, whilst Bardas Phokas exploited his relationship with the Iberian prince David of Tao – forged while he was doux of neighbouring Chaldia – to win him to Basil II’s cause.101 In addition to the title of kouropalat¯es, David received substantial territorial concessions, including the districts of Karin and Apahunik‘, recently prised from Arab control. The personal ties with Bardas Phokas which caused David to fight for Basil II later prompted him to join Phokas when he rebelled against Basil in 987. All three survived these confrontations. Gregory Taronites, doux of Thessaloniki and magistros, fought against Samuel of Bulgaria (987/988–1014) after 991 and was killed in 995.102 Zap‘ranik manglabit¯es was charged in 983 by Basil II and Constantine with transporting a relic of the True Cross from Constantinople to the monastery of Aparank‘.103 David kouropalat¯es retained possession of all the lands granted to him previously although these now reverted to the emperor after his death.104 It is striking, however, that neither Gregory nor Zap‘ranik remained in their ancestral districts and that David continued to exercise authority only in the knowledge of inevitable imperial intervention. Contemporary relations between the churches reveal a similar pattern of increased engagement. As Byzantium pushed eastwards, and significant numbers of Armenians came, or were transferred, within its borders, the respective hierarchies increasingly overlapped. An exchange between Metropolitan Theodore of Melitene and Samuel of Kamrjadzor, responding at the behest of Catholicos Khach‘ik I (973–92), confirms that confessional tensions were developing at a local level.105 Another exchange, between Khach‘ik I and the metropolitan of Sebasteia, occurred in 989.106 Complaints of oppression and torture in Sebasteia were combined with 100 101 102 103

ME, I.19-20, ed. Melik‘-Adamean and Ter-Mik‘ayelean, pp. 24–32; tr. Dostourian, pp. 29–33. ST, ed. Malkhaseants‘, pp. 191–2; French tr. Macler, pp. 56, 59–60. Skyl., ed. Thurn, p. 341; French tr. Flusin and Cheynet, p. 285, n. 121. Gregory of Narek, Discourses, ed. Awetik‘ean, pp. 9–36; Mah´e (1991); Gregory of Narek, Book of lamentations; tr. Mah´e and Mah´e, pp. 78–83. 104 Skyl., ed. Thurn, p. 339; French tr. Flusin and Cheynet, p. 283, n. 108; ST, ed. Malkhaseants‘, p. 275; French tr. Macler, p. 162. 105 Book of letters, ed. Izmireants‘, pp. 302–22; French tr. Garso¨ıan, pp. 550–79. 106 ST, ed. Malkhaseants‘, pp. 201–44; French tr. Macler, pp. 76–123.

8. armenian neighbours (600–1045)

359

observations that the Armenian bishops of Sebasteia and Larissa, and other priests, had removed themselves from the Armenian church and accepted Chalcedon. Yet neither of these sees had previously been described or treated as Armenian. By contrast eleven new suffragan bishops under the metropolitan of Trebizond had been created by the 970s, including those of Mananalis, Oulnoutin and Basean, confirming a simultaneous extension eastwards by the imperial church.107 This fluidity was recognised by contemporaries. Sargis was appointed catholicos of Armenia in 992 at a council convened by King Gagik I Bagratuni (‘the Great’) (989–c. 1017) at which there were bishops ‘from this country of Armenia and from the side of the Greeks’.108 Little is known about the contemporary actions or attitudes of leading members of the Bagratuni and Artsruni houses. Significantly, however, the deposit of the relic of the True Cross at Aparank‘ during Easter 983 was attended by the three Artsruni brothers then ruling Vaspurakan, Ashot-Sahak, Gurgen-Khach‘ik and Sennacherim-John. Their presence at this isolated, mountainous site so early in the year for the arrival of an imperial donation implies respect for – and close relations with – Byzantium. Gregory of Narek asserted in his description of the ceremony that the divine will is clear: it is that the empire of the Romans, spread out like the sky across the vast surface of the whole world, will gather in its ample bosom innumerable multitudes, as a single flock in a single place, a single synod and a single church, the one bride in the bridal chamber, the one beloved in the single dwelling place . . . the one spouse under the one tent of the Covenant.109

