Roman Emperors in Context: Theodosius to Justinian 9780367680763, 9780367680756, 9781003134084

Table of contents :
Cover
Half Title
Series Page
Title Page
Copyright Page
Contents
List of figures
Acknowledgements
List of abbreviations
Introduction: historicising imperial business
1 Reinventing Constantinople: Theodosius I’s imprint on the imperial city
2 Dynasty and aristocracy in the fifth century
3 Dynasty and ethnicity: emperor Leo I and the eclipse of Aspar
4 Leo I and the palace guard
5 The imperial reigns of Leo II
6 Ariadne Augusta: shaping the identity of the early Byzantine empress
7 Justinian under Justin: reconfiguring a reign
8 Justinian, Theodora and the Church of Saints Sergius and Bacchus
9 Justinian the ‘sleepless emperor’
10 Justinian’s Constantinople
Bibliography
Index

Citation preview

Roman Emperors in Context

Roman Emperors in Context: Theodosius to Justinian brings together ten articles by renowned historian Brian Croke. Written separately and over a period of fifteen years, the revised and updated chapters in this volume provide a coherent and substantial story of the change and development in imperial government at the eastern capital of Constantinople between the reigns of Theodosius I (379–95) and Justinian (527–65). Bookended by chapters on the city itself, this book is based on a conviction that the legal and administrative decisions of emperors have an impact on the whole of the political realm. The fifth century, which forms the core of this book, is shown to be essentially Roman in that the significance of aristocracy and dynasty still formed the basic framework for political advancement and the conduct/conflict of political power around a Roman imperial court from one generation to the next. Also highlighted is how power at court was mediated through military generals, including major regional commanders in the Balkans and the East, bishops and bureaucrats. Finally, the book demonstrates how the prolonged absence of male heirs during this period allowed the sisters, daughters, mothers and wives of Roman emperors to become more important and more central to imperial government. This book is essential reading for scholars and students of Roman and Byzantine history, as well as those interested in political and legal history. Brian Croke is Honorary Associate in Ancient History at the University of Sydney and Expert Adviser in Education. He has published several monographs and over 100 articles on various aspects of Roman and Byzantine history and modern historiography. His publications include Christian Chronicles and Byzantine History, 5th–6th Centuries (1992), The Chronicle of Marcellinus (1995) and Count Marcellinus (2001).

Also in the Variorum Collected Studies series BRIAN CROKE

Roman Emperors in Context Theodosius to Justinian (CS2000)

STEPHEN KNIGHT

Medieval Literature and Social Politics Studies of Cultures and Their Contexts (CS1099)

EKMELEDDIN İHSANOGLU

Studies on Ottoman Science and Culture (CS1098)

DAVID S. BACHRACH and BERNARD S. BACHRACH Writing the Military History of Pre-Crusade Europe Studies in Sources and Source Criticism (CS1097)

PAMELA M. KING, edited by Alexandra F. Johnston Reading Texts for Performance and Performance as Texts Shifting Paradigms in Early English Drama Studies (CS1096)

FELICE LIFSHITZ

Writing Normandy Stories of Saints and Rulers (CS1095)

STEPHEN GERSH

Metaphysics and Hermeneutics in the Medieval Platonic Tradition (CS1094)

MICHAEL HESLOP

Medieval Greece Encounters Between Latins, Greeks and Others in the Dodecanese and the Mani (CS1093)

T.A. BIRRELL, edited by Jos Blom, Frans Korsten and Frans Blom Aspects of Recusant History (CS1092)

PAMELA NIGHTINGALE

Mortality, Trade, Money and Credit in Late Medieval England (1285–1531) (CS1091)

SARAH CARPENTER, edited by John J. McGavin and Greg Walker Early Performance: Courts and Audiences Shifting Paradigms in Early English Drama Studies (CS1090)

For more information about this series, please visit: www.routledge.com/ Variorum-Collected-Studies/book-series/VARIORUM

Roman Emperors in Context Theodosius to Justinian Brian Croke

VARIORUM COLLECTED STUDIES

First published 2021 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business This edition © 2021 Brian Croke The right of Brian Croke to be identified as author of this work has been asserted by him in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Names: Croke, Brian, author. Title: Roman emperors in context : Theodosius to Justinian / Brian Croke. Description: Abingdon, Oxon ; New York : Routledge, 2021. | Series: Variorum collected studies | Includes bibliographical references and index. Identifiers: LCCN 2020053527 (print) | LCCN 2020053528 (ebook) | ISBN 9780367680756 (hardback) | ISBN 9781003134084 (ebook) Subjects: LCSH: Emperors—Rome—Biography. | Rome—History— Theodosians, 379–455. | Rome—History—Empire, 284–476. | Byzantine Empire—History—Justinian I, 527–565. Classification: LCC DG322 .C76 2021 (print) | LCC DG322 (ebook) | DDC 937/.090922—dc23 LC record available at https://lccn.loc.gov/2020053527 LC ebook record available at https://lccn.loc.gov/2020053528 ISBN: 978-0-367-68075-6 (hbk) ISBN: 978-0-367-68076-3 (pbk) ISBN: 978-1-003-13408-4 (ebk) Typeset in Times New Roman by Apex CoVantage, LLC VARIORUM COLLECTED STUDIES SERIES CS2000

E. A. J. MAGISTRO OPTIMO ET H I S TO R I C O D I S E RT I S S I M O

CONTENTS

List of figures Acknowledgements List of abbreviations

ix x xi

Introduction: historicising imperial business

1

1 Reinventing Constantinople: Theodosius I’s imprint on the imperial city, originally published in From the Tetrarchs to the Theodosians (2010), 237–60.

6

2 Dynasty and aristocracy in the fifth century, originally published in Cambridge Companion to the Age of Attila (2014), 102–27.

29

3 Dynasty and ethnicity: emperor Leo I and the eclipse of Aspar, originally published in Chiron 35 (2005), 147–203.

51

4 Leo I and the palace guard, originally published in Byzantion 75 (2005), 117–51.

108

5 The imperial reigns of Leo II, originally published in Byzantinische Zeitschrift 96 (2003), 559–76.

134

6 Ariadne Augusta: shaping the identity of the early Byzantine empress, originally published in Christians Shaping Identity from the Roman Empire to Byzantium (2015), 293–320.

153

7 Justinian under Justin: reconfiguring a reign, originally published in Byzantinische Zeitschrift 100 (2007), 13–56.

169

vii

CONTENTS

8 Justinian, Theodora and the Church of Saints Sergius and Bacchus, Dumbarton Oaks Research Library and Collection, Trustees for Harvard University, originally published in Dumbarton Oaks Papers 60 (2006), 25–63, edited by Alice-Mary Talbot.

207

9 Justinian the ‘sleepless emperor’, originally published in Basileia. Essays in Imperium and Culture (2011), 103–8.

246

10 Justinian’s Constantinople, originally published in Cambridge Companion to the Age of Justinian (2005), 60–86. Bibliography Index

253 276 306

viii

FIGURES

2.1 2.2 2.3 2.4 2.5 2.6

Family of Theodosius I Family of Valentinian III Family of Marcian Families of Leo and Zeno Family of Aspar Family of Ostrogothic King Theodoric

ix

31 33 36 39 40 42

ACKNOWLEDGEMENTS

For both scholarly assistance and personal encouragement over four decades now I remain grateful to Elizabeth Jeffreys and Roger Scott who have helped critically refine most of these studies in their original form and whose guidance assisted the broader research that led to them. Also deserving acknowledgement are Geoffrey Greatrex, Averil Cameron and Michael Maas who have had a constructive influ­ ence at some stage on several of the studies collected here and Alanna Nobbs who introduced me as an impressionable teenager to the world of the emperors from Theodosius to Justinian. This book is dedicated to my inspiring first teacher of ancient history and life-long mentor, Edwin Judge (Macquarie), now healthily and productively surging towards his century. On a monument erected in Rome’s Forum of Trajan in 402, Q. Memmius Symmachus could have had E. A. Judge in mind when honouring his father-in-law, Nicomachus Flavianus – ‘historico disertissimo’.

x

A B B R E V I AT I O N S

ACO CA CCh Chron.Const. CIL CTh CJ CFHB CLRE Const. Porph. CSCO.Scr.Syr. CSEL FHG HE MGH.AA PBE PG PL PLRE 1 PLRE 2 PLRE 3 PO

Acta Conciliorum Oecumenicorum, ed. E. Schwartz, vols. 3 (Berlin 1940) and 4.2 (Berlin 1914). Collectio Avellana, ed. O. Günther, Vienna 1898, vol. 2 (CSEL 35) Corpus Christianorum. Series Latina (Turnhout 1953–) Chronica Constantinopolitana Corpus Inscriptionum Latinarum (Berlin 1863–). Codex Theodosianus, ed. T. Mommsen (1904–5). Berlin. Codex Justinianus, ed. P. Krüger, Corpus Iuris Civilis 2, Berlin 1928 rp. 1964). Corpus Fontium Historiae Byzantinae Consuls of the Later Roman Empire, 1987. eds. K. A. Worp, R. S. Bagnall, Alan Cameron, S. R. Schwartz, Atlanta Constantine Porphyrogenitus Corpus Scriptorum Christianorum Orientalium, Scriptores Syri (Paris 1903–). Corpus Scriptorum Ecclesiasticorum Latinorum (Vienna 1866–). Fragmenta Historicorum Graecorum, 4 vols. (ed. C. Müller, Paris 1848–51). Historia Ecclesiastica Monumenta Germaniae Historica. Auctores Antiquissimi Prosopography of the Byzantine Empire, vol. 1. ed. J. R. Martindale (at www.pbe.kcl.ac.uk/). Patrologia Graeca (ed. J. P. Migne, Paris 1857–86). Patrologia Latina (ed. J. P. Migne, Paris 1844–64). The Prosopography of the Later Roman Empire, vol. 1 (1970). eds. A. H. M. Jones, J. Morris and J. R. Martindale, Cambridge. The Prosopography of the Later Roman Empire, vol. 2 (1980). ed. J. R. Martindale. Cambridge. The Prosopography of the Later Roman Empire, vol. 3 (1992). ed. J. R. Martindale. Cambridge. Patrologia Orientalis (1907–).

xi

A B B R E V I AT I O N S

RE SEC

Realencyclopädie der klassischen Altertumswissenschaft, ed. G. Wissowa et al. (1893–1978). Stuttgart. Synaxarium Ecclesiae Constantinopolitanae, ed. H. Delehaye (1902). Brussels.

xii

INTRODUCTION Historicising imperial business

The present turns into the past one day at a time. Some days can be momentous or catastrophic with sudden and enduring impact. Most days, however, are domi­ nated by the routine of personal and civic life carried out within particular family and cultural traditions and within particular social and political structures. Thus it was at Rome and, from the 330s, also at New Rome or the ‘City of Constantine’, Constantinople. As the days of the Roman empire slowly turn into months and years, then decades and centuries, life’s routines, structures and traditions change and develop, punctuated by occasional threatening moments such as the Goths causing havoc in Rome in 410, the sight of the Huns within bowshot of the walls of Constantinople in early February 447, a massacre in the hippodrome in Janu­ ary 532 or the bold usurpation of Phocas on 23 November 602. As earlier forms of Roman thinking and behaving are eclipsed, novel forms emerge, but they are normally imperceptible at the time, at least during any individual’s relatively short lifetime. Mental change comes slowly. In the Roman world and the succeeding centuries, the pace of change was nothing compared to what we experience and expect in the 21st century. The historian’s task is to observe, delineate and evaluate all these changes retrospectively. To do so, historians generally operate solo rather than in teams. Accordingly, their own personal background, education and predilections fash­ ion their own methods and habits for making their work more manageable. One essential historiographical requisite is to mark out and label a particular patch of space and time in which to work and within which change can be both plotted and explained. For this book, the space and time encompassed is the broad period now labelled ‘Late Antiquity’. There can be no doubt that one of the great intel­ lectual revolutions of the 20th century was the invention and expansion of ‘Late Antiquity’ as a single period from the 3rd to the 8th century, pinned to a loose geographical canvas stretching from Ireland to modern Iran.1 ‘Late Antiquity’ is 1 The character and popularity of ‘Late Antiquity’ are more or less attributable to one remarkable scholar, Peter Brown (Oxford, then London, Berkeley and Princeton), drawing inspiration from two other, no less remarkable, scholars, Arnaldo Momigliano (Oxford/London/Pisa/Chicago) and Henri Marrou (Lyon, Paris). In general: Bolgiani (1980), 535–87; Brown (1997), 5–90; James (2008),

1

INTRODUCTION

outward looking and expansive. It focusses on mainly long-term religious, social and cultural transformation. It embraces all the Roman empire’s constituent cul­ tures and languages, as well as its neighbouring cultures, leaving behind the nar­ rower focus of previous scholarship and understanding of this period as the decline and fall of urban Greco-Roman civilisation. This was the inward looking ‘Later Roman Empire’ of Europe, the Middle East and North Africa, characterised by economic decline, personal and official corruption, falling standards of education and literature, barbarian invasions and so on. Until the 1970s, the ‘Later Roman Empire’ was a category reflecting underlying Roman political, military and legal structures, as demonstrated by the imperial court and administration.2 It followed logically and progressively the ‘early’ then the ‘middle’ Roman empires. Courses and textbooks in ancient history concluded with the ‘Later Roman Empire’, if not sooner, while in medieval history they began with the ‘Later Roman Empire’, if not later. By contrast, ‘Late Antiquity’, not to mention its recent and more bothersome offspring ‘Late Late Antiquity’, draws its organisational logic mainly from reli­ gious and cultural imperatives alone, both Roman and non-Roman in tradition and character.3 It has no designated predecessor period except, by implication, the whole of ‘antiquity’. Historians have not yet felt the need for an entrenched ‘middle antiquity’ or ‘early antiquity’ although it would be foolish to rule them out in the future. Broadly speaking, ‘Late Antiquity’ also side-lined what used to be separate self-evident descriptors for both east and west from the 5th to the 8th century, namely ‘Early Byzantine’ and ‘Early Medieval’ respectively, thereby creating lasting problems for both, as well as allowing ‘Early Christianity’ to project itself forward by another few centuries.4 To all of them was added what was called ‘pre-Islamic’ and ‘Early Islamic’, although these were never concepts that the Islamic world itself embraced.5 Given its geographical and chronological scope, however, ‘Late Antiquity’ is certainly much tidier as a single label than the plethora of labels otherwise required. Moreover, it makes perfect sense for tracing over both time and place significant religious, cultural and social trends and habits of mind. Its achievements are considerable.

2 3 4 5

20–30; Brown (2011), 6–22; Johnson (2012); Marcone (2020), 1–11. For the Momigliano dimen­ sion: Cracco Ruggini (1989); Marcone (2003). For the Marrou dimension: Vessey (1998); Markus (2009); Hunt (2018), 268–73. The origins of ‘Late Antiquity’ lie further back, and in art history (Elsner [2002]; Westbrook [2007]; Liebeschuetz [2004]; Cracco Ruggini [1993], xxxiii–xlv; Marcone [2000], 318–34), but the category was used more broadly in Germany (Delbrück [1933]; Gelzer [1927]; Kornemann [1949]), hence the otherwise curious use of the German ‘Spätantike’ in the seminal work on Augustine by Marrou (1949). Typified by Jones (1964), illuminated by Averil Cameron (2008); Liebeschuetz (2008). Fowden (2014). ‘Early Byzantine’: Averil Cameron (2016); ‘Early Medieval’: Heather (1997). The New Cambridge History of Islam edited by Robinson (2010) begins with a section ‘The Late Antique Context’, replacing ‘The Rise and Domination of the Arabs’ in the original Cambridge History of Islam edited by Holt et al. (1977).

2

INTRODUCTION

While Late Antiquity was being fully formulated and promoted in the 1970s and 1980s as a period of transformation and change, rather than decline, then can­ onised with its own dedicated journals, institutes and handbooks in the 1990s and 2000s, a parallel movement was impinging on the way history was formulated and presented. The so-called linguistic turn in particular created, among other things, a heightened awareness of how all historical accounts are constructed and how new perspectives can result from more careful attention to the different processes of selectivity and emphasis while using the very same sources of information.6 So too, we now see more clearly that all historical periods are themselves essen­ tially constructed or deconstructed or reconfigured by historians.7 As for the ‘long fourth century’ or ‘long fifth century or ‘long sixth century’ or even ‘long seventh century’ for that matter,8 all it demonstrates is the further artificiality of a closed 100-year period. A lazy form of historical periodisation results. Still, periodic cat­ egories remain necessary and useful although historians are now more sensitive to the value judgments and implicit teleology in the retrospective subdivision of periods into ‘early’, ‘middle’ and ‘late’, or ‘high’ and ‘low’, or ‘pre’ and ‘post’. ‘Late Antiquity’ is no exception. Even though the intrinsic value and coherence of ‘Late Antiquity’ have come to be taken for granted – and sometimes passionately protected – it is no surprise that such a periodisation is now itself subject to fresh and regular critique.9 In particular, the very coherence of ‘Late Antiquity’ is con­ tested because it minimises the fact that the Roman empire did actually disappear as a political entity in the west and not without violence.10 Rome still matters. The chapters that constitute this book might be judged as outmoded by the deep-rooted criteria for Late Antiquity because they are preoccupied with insti­ tutions, politics and political events. They are focussed essentially on Roman emperors (including one empress) from Theodosius I (reigned 379–95) to Justin­ ian (527–65), as well as their court, their administration and the urban context of the imperial capital of Constantinople where they lived and ruled. Not to take them seriously, however, would be a mistake, even though Late Antiquity’s focus on religious, social and cultural change across cultures and over centuries has marginalised the study of political, institutional and governmental change at the same time. This consequence is no secret. Despite being justifiably regarded as the progenitor of ‘Late Antiquity’ in the modern world, Peter Brown himself has long recognised that his original model of Late Antiquity has its limitations, in so far as it has excluded the state and its governmental administration.11 Subsequently, 6 7 8 9

Candidly noted by Elton (2018), 3. Clark (2004); Martin and Miller (2005). MacMaster (2017). Kaldellis (2019b), 78, even proposes a ‘long Byzantium’. Giardina (1999), 157–80; Averil Cameron (2002), 165–91 and (2003), 3–21; O’Donnell (2004), 203–13; Athanassiadi (2006), 311–24, esp. 313–22; Marcone (2008), 4–19; Tougher (2017). 10 Mitchell (2014), 6–9 and the twin tomes of Heather (2005); Ward-Perkins (2005). A subsidiary, less productive debate has been around whether the empire actually ‘declined’ or not (Liebeschuetz [2001], 1–11. 11 Brown (1997), 24.

3

INTRODUCTION

other experienced scholars working within the current paradigm of Late Antiq­ uity have expressed concern for the way that Late Antiquity privileges religious, cultural and social change, at the expense of political, administrative and even economic developments occurring at the same time.12 Obviously some rectifica­ tion is overdue, and it has begun to appear.13 Various methodological and historiographical convictions underpin or emerge from these studies. One is that, ‘Late Antiquity’ aside, the period from Theodosius to Justinian is a crucial one. Yet it is arguably under-studied and ill-understood. The culture of Roman government at Constantinople, how power is used and transmit­ ted across generations, is a key focus here. So too are the traditional explanatory assumptions that have lain untouched while the surrounding cultural and religious change has been intensively studied and revised: for example, Theodosius I as the great warrior emperor transforming the west, the court of Leo I (457–474) as the battleground between native Gothic and Isaurian bands, the reign of Justin I (518–27) as nothing more than the prelude to that of Justinian. While the period covered by this book has been caught up traditionally in the quest to establish where ‘Roman’ ends and ‘Byzantine’ begins, this is increasingly seen as a false dichotomy. Roman it is, Roman it continues to be.14 ‘Early Byzan­ tine’ would be perfectly understandable too, not only because it brings into play the preoccupation with social and religious behaviour, belief and change. It also allows for how politics at the imperial court intersects with family and aristocracy at the imperial city of Constantinople from the 4th to the 7th century, as well as the lands it ruled over at different points from the Danube to the Euphrates. The wider context of Constantinople reveals a place where the hard-headed activities of emperors trying to stay in power through dealing with generals and clergy, senators, bureaucrats and the people count for as much as the overt transforma­ tions of religion and culture.15 In recent years, fresh attention has been paid to the changing literary and his­ toriographical contexts of early Byzantine texts, while other research dimensions have also been developing rapidly. For example, the brisk growth of new archae­ ological material from surveys and excavations has, in turn, accentuated study and understanding of texts as different as the Notitia Dignitatum and Procopius’ Buildings. One of the more exciting and fruitful developments in the archaeologi­ cal study of early Byzantine history has been the growth of Byzantion itself into the imperial capital of Constantinople. Even in specialist works, however, the early Byzantine city tends to be treated as a synthetic or organic whole, often without due regard for its particular and distinct periods of development.16 Hence, 12 E.g. Giardina (1999); Averil Cameron (2003), 8 and (2005), 72; Mitchell (2014), 5; Rebenich (2009), 91; Kaldellis (2019b), 51. 13 Tougher (2019); Manders and Slootjes (2020). 14 As emphasized in recent years by Kaldellis, culminating in Kaldellis (2019a). 15 Pfeilschifter (2013); Begass (2018). 16 E.g. Dark and Özgumüs (2013), but illuminated by Crow (2019); Kaldellis (2019b), 50–4.

4

INTRODUCTION

the value of in-depth focussed snapshots of the city at two key junctures in its growth – late 4th century (Chapter 1) and mid 6th century (Chapter 10). The foregoing chapters were all researched, written and originally published between 2001 and 2015, either emerging of their own accord (Chapters 3, 4, 5, 7, 8) or in response to various invitations (Chapters 1, 2, 6, 9, 10). Inevitably, some parts have been contested–or contradicted–at different points by other schol­ ars. Now they appear in a revised form that acknowledges appropriate criticism and error, which occasionally has required an explanation, reply or correction. In some cases, new editions of texts, new translations and relevant new research have demanded inclusion. Most of the original articles are therefore superseded by these chapters. Further, some of these studies may feel somewhat slow-moving and detailed. That can be explained as a result of the need to balance the treat­ ment of a specific topic with unpacking the historiographical presuppositions that underlie modern interpretation of larger themes, such as the reigns of Theodosius I (Chapter 1), Leo I (Chapters 3, 4, 5 and 6) and Justin I (Chapters 7 and 8). In the case of Leo and Justin, their particular shape and emphasis may also be explained by the fact that they originally grew out of two larger projects, provisionally enti­ tled The House of Leo and Justinian’s People, which will hopefully be completed in due course.

5

1 REINVENTING CONSTANTINOPLE Theodosius I’s imprint on the imperial city*

Constantinople was established and embellished as a new Roman imperial city by the emperor Constantine I in the 320s and 330s. He provided ‘New Rome’, as he called it, with a solid defensive wall stretching from the Sea of Marmara to the Golden Horn, inside which he created a splendid and spacious new metropolis with an oval shaped forum, an imperial palace, baths, and porticoed thoroughfares, plus a local senate.1 His efforts to consolidate and promote his new city met with mixed success and over the succeeding decades its status and fortunes advanced only haltingly. Rather than become a new imperial home as Constantine had intended, Constantinople was treated by his successors more as a transit camp as they progressed back and forth from Gaul and Italy to the eastern limits of the empire in Mesopotamia and its northern limits on the Danube. Between the death of Constantine in 337 and the accession of Theodosius I in 379 emperors spent a total of just twenty-seven months in the city, or an average of less than two weeks per year.2 Arguably the longest single stretch was from September 365 to May 366, the duration of the unsuccessful usurpation of Procopius, who was proclaimed emperor at Constantinople by capitalising on his family connection with the city’s founder and on local eagerness for a resident ruler. It is one of the many fourth-century episodes that John Matthews was the first to penetrate and elucidate effectively.3 Procopius’s usurpation nearly terminated the reign of Valens. This explains why the victor remained suspicious of Constantinople and unforgiving. Valens went on to pass most of his fourteen imperial years at Antioch, but in 378 he threatened to plough Constantinople into the ground on his return from trouncing the Goths.4 * This chapter appeared in its original form under the same title in McGill et al. (2010). It is reproduced here, with minor amendment, by permission of Cambridge University Press. 1 Fundamental are Dagron (1974); Mango (1985); Berger (1987); Grig and Kelly (2012), 6–13. On the development of the senate in particular: Heather (1994); Vanderspoel (1995), 51–70; Errington (2006), 148–61. 2 Based on calculations available in Dagron (1974), 77–86. 3 Matthews (1989), 191–203; Lenski (2002), 68–115. The original version of this chapter appeared in a collection honouring my doctoral supervisor and mentor, John Matthews (Oxford and Yale). 4 Socrates, HE 4.28.5.