His support for Basil II seems unequivocal. David kouropalat¯es of Tao died on Easter Sunday, 31 March 1000. Two sources allege that he was poisoned when receiving the eucharist, although one adds that he survived this attempt and was smothered instead.110 Arguably this reflects a confessional spin, since David ‘died’ in a spiritual sense when taking wine mixed with water in the eucharist. Basil II was quick to take advantage.111 He marched north from Tarsus, meeting and rewarding several prominent princes, including Sennacherim-John of Vaspurakan. He then moved east to the plain of Vagharshapat, but Gagik I ‘the Great’ failed to attend, ‘reckoning it a diminution’, and Basil thereupon returned via Ult‘is in Tao and Theodosioupolis to Constantinople. Gagik 107 108 109

NE, no. 9, pp. 296–306 (text). ST, ed. Malkhaseants‘, p. 259; French tr. Macler, p. 144. Gregory of Narek, Discourses, ed. Awetik‘ean, p. 11. A colophon of Gregory expresses identical sentiments: Gregory of Narek, Book of lamentations; tr. Mah´e and Mah´e, pp. 777–8. 110 Arist., ed. Yuzbashian, pp. 22–3; French tr. Canard and Berb´ erian, pp. 2–6; ME, I.33, ed. Melik‘Adamean and Ter-Mik‘ayelean, p. 44; tr. Dostourian, p. 39. 111 ST, ed. Malkhaseants‘, pp. 275–8; French tr. Macler, pp. 162–5; Arist., ed. Yuzbashian, pp. 23–4; French tr. Canard and Berb´erian, pp. 3–6.

360

the midd le empire

may have viewed David’s death as an opportunity to revive Bagratuni hegemony, an ambition that submission to Basil II would have compromised, if not thwarted; other princes had been compelled to lead or contribute large numbers of troops for operations against Bulgaria. Alternatively he may have been influenced by ecclesiastical opinion; both Catholicos Khach‘ik and his successor Sargis I (992–1018) were steadfast in their opposition to the imperial church. Whatever the cause, Basil II was prepared to consolidate his gains and bide his time. After more than a century of regular dealings with Armenian princely houses, Byzantium was keenly aware that times of political flux after the death of the leading prince offered the best opportunity for direct intervention, as the rival claimants looked for outside support. Basil could afford to wait. When George I (1014–27) succeeded his father Bagrat III as king of Georgia in 1014, Basil II asserted his claim to certain districts previously ceded to David of Tao and then Bagrat.112 George rejected this claim and resisted an attempt to occupy them. Basil waited until Bulgaria had been pacified. In 1021 he travelled east, expecting to receive George’s submission; but George did not attend. Further negotiations failed and both sides took up arms. Although there is no evidence that any Armenian princes joined George in defying Basil II, he had arbitrated between John-Smbat III and Ashot IV Bagratuni (‘the Brave’) following the death of their father, Gagik I ‘the Great’, probably in 1017, and had intervened in their subsequent confrontation.113 Arguably John-Smbat now saw an opportunity to gain imperial backing. In January 1022, Catholicos Peter I (1019–58) attended upon Basil II at his winter quarters in Trebizond, bringing with him a will from John-Smbat III appointing him as his heir.114 This underpinned the Byzantine claim to Ani after his death in 1041. John-Smbat and Ashot were therefore pulled back into the imperial orbit indirectly through the conduct of King George I of Georgia. SennacherimJohn Artsruni, however, exchanged his ancestral lands of Vaspurakan for territories in Cappadocia, including the cities of Sebasteia and Larissa, after being attacked by Turkish forces from Azerbaijan. Although conventionally dated to 1016 or early 1017, it may have occurred as late as 1021. After the collapse of a rebellion by Nikephoros Phokas and Nikephoros Xiphias in late summer 1022, it is significant that Basil II campaigned beyond Vaspurakan, attacking the city of Her.115 112 Arist., ed. Yuzbashian, p. 25; French tr. Canard and Berb´ erian, p. 7. Bagrat III became the ruler of Kartli in 975 and Abkhazia three years later. Under his direction the kingdom of Georgia was established between 1008 and 1010. 113 Arist., ed. Yuzbashian, p. 27; French tr. Canard and Berb´ erian, p. 10; ME, I.10, ed. Melik‘-Adamean and Ter-Mik‘ayelean, pp. 12–14; tr. Dostourian, pp. 22–3. 114 ME, I.50, ed. Melik‘-Adamean and Ter-Mik‘ayelean, pp. 56–8; tr. Dostourian, p. 46. 115 Arist., ed. Yuzbashian, p. 38; French tr. Canard and Berb´ erian, pp. 23–4; ME, I.51, ed. Melik‘Adamean and Ter-Mik‘ayelean, p. 58; tr. Dostourian, p. 47. See also, p. 696.