6

R E I N V E N T I N G C O N S TA N T I N O P L E

Instead, Valens died in battle with the Goths on 9 August and the city was spared his wrath. By contrast, his successor Theodosius I spent the vast majority of his reign in, and around, Constantinople from the day he first entered the imperial capital on 24 November 380 to August 394, when he left it for what turned out to be the last time.5 In doing so, he set the pattern for generations of Byzantine emperors who succeeded him. This elementary fact is not so evident, however, in most modern accounts of Theodosius where the prevailing picture is that of an emperor forever on the move between East and West, in between which his legislation marks a decisive advancement for Christians and defeat for pagans everywhere.6 Modern accounts tend to concentrate on his years at Milan (388–91), highlighted by his contests with Bishop Ambrose plus his military encounters with barbarians and with the usurpers Maximus (383–8) and Eugenius (392–4). One of their overriding historiographical preoccupations is determining whether Theodosius deserves the soubriquet ‘Theodosius the Great’.7 Except for an original and illuminating chapter of John Matthews’s Western Aristocracies focused on the emperor’s Western courtiers in an Eastern environment and their influence on religious life,8 too little attention has been paid to Theodosian Constantinople and the key role played by Theodosius in transforming the city into an imperial and Christian capital.9 What has obscured the centrality of Constantinople to the reign of Theodosius is an undue dependence on balancing the hostile contemporary account of Eunapius10 with the fifth-century church historians (Philostorgius, Socrates, Sozomen, Theodoret) who wrote from the perspective of the established Christian empire in which Theodosius is assigned a key part.11 Other factors are reliance on purely Western sources such as Orosius and Rufinus, as well as the habit of treating ‘Theodosian Constantinople’ as a single period from 379 to 450 (as do Janin [1964] and Bassett [2004], for example), thereby failing to distinguish properly the very different contributions to the city’s growth of Theodosius I, his son Arcadius, and his grandson Theodosius II. Evaluating the impact of Theodosius I on Constantinople requires devoting greater attention than previously given to the ceremonial and monumental aspects of the city from 380. Doing so sharpens our appreciation of how pivotal the city was to

5 Theodosius was emperor for sixteen years (less two days), or 192 months. Of these, he can definitely be assigned to Constantinople for a total of 109 months (57 percent) or an average of over six months per year – but probably more if documentation permitted greater precision. 6 Typically, Williams and Friell (1995). 7 The quest began with Stein (1959), 191–2; Jones (1964), 169, then led to dedicated chapters in Lippold (1980), 129–35; Leppin (2003), 229–40. Note also Williams and Friell (1995), 171; Ernesti (1998), 475–8; cf. Hebblewhite (2020), 2. 8 Matthews (1975), 101–45. 9 Recent notable exceptions are Leppin (2003), 188–201, but focussed almost entirely on Theodosius’ building activity; Errington (2006), 143–6, focused on administrative and legal aspects; and previously Dagron (1974), 85–95. Now there is Matthews (2012); Hebblewhite (2020), 161–8. 10 Cf. Ernesti (1998), 469–74. 11 Brown (1995), 3–26; Leppin (1996), 105–21; Errington (1997b), 398–443.

7

R E I N V E N T I N G C O N S TA N T I N O P L E

his whole reign. It also highlights how sudden and significant these years were in shaping the urban contours of Byzantine court and civic life for centuries to come.

Theodosius meets Constantinople Theodosius was an experienced general and the son of a general. He was proclaimed at Sirmium in January 379 but had scarcely had time to assemble his own imperial court when he was forced to set out against the Goths for the campaign season of that year. It was a hard fought and not overwhelmingly successful operation for the Romans. With hostilities ended, Theodosius and the imperial retinue wintered at Thessalonica (379/380), during which the emperor fell dangerously ill and was baptised.12 Mindful of Valens’s recent attitude, and fearful that the new emperor from Spain might choose to make Rome-oriented Thessalonica his permanent capital, the senate of Constantinople sent an urgent delegation to persuade Theodosius to come to them instead.13 The senate’s principal advocate was its most distinguished member, the philosopher and rhetorician Themistius, who was an experienced petitioner and confidant of emperors. The argument he put to Theodosius was that ‘the great city’ should ‘receive her protector as soon as possible and meet him before the rest of the east, and . . . that all the gifts which your forefathers decreed might remain secure for her.’ Urging Theodosius to consider increasing the senate’s influence and prestige as his key task, Themistius concluded: ‘If you, O divine eminence, should dedicate such prizes of victory to the great senate, then truly shall your city be a second Rome’.14 Despite a further unsuccessful encounter with the Goths in 380 Theodosius made his triumphal entrance into the ‘second Rome’ on 24 November 380.15 It was probably the first time he had ever seen the city and he would have been welcomed in style, although to the locals he was a virtual ‘blank page’.16 The imperial adventus would have involved a formal reception outside the city walls that was then repeated inside. The prefect of the city, Restitutus, would have led the local delegation, followed by acclamations and formal panegyrics by the city’s most renowned orators.17 Themistius must have been foremost among the group, although no such oration survives. Likewise, the praetorian prefect of the East, Neoterius, appears to have been there too.18 12 Leppin (2003), 35–54; Errington (1996a), 438–53. 13 Errington (1997a), 23 and (2000), 893–4. 14 Themistius, Oration 14.5 (183b–184a, Maisano [1995], 540–3), trs. Heather and Moncur (2001), 229–30. 15 McCormick (1986), 41; Errington (1996b), 10. 16 The striking phrase of Leppin (2003), 58, cf. Errington (1997a), 22. 17 Restitutus (PLRE 1, 764 [‘Restitutus 1’]) is most likely to have been Prefect at the time since his successor Pancratius (first attested as Prefect on 30 July 381) was still comes rei privatae on 20 September 380 (PLRE 1, 664 [‘Pancratius 4’) and possibly arrived in Constantinople with Theodosius. 18 PLRE 1, 623 (‘Flavius Neoterius’).

8

R E I N V E N T I N G C O N S TA N T I N O P L E

Theodosius had spent most of his adult life on campaign every summer, living with his soldiers and wintering in nearby camps and towns. Now he was taking possession of a relatively new and marvellous metropolis but as an already baptised Christian. Martial emperors may not yet have spent sufficient time at Constantinople to establish a ceremonial, cultural, and administrative dominance over the city, but it was always an imperial capital. As early as January 381, however, it appears that Theodosius had deliberately shifted the focus of imperial life and purpose from military exploits where he had become increasingly unsuccessful to civic and urban development,19 beginning with the palace overlooking the Bosporus that had always needed to be maintained. The emperor, his army and courtiers could require the palace at short notice, especially a new emperor created locally, as Procopius was in 365.20 When Theodosius had first settled into the palace, he evidently found that some of the staff had outlived their usefulness so had them pensioned off, but he was later forced to protect them from local financial obligations.21 Theodosius arrived in the city accompanied by his family, including his wife, Flaccilla, and his infant son Arcadius; a large retinue of imperial guards; and a group of essential courtiers such as Florus, the magister officiorum who was to succeed Neoterius as praetorian prefect of the East within a few weeks.22 His entourage also included a significant posse of relatives and family friends, in fact ‘an entire clan moved in with him to dominate the court life of Constantinople’,23 as Matthews put it. Once there he would soon have encountered the remnants of earlier imperial households, including the surviving wives of former emperors, namely, Domnica, the spouse of Valens; Charito, the spouse of Jovian; and perhaps even Procopius’s wife, Artemisia, who had been reduced to undignified and sightless penury.24 Presumably, their imperial children were still alive too. Carosa and Anastasia were the daughters of Valens and Domnica,25 while Varronianus, the son of Jovian and Charito, had been an infant consul in 363 and was still only a young man in 380. Charito and Varronianus lived in fear of their lives, perhaps from those who saw him as a potential usurper, like Procopius, especially since he had once been a designated imperial heir (nobilissimus).26 Otherwise, the city’s memory of court and imperial life was selective and patchy. 19 Proclaimed by Themistius Oration 15 (Maisano [1995], 546–77) but explained by Heather (2010), 189–91. 20 In 387 Libanius hoped that Theodosius and his court might still come to the home of his predecessor at Antioch (Orations, 20.46–7). Presumably, the imperial palace there was still being kept ready. 21 CTh. 6.35.11 (3 February 381). 22 PLRE 1, 367–8 (‘Florus 1’). 23 Matthews (1975), 111. 24 John Chrysostom, ad viduam, 4 (PG 48.604), with PLRE 1, 201 (‘Charito’); 265 (‘Domnica’); 111–12 (‘Artemisia’). 25 PLRE 1, 58 (‘Anastasia 2’); 182 (‘Carosa’). 26 Philostorgius, HE 8.8, also hinted at by Themistius addressing Theodosius in 383 (Oration 16.8, 204c [Maisano (1995), 590–1], trs. Heather and Moncur (2001), 271.

9

R E I N V E N T I N G C O N S TA N T I N O P L E

Constantinople meets Theodosius The Constantinople that Theodosius first encountered was bustling and boisterous. It was also disjointed and discordant. Not long before Theodosius set foot in the city, Gregory Nazianzen, who had been there for over a year guiding the small and alienated Nicene community lauded the city for having ‘walls and theatre and racecourses and palaces, and beautiful great porticoes and that marvellous work the underground and overhead river [aqueduct of Valens] and the splendid and admired column [of Constantine] and the crowded marketplace and a restless people and a famous senate of highborn men’.27 At the same time, the city was a cultural and intellectual centre that attracted distinguished teachers and ambitious students such as Jerome.28 The populace, estimated at around 250,000 at this time, was fed by the fertile fields of Egypt with grain transported to Constantinople every year, stored in large warehouses, and carefully distributed, so Gregory could greet the advancing grain fleet in 380 as a most pleasant sight.29 Just before that, the Goths who approached the city were awestruck by its size and beauty, the housing blocks they could see jutting out in the distance above the walls, and the evidently dense population.30 In fact, shortly after his own arrival Theodosius welcomed the Gothic king, Athanaric, and just a few days later he was leading the king’s funeral procession through the city. On encountering Constantinople Athanaric exclaimed: ‘I have seen what I have often heard of though I did not believe it’.31 Within days of Theodosius’s entry Gregory noted the crowded urban environment: ‘The marketplaces were full, the colonnades, streets, every place, two and three storey houses were full of people leaning out, men, women, children, the very aged’.32 It was a similar scene evoked not long after by Gregory of Nyssa in his snapshot of the funeral procession at Constantinople in mid-381 for Bishop Meletius of Antioch. He describes ‘the people in their myriads, so densely crowded together as to look like a sea of heads, became all one continuous body, and like some watery flood surged around the procession bearing his remains.’ As the procession jostled along into the darkness, you could see ‘how the streams of fire, from the succession of lamps, flowed along the unbroken track of light, 27 Themistius, Oration 33.6 (Maisano [1995], 984–7). 28 Jerome was attracted to Constantinople by Gregory Nazianzen (Kelly [1975], 69–71), who considered the city to be ‘distinguished by the eminence of its rhetorical and philosophical teachers’ (Orations, 43.14) when his friend and fellow student Basil went there, just as it was adorned earlier by pagan luminaries such as Maximus (Eunapius, VS 477 [Wright 442–3]) and Libanius (Eunapius, Lives of the Sophists, 495 [Wright 520–1]). 29 Themistius, Oration, 34.7 (Maisano [1995], 998–1001). Eunapius, on the other hand, considered this a symptom of the pampered rapaciousness of Constantine’s city (Eunapius, Lives of the Sophists, 462 [Wright 382–3]). 30 Ammianus Marcellinus, res gestae 31.16.7. 31 Jordanes, Getica. 143. 32 Gregory Nazianzenus, Poems 2.1.1331–5, trs. White (1996), 114.These are the insulae, apartment blocks, specified in the Notitia being studied by Matthews (2012), 81–115.

10

R E I N V E N T I N G C O N S TA N T I N O P L E

and extended so far that the eye could not reach them’.33 Then Meletius’s body was transported overland to Antioch but Theodosius had requested, contrary to custom, that his casket be solemnly received in every city, like the arrival of a governor or emperor, and accompanied by singing of psalms just as the traditional civic adventus involved laudatory choirs.34 Almost instantly the Spanish emperor had created a new ceremonial model by fusing secular and religious practice at Constantinople. The hippodrome and theatre were major attractions for the people of latefourth-century Constantinople. News of any imperial victory, however marginal, was celebrated in the hippodrome. The announcement of imperial military success on 17 November 379, for instance, involved races,35 and the victory spectacle was repeated in 380.36 Doubtless one of Theodosius’s first public displays was a triumph in the hippodrome. Other celebratory occasions were more measured. The birthday of Constantine’s city in 330 was commemorated on 11 May each year with a colourful cavalcade through the main street (Mese) and around the hippodrome. The central feature was a statue of Constantine carefully mounted in a chariot, which progressed with attendants carrying candles and dressed in short white cloaks. They entered the hippodrome through the starting gates and carried the statue to the Stama, opposite the normally empty imperial box, where the emperor’s representation could be acknowledged and could then preside symbolically over the anniversary races.37 Yet, such displays of civic solidarity and unity at Constantinople masked the fact that the locals were still scarred by fear of the Goths. War widows were still in mourning.38 Above all, the Constantinople that greeted Theodosius was deeply fissured by competing religious allegiances, whose passionate tone is captured in Gregory of Nyssa’s contemporary observation that when asking the price of some goods or even the price of bread one is likely to get into an argument about whether the son is ‘begotten or unbegotten,’ or is of the same substance as the father or not.39 This religious ferment had been exacerbated by the law the emperor had issued at Thessalonica the previous winter in which he prescribed the Orthodox Catholic religion as being support for the doctrinal tenets of the bishops of Rome and Alexandria.40 The Arians, or Homoeans, used to enjoying imperial support under Constantius II and Valens, were dominant, and they held the city’s main churches, those of the recently consecrated Holy Apostles and Saint Eirene in particular. 33 Gregory of Nyssa, In Meletium (Spira [1967a], 456). 34 Sozomen, HE 7.10.5, with Geyssen (1998), 53–4; Van Nuffelen (2012). 35 Marcellinus, Chron. 379.2 (MGH.AA, XI 60), Chron.Const. 379.3 (MGH.AA., IX 243) with McCormick (1986), 41–2. 36 Chron.Const. 380.1 (MGH.AA., IX 243). 37 Malalas, Chron. 13.8 (Thurn 246–7); Parastaseis, 5 (Preger [1901], 21.8–17, and 38; (Preger [1901], 42.7–14); Patria 2.87. (Berger [2013], 108–11), with Bauer (2001), 32–3. 38 John Chrysostom, ad uiduam 4–5 (PG 48.604–6). 39 Gregory of Nyssa, de deitate filii et spiritus sancti (PG 46.557–8). 40 CTh. 16.1.2 (28 February 380).

11

R E I N V E N T I N G C O N S TA N T I N O P L E

Then there were the Apollinarians as well as the followers of Macedonius, who held their own church, the adherents of the strict Novatian, and Eunomius’s supporters, known as ‘Anomoeans’, who gathered with him in various welcoming mansions.41 The orthodox Nicene congregation nurtured by Gregory Nazianzen was centred on the chapel of Anastasia in the portico of Domninus. The community was not huge and the chapel was just a large reception room in the mansion of Constantine’s praetorian prefect, Ablabius.42 It was around this time that the Western pilgrim Egeria passed through the city, where she reported that ‘when I had arrived there, I went through all the churches – that of the Apostles and all the martyr-memorials, of which there are very many’.43 Faced with this plethora of church practice and belief that had produced a range of competing congregations centred on particular churches and a deeply factionalised community, Theodosius’s instinct was to bring them all together and let them find their common ground. Theodosius was himself prepared to listen to what the local religious leaders and disputants had to say, which encouraged palace officials to believe they could dissuade the emperor from his orthodox position. Further, his doctrinally resolute wife, Flaccilla, insisted that he keep away from Eunomius in case he unduly swayed the emperor.44 No less disconcerting to the empress was perhaps his preparedness to allow the Homoean bishop Demophilus to repent his heresy.45 However, the Homoeans and Demophilus were ousted from their pre-eminent position in November 380, but it took a highly charged demonstration of brute force to install Gregory as his replacement in the Church of the Holy Apostles.46 By May 381, all the bishops of the east had assembled at Constantinople at Theodosius’s invitation, and in the Church of Saint Eirene they produced an immediate settlement, although further synods had to be held in 382 and 383 before a new doctrinal equilibrium emerged, culminating in a formal hearing before the emperor in the palace in which he considered the written depositions of the bishops of each party before opting for the Nicenes.47 Even then the situation was delicate, and a backlash always lingered just below the surface. The false rumour of Theodosius’s defeat by Maximus in 388 was enough to incite Arians to burn down the house of the bishop Nectarius.48 One of the momentous decisions of the council of 381 (canon 3), however, was the pronouncement that Constantinople should hereafter take ecclesiastical precedence after Rome because it was the ‘New Rome’.49 Socrates, HE 5.20.4; Sozomen, HE 7.17.1. Van Dam (2002), 139–41. Peregrinatio Aetheriae, 43. Sozomen, HE. 7.6.3. As argued by McLynn (2010), 218–20. Gregory Nazianzenus, Poems 2.1.1325–51, trs. White (1986), 114. Socrates, HE. 5.10.21–8. Bishops such as Amphilocius of Iconium clearly had privileged access to the imperial palace under Theodosius (Sozomen, HE. 7.6.4–7; Theodoret, HE 5.16). 48 Sozomen, HE 7.14.5. 49 Errington (2006), 229. 41 42 43 44 45 46 47

12

R E I N V E N T I N G C O N S TA N T I N O P L E

With the emperor and his court established at ‘New Rome’, at least for the time being, the orthodox community consolidated its position despite the religious indifference of most imperial officials and palace staff,50 as well as the need to dispense with the services of those who insisted on adhering to their support for Eunomius.51 They achieved this by a doctrinal clearing of the urban liturgical space (churches, squares, streets) and by expanding their control over it. The key churches were now in orthodox hands, with heretics relegated beyond the city52 and their ministers flushed out of hiding places.53 Then the local orthodox liturgical forms and calendar expanded to fill the space but were linked to the life and fortunes of the city as a whole. By the time Theodosius arrived in Constantinople, the city was used to night vigils and processions through the streets as well as communal singing and chanting on the great feasts, but they were dominated by the Homoeans. Now the orthodox appropriated them.54 The so-called stational liturgy, developed at Jerusalem and imported to Rome, took on a distinctive imperial character at Constantinople as it centred around local events and new local spaces such as the Forum of Constantine.55 So too, many key festal days were already part of the annual rhythm of urban life. Apart from Easter, there was Epiphany on 6 January, and on 25 December 380 Theodosius witnessed the first celebration of Christmas at Constantinople.56 Certain saints’ days were also now part of the local calendar: Cyprian, for example, and even Athanasius, as well as the local Constantinopolitan martyrs Acacius (8 May) and Mocius (11 May).57 Other local anniversaries had also begun to be celebrated: Constantine and Helena on 21 May,58 the dedication of the Church of Saint Menas on 21 September,59 the transfer of relics of Andrew, Luke, and Timothy to Constantinople and their deposition in the imperial Church of the Holy Apostles,60 and the military martyr Theodore on the first Saturday in Lent.61 Indeed, the only extant sermon of Bishop Nectarius (possibly mid-390s) was delivered on the feast of Theodore and mentions that the annual commemoration is now well established

50 51 52 53 54 55 56 57 58 59 60 61

A particular complaint of Gregory Nazianzen (Orations, 42.26, cf. 36.11–12). Philostorgius, HE 10.5. CTh. 15.5.6, cf. 16.5.12 (3 December 383). CTh. 16.5.13 (21 January 384). Baldovin (1987), 184, 257; Jacobs (2012), 145–8. Baldovin (1987), 211–12, 233, 263. Gregory Nazianzen, Orations, 38 (Christmas); 39–40 (Epiphany 381). Gregory Nazianzen, Orations 24.1 (Cyprian, on 2 October 379); 21.5 (Athanasius, on 2 May 379); Typicon CP, Mateos (1962–3), vol. 1, 285.19–21 (Acacius), 291.3–4 (Mocius). SEC 700.31–7. Typicon CP, 21 September (Mateos 1962–3), vol. 1, 40.26–7. For the church itself: Janin (1969), 333–5. Details in Burgess (2003), 28–34. Typicon CP (Mateos 1962–3), vol. 2, 20.1–2, also on 17 February (Mateos 1962–3), vol. 1, 234.8– 11. The annual liturgy was celebrated in the church of Saint Theodore in the quarter ta Sphorakiou (Janin [1969], 152–3).

13

R E I N V E N T I N G C O N S TA N T I N O P L E

at Constantinople.62 To these feasts was added immediately a commemoration on 3 August of the local Council of 381.63 All these new annual feasts still survived in the eleventh-century Byzantine liturgical calendar, and the number of such local feasts quickly grew under Theodosius.

Imperial ritual and ceremonial The presence of the emperor and his court also refocused imperial ritual and ceremonial in and around the palace itself, where to date it had only been witnessed spasmodically. Immediately, the established rite of the statue of Constantine being honoured in the hippodrome on Constantinople’s foundation anniversary on 11 May was halted.64 It could now be replaced with a real flesh-and-blood emperor keen to stamp his own claim on the city. Major imperial milestones called for special celebrations. Imperial births had been a relatively rare event in recent decades. At Constantinople, however, Theodosius fathered five children born into the purple: Honorius 384,65 Pulcheria 385,66 Gratian 388,67 Galla Placidia 392,68 and John 394.69 In later Byzantine times such imperial births triggered several days of ceremony inside the imperial palace and in the hippodrome. Having assembled fifty members of each faction on the fourth day after the birth, the imperial chamberlain (praepositus) would then address them as follows: ‘Our sacred emperor requests that, in accord with standing traditional custom and ancient practice you assemble tomorrow and proclaim the name of the infant born in the purple’. It is possible that this ‘ancient practice’ was initiated or took its essential Byzantine shape in the 380s and 390s.70 In any event, each year from the elevation of Arcadius in 383 to 394 the imperial birthdays of both Theodosius (born 11 January) and Arcadius (date unknown) were celebrated. In 393 and 394, after Honorius (born 9 September) had also become emperor, there were three imperial birthdays, although the only one Honorius actually celebrated at Constantinople was on Friday, 9 September 393. Theodosius decreed in 389 the annual celebration of both the emperor’s birthday and his dies imperii, the anniversary of his accession on 19 January 379 (plus that of Arcadius on the same day after 383, and Honorius on 23 January in 393 and 394).71 He was later forced to declare that Sunday was sacred and a day the Nectarius, de festo S. Theodori (PG 39.1832D). Typicon CP, 3 August (Mateos [1962–3], vol. 1, 358.17–18 [H]. Parastaseis, 5 (Preger 21.11), Patria 2.87 (Berger [2013], 108–11). PLRE 1, 442 (‘Fl. Honorius 1’). PLRE 1, 775 (‘Pulcheria’). Not in PLRE 1, but see Rebenich (1985), 372–85). PLRE 2, 888–9 (‘Aelia Galla Placidia 4’). Not in PLRE 1, cf. Rebenich (1985), 376. Const.Porph., On the Ceremonies 2.21 (Reiske 615.17–619.16, trs. Moffatt and Tall [2012], 615–19). 71 CTh. 2.8.19 (7 August 389). 62 63 64 65 66 67 68 69 70

14

R E I N V E N T I N G C O N S TA N T I N O P L E

hippodrome should be closed, but not if the emperor’s birthday fell on a Sunday.72 Theodosius was obviously anticipating his next birthday on Sunday, 11 January 393. To judge from earlier and later examples, each of these events was accompanied by games and celebrations in the hippodrome and in other public spaces. In later Byzantine times the centrepiece of the memorial day was a palatial banquet, which may already have been the case in Theodosius’ time.73 Imperial anniversaries also gave rise to orations such as those Themistius delivered on 19 January 380 (Or.15) and 19 January 384 (Or.18).74 The sheer regularity of these celebrations in the 380s and early 390s rapidly consolidated the accompanying ritual. Thus was invented what became the Byzantine ceremonial for imperial births and their annual celebration.75 Proclaiming a new emperor was one of the most significant imperial events. In 383 at the tribunal at the Hebdomon, on the fifth anniversary of his father’s proclamation on 19 January,76 Arcadius was crowned Augustus.77 Ten years later Theodosius’s next eldest son Honorius was proclaimed there on 23 January.78 In addition, by the time the fifteen-month-old Honorius entered his consulship in 386, he already held the rare but official title of nobilissimus puer, which meant he was an emperor-designate. The title must have been conferred on him sometime in 385 at a palace ceremony similar to that recorded for the title in later times.79 The empress Flaccilla was also elevated with even greater pomp to the no less rare title of Augusta in 383, so she could have her own coinage and honorary statues, on both of which she was represented with the accoutrement of an emperor.80 In recent decades the traditional Roman ritual of proclamation had been witnessed at widely scattered points including Paris (Julian), on the Persian frontier (Jovian), at Nicaea (Valentinian), Constantinople (Valens, Procopius), Amiens (Gratian), and Sirmium (Valentinian II, Theodosius). Suddenly it was exclusively concentrated at the ‘second Rome’. Arcadius’ proclamation in 383 was the first in a long line of Byzantine emperors stretching for centuries ahead and the Byzantine empresses after Flaccilla adopted her first name ‘Aelia’. Both the ceremonial and