8. armenian neighbours (600–1045)

361

Although both Sennacherim-John Artsruni and John-Smbat III had come to terms with Basil II by January 1022, this did not deter Nikephoros Phokas from soliciting support from other family members. It is unclear, however, how far they responded to his appeal.116 In the event, Phokas was assassinated on 15 August 1022, possibly by the son of Sennacherim-John Artsruni. Basil then moved quickly, inflicting a sharp defeat upon George I on 11 September 1022 and coming to terms with him shortly afterwards. Evidently Abkhazian, Georgian and Armenian princes were still tempted to participate in a rebellion fomented in the east by a member of the Phokas family. Basil II was aware of the threat. His persistent involvement with Armenia, and the extension of the empire’s frontiers to incorporate first Vaspurakan and ultimately Ani, should be seen in the context of, and as a response to, these rebellions. During the tenth century, a large number of small ‘Armenian’ themes were created, consisting essentially of a fortress and its surrounding district.117 By contrast, the themes of Taron (966 or 967), Vaspurakan (c.1021) and Iberia (1022) were organised around existing Armenian principalities ceded to the empire. Tellingly, these were not broken up. Whilst the sigillographic evidence reveals considerable fluidity in the combination of high military commands across these themes during the eleventh century, there is presently little evidence for sustained administrative down-reach within them.118 No more than a skeleton administrative structure can be traced, suggesting that existing social and political structures continued to be employed.119 This ‘slim-line’ Byzantine presence would prove to be inadequate when faced by sustained Turkish assault after 1045.120 Basil II’s campaign of 1022 did not mark an end to military operations. In 1023 or 1024 the fortified town of Archesh on Lake Van was captured by Nikephoros Komnenos whilst nearby Perkri was taken in 1035.121 These were both granted separate thematic status but this is unsurprising, seeing that they had never formed part of Vaspurakan and had been captured from the ‘Persians’.122 Separate themes of Manzikert (after 1000) and Artzike had also been created.123 This string of small themes fulfilled a long-cherished strategic aim, expressed in the De administrando imperio, that if these kastra were in imperial control, ‘a Persian army cannot come out against 116 117 118

ME, I.51, ed. Melik‘-Adamean and Ter-Mik‘ayelean, p. 58; tr. Dostourian, pp. 46–7. LPB, pp. 264–8, 355–63; Yuzbashian (1973–4), p. 169. On Taron, see Yuzbashian (1973–4), pp. 140–54; on Iberia: K¨uhn (1991), pp. 187–204; on Vaspurakan: Zacos, ed. Cheynet, pp. 93–4. 119 DOS, IV, nos. 57.1, 75.2, 75.3, 75.4, 76.1, pp. 148, 166–9; Zacos, ed. Cheynet, nos. 37a, 37b, pp. 72–4. 120 Holmes (2001), p. 56; Holmes (2005), pp. 538–41; see also, p. 698. 121 On Archesh, see Arist., ed. Yuzbashian, p. 41; French tr. Canard and Berb´ erian, pp. 26–7; on Perkri: Arist., ed. Yuzbashian, pp. 48–9; French tr. Canard and Berb´erian, pp. 38–40; Skyl., ed. Thurn, p. 388; French tr. Flusin and Cheynet, p. 322. 122 On Archesh, see Seyrig, no. 168, p. 123. 123 On Manzikert, see DOS, IV, no. 67.1, pp. 156–7; on Artzike: Oikonomides et al. (1998), p. 44.

362

the midd le empire

Romania’.124 They also deterred Ashot IV ‘the Brave’ from expanding southwards into former Artsruni territory. The literary sources reveal almost nothing about the reigns of John-Smbat III Bagratuni and Ashot IV ‘the Brave’ between 1022 and 1041. Contemporary inscriptions and colophons, however, confirm ongoing relations with Byzantium, and the numismatic evidence is persuasive. From the reign of Nikephoros II Phokas, Armenia switched from a silver-based coinage to a gold- and copper-based coinage, using exclusively Byzantine issues. During the excavations at Ani, several thousand Byzantine copper coins were found, both loose and in hoards.125 In 1979, some 3,539 of Constantine VIII’s nomismata, equivalent to almost 50 pounds of gold, were unearthed at Nouchevan, near Dvin.126 The epigraphic evidence is no less valuable in the historical reconstruction. An inscription at Khtskawnk‘, dated 1033, refers to ‘the reign of Smbat shahanshah, son of Gagik shahanshah, who had adopted the beloved boy Sargis, during the time of the three kings of the Romans, when he received the triple honour anthypatos, patrikios, vest¯es and doux of the east’.127 Aristakes records that John-Smbat’s son, Erkat‘, died young.128 This inscription confirms that he had designated Sargis as his successor, and that Sargis had received imperial sanction. By the time of his death, however, John-Smbat III had apparently changed his mind. A colophon dates the completion of a Gospel book to 1041, ‘when Yov[h]an[n]¯es [that is, John-Smbat III] king of Armenia was translated to Christ and gave his kingdom to his nephew Gagik’.129 The complex sequence of events between 1041 and 1045, concluding with the Byzantine occupation of Ani, therefore originated in a familiar context, a time of political transition.130 Instead of developing ties with both Sargis and Gagik, however, Byzantine policy after 1022 seems to have anticipated only the succession of Sargis. Gagik’s unexpected accession thwarted these plans and with Constantine IX Monomachos (1042–55) embroiled in George Maniakes’ rebellion (see below, pp. 599–600), Gagik II Bagratuni enjoyed two years’ respite.131 In 1044, however, he was induced to visit Constantinople where he was detained and offered Melitene in return for Ani.132 Initially he refused but when the forty keys of Ani were produced, proving treachery on the part of Catholicos Peter, he abdicated and received lands in Cappadocia. Although the leaders of Ani then resolved to entrust their city either to Gagik’s brother-in-law, David Dunats‘i or to Bagrat IV, king 124 126 128 129 131

125 Mousheghian et al. (2000b), p. 38. DAI, ch. 44, pp. 204–5. 127 Kostaneants‘ (1913), pp. 17–18. Mousheghian et al. (2000a), p. 149. Arist., ed. Yuzbashian, p. 32; French tr. Canard and Berb´erian, p. 16. 130 Shepard (1975–6). Mat‘evosyan (ed.), Hishatakaranner, no. 105, p. 86–7. Arist., ed. Yuzbashian, pp. 57–8; French tr. Canard and Berb´erian, p. 46; ME, I.77–8, ed. Melik‘Adamean and Ter-Mik‘ayelean, p. 96; tr. Dostourian, p. 67. 132 Arist., ed. Yuzbashian, pp. 61–2; French tr. Canard and Berb´ erian, pp. 50–1; ME, I.84, ed. Melik‘Adamean and Ter-Mik‘ayelean, pp. 102–4; tr. Dostourian, pp. 71–2.

8. armenian neighbours (600–1045)

363

Figure 24 Part of the southern fac¸ade of the cathedral church of Ani, begun in 989 under King Smbat II Bagratuni (‘Master of the Universe’) (977–89) but completed in 1001 by Queen Katrinide, wife of King Gagik I Bagratuni. It was designed by the architect Trdat who was also commissioned to repair St Sophia in Constantinople following earthquake damage in 989. Katrinide died in 1012 and was buried in a mausoleum close to the church

of Georgia (1027–72), the approach of another Byzantine army precipitated the final surrender of the city.133 conclusion Although the relationship between Byzantium and Armenia changed repeatedly across these centuries, three particular features stand out. In the first place, relations were continuous – only the period between 790 133

Arist., ed. Yuzbashian, p. 62; French tr. Canard and Berb´erian, p. 52.

364

the midd le empire

and 830 lacks evidence for any direct contact, but this mirrors the dearth of information about any aspect of Armenia during these decades. Secondly they were multi-layered. The sources tend to focus upon high-level contacts involving the leading Armenian clerics and princes and treat these as exclusive or representative. In fact, it seems very likely that lesser lords and individual bishops were also in contact with Byzantium throughout this period, although such ties are usually hidden from view. Thirdly they were reciprocal. Byzantium was eager to secure its eastern flank and therefore sought to attract Armenian clients into its service. At the same time, Armenian princes looked to Byzantium to bolster their own status within Armenia through the concession of titles, gifts and money. In a highly competitive, militarised society, there were obvious advantages in gaining recognition from a neighbouring polity, not least in the event of attack, when Byzantium could serve as a far more effective refuge than any mountain redoubt or individual fortress. It is no coincidence that the Byzantine army – and then the state – came to be filled with men of Armenian origin or descent. That, however, is another story.134 134

Garso¨ıan (1998). See also above, pp. 272, 300.