CTh. 2.8.20 (17 April 392). Const.Porph. On the Ceremonies 1.61 (Reiske 277.1–278.4, trs. Moffatt and Tall [2012], 277). Following the date of Errington (2000), 895–6. Moffatt (1996). Kent (1993), 80. Socrates, HE 5.10.5; Sozomen, HE 7.12.2; Philostorgius, HE 10.5; Cons.Const. 383.1 (MGH.AA., IX 244), Marcellinus, Chron. 383.2 (MGH.AA., XI 61). 78 Socrates, HE 5.25, Sozomen HE 7.24.1; Philostorgius, HE, 11.2; Lib., Ep 1100, Marcellinus, Chron. 393 (MGH.AA., XI 63). 79 References in PLRE, 442 (‘Fl Honorius 1’); and CLRE, 306–7 (consulship of 386). For the later investiture ceremony: Const.Porph., On the Ceremonies I.44 (53) (Reiske 225.15–229.4, trs. Moffatt and Tall [2012], 225–9). 80 Holum (1982), 29–44. For the later ceremonial: Const.Porph., On the Ceremonies 1.40 (Reiske 202–7, trs. Moffatt and Tall [2012], 202–7). 72 73 74 75 76 77

15

R E I N V E N T I N G C O N S TA N T I N O P L E

the meaning of proclamation developed over time but its customary Byzantine core was essentially established in the time of Theodosius.81 Every fifth year, generally beginning on his dies imperii, an emperor would commemorate his anniversary with games, statues, and donatives to the soldiery of specially minted commemorative coins. Vows (vota) were discharged for the previous five years and renewed for the next five. Theodosius began his fifth year on 19 January 383 by crowning Arcadius as emperor. Thereafter the two emperors, father and son, celebrated their imperial anniversary on the same day. This duplex celebration not only amplified the imperial dignity and authority at Constantinople; it also constituted a significant dynastic statement. Beginning on 19 January 388 Theodosius marked his tenth year and Arcadius his fifth ‘with exhibitions and games’.82 This was the very occasion that gave rise to the celebrated silver dish, or missorium, of Theodosius, in which he and Arcadius are depicted in the tribunal of the imperial palace handing the codicil of appointment to some official.83 On the same day in 393 Theodosius marked his fifteenth and Arcadius his tenth regnal year. At least, that was the normal pattern. An imperial marriage was also a splendid and colourful occasion full of light and striking scents, song and merriment. We catch a furtive glance in 386 at the celebrations for the union of Theodosius’s favourite niece Serena, ‘clad in scarlet’, and his promising young general Stilicho. The city was bedecked with flowers and the Bosporus glittered with the torches carried in procession, as the empress Galla was standing in as mother of the bride ‘ordering the bridal veil beneath a weight of jewels’.84 All the more resplendent and memorable would have been the day when Galla herself had married Theodosius the previous year, or when the emperor Arcadius was married in May 395.85 Once more the ritual for Byzantine imperial marriages was evidently accelerated in the late fourth century.86 Imperial funerals were no less public and ceremonial. Gregory of Nyssa was chosen to present an encomium in honour of the young princess Pulcheria, who died in July 385. A few weeks later her mother, Flaccilla Augusta, also passed away, and again Gregory was pressed into service. In the course of his eulogistic set-piece, he depicts the funeral cortege of the empress passing through the streets of Constantinople. Both ‘strangers and residents’ wailed loudly at the sight of her 81 Const.Porph., On the Ceremonies 1.38–9 (Reiske 202–7, trs. Moffatt and Tall [2012], 202–7), with Bauer (2001), 40–2. 82 Cons.Const. 387.1 (MGH.AA., IX 244); Marcellinus, Chron., 387.1 (MGH.AA., XI 62). Numismatic data suggests that the celebrations may not actually have been synchronised (Kent [1993], 83). 83 Ernesti (1998), 133–43; Leader-Newby (2004), 27–49. The detailed iconographic argument seeking to date the missorium to 421 and identify the emperor as Theodosius II, which was developed by Meischner (1996), 389–432, is fatally flawed. Theodosius II celebrated his vicennalia in 411. To claim that emperors were cavalier about dating their vota celebrations (Meischner [1996], 425–6) is itself cavalier. 84 Claudian, de Consulatu Stilichonis, 1.80–8. 85 Chron.Pasch. 395 (565–566 Dindorf). 86 Const.Porph., On the Ceremonies 1.82 (Reiske 380.1–22, trs. Moffatt and Tall [2012], 380).

16

R E I N V E N T I N G C O N S TA N T I N O P L E

coffin draped in purple and glittering gold and carried on the empress’s litter. As it passed by ‘people of every rank and age rush out, they marvel at this sight visible to all. Enthusiastically following on foot in a great throng and giving vent to grief’.87 It was a repeat of the crowded scene of concentrated grief witnessed at the funeral liturgy for young Pulcheria. ‘I have seen a sea of men crammed together’; so Gregory depicts the scene, ‘the full temple, the vestibule, the open expanse before it, people in mourning, the nearby streets, public areas, the side streets and houses. Wherever one looks there are crowds of people as if the entire world had run together for this tragedy’.88 Similar outpouring of communal sorrow will also have accompanied the funerals of the young prince Gratian in 388 and new-born John in 394. Theodosius himself died in Milan in January 395, but his body was transported to Constantinople and laid to rest there alongside both his wives Flaccilla and Galla, who had died in childbirth as he was setting out on campaign in August 394.89 Exactly where his predeceased children were buried is not clear. By then Theodosius had systematically fixed and promoted the city as the imperial capital by bringing into it the remains of as many of his predecessors as possible. Constantine had set up his own mausoleum and was buried there. Constantius II joined him in the mausoleum in 361.90 Julian’s status as an imperial apostate could even be turned to advantage by a Christian emperor. Accordingly, Theodosius had Julian’s body brought from Tarsus, where he had been buried in 363, and at the same time he evidently transferred from Rome the body of Julian’s wife, Helena. They were reunited in the early 390s in a round porphyry sarcophagus located in a portico just outside the mausoleum of Constantine at the Church of the Holy Apostles.91 Jovian’s body had already been transferred from the place where he died in Asia Minor. Now he was reburied in the mausoleum portico, and his wife, Charito, was buried beside him when she died sometime in the reign of Theodosius.92 Since Valens’s body was never found, he required no burial, but it is not known what happened to his wife Domnica, who also probably died in Theodosius’s time. The western emperor Valentinian, brother of Valens, had died in Pannonia in 375, and his body had been brought to Constantinople in 376. Yet it was left to Theodosius in 382 to arrange for his sarcophagus to be transferred to the imperial mausoleum stoa accompanied by due ceremonial.93 The ashes of Valentinian’s son Gratian had been acquired for Milan by Ambrose, who treasured them, but Theodosius transported to Constantinople the body of Gratian’s first

87 88 89 90 91 92 93

Gregory of Nyssa, in Flaccillam (Spira [1967c], 478), with Holum (1982), 22–9. Gregory of Nyssa, in Pulcheriam (Spira [1967b], 463). Grierson (1962), 43; Johnson (1991a), 332. Ammianus Marcellinus, res gestae 21.16.20. Grierson (1962), 40–1; Kelly (2003), 591–4, contra Woods (2006), 364–71. Grierson (1962), 41–2. Marcellinus, Chron. 382.1 (MGH AA. XI 61); Cons.Const. 382.1 (MGH AA.IX 243), with Johnson (1991b), 502–3.

17

R E I N V E N T I N G C O N S TA N T I N O P L E

wife, Constantia, daughter of Constantius II. On 1 December 383 her body was solemnly placed in an imperial sarcophagus at the Church of the Holy Apostles.94 Theodosius had gone to considerable lengths to bring together into a single precinct in the imperial capital every imperial corpse he could find. By the early 390s he had retrospectively turned the Mausoleum and the Church of the Holy Apostles into a truly imperial resting place and conceptually linked the two buildings.95 This sudden concentration of imperial inheritance could only strengthen the authority and dynastic prestige of Theodosius, who had erected a solid imperial phalanx protecting the city’s Apostolic shrine. As John Chrysostom observed, ‘At Constantinople those who wear the crown think themselves fortunate to be buried not near the apostles but outside on the threshold of the basilica [of the Holy Apostles]’. He went on to explain that ‘from now on the emperors are the doormen of sinners and in their eyes and the eyes of their descendants that is no shame but an honour for their ashes’.96 By the early sixth century the mausoleum populated by Theodosius’s ‘doormen’ was full and another mausoleum was built by Justinian, which became thereafter the resting place of successive Byzantine emperors. Other imperial occasions gave rise to urban ceremonial and Theodosius legislated on 2 February 383 to specify some of them, essentially reinforcing the prescript of his predecessors that certain formal public announcements should be dignified and free: ‘whenever any of our auspicious achievements are announced, if wars should cease, if victories should arise, if the honour of the bestowal of royal vestments should be added to the calendar [that is, an imperial consulship], if the announcement of the tranquillity of peace that has been concluded is to be spread abroad, if by chance we display the imperial countenances (sacros vultus) to the eager multitudes’.97 Among other events, this advice covered victory announcements at Theodosian Constantinople in 379, 380, 386, 388, and 394, as well as the establishment of peace with the Goths in 382, which was cast as a triumph,98 and probably that with the Persians in 385/6, when peace was agreed to.99 In fact, Constantinople celebrated more triumphs and victories in the 380s than it had in all its previous life.100 Also included were imperial consulships in 380 (Theodosius), 385 (Arcadius), 386 (Honorius), 387 (Valentinian II), 388 (Theodosius), 390 (Valentinian II), 392 (Arcadius), 393 (Theodosius), and 394 (Arcadius and Honorius). Claudian expressed the hope that Theodosius would continue to enjoy such festal days.101 94 Cons.Const. 383.2 (MGH AA, IX 244); Chron.Pasch. 383 (563.2–6 Dindorf), with Grierson (1962), 6. 95 Grierson (1962), 25. 96 John Chrysostom, contra iudaeos et gentiles 9 (PG 48.825). 97 CTh. 8.11.4. 98 Themistius, Or.16.1, 199c (Maisano [1995], 580–1), trs. Heather and Moncur (2001), 265 n.189. 99 Greatrex (2000a), 35–48. 100 McCormick (1986), 41. 101 Claudian, Carmina minora 32.20–1.

18

R E I N V E N T I N G C O N S TA N T I N O P L E

The traditional imperial ceremonial surrounding births, marriages, installations, anniversaries, triumphs, and deaths began to take on a distinctly liturgical flavour at Constantinople in the era of Theodosius, although this has usually been considered a later development.102 Candlelit processions along the city’s porticoed backbone, accompanied by hymns, began to spread. The monumental and physical layout of the city facilitated the development of a close relationship with the population and continued to grow under Arcadius and later emperors. It exposed the emperor to his officials and people in new ways, while the senate and imperial court were also now in a close political and social relationship that was partly expressed in ceremonial engagement. When Theodosius finally set foot in Rome in 389 and spent several months there, the locals were struck by the emperor’s public visibility, as highlighted by Pacatus: ‘You frequently emerge [from the palace] and you show yourself to the waiting people, and being willing not only to let yourself be seen, but to be approached readily’.103 Theodosius was transposing to Rome the emerging imperial ritual of Constantinople by openly presiding at civic events and meeting with people in public. At Rome seeing the emperor out on the city streets surrounded by clergy and courtiers was a novelty. Those who could remember were used to more closeted emperors. Theodosius’s blending of secular and religious ceremonial was especially evident in his regular acquisition of the relics of saints and martyrs, beginning with the body of the exiled former bishop of Constantinople, Paul, in 381. The spiritual and ceremonial emphasis came to be placed on the actual arrival of the relics, the traditional adventus, which symbolised God’s mercy for the city and its divine destiny.104 The emperor’s direct role in cradling the holy relics placed him at the centre of this unifying new ceremonial, which brought together as common suppliants the court and clergy, aristocracy and general populace. The transfer of Paul involved Theodosius carefully carrying the former bishop’s skull in procession through the city, depositing it in the church previously occupied by the Macedonians, and formally changing the name of the church to ‘Saint Paul’.105 Theodosius was also responsible for bringing to Constantinople the remains of the African martyrs Terentius and Africanus with solemn deposition in the church of Euphemia,106 and in 391 the head of John the Baptist reached the city. Theodosius received the skull of John encased in its precious relic box, then carried it himself 102 Diefenbach (2002). 103 Panegyric on Theodosius 21.2 (trans. Nixon). This is not mere panegyrical exaggeration. Theodosius’s accessibility was also noted by Rufinus (HE 2.19) and Gregory Nazianzen who tells us that the emperor ‘treated me with respect from our first meeting both in the way he spoke and in listening most kindly’ (Poems 2.1.1307–8, trs. White [1986], 113). 104 Brown (1982), 92–3. Van Nuffelen (2012), 181–8 proposes that, in Theodosian Constantinople, there was less imperial control over ceremonies. The emperor participated separately in religious and secular processions rather than in a blended fashion as suggested here. 105 Socrates, HE 5.9.1–2, Sozomen, HE 7.10.4. For the church itself: Janin (1969), 394–5. 106 Nicephorus Callistus, HE 2.62 (PG 86.213A).

19

R E I N V E N T I N G C O N S TA N T I N O P L E

folded in his purple imperial cloak all the way out to the Hebdomon, where he deposited it in his newly built church.107 The removal of John’s head to the church at the Hebdomon enhanced its significance as a location where emperors were legitimated and presented, originally in the military ceremony of being raised on a shield.108 The new presence of John formed part of a wider process at Constantinople under Theodosius that saw the replacement of the traditional military ideology with a sanctified civilian one.109 Reinforcing this process, the emperor deliberately employed the Hebdomon as the departure point for the campaign against Eugenius in 394. There he prayed for victory in the new church, fortified by the prophecy of the holy man John of Lycopolis whose prediction Theodosius had first sought out, both in 388 and 394.110 On hearing of the death of the monk Isaac on the outskirts of the city on 26 May 383, Theodosius had his body brought to Saint Eirene for a vigil. The next day the emperor, patriarch Nectarius, along with all the clergy and people, bore Isaac’s remains to his grave, accompanied by psalms and hymns.111 These processions served to consolidate further the status and dynastic prestige of Theodosius and his family.112 There were many more of them in the time of Arcadius and Theodosius II, and they are often thought to have originated then,113 but they followed the model for relic receptions established by Theodosius I. All of these processions and receptions at Constantinople were infused with colour and pageantry and they contributed to the emperor’s concern for improving the standards of formal public dress and deportment. Senior civil and military officials were to progress through the city in fine carriages, while a fierce fine or even banishment from the senate could result for any senator not complying with the new Theodosian public dress code.114 Each new type of procession became an exemplar for the next and thus the ritual and practice of Byzantine imperial ceremonial began to be codified. An integral part of the ritual was the involvement of the patriarch and, frequently, the patriarchal churches of Hagia Sophia and Saint Eirene.115 Except for the first few months, Nectarius was patriarch for the whole of Theodosius’s reign. This continuous partnership of emperor and patriarch from 381 to 394 was crucial. Nectarius was a distinguished senator who was surprisingly

107 108 109 110

111 112 113 114 115

Sozomen, HE 7.21.5. Diefenbach (2002), 26–7. MacCormack (1981), 242. Theodoret, HE 5.24; Augustine, City of God 5.26; Cassian, Conferences, 1.24.26, Institutes, 4.23; Sozomen, HE 7.22, Rufinus, HE 11.32; Palladius, Lausiac History 35.2. John had a well established reputation and was consulted by officials from everywhere (Sozomen, HE 6.28; Sulpicius Severus, Dialogues, 1.20; Augustine, De cura pro mortuis gerendis, 21). Vita Isaacii 17 (Acta Sanctorum, Mai VII, 258). There is uncertainty about the date and identity of Isaac, but the information of the vita may well be correct (cf. Dagron [1970], 233, 245–6). Mergiali-Sahas (2001), 146. Klein (2006), 84–6. The various relevant laws are discussed in Errington (2006), 165–6. Berger (2001), 74–85.

20

R E I N V E N T I N G C O N S TA N T I N O P L E

nominated, then chosen for the position by the emperor himself. He formed a link between the ecclesial and civil aristocracy. Through the 380s and 390s he also helped define the emperor’s central role in the local liturgy and in other religious occasions such as relic processions.116 Moreover, some of these original Theodosian events and processions came to be commemorated annually thereafter. So, in the eleventh century the Byzantines still celebrated, with an annual liturgy, events such as Theodosius’s transfer of Paul’s relics on 6 November117 and the death of Flaccilla on 14 September.118 All these events provide a clear indication of the significance of Theodosius’s years at Constantinople for the formation of the Byzantine liturgical calendar.

Monumental and decorative settings Constantinople may already have been close and crowded when Theodosius arrived in 380, but he continually expanded it during his reign and improved its street plan.119 The prolonged location of the court there attracted an increased population of officials and soldiers, teachers and students, petitioners and craftsmen, both permanent and transient, which put new pressure on the city’s basic infrastructure of shelter, water, and food. So, by 393 the city prefect Aurelian had to be reminded that the senior military figures receiving public rations (annonae civicae) in exchange for building houses must meet their obligations.120 Contributing to the necessary restoration of the city’s port and water supply was another obligation that could not be avoided.121 Theodosius’s restoration consisted of extending the aqueduct of Valens further across the city122 and building a new harbour on the Marmara side of the city at the mouth of the Lycus river, which expanded capacity to unload the extra grain ships and other regular cargoes required for provisioning the city.123 Recent excavations for the new Marmaray rail tunnel have unearthed the original Theodosian harbour, which lasted for over a millennium and has already led to some spectacular finds of over thirty later Byzantine ships and

116 Bauer (2001); Berger (2001). In Milan, on the other hand, Bishop Ambrose refused to allow the emperor onto the altar in the wake of the imperial sanction for the massacre at Thessalonika in 390 (Theodoret, HE 5.18.1–19). This episode later became a talismanic one for the Byzantine tradition (e.g., Const.Porph., On the Ceremonies. 2.26 (Reiske 627.4–11, trs. Moffatt and Tall [2012], 627 and helped clarify the boundaries of imperial sacrality, as explained in Dagron [2003], 105–7, 111–13, 122–4). 117 Typicon CP, 6 November (Mateos [1962–3], vol. 1, 90.9–12). Other feasts of Paul were on 20 August (Mateos [1962–3], vol. 1, 378.22–3 and 2 September (Mateos [1962–3], vol. 1, 13.3–4). 118 SEC 46.17–21. 119 Berger (1997), 402–9 and (2001), 174. 120 CTh. 14.17.11 (26 April 393). 121 CTh. 15.1.23 (18 January 384). 122 Berger (1997), 380. The ‘Theodosian aqueduct’ was extensive enough to require management by five praetors (CTh. 6.4.30: 31 December 396). 123 Mango (1985), 121 and (2001), 25.

21

R E I N V E N T I N G C O N S TA N T I N O P L E

other objects.124 Nearby, he constructed new grain stores, the horrea Theodosiaca, which Byzantine emperors continued to inspect with elaborate ceremony each year.125 These stores enabled him to increase the imperial donation of free public grain, which was publicised on his equestrian statue erected in the Milion.126 Theodosius must also have been responsible for some modification and expansion to the imperial palace but there is no record of such work – similarly with the palace at the Hebdomon and possibly other palaces around the Bosporus. At the same time, new personal palaces were built for both Flaccilla near the Church of the Holy Apostles127 and Galla, near the imperial palace.128 Arcadius may also have constructed his own palace after 383, but again there is no documented confirmation. Theodosius was also responsible for a new hippodrome near the mansion of Eleutherios,129 and in 394 the Arcadian baths situated just north of the imperial palace were opened with great celebration for the city.130 Imperial building sponsored or executed by Theodosius reflected the emerging ideology and the new modes and expectations of court ceremonial. One of the earliest (379/80) may have been the ‘Gothic column’, which still stands in the garden of the Seraglio just beyond the Topkapi Palace.131 However, the most important and impressive of Theodosius’ constructions was the forum which bore his name, although it was also known as the Forum Tauri, perhaps best explained as being originally the property of the distinguished former prefect Taurus.132 It was located west of the forum of Constantine, further along the Mese but before the Capitol. In recent decades the forum has been the subject of regular research, mainly inspired by new excavation which has unearthed various different pieces of the forum structure and related buildings.133 Work on the forum must have

124 Rose and Aydingun (2007); Pulak et al. (2015); Kocabaş (2015). 125 Const.Porph. On the Ceremonies 2.51 (Reiske 699.6–701.17, trs. Moffatt and Tall [2012], 699–701). 126 Paras. 18 (Preger 31.19–32.2). 127 Chron.Pasch. 385 (564 Dindorf) with Janin (1964), 135. 128 Chron.Pasch. 385 (563 Dindorf) with Janin (1964), 135–6. 129 Patria 3.173 (Berger [2013], 211–12). 130 Marcellinus, Chron. 394.3 (MGH.AA., XI 64). 131 Peschlow (1992), 228; Bardill (1999a), 695. 132 PLRE 1, 879–80 (‘Flavius Taurus 3’). He was the father of two even more distinguished sons Aurelian (PLRE 1, 128–9 [‘Aurelianus 3’) and Eutychianus (PLRE 1, 319–20 [‘Flavius Eutychianus 5’]). Taurus presumably retired to Constantinople after Julian’s death in 363, when he was able to be recalled from exile for having supported Constantius against Julian in 361. The other plausible explanation of the name is that it was originally a market modeled on that at Rome known as the ‘forum of the bull (tauri)’. 133 Janin (1964), 64–8; Faedo (1982); Bauer (1996), 187–203. A useful and well-illustrated starting point for both context and the disparate excavated fragments related to the forum is Barsanti (1995). Recent discussion began with Naumann (1976), who reconstructed the forum’s large entrance arch in light of the bases and columns unearthed in excavations between 1969 and 1973. Most recently: Hurbanič (2012), 15–24.

22

R E I N V E N T I N G C O N S TA N T I N O P L E

commenced not long after Theodosius arrived in Constantinople in 380. The porticoes and entrance archways at each end were clearly grand as was the basilica erected along one side which was finished by the mid-380s.134 Questions about the exact size of the forum (larger or smaller) and whether the uncovered arch is the western or eastern entrance to the forum are still not answered satisfactorily.135 The forum was decorated with equestrian statues of both Arcadius and Honorius which may have been installed at the time the column was completed or in 393, when there was a formal dedication of a new equestrian statue of Theodosius mounted on a pedestal.136 The building and iconography of the forum anchored and echoed the emperor’s dynastic ambitions, linking back to Trajan and forward to his imperial sons. The carved spiral column that dominated the forum was constructed at the same time, with its decoration depicting the imperial victory over the Goths in 386, when Theodosius and Arcadius celebrated a triumph at Constantinople.137 Some fragments of the column were incorporated in the Ottoman baths of Sultan Bayazit II (ruled 1481–1512) and in recent times other remnants of the column have been uncovered nearby.138 The column was a deliberate imitation of that of Trajan at Rome, and the connection with Trajan was always an integral part of Theodosian publicity. It was accentuated in the forum by promoting decorative motifs of Hercules, a patron god of the Spanish emperors Hadrian and Trajan, which have been identified on the entrance arch columns.139 The inscription on the equestrian statue addresses Theodosius as a ‘second light bringing sun’, a phrase designed to evoke Constantine’s solar associations and promote Theodosius as the second founder of Constantinople to rival Constantine.140 The forum created a distinctive space unique to Theodosius and new opportunities for expanding the public ritual of the city. It became a new ceremonial stage for the emperor, a place where he could be seen in public and in context and where the voices of the people could shout appropriate acclamations. Apart from becoming a fixed station on regular processional routes, it was now the central locale for certain events such as the reception 134 Faedo (1998), 327. 135 Mango (1985), 45 estimated the size of the forum as 120 meters by 110 meters, being equivalent to that of Trajan’s forum at Rome. However, Berger (2000), 167–8 has argued strongly for a smaller forum (55 meters by 55 meters) and that the surviving part of the arch of Theodosius was the western, not the eastern, entrance, as Naumann (1976) had proposed in his reconstruction. 136 Chron.Pasch. 393 (565 Dindorf). Statues: Parastaseis, 66 (Preger [1901], 64.20–1), Patria 2.47 (Berger [2013], 82–3), 2.63 (Berger [2013], 92–5). Arcadius later added an equestrian Theodosius II to the collection: Patria 2.38 (Berger [2013], 74–5). 137 Zosimus 4.35.1; Cons.Const. 386.1 (MGH.AA. IX 244); Orosius 7.34.9; Marcellinus, Chron. 386.1 (MGH.AA. XI 62). For the erection of the column: Theophanes, Chron. AM 5878 (de Boor 70). 138 Sande (1981), 1–78, which presents and analyzes five fragments discovered in 1973 and their iconography. 139 Faedo (1998); Barsanti (1995), 20–1; Taddei (2019), 43–58. 140 Anthologia Planudeana. 16.65 with Leppin (2003), 201.

23

R E I N V E N T I N G C O N S TA N T I N O P L E

of barbarian envoys and the emperor himself when he returned from the west.141 The Persian envoys in 384 were probably received there too.142 In addition to those statues in the Forum Theodosii, there were many others erected by, or for, Theodosius throughout the city: at the Chalke, the entrance vestibule of the imperial palace,143 the Great Church,144 and the hippodrome,145 while at the Basilica cistern could be found a bronze statue of the seated emperor.146 There was also the silver one on the column in the Augusteon.147 In addition, a bronze equestrian one was located at the Milion148 where Theodosius also set up statues of Hadrian and Trajan,149 which were possibly erected elsewhere in the second century and re-appropriated by Theodosius to form a dynastic set. Statues could even have their own sacred function in representing the sacred presence of the emperor. In fact, Theodosius announced in 386 that asylum could be claimed by fleeing to the ‘statues of the emperors’.150 As for churches, while Theodosius promoted their use and integrated them more fully into the public and imperial life of the city, he was a limited church builder. Certainly he was responsible for the Church of John the Baptist at the Hebdomon which was evidently proposed to him by Rufinus;151 as well as that of the ‘Holy Notaries’ (Marcian and Martyrius, executed 341)152 and the Church of Saint Mark near the Forum of Theodosius.153 Also attributed to him, probably incorrectly, is the Church of the Virgin on the property of Eugenius.154 All this imperial construction – columns, forum, statues, churches – consolidated and amplified the emperor’s status within the city and reflected his imperial power, patronage, and dynastic intent in a traditional way. It also complemented and enhanced the ceremonial by exhibiting the imperial countenance in a fashion that bound it to the closely pressing populace of the city. It was now a

141 Details in Bauer (1996), 379–94. 142 Marcellinus, Chron. 384.1 (MGH.AA. XI 62); Orosius 7.34.8; Socrates, HE 5.12; Cons.Const. 384.1 (MGH.AA. IX 244); Chron.Pasch. 563 (Dindorf). 143 Paras. 77 (Preger 70.5–6); Patria 2.47 (Berger [2013], 82–3) with Bauer (2001), 165. 144 Parastaseis, 11 (Preger [1901], 26.15) 145 Parastaseis, 19 (Preger [1901], 32.3–4); Patria 2.76 (Berger [2013], 100–1, 104a); (Berger [2013], 124–5). 146 Parastaseis. 74 (Preger [1901], 68.2–3) 147 Marcellinus, Chron. 390.3 (MGH.AA., XI 62); Paras. 68 (Preger [1901], 65.14–15). 148 Parastaaseis 18 (Preger [1901], 31.19–32.2). 149 Cedrenus, Chron. (Tartaglia 557.2–3). Theodosius is the most obvious candidate for this particular statue set. 150 CTh . 9.44.1 (6 July 386). 151 Patria 3.145 (Berger [2013], 198–201); Chron.Pasch. 391 (564 Dindorf); Sozomen, HE 7.21.5; Theodore Anagnostes, HE 268 with Berger (1987), 514–15, 743–4. 152 Patria 3.188 (Berger [2013], 216–17) with Janin (1969), 391–2. 153 Patria 3.199 (Berger [2013], 218–19) with Janin (1969), 339. 154 Patria 3.21 (Berger [2013], 146–7). He is also often thought to be responsible for the reroofing of Hagia Sophia, although this may be the work of his grandson Theodosius II (Dagron [1984], 196–7 [translation of Descriptio], 213, 274).

24

R E I N V E N T I N G C O N S TA N T I N O P L E

shared ritual in a shared space. At the same time, Theodosius and his court officials along with local dignitaries and aristocrats were increasingly attracted by the new forms of power and patronage embedded in the local holy men, along with the relics of saints and martyrs. There was a sudden growth in monasteries and martyria at Constantinople encouraged and promoted by Theodosius.155 The emperor himself was responsible for bringing the monk Dios from Antioch and settling him in a local monastery.156 Prior to that, the city’s first monk, Isaac, was provided for by Theodosius’s general Saturninus on his own property which was near the city wall at Helenianae and neighboured that of his fellow-general Victor. Only recently they had both been leading Roman troops against the Goths. Theodosius used to visit Isaac there,157 as well as his successor in the monastery Dalmatius, a former palace guardsman of Theodosius.158 The way was paved for development of martyria by the emperor himself who decreed in 381 that burials should be outside the city and that any sarcophagi resting above ground should be relocated there.159 In 386 he further pronounced that in the case of a local saint or relics a martyrion could be built over the grave.160 This is what happened in the case of Eusebia’s shrine of the forty martyrs161 and probably that of the Persian trio put to death by Julian in 362, namely, Manuel, Sabel, and Ismael.162 Theodosius’s praetorian prefect Rufinus built a martyrion of Peter and Paul on his estate at Drys near Chalcedon, incorporating relics he had brought from the west in a solemn deposition.163 The city prefect Aurelian, however, built a shrine in anticipation of receiving the relics of the martyr Stephen but they never eventuated for him. Instead he had to content himself with the temporary relocation of the body of Isaac there in 383.164 Noble women, no less than men, were also patrons of holy persons and objects in Theodosius’s time. Martyria were constructed in the Lycus valley by Alexandria and Gregoria, otherwise unknown but possibly the wives of leading imperial courtiers.165 Another monastery was founded at the time by a western woman named Domnica who came to Constantinople via Alexandria and was baptised by Nectarius.166 155 156 157 158 159 160 161 162 163

The city’s martyria are listed in Maraval (1985), 402–10. Patria 3.193a (Berger [2013], 216–17) with Janin (1964), 343 and (1969), 97–9. Vita Isaacii, 14–15 (Acta Sanctorum, Mai VII, 256–7), with Janin (1964), 422. ACO 1.1.2 [66], 65.27 with Saradi (1995), 91. CTh. 9.17.6 (30 July 381) with Dagron (1977), 13. CTh. 9.17.7 (26 February 386). Sozomen, HE 9.2.1–10. Patria 3.190 (Berger [2013], 216–17). Callinicus, vita Hypatii 8.4–5. The monastery and its location became known as ‘Rufinianae’ (Janin [1975], 36–7). 164 Vita Isaacii 18 (Acta Sanctorum, Mai VII, 258). 165 Patria 3.193 (Berger [2013], 216–17) with their historicity doubted by Janin (1969), 18, 80 and 100–1. 166 SEC 337.21–33 with Janin (1969), 100, who is inclined to dismiss the possibility on the, at least questionable, grounds that it is difficult to imagine women founding monasteries at Constantinople so soon after men, that is, Isaac and Dios.

25

R E I N V E N T I N G C O N S TA N T I N O P L E

The single event that gave rise to a new burst of imperial building activity was the victory over the usurper Maximus and his son Victor in 388, which was duly celebrated in Constantinople. Indeed, the day became firmly fixed in the Byzantine calendar thereafter.167 Although Theodosius did not return to the east immediately and remained away from Constantinople for nearly four years (August 387–July 391), the city still had a resident emperor, the teenage Arcadius. So the ceremonial routine of the city and court continued uninterrupted, and the task of immortalising Theodosius’ new success proceeded apace in the usual fashion. Outside the city was built the magnificent Golden Gate, a ceremonial portal for the victorious emperor as he approached the capital once more and it may well have been completed in time for his return to Constantinople in 391.168 By then, too, there was erected another famous memorial that is still standing in situ, the obelisk in the hippodrome. The obelisk rests on a base in two parts and depicts scenes of the emperor and his sons presiding at the games. Again, it was probably set up in 390 by the prefect Proculus when Theodosius was still in the west.169 The obelisk has been the subject of intensive research in recent times, but uncertainty remains about the identity of the imperial figures depicted, although there is general agreement that its iconography is essentially dynastic, expressing the concord of east and west.170 Other construction was also inspired by Theodosius’s victory in the West, namely, a possible column in the Forum Strategii.171

Conclusion Theodosius was drawn into Constantinople in 380 by the city’s senate, facilitated by the persuasive intervention of Themistius. The local aristocracy of office wanted to have the advantage of imperial proximity and the patronage they thought they should have been enjoying all along. Once there he reclaimed the city for the orthodox cause to which he was already committed by baptism and turned his imperial attention from military campaigns to protecting, improving, and

167 Procopius records its continued celebration in the 530s (Wars 3.4.16). 168 The gate was previously considered most likely to commemorate the victory of Theodosius II over the usurper John in 425 and built out of the Theodosian city wall constructed in 413. However, more recent research (Bardill 1999a) has clearly established that it was built separately as a standalone monument before the wall was built and is therefore dated to the time of Theodosius I. 169 Marcellinus, Chron. 390.3 (MGH.AA. XI 62). An earlier date (386) was proposed by Rebenich (1991). 170 Ritzerfeld (2001) emphasizes the dynastic continuity in the monument while its presentational function is addressed by Safran (1993) with access to earlier literature and excellent photographs. Effenberger (1996) mounted an elaborate technical case for claiming that the present base was not the original but was added later to compensate for a broken bottom section of the column in order to maintain the original column height from the ground. However, this has not proved convincing (Safran [1993]; Speck [1997]; Kiilerich [1998], contra Ritzerfeld [2001], 170). 171 Mango (2000), 187–8.

26

R E I N V E N T I N G C O N S TA N T I N O P L E

promoting the lives of his imperial subjects. By January 384 the senate and people of Constantinople could be confident that Theodosius was now there to stay, and Themistius could express strong satisfaction with the emperor’s investment in the city and the way it was growing. Indeed, it was now difficult to say whose contribution was the greater: how Constantine created a new city out of Byzantion or how Theodosius had created a new city out of Constantinople. Looking ahead, Themistius concludes, Theodosius would be seen to have created a whole ‘third city’.172 Moreover, he had also now begun to entrench Constantinople’s role as the ‘New Rome’ by developing the imperial mausoleum precinct and relocating the remains of previous emperors there. The succession was guaranteed now, too, as Themistius proclaimed.173 Theodosius’s eldest son Arcadius was already crowned Augustus, and soon after his wife, Flaccilla, crowned Augusta. There would be no more uncertainty about succession as there had been on the deaths of Julian, Jovian, and Valens, nor would Constantinople ever experience an imperial vacuum again for several generations. By the time Theodosius left Constantinople to engage with Maximus in 387/8, he had already resided longer in the city than even Constantine. By then he had set about creating his own monumental forum and other trophies to decorate the city and promote himself and his family as an established imperial dynasty. The settled presence of the imperial court at Constantinople for most of the reign of Theodosius I enabled the already large city’s rapid development as a genuine ‘New Rome’. Lack of military success against the Goths in 379 and 380, combined with the peace settlement with them in 382 as well as that with the Persians in 385, produced an emperor focused on court and capital. Instead of training his troops and marching out with them each summer, as his predecessors had been obliged or chose to do, Theodosius remained in his capital. The city became his battlefield, and he even legislated to keep his sons and future emperors anchored in the city and away from the military front.174 In these years rather than later, as usually supposed, the imperial and topographical core of Byzantine public liturgy and life was first fashioned. This involved the creation of new ceremonial spaces in the city and the development of a distinctive public ritual that resulted from a fusion of stational liturgy from Jerusalem, traditional Roman imperial ceremonial, a demilitarisation of formal rites, and, as Matthews has shown, the orthodox piety of Theodosius’s family and the many western officials and relatives who accompanied him to Constantinople. It was Theodosius I who turned the emperor’s role into one of managing and mediating the imperial power through church and consistorium, senate and bureaucracy, court and urban ceremonial. These are the very elements that appear now to be such a solid and prominent feature of the reigns of his son Arcadius

172 Themistius, Oration 18.9, 222b–223b (Maisano [1995], 636–7). 173 Themistius, Oration 18.11, 224b–225b (Maisano [1995], 638–41), with Jacobs (2012), 149–50. 174 John the Lydian, On the Magistrates, 2.11 (no such law is extant).

27

R E I N V E N T I N G C O N S TA N T I N O P L E

and grandson Theodosius II.175 However, they merely followed and expanded the model of urban development and decoration established by Theodosius I who had spent most of his reign inside the walls of Constantinople. Following the final committal of Theodosius to his imperial sarcophagus in the Church of the Holy Apostles on 9 November 395 the Byzantines continued to remember on the same day each year the man responsible for the reinvention of their city as a Christian imperial capital.176

175 As in the seminal study of Millar (2006), but with no mention of the role of Theodosius I, or Arcadius, in the process. For this perspective, Holum (1982) still provides the best model. Several of the arguments advanced here are taken beyond 395 by Pfeilschifter (2013). 176 SEC 205.24–5; Typicon CP, 9 November (Mateos [1962–3], vol. 1, 96.15–16). On the second anniversary, 9 November 397, the new patriarch, John Chrysostom, could encapsulate Theodosius as someone honored ‘not because he was emperor, but because he was pious; not because he was clothed in the purple but because he had put on Christ’ (John Chrysostom, Adversus Catharos. Homilia VI [PG 63. 491]).

28

2 D Y N A S T Y A N D A R I S TO C R A C Y I N T H E F I F T H C E N T U RY *

The fifth century is regularly characterised as a period of upheaval and breakdown of Roman government and society, especially in the western provinces. Yet that is to overlook or underestimate the traditional social and political forces for continu­ ity, and land ownership as the enduring basis for real wealth. Emperors may come and go, provinces may fall in and out of imperial authority, armies may form and reform with non-Roman generals and new military leaders; but the traditional drivers of power and influence, the imperial dynasties and aristocracies of both East and West, remain. The enduring pull of the dynastic principle, the quest to formally link self and family to a centre of power and influence, still drove and shaped fifth-century Roman political life. The Roman aristocracy traditionally dominated the imperial court through tenure of the highest civilian and military offices, while the impulse to dynasty continued to link the eastern and western imperial families and courts. Even so, dynasty and aristocracy were themselves subject to changes in character and texture over the fifth century. This chapter explores the essential dynastic politics of the fifth century and the mutating Roman aristocracies, focussing on their relation to military and politi­ cal authority at the imperial court. As the fifth century progressed, the fluctuating balance of power between East and West, and between Roman and non-Roman, became key dichotomies, which were forever in tension but never resolved. In the wider context of Roman political and cultural life in the fifth century these particular themes provide key anchor points for analysing how control and influ­ ence were acquired and mediated, and how Roman government and society were developing. The actions and attitudes of the Hun king Attila in the 440s and early 450s illuminate these themes by showing that, unlike the leaders of Goths, Bur­ gundians, and Vandals in particular, he remained a barbarian warlord who never appreciated, nor fully engaged with, the dynastic and aristocratic mainstays of contemporary Roman politics and society. Therein lies the key to his ultimate political failure. * This chapter first appeared in M. Maas (ed), Cambridge Companion to the Age of Attila (2014), 102–127 and is reproduced by permission of Cambridge University Press, but with several updates and enhancements.

29

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

Dynasty Throughout the fifth century, no less than in other periods of Roman and Byzantine history, the dynastic principle regulated the quest for achieving, maintaining, and participating in the exercise of military, political, and social power.1 The imperial dynasties of East and West, often just a single family, channelled imperial author­ ity and prestige. The dynastic urge was a binding force, forever shaping the court and aristocracy around it. Being or becoming part of the prevalent imperial fam­ ily conferred particular access and influence. Often these advantages flowed from a civic or military appointment in the service of the emperor in Rome, Ravenna, or Constantinople. This new pattern of office holding and intermarriage also gave more prominence to women and especially to marrying into the imperial family. Hence the mothers, wives, sisters, and daughters of emperors became increasingly powerful themselves at this period. An imperial Augusta possessed the wealth and property, her own courtiers and household, and often her own coinage, to enable her to assert her own authority virtually independently. She may not have been able to rule in her own right but she could choose and install an emperor: Galla Placidia stimulated the elevation of her young son Valentinian III (425), Pulcheria played a role in the acces­ sion of Marcian (450), Verina sanctioned both Basiliscus (475) and Leontius (484), while her daughter Ariadne facilitated the accession of Anastasius (491).2 For what the Romans called ‘barbarians’, most of whom emerged from the higher ranks of the Roman army, access to dynastic linkage was through impe­ rial favour and patronage. Over the course of the fifth century such generals were integrated into the Roman aristocracy and imperial court, even the imperial fam­ ily. In the late fourth century the emperor Theodosius I (see figure 2.1) promoted several key generals of non-Roman blood: Bauto the Frank, for example, who died in 388 and whose daughter was brought up in Constantinople;3 Fravitta the Goth, who took a Roman wife with the blessing of the emperor;45 Stilicho the Vandal, who married the emperor’s niece Serena. On his death in 395 Theodosius was succeeded separately by his sons, with Arcadius (aged twenty-one) ruling the East until 408 and Honorius (aged eleven) the West until 423. Together they manifest the power of a dynastic tie that was not always evident, or not sufficient by itself.6 They were succeeded in turn by Arcadius’s son Theodosius II (aged six) until 450, Arcadius’s and Honorius’s nephew Valentinian III (aged five) until 455 and Mar­ cian (aged fifty-eight), brother-in-law of Theodosius II, until 457. Thus the dynasty of Theodosius I directly prevailed in both East and West until the 450s with their imperial tenure consolidated by marriage alliances with leading generals.7 1 2 3 4 5 6 7

Dagron (2003), 23. McCormick (2000), 146–48; Chapter 6 (below, 153–68). PLRE 1, 159–160 (‘Flavius Bauto’). PLRE 1, 853–8 (‘Flavius Stilicho’); 824 (‘Serena’). PLRE 1, 372–373 (‘Flavius Fravitta’). Icks (2014), 77–84. McEvoy (2013), 136–52 and (2014), 257–60.

30

31

Figure 2.1 Family of Theodosius I

cos. = consulship

Arcadia Flaccilla

HONORIUS (425-55)

Marina

Eudoxia

Arcadius

m. (1) Ataulph

m. (2) CONSTANTIUS (421)

Justa Grata Honoria VALENTINIAN III (425-55) (Figure 2)

Licinia Eudoxia (Figure 2)

Aelia Eudocia

Theodosius

m. (2) Galla

Galla Placidia

Flaccilla

THEODOSIUS II (408-50)

ARCADIUS (395-408)

Fl. Bauto (cos.385)

m. (1) Aelia Flavia Flaccilla

Pulcheria

THEODOSIUS I (379-95)

Maria m. (1)

Serena

Pulcheria

CAPITALS = Emperor

MARCIAN (450-7) (Figure 3)

Thermantia m. (2)

Eucherius

Stilicho (cos.400, 405) m.

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

. . . in the West It was the daughter of Theodosius I’s Frankish general Bauto who became Arca­ dius’ wife, the empress Eudoxia, while the Vandal general Stilicho, himself espoused to the emperor’s niece (Serena), had two daughters who were married successively to Emperor Honorius – first Maria (died 407–408) then Thermantia (died 415). In the end, neither marriage bore offspring, so if Stilicho dreamed of being an imperial grandparent like Bauto he now needed to look elsewhere. Accordingly, at one stage Stilicho sought to marry his son Eucherius to the halfsister of Arcadius and Honorius, Galla Placidia (born 388), who then lived at Ravenna. Honorius resisted this pressure and had Eucherius killed following the demise of his father Stilicho in 408. As for Galla, the emperor Honorius’s general Constantius was anxious to marry her, but was chagrined to see her become the wife of a Gothic king, Athaulf, in 414, which also demonstrated that the dynastic imperative trumped creedal alignment (she orthodox, he Arian).8 Had he not died in infancy, Theodosius (born 415), the son of Athaulf and Galla, may well have succeeded to the western imperial throne on the death of Honorius in 423 and pro­ ceeded to enjoy a long reign as the joint offspring of imperial and Gothic blood. Moreover, he may have unified or assimilated the Roman and Gothic realms.9 Not long after the unexpected death of Athaulf, Constantius was finally able to wed Galla on his consular inauguration day in 417. It was their son Valentinian III (born 419) who ascended the throne as a young boy in 425, with their daughter Honoria (born 417) remaining a prize goal for all those aspiring to an imperial link. Constantius himself became a co-emperor to Honorius in 421 and ensured dynastic continuity in the West for the Theodosian house.10 . . . in the East At Constantinople, meanwhile, Theodosius II ( figure 2.1), son of Arcadius and Eudoxia, enjoyed the longest imperial reign (408–450) of the fifth century. His elder sisters Flaccilla, Arcadia, and Pulcheria and his younger sister, Marina, frus­ trated all potential suitors by retarding the dynastic principle.11 They preferred to turn their courtly life into a monastic routine of prayer, penance, study, and charity. Theodosius II, however, married a cultured young Athenian (Eudocia), who gave birth to a daughter Eudoxia, named after her Frankish grandmother, followed by a son, Arcadius, and another daughter, Flaccilla, both of whom died young.12 In 437 the young Eudoxia married her western cousin Valentinian III, thereby reconnect­ ing both eastern and western branches of the Theodosian clan. Soon after, around

8 9 10 11 12

Demandt (1980), 620–21. Oost (1968), 133–34; Sivan (2011), 25–7. Holum (1982). Holum (1982), 96–102; Harries (2013), 67–77, 88–9; McEvoy (2019), 116–20. PLRE 2, 410–12 (‘Licinia Eudoxia’); 130, (‘Arcadius 1’); 473, (‘Flaccilla 2’).

32

Hilderic

Eudocia

33

Placidia

Figure 2.2 Family of Valentinian III

cos. = consulship

CAPITALS = Emperor

Huneric

Gaiseric

VALENTINIAN III (425-55)

Placidia

Licinia Eudoxia (Figure 1)

Areobindus

Anastasius

daughter

Proba

Proba

Irene

Juliana

Magna

Areobindus (cos.506)

Olybrius (cos.491)

Anicia Juliana

ANICIUS OLYBRIUS (472)

Probus

Probus (cos.402)

Paulus (cos.496)

ANASTASIUS (491-518) (Figure 4)

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

441, Eudocia, deserted her imperial husband, as well as the court and capital, for the spiritual delights of Jerusalem, where she lived until her death in 462.13 Her husband Theodosius had died in 450 and it was only then that Pulcheria, his lone surviving sister and now well beyond child-bearing age herself, was persuaded to marry by providing an essential dynastic link for the Roman general Marcian. With her death in 453, followed by her husband Marcian’s in 457, the eastern Theodosian dynasty expired.

Valentinian III and family, 455–518 In the West, the Theodosian household (figure 2.2) was carrying on through the daughters of Valentinian III and Eudoxia, namely Eudocia (born 438–489) and Placidia (born 440).14 From a very young age both girls were at the centre of an intense competition among the most powerful Roman generals and barbar­ ian kings. Marrying into the venerable house of Theodosius and Valentinian III bestowed not only the prestige of an imperial connection. It provided a possibility of future imperial parentage and the formation of new dynasties. No wonder it was so hard fought. By the early 450s the senior Roman general Aetius planned for his son Gaudentius one day to become the husband of young Placidia, but he failed. Her sister Eudocia, however, was espoused to the Vandal king Gaiseric’s son Huneric, who had been sent as hostage to the court of Valentinian, probably as surety for the treaty negotiated between Romans and Vandals in 442. Huneric was already married to the daughter of the Visigothic king Theodoric, but at this point she was unceremoniously disfigured and sent back to her father. An imperial connection clearly offered more opportunity. Such bonds were not seen as unusual simply because they were between barbarians and Romans. Rather, they were tra­ ditional arrangements between two families striving to attain or retain power and influence.15 Court connections were the main reward. . . . in the West On Valentinian’s death in 455 the political fortunes of the West would have been very different if the emperor had had a male heir to succeed him.16 Instead, his wife Eudoxia and her daughters (figure 2.2) were peremptorily coerced into new marriages by the new emperor, Petronius Maximus. Eudoxia herself was taken in marriage by Maximus, and her other daughter Placidia was betrothed to Maxi­ mus’s son Palladius, instead of the general Majorian as her mother had evidently planned.17 Eudoxia and Placidia provided dynastic legitimacy for Maximus and 13 14 15 16 17

PLRE 2, 407–8 (‘Eudocia 1’). For date and context, Alan Cameron (1982a), 258–63. PLRE 2, 407–8 (‘Eudocia 1’); 887, (‘Placidia 1’). McEvoy (2013), 266–7. Oost (1968), 247. Oost (1964), 27–8.