CHAPTER 9

CONFRONTING ISLAM: EMPERORS VERSUS C A L I P H S ( 6 4 1 – c. 8 5 0 ) walt e r e . kae gi

introduction 1 Two features characterise Byzantine–Muslim relations between the seventh and ninth century: a finely tuned link between domestic strife and the external fortunes of war and diplomacy; and the fitful involvement of both polities’ leaders with their armed forces, without exercise of personal command. The Arabs’ dramatic conquest of Byzantium’s eastern territories in the 630s was followed by four further periods of Muslim expansion; by gradual stabilisation; and then by Byzantine strengthening and eventual territorial recovery. The four periods of Muslim expansion were all brought to an end by bouts of civil war (fitna) among the Muslims, the first lasting from 656 until 661. The second expansionary period under the Sufyanid Umayyad caliphs was followed by almost ten years of civil war, from 683 until 692; the third, under the Marwanid caliphs – the final branch of the Umayyad dynasty – was broken by infighting for some two years between 718 and 720, only to be followed by a twenty further years or so of aggressive campaigning. The violent replacement of the Umayyads by the Abbasids in the mid-eighth century owed nothing to Byzantium, nor did it halt military and diplomatic interaction between the two polities; but it did transform Arab–Byzantine relations. the parameters of co nflict The most vulnerable period for Byzantium came immediately after the disastrous battle of the river Yarmuk in 636, during the imperial succession crisis triggered by Heraclius’ death in 641 and in the earliest years of his successor Constans II (641–68) (see above, pp. 230–1). After the withdrawal of their armies from Syria and northern Mesopotamia, the Byzantines had managed to regroup by the late 630s and early 640s and create new Anatolian defences, taking advantage of the Taurus mountains and key fortified points in the interior. Although limited truces had previously been struck 1

I should like to thank Paul Cobb for his advice and clarification on a number of points.

365

Constantinople

CORSICA

(sieges 674–8,716–18)

SARDINIA

Carthage

SICILY

Phoenix

(903)

Taormina Syracuse

(698)

(battle 654/655)

(879)

Kairouan Jerba

CRETE

Rhodes

(battle 654) CYPRUS

Yarmuk

Alexandria (battle 636) (645)

(635)

Damascus

Jerusalem (638)

al-Qadisiyya (battle 636)

Muhammad Abu Bakr (632–4) Umar I (634–44) Uthman (644–56) and Ali (656–61) Umayyads (661–750) 750–850 Key battles and sieges (654)

0 0

Dates when key cities fell 1000

500 500

1500 1000

2000 km 1500 miles

Map 18 The expansion of Islam 632–850

9. confronting isl am (641–c. 850)

367

with the Arabs, no formal, linear frontier was ever established and hostilities persisted. Fortunately for the Byzantines, the Muslims had priorities elsewhere. They needed to consolidate their vast territorial gains in Syria and northern Mesopotamia and to complete their conquest of Egypt, both more attractive and easier goals than the continued seizure of territory in Byzantine Anatolia. By the time that Mu‘awiya (661–80) emerged as caliph, the Muslims had missed their chance of outright conquest of Byzantium: the empire was stabilising, as Constans II rapidly gained military experience and judgement, and developed his defences against the Arab threat from Syria. Caliph ‘Umar (634–44) reportedly believed that the Muslims needed to consolidate their territorial expansion of the 630s before pursuing further conquests at the expense of Byzantium. Tradition has it that during his caliphate ‘Umar restrained Mu‘awiya from attempting an invasion of the island of Cyprus.2 This is plausible, and consistent with the well-known story that ‘Umar also tried to restrain the very able military commander ‘Amr ibn al-‘As from invading Egypt.3 Mu‘awiya only succeeded in implementing his far more aggressive policy towards Byzantium after ‘Umar’s death, prompted no doubt in part by a calculation of his own interests, as well as of the advantages he believed such a policy would gain for Islam. Political, topographical and logistical impediments combined with Byzantine military resilience to halt major Arab advances into Anatolia in the seventh century, even though the Arabs made significant territorial conquests in the central and western Mediterranean at Byzantium’s expense. They initially used a combination of force and diplomacy to overcome the Byzantine defences, being prepared to engage in fierce combat, while also negotiating separate terms with both local civilians and military commanders. However, these tactics ceased to be effective once the Muslim armies tried to penetrate and establish permanent control north of the Taurus and Anti-Taurus mountains in Asia Minor. It is difficult to define the style of Arab–Byzantine warfare in the seventh century. Muslim methods involved a broad conformity to Islamic principles, including the spreading of the faith by force, together with use of combined military and political initiatives. The Muslims would divide their opponents both on and off the battlefield, identifying those willing to conclude separate peace terms and then allowing them to negotiate their submission, thus reducing the likelihood of costly, bloody resistance. However, using political pressure to control the Byzantine civilian population was not inconsistent with fighting Byzantium’s forces and demolishing their strong points, destroying their opponents’ equilibrium. There tended to be 2 3