34

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

Palladius, as Pulcheria had done for Marcian in 450.18 Aetius’s imperial aspira­ tions for his own son had perished with his murder the previous year (454) and may even have contributed to the imperial anger that resulted in his death. This sudden dynastic crisis in 455 was resolved when Gaiseric’s Vandals mounted an expedition and took Rome, had the new emperor Petronius Maximus expelled, then escorted Eudoxia and both her daughters off to Carthage. The son of the Van­ dal king (Huneric) and the daughter of the Roman emperor (Eudocia) were now finally united. Their offspring Hilderic (born 456) could have become a Roman emperor and instantly reconciled and reunited Vandal Africa with the Roman empire. Instead, Vandal succession custom prevented him becoming king until late in life (523–530), by which time he was a ‘special friend’ of the Roman emperor Justinian (527–565). Hilderic was never a serious imperial contender because, despite his imperial bond, he was born and raised in Carthage rather than in Rome or Ravenna. . . . in the East Some years later (462) Hilderic’s sister-in-law Placidia and his mother-in-law Eudoxia, widow of Valentinian III since 455, left his Vandal court at Carthage for Constantinople, where they became the only Theodosian residents in the eastern capital. Indeed, until recently (457), the Theodosian dynasty was the only impe­ rial family most of the living inhabitants of Constantinople had ever known. Their very presence once more revived and promoted their potent dynastic status. Mean­ while, following the earlier example of her grandmother and namesake, Eudocia left Africa for Jerusalem in 471 and died there shortly after. The dowager empress Eudoxia lived on at Constantinople into the 490s, while her other daughter, Pla­ cidia, was already married to a senator of distinguished lineage, Anicius Olybrius, thereby making him a potential emperor of either East or West.19 The first opportunity arose in 461 on the death of the western emperor Majorian, when the Vandal king Gaiseric strongly supported Olybrius’s right to the throne. Gaiseric was now the father-in-law of Eudocia, sister of Olybrius’s wife. Whether Gaiseric renewed Olybrius’s claim on the death of the emperor Libius Severus in 467 is not known. Instead, the eastern court appointed Anthemius (figure 2.3), hus­ band of the only child of the late emperor Marcian and who might have expected to succeed his father-in-law at Constantinople ten years earlier.20 The dynastic authority of Anthemius was further cemented by marrying his daughter Alypia,

18 Demandt (2007), 257. 19 PLRE 2, 796–798 (‘Anicius Olybrius 6’), see also Clover (1978), 169–96, here 192–5. The pro­ posal that Olybrius was the son of Petronius Maximus who organised the marriage with Placidia when Maximus was emperor in 455 (Mommaerts and Kelley [1992], 119), has been convincingly rejected by Gillett (2003), 88 n.11. 20 PLRE 2, 96–8 (‘Anthemius 3’). This appointment was an overt symbol of reunification of eastern and western courts: O’Flynn (1991), 125.

35

36

cos. = consulship

CAPITALS = Emperor

Alypia

Aelia Marcia Euphemia

Procopius

Procopius Anthemius (cos.515)

ANTHEMIUS (467-72)

m. (2) Pulcheria (Figure 1)

Anthemiolus

MARCIAN (450-7)

Figure 2.3 Family of Marcian

Ricimer

m. (1) ??

Romulus

??

m. (1) ??

Leontia

LEO I, 457-74 (Figure 4)

Marcian (cos.469)

Procopius

Procopius

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

who was daughter of one emperor (Anthemius) and granddaughter of another (Marcian), to his all-powerful barbarian general Ricimer.21 Despite the new fam­ ily bond, before long relations between Anthemius and Ricimer deteriorated and Ricimer revolted. So, in 471 Olybrius (figure 2.2) was sent to Rome by the eastern emperor Leo I to mediate the hostilities between them. He failed to achieve that, but on Anthemius’s death the following year Olybrius unexpectedly found himself emperor at last, being proclaimed such by Ricimer. At this point the Theodosian dynasty might have been firmly restored in the West once more, had Olybrius not suddenly died six months later. Thereafter no real capacity for dynastic continuity in the West was possible. The next few years saw a series of brief and insecure reigns: Glycerius, sponsored by the Burgundian general Gundobad in 473;22 Nepos, husband of the empress Verina’s niece, who was sponsored by the eastern emperor Leo I in 474 to reassert authority over the West;23 and Romulus, sponsored in 475 by his father Orestes, a Roman general and former secretary to the Hun king Attila.24 All three were deposed and exiled. After 476 Gothic kings replaced emperors altogether at Rome and Ravenna.

The house of Leo, 457–518 Meanwhile at Constantinople, with the disruption to the Theodosian household occasioned by the death of Marcian in 457, real power now lay with the gener­ als, and Aspar was foremost among them. He secured the throne for a favoured military associate, Leo I (figure 2.4), as he likely did for Marcian in 450.25 As a barbarian, Aspar was ineligible to be emperor himself, but, like Bauto and Stilicho decades before, he could become an emperor’s father-in-law thereby potentially securing his family’s imperial connection for generations. In the early years of Leo I’s reign, Aspar consolidated his influence by securing a marriage commitment for his son (Patricius) and Leo’s daughter (Ariadne). Over time Leo’s independent power grew, tension emerged with Aspar, and, when it was finally clear that Leo was intent on founding his own imperial dynasty, Aspar and his family were liq­ uidated. As with the demise of Aetius in 454, the murder of Aspar and sons in 471 was triggered by the emperor’s desire for greater dynastic autonomy.26 Leo had his own plans. The marriage in 467 of his younger daughter Leontia with Marcian, grandson of Emperor Marcian, and son of Anthemius, whom he soon sent as emperor to the West, was a deliberate act of dynastic continuity. It was Leo’s elder daughter Ariadne, as the wife of successive emperors Zeno (474–491) and Anastasius (491–518), who represented the dynasty of Leo until her death 21 22 23 24 25 26

PLRE 2, 942–945 (‘Fl. Ricimer 2’), with MacGeorge (2002), 262–7. PLRE 2, 514 (‘Glycerius.’). PLRE 2, 777–778 (‘Iulius Nepos 3’), with Salzman (2017). PLRE 2, 949–950 (‘Romulus Augustus 4’). R. Burgess (1993-4), 47–68. Croke (2005a), 147–203; Chapter 5 (below 51–107).

37

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

in 515, having inhabited the imperial palace at Constantinople for nearly sixty years.27 If the venerable Ariadne had a rival, it would have been Anicia Juliana (figure 2.3), daughter of Olybrius and the Theodosian Placidia, born in around 461–2. Her grandmother, the dowager empress Eudoxia, embodied her imperial power and image in the churches she constructed during her latter years at Con­ stantinople, especially St Polyeuktos, while her mother Placidia publicly consoli­ dated the prestige of the imperial Theodosian family. Juliana’s imperial lineage was proclaimed for all to see in the church of St Polyeuktos as the ‘bright light of blessed parents sharing their royal blood in the fourth generation’, that is from her great-grandmother Eudocia to Eudoxia to Placidia to Juliana.28 In 478 Leo I’s son-in-law, the Isaurian emperor Zeno (figure 2.4), saw an opportunity to associate his regime more closely with Roman imperial heritage by arranging a marriage between Anicia Juliana and the Goth Theodoric, then an eastern general, leader of his Gothic clan and later king of Italy.29 Again, there would be potential to unite East and West within the authority of a single impe­ rial family resonant with the Theodosian name. He failed, perhaps overruled by Ariadne. Instead, Juliana married into a more distinguished aristocratic family of barbarian lineage. Her husband Areobindus was the grandson of the generals and consuls of 434, the Alan Aspar and the Goth Areobindus (figure 2.5). The emperor Zeno and his father-in-law Leo I had effectively seen off the dynastic aspira­ tions of Aspar in 471, but they had now resurfaced in the marriage of Juliana and Areobindus. In the end, it was not Aspar’s son who formed a link with a woman of royal Roman blood, but his grandson. Juliana’s husband himself became the distinguished consul of 506 but was preceded in the honour by their son Anicius Olybrius (consul 491), who married Irene, niece of the emperor Anastasius. Then, in 512, Areobindus was being promoted as a replacement emperor for the aged and unpopular Anastasius. By the early sixth century the combined dynasties of Theodosius (through Juliana) and Aspar (through Areobindus) had joined with the dynasty of Anastasius (through Irene). In short, through the fifth-century East the ties of blood and office were woven more tightly and narrowly from generation to generation. The whole of the fifth century was dominated by two overarching imperial houses, those of Theodosius (379–457) and Leo (457–518), while in the West the Theodosian authority lasted until 455, followed by a series of unrelated emperors, mainly representing chang­ ing local power bases between Rome and Ravenna until 476. While unsuccessful, compared to the West, regular power struggles and threats of usurpation in the East are explained by the need to contest dynastic access. From the 430s nearly all imperial usurpers, both those who succeeded and those who failed, were related to the imperial family, principally Basiliscus (475), Marcian (479), Leontius (484), and Longinus (491), as well as the young Basiliscus (476), briefly recognised as 27 Croke (2015), below pp. 153–68. 28 Anthologia Palatina 1.10.6–8, see also Mary Whitby (2006), 159–88. 29 Malchus, fr. 18.3 (Blockley 432–3).

38

39 Leontia

Basiliscus, Caesar 476-7

ANTHEMIUS

(Figure 3)

Figure 2.4 Families of Leo and Zeno

cos. = consulship

??

Armatus (cos.476)

m. (2) Marcian (cos.469)

Marcus, Caesar 475-6

BASILISCUS (475-6)

CAPITALS = Emperor

children

Aelia Zenonis

Lallis

ZENO (474-91)

LEO II (474)

m. (1) Julius Patricius (cos.459)

Aspar (cos. 434) (Figure 5)

m. (2) Ariadne

LEO I (457-74)

Infant son

ANASTASIUS (491-518) (Figure 2)

Aelia Verina

m. (1) Arcadia

??

Longinus

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

Julius Patricius (cos.459)

40

Figure 2.5 Family of Aspar

cos. = consulship

CAPITALS = Emperor

Leontia

LEO I (457-74) (Figure 4)

Herminiricus (cos.465)

Fl. Ardabur Aspar (cos.434)

Fl. Ardabur (cos.427)

Godisthea

Ardabur (cos.447)

Areobindus (cos.506)

OLYBRIUS (472)

Anicia Juliana (Figure 3)

Placidia

daughter

ZENO (474-91)

Dagalaiphus (cos.461)

Areobindus (cos.434)

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

Caesar by Zeno. He was the son of Armatus, who was the nephew of empress Verina and her brother Basiliscus.30 In the fifth-century West, by contrast, Romans and barbarians occasionally intermarried but only rarely did kings and military leaders of other nations seek to form closer and more integrated relations with the Roman court. As Roman generals became more powerful they took the imperial throne: Avitus (455) and Majorian (457) in the West; Marcian (450), Leo (457), and Zeno (474) in the East. Those whose ethnic background barred their way to imperial authority became the power behind the throne instead: Areobindus, Arda­ burius, and Aspar in the East; Ricimer and Gundobad in the West. Their goal was to link their family with that of the imperial dynasty. More independent barbarian leaders such as Athaulf, Gaiseric, Theodoric I the Visigothic king in Gaul (418–451), and later Theodoric the Ostrogoth king in Italy (figure 2.6), and the Burgundian Gundobad, generally preferred to avoid Roman aristocratic families by building their dynastic connections across nations. Only rarely (Athaulf in 414, Huneric in 442, Ricimer in 467) did they seek to join with the imperial household. Instead, Theodoric I the Visigoth married his daughter to the Suevic king Rechiarius in the 440s,31 while a half-century later the Ostro­ gothic Theodoric, although virtually a Roman emperor and immersed in Roman court culture from a very young age, still looked outwards. He married his sister Amalafrida to the Vandal king Thrasamund, one daughter (Ostrogotho Areagni) to the Burgundian King Sigismund, another (Theodegotha) to the Visigothic king Alaric, and yet another (Amalasuintha) to Eutharic.32

Honoria and Attila: dynasty thwarted Of all the identifiable Roman-barbarian marriages across East and West, not one involved a Hun, male or female.33 As Hun king in the 440s and early 450s Attila never showed any interest in seeking a marriage alliance with the Roman court for any of his sons. He never sought to compete for the daughters of Valentinian III with Roman generals such as Aetius and Majorian, distinguished aristocrats and imperial officials such as Petronius Maximus, or other kings such as the Van­ dal Gaiseric. Had he done so he may have been successful. Even so, at one time he suddenly found himself faced with the enticing prospect of an alliance with Honoria, the younger sister of Valentinian III. The exact roles of both Attila and Honoria in this affair are not entirely clear. By the time Attila reached the height of his power Justa Grata Honoria (born 417) was an imperial princess in her early thirties with her own household and her own coinage, an Augusta of considerable standing, like her mother Galla Placidia.

30 31 32 33

Croke (1983), 81–91. PLRE 2, 935 (‘Rechiarius’). PLRE 2, 63–4 (‘Amalafrida’); 138–9 (‘Areagni’); 1068 (‘Theodegotha’); 65, (‘Amalasuintha’). Blockley (1982), 63–79.

41

42

Ostrogotho Areagni

Amalafrida

Amalasuentha

Theodoric (cos.484)

Figure 2.6 Family of Ostrogothic King Theodoric

cos. = consul

Bold = tribal king

Sigismund

Gundobad

Burgundians

Thrasamund

Vandals

Theodemer

Valamer

Eutharic (cos.519)

Audefleda

Ostrogoths

Theodegotha

Theodimund

Vidimer

Alaric II

Euric

Theodoric

Visigoths

daughter

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

In 449, so it would appear, Honoria was disenchanted by the prospect of a marriage organised by her brother to a distinguished Roman senator Flavius Bas­ sus Herculanus. Consequently, she sought the help of Attila by sending a trusted eunuch from her household named Hyacinthus, who carried her signet ring as a symbol of good faith. Attila took her request as a marriage offer and insisted on it being finalised. After all, Attila was a Roman general, having been granted the title of magister militum by Theodosius II,34 and it was not unusual for Roman imperial women such as Honoria to be betrothed to Roman generals. Apart from the example of Theodosius’s niece Serena, wife of Stilicho, Honoria’s own mother Galla had married the general Constantius. Before long (450) her aunt Pulcheria was to marry Marcian. Attila was also a tribal leader based outside the boundaries of the Roman empire but linked to it by imperial tribute and the routine channels of diplomacy. Again, Roman princesses were not beyond the expectation of lead­ ers in Attila’s position. Honoria’s mother, Galla Placidia, had earlier married the Gothic leader Athaulf in 414, and her young niece Eudocia was already amicably betrothed to the son of the Vandal king Gaiseric. Evidently, Theodosius II found nothing impossible or unprecedented about a union between Attila and Honoria, so he endorsed it, not least as a means of paci­ fying Attila, and he advised Valentinian III accordingly.35 Attila’s advisers will have known that Roman law also required a future wife to provide a dowry, which may have involved designated Roman territory and that Honoria had legal control over her own property.36 Certainly that would have appealed to Attila. He probably had in mind, or was informed about, the recent precedent of Theodosius II, who acquired sovereignty over some of the western provinces of Illyricum through the union of his daughter Eudoxia with the young Valentinian III in 424 (married 437). The territorial price paid to settle the pre-nuptial deal between the eastern and western courts, which had been negotiated by Galla Placidia, still rankled with the West a century later. At Ravenna in 533 Cassiodorus could explicitly claim that Galla had ‘purchased a daughter-in-law for herself by the loss of Illyricum and the union of rulers was accomplished through the lamentable division of provinces,’37 while shortly after, at Constantinople, the emperor Justinian could associate reor­ ganising the administration of Illyricum with the ‘age of Attila’.38 Attila’s expec­ tation, if not Honoria’s offer, might have been for settlement of Huns in certain Illyrian provinces, similar to the current arrangements for Goths and Burgundians in Gaul, Sueves in Spain, and Vandals in Africa.

34 Priscus, fr. 11.2 (Blockley 278–279). Contra PLRE 2, 182 (‘Attila’), he was appointed ca. 447 by Theodosius II, cf. Gračanin (2006), 51–2; Meier (2017), 49–51. 35 John of Antioch, fr. 223 (Mariev 404–5), see also Kelly (2008), 181 and especially Meier (2017) arguing for a distinctive eastern imperial approach to the ‘Honoria affair’. 36 For legal details and documents: Evans Grubbs (2002), 89–128. 37 Cassiodorus Variae 11.1.9, trs. Bjornlie (2019), 411, cf. Jordanes, Romana 329 (the ‘whole of Illyricum’); Oost (1968), 185, 244–5; Gračanin (2006), 54–8. 38 Justinian, Novel 11.1. This position elucidated in detail in Sarantis (2019).

43

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

Nonetheless, notwithstanding Theodosius II’s consent, when Honoria’s brother Valentinian III discovered this overture to the Hun king, he was so enraged that he had the messenger Hyacinthus immediately killed and it took the pleading of their mother Galla for his sister Honoria herself to be spared. Attila persisted with his request for Honoria’s hand, along with an imperial dowry consisting of ‘half the western empire’, only to be rebuffed. Honoria was already married, so he was informed, and Valentinian flatly refused to permit any union with Attila. Attila then threatened to go to war with the Romans on her behalf, but events overtook him. The new eastern emperor Marcian supported Valentinian’s opposition to the union, and by the time Attila invaded Italy in 452 the demand for Honoria and her dowry appears to have receded. Nothing more is heard of her, while Herculanus enjoyed the prominence of a consulship in that year, presumably highlighting his imperial connection. Whether Honoria saw herself as marrying Attila, as her mother had once married Athaulf, with a view to producing a male Theodosian heir, or whether she proposed that Attila would share imperial power in Gaul,39 or whether Attila was seeking to replace Aetius as the chief imperial general at the western court through her, can only be speculation. In any event, it was an unplanned opportunity presented to Attila, but he was unable to take advantage of it. The potential power and prestige of a Roman dynastic link passed him by. As a result, Hun blood never ran in the veins of the extended imperial family and the command of Attila over his vast and destructive warrior confederation was lost in a single generation. A decade after Attila’s death his hegemony and his military legacy had virtually disintegrated. Subject nations such as the Gepids and Goths soon began to muscle their way out of Hun sub­ mission as Attila’s sons squabbled and forfeited authority when they divided up the various tribes between themselves.40 The former subject nations now wanted what the Visigoths, Sueves, Burgundians, and Vandals already had: security of place and sustenance, and the ability to be part of the Roman Empire and imperial service. Some might even dare to dream of a link with the imperial dynasty or blending into the imperial aristocracy.

Aristocracy By the turn of the fifth century the aristocracy of the western part of the Roman empire was being formed and perpetuated by imperial and civic appointments, traditional wealth acquired principally but not always by land tenure, common cultural habits, and the routines of an elite life. At Rome, the particular political role of the senate defined aristocratic status and activity even more narrowly.41 Even so, senators could be provincials and by now even military men. In the East, the imperial aristocracy was more recent and more diverse. Around the imperial 39 As proposed by Bury (1919), 12, cf. McEvoy (2013), 239, 260, 293. 40 See Jordanes, Getica 259, with Heather (2005), 358–66. 41 Salzman (2002), 19–68.

44

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

court at Constantinople had emerged an aristocracy of office rather than one pri­ marily of blood and heritage. There, more than in the West, a new governing and cultural elite was forged through wealthy local landowners and merchants, together with the notables of provincial towns who had acquired influence and office in the capital. Greek speakers from Syria and Palestine, Egypt, Greece, and Asia Minor, along with some Latin speakers from the Danubian and Balkan provinces, intermarried and progressively consolidated their hereditary position. Generals of both Roman and non-Roman origin were also assimilated into the eastern aristocracy but political integration did not always mean cultural integra­ tion.42 From the late fourth century, the aristocratic households of both East and West became Christian, and their public and private religious practice became a new and important facet of aristocratic life. Their wealth remained in the land they owned which could be spread across a number of provinces.43 Diversification The period from 395 to about 440, which encompassed the reigns of the emperors Arcadius and Honorius (the sons of Theodosius I), and at least the childhood years of Theodosius II and Valentinian III, led to the formation of more socially and ethnically open aristocratic households, as well as a reformation in the authority of the imperial court and how it was presented both publicly and privately. Under­ pinning this transformation and expedited by the reign of child-emperors was the gradually entrenched influence of generals and bureaucrats.44 For the rest of the fifth century these trends were solidified. Further, the rationale for aristocratic political contests and rivalry was not so much Christian against pagan, or Roman against barbarian, as so often described. Rather it was the traditional dynamic of Roman politics: shifting alliances based on sponsorship centred around the impe­ rial court. In the West, emperors like Valentinian III promoted the long-established Roman senatorial aristocratic families,45 while the urban elites from the affluent cites of the eastern provinces competed for imperial favour and patronage. In both East and West religious practice became an integral component of political life with the growth of episcopal power providing a new outlet for the local aris­ tocracies from which most bishops now emerged.46 By 425 the broadly common Christian culture and way of life had seeped into the fabric of government and the civilian aristocracy.47 At Constantinople the notion of the populace as a single congregation explains the importance of aristocratic involvement in the public ritual and the 42 43 44 45 46 47

Heucke (1997), 45–54; McEvoy (2016), 483–511. Begass (2016), 462–82. McEvoy (2010), 151–92. Humphries (2012), 174–9; van Nuffelen (2012), 183–200. Rapp (2005), 188–95. An illustrative example: Elton (2009), 133–42; Millar (2006), 192–234.

45

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

newly liturgified ceremonial in the capital, which began with Theodosius I, then expanded over these decades.48 It also explains the deep feeling aroused by threats to doctrinal orthodoxy and unity, especially after the councils of Ephesus in 431 and Chalcedon in 451, as well as the presence of senior officials and aristocrats at the church councils.49 Securing the blessings of God’s favour had become an essential part of imperial rhetoric and ideology, so aristocrats could be seen devel­ oping alliances with bishops but also vying for the patronage of holy men and women, as the former praetorian prefect Cyrus petitioned Daniel the Stylite.50 Most of the leading generals were also increasingly engaged at the intersection of political and ecclesiastical power. At the same time, in traditional Roman fashion the oligarchy provided a pathway into high office and status for a whole family. The empress Eudocia, for example, paved the way for her uncle Asclepiodotus to become consul in 423, while one of her brothers (Gessius) became praetorian prefect of Illyricum and another (Valerius) was vicar of Thrace.51 This rapid tran­ sition in the character of the Roman aristocracy in both East and West collapsed the differences between Romans and others. Aristocracy and dynasty mainly sub­ sumed ethnicity.52 Consolidation A barometer of these changes in the political and cultural expressions of elite power and prestige in the fifth century was the consulship and the annual holders of this routine symbol of imperial unity. For centuries already, and another century yet, each Roman year was officially labelled for posterity by the names of the two consuls whose period of office it covered. The annually appointed consuls were announced by each imperial court in Constantinople and Rome, or Ravenna. In the fifth century the consulship was still the pinnacle of any political career, with the honour monopolised by the emperor himself (eighteen times for Theodosius II, for instance), his relatives, and the most influential senators, courtiers, and gener­ als.53 Through marriage and family alignment in particular the civic and military leadership of the East grew closer together, less so in the West.54 Now, however, it was increasingly likely that a consul could be a non-Roman who had risen to supreme military authority: Bauto (385), Stilicho (400, 405), Fravitta (401), Plinta 48 49 50 51

Croke (2010), 241–64; Chapter 1 (above 6–28). Delmaire (1984), 141–75. V.Dan Styl.31. PLRE 2, 160 (‘Ascelpiodotus 1’); 510–11 (‘Gessius 2)’; 1144 (‘Valerius 4’). It is also possible that Rufinus, the praetorian prefect of the East in 431 and recorded as a ‘relative’ of Theodosius II, was a brother of Eudocia (PLRE 2, 953 [‘Rufinus 8’]). 52 Croke (2005a), 147–203; Börm (2010), 183. Likewise, the idea of hardened pro- and anti-Gothic factions during the revolt of the general Gaïnas in 400 has now evaporated (Alan Cameron [1993a], cf. McEvoy [(2016]). 53 Delbrueck (2009); Ravegnani (2006); Sguaitamatti (2012). 54 Demandt (1980), 622–7; Lee (2013), 105–6.

46

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

(419), Areobindus (434), Ardaburius (427) and his son Aspar (434) and grandsons Ardaburius (447) and Ermanaric (465), Zeno (448), Ricimer (459), and Theodoric (484). Although the social and ethnic origins of consuls became more diverse dur­ ing the fifth century, the diversity became less obvious. An increasingly common aristocratic ethos and style meant the way emperors presented themselves publicly as consuls became less distinguishable from the ways other consuls presented themselves in the same role. The demands of iconography and ceremonial sup­ pressed such differences.

Attila and aristocracy As with dynasty, the way the Hun king Attila related to Roman officials and courtiers provides a useful insight into how the tension between Roman and nonRoman in the aristocratic ranks played out. Although endowed with the status of a Roman general by the emperor Theodosius II, Attila never sought a western or eastern consulship for himself or any of his family or his Roman associates. Yet he fully understood the significance of the consul’s position and power. In 449 Attila insisted to the envoy Maximinus, and his companion the rhetor Priscus, who composed a famous account of the embassy,55 that he would only deal with the most senior ex-consuls Nomus, Anatolius, or Senator.56 In fact, the various repre­ sentatives of the eastern and western Roman courts who met with Attila over the years are virtually all known and together they provide a clear insight into Attila’s political style and character, as well as how power was brokered between Romans and Huns. They represent a narrow channel of engagement with the Roman gov­ ernment and imperial courts and aristocracy. Almost all these encounters occurred outside Roman territory. . . . in the East The first Roman envoy Attila encountered was likely Plinta, who was accompa­ nied by the emperor’s chief legal courtier, Epigenes.57 Plinta was a Goth but for almost twenty years (419–438) was one of Theodosius’s most senior generals.58 He had been consul himself in 419. By now quaestor, Epigenes had enjoyed a dis­ tinguished legal career culminating in his recent appointment to the commission established to compile Theodosius’s new legal code.59 The next official occasion when Attila would have encountered imperial authority was in 441 when he and

55 PLRE 2, 743 (‘Maximinus 11’). 56 Priscus, fr. 13 (Blockley 282–283), and Priscus, fr. 11.2 (Blockley 246–247). Theodosius was definitely consigning diplomatic relations with the Huns to a lower level than would be the case with Persia, for example, cf. Mathisen (1986), 43–5. 57 Priscus, fr. 2 (Blockley 224–227), following the chronology of Zuckerman (1994), 159–63. 58 PLRE 2, 892–893 (‘Fl. Plinta’). 59 PLRE 2, 396 (‘Epigenes’).

47

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

his brother Bleda negotiated a settlement with Aspar.60 Like Plinta and Epigenes, Aspar was a powerful and experienced figure at court, as noted above. He had been a trusted general for a long time and by 447 was magister militum in Illyricum,61 having been consul in 434. Here are two perfect examples of how a Goth and an Alan could now secure the highest imperial office through military service and the patronage that flowed from that. Following the Hun attacks through Roman Illyricum and Thrace in 447, the aristocrat Senator, consul in 436, was despatched as an envoy to Attila.62 Over coming months, extending into years (447–450), the successive embassies exchanged between Romans and Huns involved in 447 the long-serving general and consul for 440, Anatolius, along with the Thracian gen­ eral Theodulus,63 then that of Maximinus and Priscus in 449. Next, as Attila had requested, came Anatolius again and the distinguished general and consul for 445, Nomus.64 Finally, in 451, the emperor sent the senior court general Apollonius to negotiate with Attila.65 By now it was clear that the emperor Marcian was termi­ nating the tribute payments to the Huns and spurning them at every opportunity. Attila therefore refused to meet the imperial envoy.66 . . . in the West The western court of Valentinian also deployed aristocratic officials whom Attila had come to know and trust. On one occasion (ca. 445), the imperial envoy was a court official named Cassiodorus, whose eloquence subdued the ferocity of the Hun king. He was an associate of the senior general Aetius, whose son Carpilio accompanied Cassiodorus to Attila, and grandfather of the famous sixth-century orator and administrator for the Gothic king Theodoric.67 The Cassiodorii provide a timely example of a provincial family which emerged in the fifth century to become part of the governing aristocracy of Italy, while the increasing impor­ tance of the provincial aristocracy to the central government is also illustrated by the next known legation, in 449, which consisted of members of the local civil and military elite of Attila’s neighbouring territory inside the Roman Empire, the provinces of Pannonia and Noricum. They were Promotus, the governor of Noricum, and the general Romanus,68 Attila’s secretary Orestes, who was accom­ panied by both his father Tatulus and his father-in-law Count Romulus,69 and 60 61 62 63 64 65 66 67 68 69

PLRE 2, 164–169 (‘Fl. Ardabur Aspar’). Zuckerman (1994), 171. Priscus, fr. 9.2 (Blockley 236–237). PLRE 2, 84–6 (‘Anatolius 10’); 1105–1106 (‘Theodulus 2’). PLRE 2, 785–6 (‘Nomus 1’), see also Croke (1981), 159–70. PLRE 2, 12 (‘Apollonius 3’). Hohlfelder (1984), 54–69. PLRE 2, 264 (‘Cassiodorus 2’). PLRE 2, 926 (‘Promotus 1’); 946–947 (‘Romanus 2’). PLRE 2, 1055 (‘Tatulus’); 949 (‘Romulus 2’). Tatulus and Romulus were the grandfathers of the emperor Romulus Augustus installed by his father Orestes in 476, Kos (2008), 439–49.

48

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

another of the secretaries provided for the Huns by Aetius, namely Constantius.70 The most important western legation was that sent by Valentinian in 452, when Attila’s attention had turned to Italy, following his defeat the previous year by the combined forces of Aetius and the Goths. Near Mantua, the embassy succeeded in turning Attila away from Italy. Pope Leo’s participation in the embassy has resulted in him acquiring exclusive credit for taming the rampaging Hun king and their direct confrontation became renowned in word and image. It is too often forgotten, however, that Leo was accompanied by two senior local aristocrats, namely the experienced ambassador and now praetorian prefect Trygetius71 and the consul of 450, Gennadius Avienus, whom Sidonius Apollinaris later consid­ ered the most distinguished man in Rome after the emperor.72 If the channel of engagement between the Roman courts and Attila was limited and narrow, all the more so was the reverse channel, that between Attila and the Romans. During Attila’s reign there were three Roman emperors, Valentinian III in the West and in the East Theodosius II, followed by Marcian, but the Hun king never came face to face with any of them. Dealings with emperors and their courts he left to his associates who were expert in Roman ways and Roman bureaucracy, men such as Constantius and Orestes. In doing so Attila was merely following the model of Hun kings before him: Charaton, for instance, whom the historian Olympiodorus met in 41273 and Rua, who dealt with Roman envoys in 422 and 435.74 The only Huns who ever travelled to Constantinople or Rome or Ravenna were few and had limited impact: Scottas, one of the leading Hun warriors, in 447;75 Edeco, a highly regarded soldier and one of Attila’s personal bodyguard, in 449;76 Eslas, twice in 449, who had represented King Rua in negotiations with the imperial court in 435;77 and finally a senior Hun ruler named Berich.78 Notwith­ standing the limitations of surviving documentation, there is not a single record of any of these Hun envoys ever actually negotiating with Emperor Theodosius II, nor of any comparable engagement with Valentinian III. Protocols of embas­ sies would normally have at least required some ceremonial reception, such as Edeco having the Huns’ letter read out in a formal palace audience in the presence of Theodosius.79 Attila never permitted or encouraged the most illustrious of his own colleagues to become part of the Roman aristocracy. He never even forged close personal or family links with Roman aristocrats and the most senior imperial officials, East or West. 70 71 72 73 74 75 76 77 78 79

PLRE 2, 319 (‘Constantius 7’). PLRE 2, 1129 (‘Trygetius’). PLRE 2, 193–194 (‘Gennadius Avienus 4’), see also Wessel (2008), 45–7. PLRE 2, 283 (‘Charaton’). PLRE 2, 951 (‘Rua’). Priscus, fr. 9.3 (Blockley 239), see also PLRE 2, 983 (‘Scottas’). PLRE 2, 385–386 (‘Edeco’). PLRE 2, 402 (‘Eslas’). PLRE 2, 225 (‘Berich’). Priscus, fr. 11.1 (Blockley 242–230), enriched by the reconstruction of Kelly (2008), 120–6.

49

DYNASTY AND ARISTOCRACY IN THE FIFTH CENTURY

Conclusion On the very first day of the year 400 two men celebrated the ceremonial com­ mencement of their consulship, Stilicho at Rome and Aurelian at Constantinople. Both had been strong supporters of Theodosius I and owed their preferment to him. Stilicho was a military man and Aurelian the consummate civil administrator who had been prefect of the city (393–394) and was to be twice praetorian pre­ fect of the East (399, 414–416).80 Stilicho had secured his influence by marrying into the imperial family, while Aurelian’s civic administrative ability led to his high honours, wielding authority for Theodosius I’s young son Arcadius and his even younger grandson Theodosius II. Aurelian’s son Taurus imitated his father’s double prefectures (433–434, 445) and was consul in 428.81 A century later, on 1 January 500, two very different men entered on the same office, but both were at Constantinople: the Dalmatian Hypatius owed this honour to his kinship with the reigning emperor Anastasius, and then went on to hold senior generalships for another thirty years.82 The career of the Phrygian Patricius had reached a peak but he remained an influential general until the 520s.83 The West was now ruled by the Gothic king Theodoric at Ravenna, and no consular candidate from the western aristocracy or royal family was advanced for recognition at Constantinople. In the profoundly different worlds of 400 and 500, separated by enormous politi­ cal, social, and cultural change in a diminished Roman Empire, their respective consuls illustrate the enduring power of dynasty and aristocracy throughout the fifth century.

80 81 82 83

PLRE 1, 128–129 (‘Aurelianus 3’). PLRE 2, 1056–1057 (‘Fl. Taurus 4’). PLRE 2, 577–581 (‘Fl. Hypatius 6’). PLRE 2, 840–842 (‘Fl. Patricius 14’).

50

3 DYNASTY AND ETHNICITY Emperor Leo I and the eclipse of Aspar*

There was absolute uproar at Constantinople in 471, sometime after June, when news spread of a bloodbath inside the imperial palace. The polygamous, longserving and senior Roman general Aspar, together with two of his senatorial sons, was invited to a meeting of senators but stepped into an ambush. Aspar and his eldest son Ardaburius, an experienced general himself, were assassinated by the palace eunuchs on instruction from the emperor Leo I. Patricius, younger son by Aspar’s second wife, was seriously wounded but managed to escape. He had only recently been married to the emperor’s daughter Leontia, and designated as Leo’s imperial successor by acquiring the title of Caesar. Herminericus, Aspar’s youngest son by his Gothic third wife, was absent from the city at the time.1 For an emperor to move in such a fashion on his new son-in-law and chosen successor he must have been desperate. In 471 someone of Aspar’s status as the most senior dignitary of the Roman state naturally had many allies at court, and in the imperial capital. One of them, a Roman military officer named Ostrys, who was stationed near the city, suddenly turned attacker at the palace, was surrounded by the guards but managed to escape, making off with Aspar’s concubine. The emperor was rescued from any further danger by the timely arrival of two other key generals, his brother-in-law (Basiliscus) arriving from Herakleia and his other son-in-law (Zeno) arriving from Chalcedon.2

* This chapter first appeared under the same title in Chiron 35 (2005), 147–203 and is reproduced here in a revised form by permission of the Ancient History and Epigraphy Commission of the German Archaeological Institute (Munich) and CH Beck. Despite the title, this paper has been criticised for underplaying, denying even, the influential role of the key participants’ ethnicity compared to their dynastic connections and their actual positions as generals in charge of trained and experienced polyethnic Roman armies. A fuller explanation of the limited role of ethnicity in Roman political episodes in the 5th century East is therefore embedded in this version, along with other modifica­ tions and additions. Debate will continue. 1 The essential records of these events, in order of composition, are Candidus, frag.1 (Blockley 466– 7) = Photius, Bibl. 79; Marcellinus, Chron. s.a. 471 (MGH AA XI, 90); Malalas, 14.40 (Thurn 294); Jordanes, Romana 338; Procopius, Wars 3.6.27; Vict.Tonnonensis, Chron. s.a. 470 (ed. Hartmann, CCh 173a, 35 [12] (= MGH AA XI 188); Evagrius, HE 2.16; Chron Pasch. 596.17–597.9. 2 Malalas, Chron.14.40 (Thurn 294–5); Theophanes, AM 5964 [de Boor 117–18]).

51

DYNASTY AND ETHNICITY

The murder of Aspar and his sons in 471 is normally portrayed as marking a decisive turning point in the fortunes of Roman imperial power in the fifth centu­ ry.3 On the one hand, it removed the political domination of Aspar’s family that had extended for just on half a century, from the early 420s through the reigns of Theodosius II, Marcian and Leo I. On the other, it paved the way for a new but less comprehensive kind of domination from another quarter by consolidating the influence of Zeno and his ethnic allies. The ‘butchery’ of Leo is traditionally cast as a battle for ethnic supremacy at the imperial court, the replacement of the ‘Gothic, or German/Gothic, faction’ with the ‘Isaurian faction’.4 The old but still influential study of Brooks claims that from the mid-460s ‘there were two factions at the court of Constantinople, the Isaurian and the barbarian, which for convenience we may call the Gothic faction. For the next twenty years the history of the empire turns upon the struggle between these factions’.5 In an extended metaphor, Ostrogorsky claimed that ‘in its attempt to find relief from German pressure, the suffering empire had swallowed the Isaurian antidote. This worked, but it was an over powerful dose, and the body of the empire was correspondingly affected’.6 More recently, this approach is echoed by Alexander Demandt who reverses events by claiming that under Aspar’s leadership an Alan–German fam­ ily alliance was forged in opposition to the Roman–Isaurian one which prevailed under the emperors Leo and Zeno.7 More specialised studies, focussed on Aspar and his individual career, are less extravagant but tend to both overestimate his role and to isolate him from the wider political context. Typically, J. B. Bury concluded that the vanquishing of Aspar by Leo was ‘an important act in the long struggle against the German danger in the East’ and that it was ‘the price which had to be paid for the defeat of the German generals who sought to appropriate the Empire’.8 This rather grandiose sentiment is echoed in the more recent contention that ‘Leo had broken the Germans’ hold on the govern­ ment. That delicate and difficult task had been necessary if the eastern empire was to prosper, and perhaps even if it was to survive’.9 On any reconstruction political developments in the years and months leading up to Aspar’s murder at the court of Leo in 471 were fast-moving and complex 3 The most detailed accounts are older ones: Brooks (1893), 212–14; followed by Seeck (1920), 369– 71; Bury (1923a), 320; Ensslin (1925), 1947–62; Stein (1959), 361; then came a sober dissenting voice in Jones (1964), 224. The most authoritative recent works demonstrate the same interpretative pattern: Averil Cameron (1993b), 30; Treadgold (1997), 150–6; Demandt (1989), 185–7; and, albeit more nuanced, Lee (2000), 45–9. The most cogent interpretation is to be found in Snee (1998), 157–86, and the most recent in Crawford (2019), 85–92. 4 Vernadsky (1941), 38–75; Scott (1976), 59–69. 5 Brooks (1893), 212. 6 Ostrogorsky (1969), 63, cf. Vasiliev (1958), 104: ‘The Emperor decided to free himself of Germanic power and with the aid of a number of warlike Isaurians quartered in the capital killed Aspar and part of his family, dealing a final blow to Germanic influence at the court of Constantinople’. 7 Demandt (1970), 771. 8 Bury (1923a), 316, 318, 320. 9 Treadgold (1997), 155–6.

52

DYNASTY AND ETHNICITY

with many crosscurrents of treachery. Even if the extant historical records were complete and abundant it would be challenging enough to establish precisely what was happening in the period from 457 to 471, and how to interpret it. A distin­ guished scholar once noted in passing, about this very period, that ‘fortunately it is not our task to straighten out the confused chronology of those years’.10 In fact, the records are few, fragmentary and tendentious which explains the ensuing problems of confused chronology and the frustration of having to concede that ‘a debate rages over the interpretation of meagre information’.11 This chapter sets out to examine the period from 457 to 471 by suggesting that, despite ‘meagre information’, there is need for a fresh interpretation including an attempt ‘to straighten out the confused chronology of those years’. It is designed to bring into a single interpretative framework a range of recent research, but is focussed on a detailed reconsideration of certain key episodes: (1) the circum­ stances surrounding Zeno’s uncovering of Ardaburius’ treachery with the Persian king in 465, and its implications for Zeno’s position and career; (2) Aspar’s role in the failed campaign against the Vandals in 468; (3) the purpose and consequence of Zeno’s appointment as magister militum for Thrace which must be dated to 469, not 467; and (4) the actions of Aspar and Leo during Zeno’s absence from Constantinople, as magister militum per Orientem in Antioch, between mid-469 and mid to late 471. The misunderstanding of these episodes, exacerbated by the usual resort to simplistic and polarised ethno-factional explanations, has obscured the underlying developments and the shifting allegiances of generals and impe­ rial officials. Recent research on the subsequent reign of the emperor Zeno has swung attention away from factional explanations. What once looked like a clash of Isaurian factions during Zeno’s sole reign (474–91) turns out to be simply divi­ sions between two families and their political allies.12 Was it factional or some other form of conflict which underpinned the various political goals and actions of generals and courtiers during the reign of Leo I? Was Leo’s ‘butchery’ of Aspar and sons merely the culmination of a rivalry between two self-conscious ethnic groups struggling to assert superiority at court? Living and researching in an era when ethnicity is considered an essential element of anyone’s personal identity, was it so central as to be a clear and prime motivator of human behaviour in the fifth century, as sometimes argued?

The families of Aspar and Leo On the death of the emperor Marcian on 27 January 457 there was a hiatus. No immediate appointment was made to the vacant throne. The army did not clamour for any new emperor, the senate did not have a preferred candidate in waiting. Eleven days were to pass before another emperor was announced on 7 February. 10 Maenchen-Helfen (1973), 168 n.837. 11 Clover (1978), 174. 12 W. Burgess (1992), 874–80; Elton (2000a), 393–407 (important study).

53

DYNASTY AND ETHNICITY

The unusually lengthy interlude suggests not only indecision about who might actually perform the coronation rite, there being no western emperor and no living member of the eastern imperial family at the time,13 but also a power struggle. In any such struggle the dominant player would be the senior court general Aspar, as he had been on the death of Theodosius II in 450 who was survived by his pow­ erful and protective sister Pulcheria.14 Certainly there were eligible candidates, foremost among them being Anthemius, the son-in-law of the previous emperor Marcian and grandson of an eminent courtier of Theodosius II. Indeed, it was widely anticipated Anthemius would be the new emperor although no Theodo­ sian blood ran in his veins. Perhaps he so hoped himself but lacked the necessary ‘cupido imperii’, as a later panegyrist politely excused his claim.15 Still, Anthe­ mius was an experienced general, in fact sharing power at the time with Aspar who was the other court general (magister militum praesentalis) but the senior of the two.16 For whatever reason, Aspar would not countenance the elevation of Anthemius.17 The senate deliberated, so we are told, and the choice fell not on any living member of the imperial families of Theodosius or Marcian after all. Instead, it was resolved to make a new dynastic start. Aspar could possibly have seized the throne for himself. After all, he had been a senior Roman general and courtier for over thirty years and consul over twenty years ago (434). Choosing to back his son Ardaburius would not be a surprise either. He too was a very experienced Roman general, then magister militum per Orientem, and had been consul ten years ago (447). At a church council in Rome thirty years later it was reported by the Gothic king Theodoric that Aspar was actu­ ally offered the throne by the senate in 457 but declined with the enigmatic obser­ vation that ‘I fear I would launch an imperial tradition’.18 Instead, he chose Leo, the comes et tribunus Mattiariorum, the tribune from a regiment of the magister 13 Avitus had been deposed on 17 October 456 but his successor Majorian was not proclaimed until 1 April 457. The daughter of Theodosius II, Licinia Eudoxia (PLRE 2, 410–12), an imperial widow now, having been the wife of Valentinian III (died 455), was living in abducted exile in Vandal Africa. 14 It is very likely that Aspar was overwhelmingly powerful at court in order to influence the appoint­ ment of Marcian as emperor in 450, as argued by R. Burgess (1993–4). For contrary views: Chew (2006); Pigonski (2018). 15 Sidonius Apollinaris, Carmina 2.210–12. 16 For Anthemius as magister militum praesentalis: Demandt (1970), 777. His tenure of this position is overlooked in the list of magistri in PLRE 2, 1290. 17 PLRE 2, 97 (‘Anthemius 3’), following Seeck (1920), 355–6. 18 Contained in the record of a council held in Rome in 501 (Acta synodorum habitarum Romae, 5 [MGH AA 12, 425]): ‘Aliquando Aspari a senatu dicebatur ut ipse fieret imperator, qui tale refertur dedisse responsum: timeo ne per me consuetudo in regno nascatur’. It is not clear exactly what Aspar had in mind on this occasion, nor whether it occurred after the death of Theodosius II when Marcian was chosen or after the death of Marcian, although the latter seems more likely (Stein [1959], 353–4). He probably meant being an Arian above all (Procopius, Wars 3.6.3, cf. Demandt [1970], 770; von Haehling (1988), 97–103 is unnecessarily doubted by Kaegi (1981), 26 n.33). Remembering that Theodoric was a virtual hostage at the imperial court throughout the 460s, there is every likelihood that he heard Aspar tell this story at first hand.

54

DYNASTY AND ETHNICITY

militum praesentalis stationed at Selymbria (Silivri),19 just a day’s march from Constantinople. It was an unusual choice. Leo does not appear to have had any prior imperial connections. He was of Bessan stock, born and raised in the Balkans and in his mid-fifties at the time.20 Aspar, who was not much older than Leo, must have had some close connection with him. Evidently Leo had once been in charge of Aspar’s considerable estates, his curator,21 and was now heading up one of Aspar’s main military units. He was a relatively senior and experienced military figure, no humble soldier plucked from obscurity as he has sometimes been por­ trayed.22 Perhaps Leo and Aspar had served together in the army over many years. Whatever inspired the choice, Leo was emperor because of Aspar and would natu­ rally be expected to comply with Aspar’s preferences and plans.23 Aspar’s ultimate power was his capacity to manage the imperial succession as he had done on the death of Marcian in 457, having earlier played a key role in the elevation of his former domesticus Marcian following the death of Theodosius II in 450. If Aspar were to maintain and extend his considerable power and influence, however, it was essential to retain control of the succession. The coronation of the new emperor took place with due ceremony on 9 Febru­ ary 457 and its conduct underlined the extraordinary power of Aspar. There are two surviving accounts of Leo’s coronation: one is preserved in Constantine Por­ phyrogenitos’ Book of Ceremonies;24 the other preserved as an interpolation in a 19 Const Porph, On the Ceremonies 1.91 (Reiske 411.4, trs. Moffatt and Tall (2012), 411). Leo was probably tribune of the Matiarii seniores rather than the Matiarii iuniores although it is not possi­ ble to be definite, cf. Lammert (1956), 2322–3; Hoffmann (1969), 488; Salamon (1994), 188 n.38. The regiment was one of the six ‘legiones palatinae’ under Aspar’s command (Notitia Dignitatum Or.vi.42, ed. Seeck [1876], 17). 20 There are various different descriptors for Leo (references in PLRE 2, 663 [‘Leo 6’). Together they point to someone from a military background and indicate geographical (Dacia, the province in Illyricum) rather than ethnic (Bessan) origin, even though the Bessi were a Thracian nation. Leo is generally called ‘Thracian’ by contemporaries. In any event, to call someone ‘Bessan’ (or whatever) at this stage was arguably no different from modern descriptors of political leaders, those which refer to someone as being of ‘Hungarian background’, ‘Chinese origin’, ‘Russian descent’ and so on. 21 Theophanes AM 5961 (de Boor 116.), Zonaras 13.25.35 (121.13–122.2 Büttner-Wobst) with Jones (1964), 221. 22 E.g. Blockley (1992), 71; Lee (2000), 46: ‘A relative non-entity of Balkan origin . . . the unknown Leo’. 23 Jordanes, Romana 335 (‘Asparis patricii potentia’); Procopius, Wars 3.5.7 (:Asparoj evj tou/to auvto.n katasthsame,nou) cf.3.6.3; Priscus, frag. 19 (Blockley 305/6) = Suda A 3803 (auvtokeleu,stw| gnw,mh|); Candidus, frag. 1 (Blockley 464/5) = Photius, Bibl. 79: th.n basilei,an spoudh/| :Asparoj evgceireisqei,j . . . th.n avna,rrhsin dia. tou/ :Asparoj Le,ontoj) cf. Ensslin (1925), 1957 (Aspar as ‘Kaisermacher’). He soon appears in the letters of Pope Leo as the ‘illustrious patrician’, a sort of co-ruler (Epp., 149.2, 150, 153.1), cf. Seeck (1896), 608; Demandt (1970), 771; Vernadsky (1941), 57 (‘Mitherrscher’). 24 Const.Porph., On the Ceremonies, 1.91 (Reiske 414.11–13, trs. Moffatt and Tall [2012], 414), derived from Peter the Patrician, with the remarks of MacCormack (1981), 243–6. Aspar’s role is not analysed in Lilie (1998), 395–408, which seeks to separate the original account of Peter from later editorial accretions. It was Aspar’s Arianism which made him an unsuitable person to

55

DYNASTY AND ETHNICITY

12th-century Hebrew manuscript (Paris.Hebr. 1280) but clearly deriving from a Greek original.25 Following his acclamation by the army at the Hebdomon out­ side Constantinople Leo undertook a lengthy processional route through the city to the imperial palace. According to the Book of Ceremonies account there was a key role for Aspar in the ceremony. After an imperial costume change at the Helenianae palace not far inside the city Leo boarded the imperial carriage for the next part of the journey. Joining him in the vehicle was the ‘foremost patrician’ (o‛ prw/toj patri,kioj), Aspar at that time, who kissed the emperor’s hand on board­ ing.26 When the imperial carriage reached the Forum of Constantine Leo alighted to greet the senate and the Prefect of the City before being offered the gold crown (modiolus) by the head of the senate (o‛ prw/toj tw/n suglhtikw/n), Aspar again.27 The remainder of this ceremony in the forum, the imperial palace and the hippo­ drome involved much reciprocal acclamation. Aspar was never far from the side of his new imperial protégé. It would have been obvious what the relationship was between the two. Still, it would be interesting to know how Aspar was portrayed in the mosaic depicting Leo’s accession which was set up in the new headquarters of the Praetorian Prefect at Constantinople in about 470.28 Aspar’s power to make and, if required, unmake emperors in the 450s, which was so manifest during Leo’s coronation ceremony, had developed as part of the consolidation and perpetuation of a new military aristocracy. The senatorial land­ owners of the fourth century had been forced to compete for influence and office with new generations of professional soldiers, many of them of barbarian origin who were soon passing their authority to the next generation.29 Aspar belonged to the most powerful of these new dynasties, the Ardaburii. Aspar’s family, report­ edly of Alan descent,30 emerged during the reign of Theodosius II and united

25

26 27 28 29 30

crown the emperor, thereby establishing the precedent of patriarchal crowning (argued by Ensslin [1942/6], 370). This detailed but stylised account of an imperial coronation at Constantinople, but without ever naming the emperor or any of the officials, needs to be treated with due caution. It could merely be a composite and generic description. However, a case has been made for seeing it as the eye­ witness record of a specific coronation, namely that of Leo I: del Medico (1955), 43–75 (text at 50–3, followed by detailed commentary). Del Medico concludes (75) that this coronation account represents ‘l’impression d’un homme de la rue, d’un juif perdu dans la foule des chrétiens et de païens qui assistaient au passage du cortège impérial’. Aspar is also called ‘first patrician’, that is the most senior of all the patricians, by Marcellinus (Chron. s.a. 471, MGH AA XI 90: ‘Aspar primus patriciorum’). Aspar was the most senior senator (‘princeps senatus’), as noted by Malalas, Chron. 14. 40 (Thurn 371.10): :Aspara to.n patri,kion w`j kai. prw/ton sugklh,tou cf. Chron Pasch. 596–7 (Dindorf). John Lydus, On the Magistrates 2.20. The building, if not the mosaic itself, was commissioned by the Praetorian Prefect Constantine (PLRE 2, 312–13, [‘Constantinus 8’]) and named after the emperor. Demandt (1980), 609ff; Heather (1991), 262. Seeck (1896a), 607, with, on the ethnic background, Alemany (2000), 82–4. It is possible, however, that Aspar’s family was of African rather than Alan origin, cf. Vattioni (1980), 191–4 (focussed on the Aspar mentioned by Sallust [Jug. 108.1: ‘quidam Aspar nomine’] as an envoy of Jugurtha, but not discussing the late Roman general of that name).