al-Baladhuri, Futuh al-buldan, ed. de Goeje, p. 152; tr. Hitti and Murgotten, I, p. 235. al-Baladhuri, Futuh al-buldan, ed. de Goeje, p. 212; tr. Hitti and Murgotten, I, p. 335.

368

the midd le empire

close consultation, as far as was practicable, between field commanders and the highest Muslim leadership. The Arabs generally tried to avoid positional warfare, such as slow-moving sieges, except when they were forced onto the defensive; their strategy was to drive their opponents into decisive battle, with an eye to exploiting military victories to the full. The Byzantines, for their part, tended to avoid the risk of major land battles after their defeat at Yarmuk, preferring to seek refuge in fortified positions. They made cautious efforts to identify, cut off, attack and destroy smaller detachments of Muslim raiders, using relatively modest-sized mobile units. Concepts of holy war and crusade did not dominate Byzantine warfare between the seventh and ninth century.4 It is similarly hazardous to superimpose later concepts of jihad onto seventh- and eighth-century Arab– Byzantine warfare, as both jihadi practices and concepts in this period are poorly documented. They become better documented with the expansion and stabilisation of the Muslim territories, when communal obligation to perform jihad was increasingly focused on the frontier regions, and it was in areas such as northern Syria and upper Mesopotamia that the most zealous Muslim soldiers tended to be concentrated. Areas further back from the frontiers, although theoretically supportive of military expansion, were in practice less involved on a daily basis. It became increasingly difficult to engage the whole Muslim community actively in the process of jihad. Between the late 630s and 650s both empire and caliphate periodically created zones of devastation between their territories. Local inhabitants would occasionally be allowed to stay, but only if they agreed to act as informers and refused to help the enemy. However, these more charitable arrangements do not appear to have been successful: both powers expelled inhabitants they regarded as hostile, leaving either a total wasteland, or settling their own armed troops and loyal populations. ‘Umar reportedly wanted to create at least a temporary zone of destruction between Byzantium and the caliphate, just as Heraclius (610–41) had done in the remaining imperial territories after the Arabs overran Syria. According to al-Ya‘qubi, whenever ‘Umar spoke of the Byzantines, he voiced the hope that God would ‘turn the passes between us and them into burning coals; this side [of the passes] for us and what is behind [the passes] for them’.5 The ninthcentury historian al-Baladhuri reports that ‘Umar ordered that Arabissos be destroyed and its inhabitants forcibly removed, after learning of their refusal to give information on Byzantine troop movements to the Muslims, while continuing to act as informants for the empire.6 The inhabitants of 4 On the question of these concepts’ existence in Byzantium, see Oikonomides (1995); Kolbaba (1998); Dennis (2001b); Stephenson (2007). 5 al-Ya‘qubi, Ta’rikh, II, pp. 178–9. 6 al-Baladhuri, Futuh al-buldan, ed. de Goeje, pp. 156–7; tr. Hitti and Murgotten, I, pp. 241–2; on Arabissos see Hild and Restle (1981), pp. 144–5; Kaegi (1992), p. 244; MS, X.21, ed. and French tr. Chabot, II, p. 359; al-Tabari, Ta’rikh, ed. de Goeje et al., I, p. 2349; tr. Friedmann, XII, p. 134.