56

DYNASTY AND ETHNICITY

almost all the Eastern generals, irrespective of ethnicity, in a single house. His father Ardaburius was an outstanding general who had distinguished himself dur­ ing the war against the Persians in 421/2 and who was connected by marriage to Plinta (cos.419), an Arian Goth and magister militum praesentalis, like Aspar, and the most powerful man at court.31 Ardaburius was consul in 427 and in 434 his son Aspar shared the consulship with Areobindus, a Goth who, like Aspar’s father, had come to notice during the ‘great war’ with Persia and later acted as magister militum per Orientem.32 Areobindus became patrician and consul (434), then the father, grandfather and great-grandfather of consuls (461, 491, 506). He was, in effect, a Roman noble of the highest rank and status. So by the 430s we find the Roman generals fully integrated into political and social life at Constantinople. They held high office and celebrated it in the time-honoured Roman way. They sought to advance their family and their personal interests by strategic marriage alliances with other aristocratic families, both military and civilian. Generals of Gothic background did not constitute a self-styled faction or cabal with its own overriding ethnic agenda. Ethnicity, however defined, was less important as a determinant of political behaviour than the competition for wealth, power and office-holding. Increasingly too, ethnicity was less significant than religion. Both Aspar and Ardaburius were Arians, even though the divisive theological issues of the day were mainly not those of Arius but more recent ones involving the rela­ tionship between the human and divine natures in Christ, which successive church councils (Ephesus 431, 449; Chalcedon 452) had tried to resolve. Besides his three sons, Aspar had two daughters.33 Although there is no indica­ tion of the identity of their spouses, we can be confident that he will have ensured they were favourably placed from his point of view. The same expectation also fell 31 Sozomen, HE 7.17.14: dunastw,tatoj to,te tῶn evn toi/j basilei,oij gegonw,j. It is not known how Plinta achieved such influence and authority but presumably his military skill had originally shone out. Since the relationship between Plinta and Ardaburius is likely to have been by marriage we see instantly how two Roman generals united their families, as Roman aristocrats had always done. Plinta’s daughter was probably Aspar’s first wife and therefore the mother of the younger Ardaburius (Seeck [1896], 606, ‘Ardabur 1’); Vernadsky (1941), 44, cf. PLRE 2, 892 (‘Fl. Plinta’), while Plinta’s son Armatius had married a woman of great wealth and distinguished pedigree (Pris­ cus, frag.15.4 (Blockley 299–300) = Exc.de Leg.Rom. 5, with PLRE 2, 148 (‘Armatius’). Armatius had won honours in the 440s fighting in Africa but had died soon after. His aristocratic wife was then espoused, at the suggestion of the emperor Theodosius II himself, to Attila’s secretary Con­ stantius (PLRE 2, 319 [‘Constantius 7’]). Both Plinta and Ardaburius remained Arians even though the imperial court was intensifying its campaign against heretics at the time, and both dealt with Attila’s Huns in the 440s. 32 On Areobindus: PLRE 2, 145–6 (‘Fl. Areobindus 2’); Demandt (1970), 752–3. Aspar himself was already a magister militum too and continued the effective influence of his father (PLRE 2, 164–9 [‘Ardabur Aspar’]); Demandt [1970], 748ff). His consulship was commemorated in a famous silver dish, now in the Museo Nazionale del Bargello in Florence where the consul Aspar is repre­ sented like any other senatorial aristocrat, seated between the effigies of Rome and Constantinople: Painter (1991). 33 Candidus, frag. 1 (464–5 Blockley) = Photius, Bibl. 79. That neither daughter is ever heard of otherwise may indicate that neither survived to marriageable age.

57

DYNASTY AND ETHNICITY

on his son Ardaburius, depicted as a young Roman aristocrat on Aspar’s consular dish in 434. It is therefore unlikely that he would have married a Cilician seer named Anthusa.34 The silver chalice at Dumbarton Oaks which bears the inscrip­ tions of Ardaburius and Anthusa is probably that of husband and wife. However, a different Anthusa is involved here. Indeed, it has been argued that she was the daughter of Illus, a leading but volatile figure in the reign of Zeno in the 470s and 480s.35 In that case we would have testimony to the unification of a Gothic family and an Isaurian family at precisely the time when Goths and Isaurians are construed as fighting each other for imperial attention. Given the lengthy time the Ardaburii had enjoyed office and influence, it should hardly surprise that over a quarter of a century later, on the accession of Leo, their control was paramount. Aspar could hardly have been in a more powerful position than he was in 457. By now Aspar had acquired a second, then a third, wife which brought him into relationship with the royal families of the Goths then located at different places on Roman soil. In particular, in the 450s Aspar had married into the same family as Theodoric Strabo, leader of a large group of Goths which had been settled in Thrace since the 420s and with whom he had long had dealings. On Aspar’s death in 471 Strabo reacted promptly and demanded the legacy of Aspar for himself. Exactly what this relationship was, and when Aspar married into the family of Strabo, are not certain. Our best guess is that he married the sister of Strabo’s mother or father, Triarius, so that Strabo was Aspar’s nephew by marriage, with their son Herminericus being born around 445–450, that is, in the reign of Theodo­ sius II when Attila’s Huns were threatening the imperial capital.36 In other words, Aspar formed an alliance with a high-ranking Gothic woman in the generation before Strabo’s rise to power. Aspar obviously had long had dealings with the Goths and with the family of Strabo their current leader. The alliance between Aspar and Theodoric Strabo was not newly confected during the reign of Marcian, 34 PLRE 2, 100 (‘Anthusa 1), with Demandt (1986), 113–17. Demandt links the chalice to the story of Anthusa the cloud-seer told by Damascius (Photius, Bibl. 242. 69) but to do so he has to make the unlikely assumption that Ardaburius took a leading part in the campaign against the Vandals in 468 (stratiwtikh,n tina avrch.n). A further difficulty with Demandt’s identification of Ardaburius as the husband of the seer is that in describing the death of Aspar and sons no connection is made, as one might have expected, with Anthusa’s husband: e[peita to.n h`gemo,na tw/n Go,tqwn :Aspara basileu.j Le,wn evdolofo,nhsen kai. pai/daj. 35 Scharf (1993), 213–23 arguing that the Cilician seer was the spouse of the general Damonicus (PLRE 2, 344–5) and that Anthusa, daughter of Illus (PLRE 2, 586–90), was the second wife of Ardaburius with a wedding sometime around 469/70. This looks problematic (as already observed by Feissel [1999], 15). 36 Heather (1991), 255 n.40 (mother); Demandt (1986), 114 (father). As for the date of the union between Aspar and the relative of Strabo, there is no direct indication. Since Hermeniricus was the offspring of that union it might help to consider when he could have been born. The only possible clue is to speculate from the date of his consulship. A consul in 465, even the youngest son of the most powerful man in the state, is not likely to have been a mere child. If we assume Hermeniricus was somewhere between 15 and 20 years old at the time, but recognising that he could have been younger or older than that, then he will have been born around 445–50, but it could have been much earlier.

58

DYNASTY AND ETHNICITY

or that of Leo, in order to bolster his authority with Gothic warriors. The Gothic foederati had formed part of the Roman army led by Aspar against Attila and the Huns in the 440s. When Leo became emperor in 457 Gothic soldiers had long been an integral part of the Roman military establishment, but so were Isaurians. Like Illyrians, (inhabitants of Roman Illyricum), Isaurians (inhabitants of Roman Isauria) were considered a particularly valiant and hardy people well-disposed to army life and action. Like the Illyrians, Isaurians were disproportionately represented in the ranks of the Roman army and in its leadership wherever the army was stationed37. There were even specific Isaurian units in the army but they also included many individual soldiers from other ethnic backgrounds and cultures.38 Nor were all Isaurian soldiers in those units. With so many soldiers of Isaurian origin it was inevitable that there should be a significant number of Isaurian officers. Even in the 440s there were Isaurians in the highest military and civil positions.39 The most famous of them was the magister militum Zeno. This Zeno, like Aspar and Areobindus, had acquired the highest offices and dignities in the Roman state as consul in 448 and patrician. Further, he had married a certain Paulina who was of sufficient wealth and aristocratic prominence to have a district of Constantinople named after her mansion (ta. Pauli,nhj). Her recently discovered pavement verse to her husband Zeno evokes a cultured partnership.40 Just as other such mansions attracted the owner’s family and supporters, Paulina’s may have become a centre for notable and aspiring young aspirants from Isauria.41 So, while this general’s later namesake Zeno may have become the most prominent of the Isaurian offi­ cers in the Roman army of the 460s he will not have been the only one. Nor was a Roman military officer of Isaurian background unusual by then. Leo’s elevation in 457 also opened up unforeseen opportunities for his own family and that of his wife Verina, although neither could boast an aristocratic or traditional office-holding household.42 Leo’s only family seems to have been a sis­ ter Euphemia whom he used to visit every week at Constantinople and who later dedicated a statue to her imperial brother,43 while Verina had a brother Basiliscus44 37 Elton (1996), 394. 38 Elton (1996), 133 and (2000b), 295. 39 Elton (2000a), 396. One such is the comes Longinus who was sent to assist the bishop of Tyana in c.450 (PLRE 2, 687 [‘Longinus 1’]), cf. Lenski (1999), 426, 451–2. 40 Şahin (1991), 155–63 (text at 156), with Feissel (1999), 9–11. This inscription was set up in Seleu­ cia, demonstrating thereby that Paulina was honoured both in the imperial capital and in Isauria. 41 Explained in Feissel (1999), 9–11. Given the tight marriage connections of the military and civil office-holding families at this time, it is possible that this Paulina was either the daughter or sister of the magister officiorum Paulinus, the discredited intimate friend of Theodosius II (PLRE 2, 846–7 [‘Paulinus 8]). 42 Salamon (1994), 187–8; Leszka (1998), 128–36. 43 Patria Constantinopoleos 2.31 (Berger [2013], 70–1]) with PLRE, 422–3 (‘Euphemia 3’). 44 PLRE 2, 212–14, Fl. Basiliscus 2, with the detailed study of. Salamon (1994), 179–96. The exchange between Krautschick and Brandes concerning the relationship between Basiliscus and Odoacer is not conclusive: Krautschick (1986), 344–71; Brandes (1993), 407–37; Krautschick

59

DYNASTY AND ETHNICITY

and at least one other sister. Their sister’s husband, named Zuzus, appears to have come from Thrace, like Leo.45 A younger relative whose career was to eventu­ ally prosper significantly was Armatus, possibly the son of Zuzus.46 In addition, in choosing to support Leo as emperor in January/February 457 Aspar had suc­ ceeded in terminating the entrenched dynasty of Theodosius. Yet, there were still strong claimants who might have expected to succeed Marcian, particularly (as already noted) the patrician Anthemius who was the husband of Marcian’s daugh­ ter Euphemia.47 Then there was Anicius Olybrius now the husband of Placidia, granddaughter of Theodosius II.48 He was a distinguished senator at Rome but had fled to Constantinople in 455 to the protection of Marcian’s court. There he awaited the return from exile in Vandal Africa of his wife Placidia and his mother­ in-law Eudoxia Augusta, wife of the western emperor Valentinian III. Olybrius also had links with the other living Augusta resident in Jerusalem, namely Eudocia the wife of Theodosius II. The power and prestige of Anthemius, Olybrius and the Theodosian women, even in their absence from the imperial capital, could not be suppressed or dismissed lightly. For three generations already, Aspar had been at the centre of a series of devel­ oping marriage and other alliances. Opportunities had presented themselves, and he had exploited them. Aspar’s authority relied on securing the throne for his favourite, as he had done in 450 and 457. It was a long-term venture not something to be considered only on the emperor’s death. Theodosius II had no son, Marcian had no son. He did have a daughter but, as Marcian himself proved, the power of the throne could only devolve on men irrespective of the lineage, prestige and influence of imperial women.49 Since Leo had but a single daughter at his acces­ sion, the young Ariadne, it is the men in her later life who would be of prime importance. If, as an Arian, Aspar could not be emperor himself he could surely be the father of an emperor. This seems to have been Aspar’s natural strategy for extending his influence and that of his family for the next generation. Rather than take any chance, Aspar made Leo pledge at his accession (457) that the hand of his daughter Ariadne would be reserved for Aspar’s son Patricius. At the same time, Aspar and other senior officials then present in Constantinople were required by the emperor to swear an oath that they would never plot against him, with the emperor retaining the text of the oath.50 Stability appeared to beckon.

45 46 47 48 49 50

(1995), 332–8. Were two such prominent political leaders so closely related it is odd that such a central fact should not remain more prominently noticed even in the scanty records surviving from the period. PLRE 2, 1207 (‘Zuzus’), with Salamon (1994), 188–9. PLRE 2, 148–9 (‘Armatus’). PLRE 2, 96–8 (‘Anthemius 3’). PLRE 2, 796–8 (‘Anicius Olybrius 6’). Olybrius had made a special trip to Carthage to marry Pla­ cidia not long before Leo’s accession according to the careful argument of Clover (1978), 192–4. Holum (1982), 1–5. Const. Porph., On the Ceremonies, I.91 (Resike 416, tr. Moffatt and Tall [2012], 416).

60

DYNASTY AND ETHNICITY

Aspar’s ascendancy, 457–465 As customary, the new emperor took the consulship in the year following his acces­ sion (458), so 459 became the earliest opportunity that this still prestigious and coveted honour could be conferred on Aspar’s son Julius Patricius.51 Aspar was already hoping for more, namely for Patricius to be named Caesar which would formally and publicly secure a right of succession, to be cemented by an already agreed marriage as soon as practicable with the emperor’s daughter.52 Yet, as Ari­ adne grew up, Leo procrastinated in the hope of keeping the throne within his own family. Meanwhile, Aspar and Ardaburius retained their military commands while another supporter of Aspar, Vivianus, became Praetorian Prefect of the East.53 The continuing influence of Aspar is also evident in the appointment of Dagalaiphus as consul for 461. Dagalaiphus was well known to Aspar as the son of his consular colleague in 434, Areobindus. More significantly, the new consul was the husband of Aspar’s only grandchild, Godisthea, the daughter of Ardaburius and Anthusa.54 This strong link further bound the military elites together and equally reinforced the position of Aspar who remained in the ascendency, asserting authority and influence in a way that had not been possible under Marcian. The son of Dag­ alaiphus and Godisthea, named Areobindus Dagalaiphus, would become consul in 506, having married the granddaughter of Valentinian III, Anicia Juliana.55 In other words, thirty years after Aspar’s death, his great-grandson had married into the imperial line in an uninterrupted dynastic succession. Back in 457, Aspar did enjoy considerable influence under Marcian who, like Leo, had been his former employee. By the end of 462, the consulship of Vivianus who was another likely nominee of Aspar, Leo was surrounded by Aspar’s family and associates. How­ ever, his thoughts were already clearly focussed on his own succession. Leo and Verina had two daughters, one (Ariadne) born before his elevation in 457 and one (Leontia) after, but they were anxious for a son.56 So anxious was Leo that in 462 (late July/August) he sought divine intervention through the holy man Daniel, perched on his column at Anaplus up the Bosporus from Constantinople. Leo asked Sergius, the disciple of St Simeon who had attached himself to Daniel, to advance his request. Verina conceived and an heir was born in the purple on 25 51 PLRE 2, 842–3 (‘Iulius Patricius 15’). 52 Brooks (1893), 210, cf. Zonaras, 14.1 (122.3–5 Büttner-Wobst), who claims that now that Leo was emperor Aspar was strongly urging him to appoint one of his sons as Caesar ‘in accordance with his promise’ (kata. th.n u`po,scesin). Patricius possibly rode beside Leo in the coronation procession in February which would have been a powerful public signal of imperial endorsement and expecta­ tion (suggested by Del Medico [1955], 61 on the basis of the text in Paris.Hebr. 1280). 53 PLRE 2, 1179–80 (‘Fl. Vivianus 2). 54 PLRE 2, 516 (‘Godisthea’), with Demandt (1986), 114. 55 PLRE 2, 143–4 (‘Fl. Areobindus Dagalaiphus Areobindus 1’). 56 PLRE 2, 667 (‘Leontia 1’), puts her birth after Leo became emperor. Since both daughters were tutored by the grammarian Dioscorus they must have been relatively close in age. Dioscorus would have been their tutor in the early to mid-460s, cf. Kaster (1988), 272–3; PLRE 2, 367–8 (‘Dioscorus 5’); Mathisen (1991), 209.

61

DYNASTY AND ETHNICITY

April 463.57 So relieved and pleased was Leo that he had the foundations laid for a third column for Daniel.58 The young boy would grow up to be emperor after his father. The succession was now secure. There will have been considerable rejoic­ ing and public ceremonial at Constantinople surrounding this rare announcement of an imperial birth, followed at some stage by the baby’s baptism. Aspar’s reac­ tion to the news is not recorded. Giving birth to an heir also increased the prestige of Verina who was now entitled to the honour of Augusta and of having her own imperial coinage.59 Five months later, however, the rejoicing and expectation were displaced by shock, sadness and the reassuring ritual of a princely funeral. The infant son of Leo Augustus and Verina Augusta had died. The horoscope cast on his birth gave him no chance.60 Suddenly, dynastic uncertainty confronted Leo once more. If Leo was never to have another son, as he presumably now realised, then his succession would have to be effected through one of his daughters after all. Ari­ adne and her younger sister Leontia instantly became important political prizes for ambitious aristocrats, although Ariadne had been long promised to Aspar’s son. Aspar was still in control. An unmarried daughter was a potential empress, a new Pulcheria whose memory, authority and example were always promoted by Leo and Verina.61 In choosing her consort, Ariadne could assert the authority and con­ tinuity of Leo’s family through her imperial blood. Aspar was doubtless growing impatient and possibly reasserted his request for Patricius to be named Caesar. By now Leo had also to take account of the aspiration of the surviving Theodosians. Anthemius seems to have been co-operative enough. The stance of Olybrius, how­ ever, was less certain. His consulship in 464 may be interpreted as both a state­ ment of recognition by Leo and a concession to his influence. Theodosius II’s wife Eudocia had died at Jerusalem in 460 but their daughter Eudoxia had returned to Constantinople with her daughter Placidia, wife of Olybrius in 462. Eudoxia may have died herself soon after but the prestige of her family survived in the house­ hold of Olybrius and Placidia at Constantinople. Moreover, the families of Aspar and Theodosius II were eventually fused, as noted above, with the marriage of Aspar’s great-grandson (Areobindus) and Theodosius’ great granddaughter (Ani­ cia Juliana), the daughter of Olybrius and Placidia. Olybrius’ colleague as consul in 464 was the experienced general Rusticius, magister militum per Thraciam, now towards the end of his career. During the 57 Dagron (1982), 271–5. It may be that the young child was immediately created Caesar in order to designate his right of succession, that is, if there is any truth in the statement to that effect by Severus, patriarch of Antioch, in one of his letters (Letter 65 in Brooks [1915]). 58 v.Dan.Styl. 38 59 The coinage of Verina Augusta must have begun at this time as well (Kent [1994], 101: ‘early in the reign’), though she was not necessarily Augusta in 457 as an automatic entitlement on Leo’s accession, as proposed by Grierson and Mays (1992), 170–1. 60 Published in Cumont, apud Pingree (1976), 147–8. Pingree erroneously argues that, fearing an imperial rival, this horoscope was cast for Zeno who was already betrothed to Ariadne by 463. 61 Holum (1982), 227–8.

62

DYNASTY AND ETHNICITY

consulship of Olybrius and Rusticius, on the feast of St Mamas (2 September), a massive conflagration broke out at Constantinople which destroyed large parts of the city and led to considerable disruption.62 Aspar is credited with playing a conspicuous role in saving life and property during the blaze. By contrast, Leo took fright and relocated the imperial court up the Bosporus near the shrine of St Michael at Anaplus, and a new palace was quickly established, presumably in an imperial villa at St Mamas.63 It was at precisely this time, perhaps earlier in the year (464) that the consul Rusticius died and was replaced as general in Thrace by the emperor’s brother-in-law Basiliscus.64 This appointment arguably marks the first step in the emergence of Leo’s own family into positions of power and influ­ ence. It may also be construed as the first step in challenging the domination of Aspar by the promotion of officers more closely aligned to the interests and fam­ ily of Leo. Aspar was unlikely to easily forego his long-established influence and his quest to bring the imperial succession within his own family. Accordingly, the following year (465) the consulship was held by his youngest son Herminericus, while Aspar and Ardaburius still retained the central military commands which formed the basis of their power. Aspar had been a magister militum since the 420s and was probably senior court general (magister militum praesentalis) since 453. As such he had thousands of trained and experienced troops at his immediate com­ mand, both stationed in Constantinople and nearby. Ardaburius had been magister militum per Orientem for over 12 years. He too was in charge of thousands of troops, both those immediately at his command in Antioch and others spread throughout the eastern forts and towns under their subordinate local commander (dux or comes). Both generals enjoyed entrenched power and security. Moreover, in a polyethnic army and a polyethnic senate at Constantinople Aspar’s ethnic heritage may not have been all that important. What was clear, however, is that he was not of the same doctrinal allegiance as the emperor but a confirmed Arian. That was the characteristic which marked him out. Neither general, father or son, was anticipating the crisis which exploded that year. What happened, as we shall see in more detail below, was that in Constanti­ nople a certain imperial functionary named Tarasis, or Zeno, produced for Leo some incriminating letters of Ardaburius which had come into his possession. A meeting of the senate (conventus) was called, presumably sometime after March 62 The fire is frequently dated to 465 (e.g. Bury [1923a], 321; Stein [1959], 358; Snee [1998], 170) but actually occurred in September 464. For details: Croke (1995), 99; Whitby and Whitby (1989), 87 n.285. 63 For the fire and its aftermath: Candidus, frag. 1 (Blockley 464–6) = Photius, Bibl. 79; Zonaras 14.14–19 (124.5–125.10 Büttner-Wobst). On the topography of St Mamas’: Janin (1964), 473–4. Perhaps Leo had initially processed in propitiation to S. Mamas for relief from the inferno, but he stayed there for six months. 64 Priscus, frag.43 (350–1 Blockley) = Suda, B 163; Theophanes AM 5956 (De Boor 113.18–19); Michael the Syrian Chron. 9.1; death of Rusticius: Zonaras 14.23 (125.24–126.1 Büttner-Wobst) cf. 4 (n.11), 180 n.4. For the date, 464: Demandt (1970), 766f; PLRE 2, 212 (Fl. Basiliscus 2’). Bersanetti (1943/4), 335; Ensslin (1925), 1953 say 463.