9. confronting isl am (641–c. 850)

369

nearby Duluk and Raban, in northern Syria, apparently honoured a similar arrangement with the Muslims.7 And while governor of Syria in 649, Mu‘awiya forced the Cypriots to stop giving aid to the Byzantines and to inform on them, an arrangement which they failed to respect.8 Byzantine and Muslim governing circles thus had an equal interest in creating zones of devastation. The resulting attempts to tighten governmental control on either side of the de facto border, to counteract the emergence of independent borderland powers, helped to strengthen state-building for both caliphate and empire. From Heraclius’ reign onwards, the Byzantines started to appoint military commanders in place of those civil governors who proved too willing to come to terms with the Muslims. Through such appointments, the Byzantines hoped to concentrate power in the hands of military leaders who were dependent on the emperor: they would therefore make no local settlements with the Muslims without having received explicit imperial authorisation and approval.9 Not all early Arab–Byzantine contacts were violent, and despite extensive military engagement, limited maritime trade, exchange and travel – especially pilgrimages – persisted. Some Christian churchmen and ascetics managed to cross the frontiers at transit points such as Cyprus, and smugglers and renegades played their part in creating a porous frontier. Diplomacy coexisted with warfare. Prisoner- and hostage-exchange was a complex challenge for both Byzantine and Muslim authorities in the seventh and eighth centuries. Diplomatic negotiations generally took place either at Damascus or Constantinople at the highest level and were conducted by the caliph and emperor – or their envoys – but ad hoc exchanges could also occur occasionally between local commanders.10 Accommodating such political realities committed neither side to any fundamental theoretical or religious concessions. Diplomatic protocol was highly formalised by the tenth century, as witness Constantine VII Porphyrogenitus’ Book of ceremonies, but it is likely that this protocol owed its origins to seventh- and eighth-century practices.

m u‘awiya versus constans ii: byz ant ium und er pressure Byzantine Anatolia quickly became the target of Muslim expeditions after the Byzantine evacuation of Syria and northern Mesopotamia in the midseventh century. Muslim historical traditions disagree on who led the 7 8

al-Baladhuri, Futuh al-buldan, ed. de Goeje, p. 150; tr. Hitti and Murgotten, I, p. 231. al-Baladhuri, Futuh al-buldan, ed. de Goeje, p. 153; tr. Hitti and Murgotten, I, p. 236. See also ibid., pp. 153–8; tr. Hitti, and Murgotten, I pp. 236–43; Ibn Sallam, al-Amwal (1968), pp. 248, 253; repr. 1986, pp. 184–5, 187–8, citing the scholar al-Awza’i as his authority. 9 Haldon (1993), pp. 1–47. 10 See Kaegi (1992), pp. 250–2; Kaplony (1996).

370

the midd le empire

earliest Arab raids through the mountain passes into Anatolia and the ‘land of the Romans’. Ibn ‘Abd al-Hakam’s Futuh misr reports the earliest Muslim expedition against Amorion in 644 (AH 23), when Constans II was still too young to be capable of developing his polity’s defences.11 These early raids penetrated deep into Byzantine territory. Mu‘awiya had already commanded an incursion into Asia Minor in 643, and he probably led another expedition against Amorion in 646. His expeditions disrupted the Anatolian interior, forcing the Byzantines into defensive countermeasures. The antecedents of later Muslim warfare and diplomacy can be traced back to Mu‘awiya’s governorship of Syria, and to the period after he became caliph in 661.12 Summer raids (saifa) began from around 640. The raids of the 640s were often launched from Mesopotamian and Syrian towns such as Homs and Antioch, and the raiders entered Anatolia by way of passes such as the one at Hadath (between Germanikeia and Melitene) and the Cilician Gates (using bases such as Mopsuestia and Tarsus once these were in Arab hands). Whether or not Mu‘awiya himself went on the important early expedition against Amorion in 644,13 he led a number of other campaigns into Anatolia at a time when Byzantine resistance was beginning to harden. As Constans II tried to fortify Byzantine cities and strongholds and to develop a coherent resistance, Mu‘awiya gained experience in how to fight and to negotiate with the Byzantines, becoming familiar with the problems and challenges of their Anatolian terrain, climate and logistics. Probably no other caliph had as much personal military experience against the Byzantines as Mu‘awiya did. Despite this, Mu‘awiya’s offensives against Byzantium did not result in any lasting Muslim conquests in Anatolia between 643 and his death in 680. Muslim raids became an almost annual event, penetrating up to 1,000 kilometres into the Anatolian plateau. They were not restricted to summer, and a winter raid would sometimes follow hard upon a summer one.14 The raids contributed to Mu‘awiya’s prestige, helping to enrich the Muslims and attracting ever more tribesmen to take part, while the Muslim casualties probably remained relatively modest. However, this persistent raiding seriously damaged the empire’s infrastructure: the Byzantines’ territories were devastated; they lost property and human lives; many were taken captive; and their commerce and agriculture were destroyed. The raids also kept Byzantium off-balance, forcing them onto the defensive and preventing them from launching major offensives of their own against Muslim Syria. 11 13

12 On Mu‘awiya, see PMBZ, #5185 and below, n. 35. Kaegi (1977). Kaegi (1977). See also the eighth-century scholar Layth bin Sa‘d’s account in al-Fasawi, al-Ma‘rifah, III, p. 307; Ibn Hajar al-‘Asqalani, Kitab al-Isabah, II, p. 533. 14 See Brooks, ‘Arabs in Asia Minor’ for a collection of translated fragments now in need of major revision.

BLACK

SEA

Constantinople

ARMENIA

Dazimon Dorylaion

Sebastopolis Sebasteia Kamacha

G A L AT I A

Amorion Akroinon

A N A T O L I A (ASIA MINOR)

Dvin

Theodosioupolis

Charsianon

L. Van

Ikonion L. Urmia

ISAURIA

CIL

IC

IA Antioch

Aleppo Chalkis

O

T

A ris Tig

IA

Homs

P

M

Tripoli

Mosul

O

s te ra

Arwad

ES

ph Eu

CYPRUS

SYRIA

Soloi

M Raqqa

MEDITERRANEAN SEA Samarra

Baghdad Basra

Yarm uk

Caesarea

Jerusalem

RU SM OU NT AI N

Damietta

Alexandria

PA L

EST

INE

Damascus

S Taranda

Melitene

Tyana

E G Y P T

US NS UR TAI TA UN O M Tarsus

Pass 0 0

Heraclea- Lulon Cybistra

100

200 100

300

-T AU

Arabissos

AN

TI

(Ceyhan)

Amida

Hadath

Cilician Gates

Germanikeia

Anazarbos Mopsuestia

Zapetra

Pyramos

Semalous (?)

400 km

AL

W - A

Samosata Raban

Duluk

ASIM THUGHUR Membij

200 miles

Map 19 Byzantium versus Islam: the zone of direct confrontation

AL-JAZIRA

372

the midd le empire

Under Mu‘awiya’s able command, the Muslims were innovative and capable of taking their opponents by surprise. Despite a lack of any Arab seafaring tradition, they embarked on combined land and naval operations, highlighting their readiness to adopt new strategies and techniques of warfare. Although the literary evidence about naval expeditions against Byzantine-controlled islands such as Arwad and Rhodes is contradictory and impossible to verify, epigraphic evidence confirms an intensification of Muslim military and naval activities, including devastating Muslim raids on Byzantine Cyprus in 649 and 650 at Soloi.15 Some 120,000 Cypriots are said to have been deported, marking a serious change in the island’s fortunes. Although the Byzantines did receive forewarning of the Arabs’ preparations for some of these naval expeditions, the raids further reduced Byzantine resources and naval capabilities in the Mediterranean, jeopardising yet more of the empire’s coastline and islands. Muslim power was proving capable of extracting financial concessions from regions hitherto beyond its reach. In 654 or 655, Mu‘awiya’s naval forces decisively defeated Constans II at the battle of Phoenix off the south-western Anatolian coast, also known as the ‘battle of the masts’. The late seventh-century ascetic Anastasius of Sinai testified to the shock of this Byzantine naval defeat.16 Ibn Abi Sarh, governor of Egypt, commanded the Arab fleet, and its crew members may have included many Christian Egyptians. The Muslims then mounted a threatening but brief and abortive combined land-and-sea operation, reaching almost to Constantinople itself.17 The seventh-century Muslim naval offensives culminated in their costly and disastrous assault and naval blockade of Constantinople from 674 to 678. The Arabs failed to plan adequately and they also encountered a new Byzantine weapon, Greek fire, which devastated their warships and inflicted heavy casualties (see above, p. 233). Mu‘awiya’s governorship of Syria and his caliphate extended Arab territorial control, with Cyprus and most of Armenia falling under Muslim influence. The period of Mu‘awiya’s ascendancy also saw larger-scale expansion in North Africa, as Muslim military pressure on Anatolia reduced the Byzantine government’s ability to reinforce and defend vulnerable positions in the empire’s western approaches. Mu‘awiya’s prestige derived from his military victories; from the fact that he received recognition from the Byzantine emperor; from his control of the holy places of Christianity and Islam; from the line of successors from the Prophet Muhammad; and also from the messianic attributes of his leadership. His aggressive, risky and unpredictable strategies challenged a number of arrogant assumptions of the Byzantines 15 16

Feissel (1987), pp. 380–1. Anastasiu