63

DYNASTY AND ETHNICITY

when Leo had returned to Constantinople from S. Mamas. Aspar, as the leading senator, that is to say, of all the senators he was consul longest ago (31 years, con­ sul 434), was present but obviously not forewarned. Herminericus, the consular brother of Ardaburius would also have been present. Ardaburius himself was at Antioch. When the letters were read out the emperor was horrified. Ardaburius’ letters revealed that he was guilty of inciting the Persian emperor to launch an attack on Roman territory. Aspar appeared surprised but was astute enough to dis­ sociate himself from his son’s treacherous dealings, offering to concur in whatever action Leo saw fit to take. The emperor immediately dismissed Aspar’s son from the position of magister militum and stripped him of the highest honour, the title of ‘patrician’.65 By any reckoning this was a huge setback for Aspar’s ambitions and influence. Ardaburius was recalled to Constantinople in disgrace. If he had any defence to offer Leo it was obviously ineffectual. Almost immediately, that is in 465, the balance of power within the court of Leo began to tilt away from Aspar. Ardaburius was replaced as general by Jordanes, the son of another high-ranking officer John the Vandal. As with the house of Ardaburius, Jordanes maintained the Arianism of his family until an encounter with Daniel, which may have paved the way for his new appointment.66 Forsaking his Arianism also helped prise Jordanes loose from the patronage of Aspar. In any case, the emperor seized the opportunity to take Jordanes to Daniel for his blessing on the new general before he departed for Antioch.67 So, Jordanes set off with the sanction of both the emperor and the holy ascetic. Following his revelation of the treachery of Ardaburius, Zeno was appointed comes domesticorum which gave him great authority and prestige within the imperial household.68 There he lived with his wife Arcadia and his young son, also named Zeno. If nothing else this position would have protected him more readily from any vengeful actions on the part of Ardaburius and his father. With the key armies of Thrace (Basiliscus) and the East (Jordanes) now under the command of Leo’s appointees, not Aspar’s, the emperor appears to have sought to further his newly won advantage. In the first 65 v.Dan.Styl. 55. There is some uncertainty about the date. Daniel says ‘about that time’ without linking it to any other datable event, except to a windstorm which provoked the emperor’s concern for Daniel’s exposure. As Daniel’s story proceeds, the Ardaburius episode occurs after the great fire (c.45) of September 464 and about the time of the violent windstorm of the ‘following year’ (cc.52–4) which would put it in 465, probably mid-late 465. Most often, however, it is dated to 466 (as in PLRE 2, 136: ‘probably in 466’, most recently in Lane Fox [1997], 190), but only because the fire of the preceding year is placed in 465, not 464, although Eduard Schwartz argued that by ‘following year’ the hagiographer meant the next indictional year beginning on 1 September 465 (1934, 180 n.1). If that is so, then this event might be dated to the period 1 September–31 Decem­ ber 465 rather than necessarily having to be placed in the following consular year (from 1 January 466). 66 v.Dan.Styl. 49. 67 v.Dan.Styl. 55. 68 PLRE 2, 1201 (Fl. Zenon 7); Lippold (1972), 154 (Zeno 17). Kosinski (2010), 65 claims that Zeno must have been put in charge of the excubitores but the comes excubitorum was a separate and narrower role.

64

DYNASTY AND ETHNICITY

instance, he decided that the consulships for 466 would be held by himself and Tatian. Flavius Tatian was from a distinguished Lycian senatorial family and had been summoned to Constantinople by Marcian to be invested with the patriciate and the Prefecture of the City. By now, however, Tatian was a relatively old man for public life but he retained the favour of Leo having acted as his unsuccess­ ful imperial envoy to the Vandals the previous year (464).69 It was probably the announcement of Tatian’s appointment as consul which aroused the indignation of the increasingly insecure Aspar.70 Photius’ summary of the first book of Candidus’ history covering these years reports that Aspar and Leo were somehow brought into conflict over Tatian and this conflict led to the promotion of Tarasicodissa (Zeno). According to Photius’ syn­ opsis, Candidus wrote ‘about Tatian and Vivianus and how Aspar and the emperor differed concerning them, and what they declared plainly to one another and how for this reason the emperor allied himself to the Isaurian people in the person of Tarasicodissa . . . (dia. tou/to h‛tairi,sato to. VIsau,rwn ge,noj dia. Tarasikodi,ssa). What is at least discernible from Photius’ précis is that the quarrel between Leo and Aspar over Tatian and Vivianus coincided with the promotion of Tarasis/Zeno and in fact helped advance it. In other words, by directly connecting (dia. tou/to) the argument between Leo and Aspar with the appointment of Tarasis/Zeno, Candi­ dus dated the episode after the great fire of September 464, so probably in 465 in the aftermath of the undoing of Ardaburius. In that case, an argument over consular appointments for the following year provides a plausible context.71 We cannot know exactly what transpired between Leo and Aspar but since Tatian was announced as consul for 466 it is a reasonable assumption that because he is men­ tioned as the cause of the disagreement, it was his appointment which precipitated the difference of opinion between the emperor and his overlord. In this event, mention of Vivianus would suggest that Aspar supported Vivianus as he had done previously. It may be that Aspar was now supporting him for a second consulship in 466 before the treachery of Ardaburius was exposed. Consequently, Leo sought to use the disclosure in order to substitute Aspar’s nomination with the unlikely but reliable Tatian, so an argument flared. On the other hand, it is possible that Leo acted first in appointing Tatian but that Aspar was opposed to this and put the case for Vivianus. Whatever the sequence, it seems most likely that in the latter part of 465 Aspar and Leo fell out over the appointment of a second consul for 466. Further, Photius’ cryptic summary of Candidus’ history implies by such strong discord that in the end there was no

69 Priscus, frag.41.2 (Blockley 346–7) = Exc.de Leg.Rom. 11. 70 Cf. Ensslin (1925), 1957. 71 Otherwise this episode is dated to 459 with the notable exception of PLRE 2, 1054 (‘Tatianus 1’): ‘The reasons for his disappearance from the Fasti are obscure. Possibly he was a victim of Aspar’. While it is assumed here, and usually by others as well, that Candidus means a quarrel involving both Vivianus and Tatian it is possible that Photius’ summary compounds two separate incidents years apart (cf. Jones [1964], 221).

65

DYNASTY AND ETHNICITY

unanimity about the appointment so that the previous announcement of Tatian’s consulship was withdrawn. Leo held the office alone in 466. All the most reliable Eastern consular lists, as well as the inscriptions and papyri, record Leo as sole con­ sul for that year. However, some remote western ones also include Tatian.72 What has evidently happened is that the consulship for 466 was originally announced as ‘Leo and Tatian’ and the news despatched abroad at the usual time, in the usual way and at the usual speed. Later, when the appointment of the elderly Tatian was withdrawn not all lists were brought up to date, especially those furthest away in Gaul. This interpretation has the benefit of explaining both the incident recorded by Candidus, and the puzzling ambiguity over Tatian as consul in 466. This was not the only disagreement between Leo and Aspar over an imperial appointment. Traces of two other clashes survive, probably deriving ultimately from the history of Priscus (directly, or through Eustathius). In the first case the two fell out over the appointment of a City Prefect of Constantinople. The City Prefect had the important role of convening the senate. Aspar had insisted that the appointee be someone of like mind and Arian belief as himself (o‛ mo,doxon auvtou/ kai. o‛ mo,frona). Leo was clearly not going to countenance an Arian City Prefect, irrespective of his other qualifications for appointment, and privately made his own choice without consulting Aspar. The magister militum was furious and confronted the emperor directly. He tugged at the emperor’s cloak exclaim­ ing: ‘Emperor, it ill befits the wearer of this garment to lie’. Leo was obviously indignant at being manhandled in such a fashion and replied sharply to Aspar, ‘And it’s not fitting that he should be constrained and driven like a slave either’.73 Zonaras reports the same exchange but in relation to an entirely separate inci­ dent, an argument over the appointment of Aspar’s son as Caesar.74 A similar conflict is certainly possible concerning the contentious issue of Leo’s continued

72 CLRE, 466–7 suggests that Tatian was a western, not eastern, consul because the only extant traces of his consulship are western so that he may have been ‘proclaimed alone at some point but never universally recognised, perhaps out of office early in the year’(ibid., 467). The records seem also to be consistent with the possibility that he was a proclaimed Eastern consul but having had his proclamation revoked before entering on the consulship, and without a substitute being proclaimed, either eastern or western. That would also explain the expectation in Egypt of another consul still to be appointed (‘et qui nuntiatus fuerit’). Egyptian scribes generally followed the early announcement of the consuls provided by the office of the Prefect of Egypt (ibid, 67–8). Begass (2018), 240 favours the CLRE approach here. 73 Cedrenus, 379.1 (593–4 Tartaglia). Speculation about which of Leo’s City Prefects this was can only be limited, given the surviving documentation. One possibility is Dioscorus (PLRE 2.367–8, ‘Dioscorus 5’), the tutor of the emperor’s daughters, another is Diapharentius (PLRE 2.358, ‘Diapharentius’). This passage of Cedrenus is conflated with that of Candidus in the detailed study of Schwartz (1934), 179 n.2, as well as by Vernadsky (1941), 59–60, so that the argument between Leo and Aspar over Vivianus or Tatian becomes one about which one should be City Prefect. Brooks (1893), 212 also links this episode not only with that described by Candidus but also to a law (CJ 1.3.26) which requires a further extension, that is, from a dispute about the City Prefecture to one about the Praetorian Prefecture. 74 Zonaras 14.1 (122–3 Büttner-Wobst).

66

DYNASTY AND ETHNICITY

procrastination in naming Patricius as his Caesar. Serious differences with Aspar over religious policy were evident from the start of Leo’s reign although Pope Leo considered both the emperor and the ‘illustrious patrician’ strong advocates of resistance to the opponents of the Council of Chalcedon.75 Aspar later tried unsuc­ cessfully to intervene with the emperor Leo over his concerted effort to accept the advice of the Constantinopolitan patriarch Gennadius and depose the bishop of Alexandria, Timothy Aelurus (‘the Cat’), who had usurped the see and had his predecessor Proterius brutally killed.76 Earlier, when Anatolius was patriarch of Constantinople, Aspar had taken the side of Amphilocius, bishop of Side, against Leo and successfully persuaded the emperor not to attack Amphilocius.77 Leo was not turning out to be the compliant emperor Aspar had expected.

The emergence of Zeno, 465/6 By the end of 465 Leo clearly felt in a stronger position for dealing with Aspar who still held the powerful position of senior magister militum praesentalis. Aspar was suddenly more vulnerable while Leo was in a position to effect more of his own appointments. That is the stage when Zeno was appointed as comes domesticorum. This important position in the imperial bureaucracy gave him responsibility for the protection of the imperial household and the emperor’s own military staff. Exploring the background to this apparently unusual appointment is crucial to understanding the subsequent relationships between Leo and Aspar. Writing towards the end of the sixth century Evagrius tells us that Eustathius of Epiphaneia much earlier in the century had narrated how Leo came to promote Tarasis (later the emperor Zeno), and why the emperor singled him out.78 Unfor­ tunately, Eustathius’ History has not survived but there do survive two earlier accounts of how Tarasis/Zeno appeared to burst upon the scene at Constantinople: one contained in the history of Candidus, as preserved in Photius’ summary, the other in the vita of Daniel the Stylite. Both were written around the same time (late 5th/early 6th century) so that they are at least contemporary with Eustathius. Candidus was in a position to be well-informed on the events he covers in the reigns of Leo and Zeno from 457 to 491 since he was secretary (u‛ pografeu,j) to the most powerful men in Isauria (tw/n evn VIsau,roij plei/ston ivscusa,ntwn), surely a reference to successive comites Isauriae, the highest-ranking Roman officials in the region stationed at Tarsus.79 So too he will have been especially familiar with 75 Leo Epp., 149, 151. 76 Theodore Anagnostes, HE.Epitome 379 (106.18–19 Hansen) cf. Theophanes AM 5952 (De Boor 112) and Evagrius, HE 2.9–11. 77 Zachariah of Mytilene, HE 4.7. 78 Evagrius, HE 2.15 79 Jones (1964), 609. Roberto (2000), 685–727, esp. 726 (but relying on attributing the relevant extracts of John of Antioch in Constantine Porphyrogenitus’ de insidiis to a close copying of Candidus). Perhaps Candidus had been secretary to either or both of the known comites Isauriae during these years, Aetius in 479 (PLRE 2, 20 [‘Aetius 4’]) and Lilingis in 491 (PLRE 2,

67

DYNASTY AND ETHNICITY

events in Isauria and with the reporting in Isauria of happenings elsewhere. In condensing into a few sentences a whole book of Candidus, the first of three and covering the years 457 to 476, Photius directly links the disagreement between the emperor and Aspar over the appointment of the elderly Tatian as consul for 466 to Leo promoting Tarasis (Tarasicodissa): ‘He [Candidus] mentions Titianus and Vivianus and relates how Aspar and the emperor disagreed over them and what they said to one another. He tells how, as a result of this, the Emperor allied him­ self with the Isaurian people through Tarasicodissa, the son of Rusumbladeotus whose name he changed to Zeno and whom he made his son-in-law after Zeno’s former wife had died . . .’.80 This very compact sentence of Photius’ précis of Candidus bears almost the entire weight of the thesis that Leo turned from depen­ dence on a ‘Gothic faction’ led by Aspar, referred to by Candidus as an ‘Alan’, to dependence on an Isaurian ‘faction’ led by Zeno. Writing in the distinctly antiIsaurian climate after Zeno’s death in 491 Candidus will have given close atten­ tion to explaining Zeno’s emergence at Constantinople thirty years earlier. In fact, it is probable that the whole thrust of his history was a riposte to the Byzantiaka of Malchus which was patently hostile to Zeno and the Isaurians.81 Candidus was not saying that the emperor’s new support for Zeno involved anything extra for Isaurians, nor that Isaurians would now be flooding into Constantinople under the emperor’s personal patronage and protection. In hindsight, however, he was identifying the point where it now appeared that Isaurians had first emerged as a powerful influence at Leo’s court. Thirty years later it could be construed that Leo had promoted Zeno as a way of rescuing the realm from the dominance of Aspar and his coterie, although it did not appear that way at the time. Indeed, Candidus evidently linked the promotion of Zeno directly, but without Photius explaining further, to the altercation between Aspar and Leo concerning the senior court offi­ cials Tatianus (cos. 466) and Vivianus (cos. 463). The second document with information on Zeno’s emergence at the court at Constantinople in 465 is the life of Daniel the Stylite. This vita was written by one of Daniel’s disciples and is based on what he saw and heard himself from Dan­ iel and from elder disciples, including the original followers of Daniel. It brings together a series of incidents and episodes in typical hagiographical style. The chronology is incidental to the purpose of the work.82 It runs as follows:

683–4 [‘Lilingis’]); or perhaps he had some official connection with Illus, as proposed by. Laniado (1991), 153–4. The comes Zeno whose epitaph has been found near Isaurian Seleucia may also be dated to this period (cf. Feissel [1999], 11). 80 Candidus, frag. 1 (Blockley 466/7) = Photius, Bibl.Cod.79. 81 Elton (2000b), 295–8. For the anti-Zeno thrust of Malchus’ history: Baldwin (1977), 91–107; Blockley (1981), 80–5; Laniado (1991), 147–50; Shahid (1989), 100–6. 82 For the chronology of the vita and its sources of information: Delehaye (1913), 225–7 and (1923), XLIII–V, LIV–LVII. Lane Fox (1997), 185–200 demonstrates the problems of trying to interpret the chronology literally and as strictly sequential.

68

DYNASTY AND ETHNICITY

About that time a certain Zeno, an Isaurian by birth, came to the Emperor and brought with him letters written by Ardaburius, who was then Gen­ eral of the East; in these he incited the Persians to attack the Roman State and agreed to co-operate with them. The Emperor received the man and recognizing the importance of the letters he ordered a Council to be held; when the Senate had met the Emperor produced the letters and com­ manded that they should be read aloud in the hearing of all the senators by Patricius, who was Master of the Offices at that time. After the letters had been read the Emperor said, ‘What think you?’ As they all held their peace the Emperor said to the father of Ardaburius, ‘These are fine things that your son is practising against his Emperor and the Roman State’. The father replied, ‘You are the master and have full authority; after hearing this letter I realize that I can no longer control my son; for I often sent to him counselling and warning him not to ruin his life; and now I see he is acting contrary to my advice. Therefore do whatever occurs to your Piety; dismiss him from his command and order him to come here and he shall make his defence’.83 The vita continues by explaining that Leo took Aspar’s advice and proceeded to dismiss Ardaburius and summon him to the imperial capital, replacing him as magister with Jordanes. This is all the documentation we now have concerning Zeno’s role in the deposi­ tion of Ardaburius and its immediate aftermath. Before examining it more closely, it is necessary to look critically at how each of the elements of this episode have been constructed by modern scholars: (A) Leo deliberately decided to counteract the intolerable domination of the ‘Germans’ at court, led by Aspar, by inviting to Constantinople an Isaurian chieftain called Tarasis. The notion that the Isaurians, and Zeno in particular, were Leo’s chosen policy instruments against Aspar and the Goths is encountered regularly, beginning with Brooks: ‘with the plan of an Isaurian alliance perhaps already in his mind . . . Leo sent for an Isaurian chieftain, Tarasicodissa of Rousomblada and gave him his daughter in marriage’;84 followed by Seeck: ‘Leo suchte in den Isaurern ein Gegengewicht gegen die Übermacht der Germanen’85 and Bury: ‘he formed the plan of recruiting regiments from native subjects . . . he chose the hardy race of Isaurian mountaineers’.86 Vernadsky went further, offering the explanation that 83 84 85 86

v.Dan.Styl. 55. Brooks (1893), 211–12. Seeck (1920), 609. Bury (1923a), 317; cf. Schwartz (1934), 180: ‘ . . . die militärische Macht des Reiches Zenon und seinen Isaurien anvertrauen wollte, um die germanische Gefahr zu bannen’.

69

DYNASTY AND ETHNICITY

Leo chose Isaurians because they were the only national group not under the influ­ ence of Aspar.87 Others have broadly followed suit.88 Most striking throughout all accounts is the frequently repeated catch-phrase which sums up this deliberate action – Isaurians and Zeno as ‘counterweight’/‘contrepoids’/‘Gegengewicht’ to Goths and Aspar.89 No less arresting is the consistent reference to Tarasis as a tribal chieftain from Isauria.90 In other words, it is assumed that (1) Leo consciously sought to counteract undue Gothic influence at court by seeking out an anti-Gothic ethnic counterbal­ ance, (2) he knew, or knew of, Zeno and knew that of all the military and civil officers at his disposal Zeno was the perfect person to be the ‘counterweight’, (3) he knew Zeno would bring such a large contingent of Isaurians, and, (4) these Isaurians would strike fear into the hearts of the Goths. A recent summary of this whole thesis proceeds as follows: ‘Leo now felt it prudent to seek a military counterweight to Aspar and his Gothic power base . . . cautiously he began to put feelers out to the Isaurian mountain peoples of Asia Minor . . . with lavish prom­ ises, and much gold, he reached a secret agreement with one of the most powerful headmen, Tarasicodissa. . . . In return Leo promised to recruit Isaurian troops on generous terms, and raise Tarasicodissa to an exalted position. Tarasicodissa accepted readily, the raiding [by Isaurians] was ended, a rival chief easily sup­ pressed, and Isaurian troops duly recruited’.91 Most of this narrative is pure fiction, it has to be said immediately. All that we know, and can say with confidence, is that Zeno was responsible for providing the evidence that convicted the magister militum of the East, Ardaburius, and that as a result of this conviction Zeno himself came to prominence and imperial favouritism. As the vita of Daniel expresses it, ‘A certain Zeno . . . came to the emperor’ (paragi,netai tij pro.j to.n basile,a ovno,mati Zh,nwn).92 That is all. There is no evidence that Leo sought him out deliberately as part of a strategy to coun­ terbalance Aspar’s power, let alone to choose him as a specifically ethnic leader. Further, there is no evidence that Zeno was summoned or came from anywhere 87 Vernadsky (1941), 60 n.4. 88 For example: Treadgold (1995), 13 n.3, 160; Averil Cameron (1993b), 30. ‘Leo tried to counteract the German influence by recruiting heavily for the army among the Isaurians . . . ’; Lee (2000), 46: ‘ . . . a longer-term plan designed to free himself completely from dependence on the Alan general . . . Leo sought to counterbalance this by drawing on manpower from Isauria . . . ’. The emperor’s imputed policy even meets with censure in Grierson and Mays (1992), 161: ‘Leo . . . had a mind of his own, and with singular ingratitude set about reducing the power of the Germanic element in the army by recruiting Isaurians in their place’. 89 Lippold (1972), 155; Demandt (1970), 187: ‘Gegengewicht’; Stein (1949), 356: ‘contrepoids’; Kaegi (1981), 27: ‘counterweight’; Bury (1923a), 317: ‘counterpoise’. W. Burgess (1992), 875: ‘Counter-force to Gothic power’. 90 Brooks (1893), 213: ‘Isaurian Chieftain’; Stein (1949), 358: ‘le chef isaurien’; Kaegi (1981), 27: ‘the Isaurian chieftain’; Cameron (1993b), 30: ‘[Zeno] was their chief’; Shaw (1990), 252: ‘a typi­ cal powerful baron of the Isaurian mountains’. 91 Williams and Friell (1999), 177. 92 v.Dan.Styl. 55

70

DYNASTY AND ETHNICITY

outside Constantinople. That is not to say it is impossible. If he came from any­ where else, it would have been from Antioch, the headquarters of Ardaburius as magister militum of the East. If Leo had decided to seek out Zeno as an Isaurian chieftain then the connection with the incriminating letters is not obvious. Indeed, it becomes less important. Yet, it is the very revelation of Ardaburius’ correspon­ dence with the Persian king that provides the context in which Zeno first emerges into public light. To ask the compelling question how it was that Zeno was in possession of such confidential and high-level correspondence from a senior Roman general to the Persian king, points to the answer. To have the documentation he must either himself have been a close associate of Ardaburius or else had access to someone within the military headquarters at Antioch. Certainly, at some stage the incrimi­ nating documents originated at Antioch where Ardaburius had been located since 453. Rather than Zeno being a member of Ardaburius’ permanent staff, having to explain himself on arriving in Constantinople, the more likely possibility is that he was actually based at Constantinople all along, at the court of Leo. In that case he would probably have been one of the protectores domestici who were actually part of the emperor’s personal military staff under the command of the comes domesticorum. Numbers of protectores were also stationed on the staff of each of the generals of Illyricum, Thrace and the East but there they remained the emperor’s men. Their role was therefore seen as a way of asserting and maintaining imperial control within each military command.93 Sometimes too protectores could be sent on a special assignment to a particular general and report back to the emperor. For Zeno to have the secret and sensitive correspondence between Ardaburius and the Persian king, and to be held responsible for producing it at court without any prior warning, the most likely position he held at the time was either that of a protector, attached to the staff of the magister militum at Antioch, or more likely one of the protectores domestici at Leo’s court. It may be that Leo had suspicions about Ardaburius and sent Zeno to Antioch to investigate and report back. Certainly Malchus is critical of Leo for his habit of using such tactics.94 Either that or, if Zeno was a protector attached to Ardaburius, he knew how to handle the incrimi­ nating letters which came his way. The emperor Leo later shared with the holy man Daniel his appreciation of the loyalty (eu;vnoia) of Zeno,95 which may indicate that Ardaburius had confidently attempted to involve Zeno in his treachery only to find that the protector’s loyalty to the emperor prevailed. Whichever position Zeno held in 465, there is not the slightest indication that he was already a favourite of Leo, nor that he was deliberately summoned to 93 Frank (1969), 94–5; Haldon (1984), 130–6, Elton (1996), 101. Begass (2018), 239 n.1596 objects that Zeno had already been in Constantinople and did not suddenly appear there. Indeed, that is implied here in the proposition that for a number of years he had been one of the imperial court’s protectores domestici. 94 Malchus, Byzantiaka frag. 3 (Blockley 408–9). 95 v.Dan.Styl. 55.

71

DYNASTY AND ETHNICITY

Constantinople to be a ‘counterweight’, least of all because he was an Isaurian chieftain as often claimed.96 That too is mere myth. Certainly he was an Isaurian, at least by birth (to. ge,noj :Isauroj) which is all that the vita (c.55) of Daniel says. There is no reason, however, to think he had followed a career any different from the many other Isaurians in the Roman army before him.97 That Zeno was more or less immediately put in charge of the protectores domestici, joined other generals on a military campaign against the Goths in the following year (467), and was subsequently made magister militum himself (469), suggest someone of military background, a typical career Roman officer. He may have been no less typical in other ways too. His wife’s name, Arcadia, suggests that he may have married into a local Constantinopolitan family. Local nobility may also be implied in the fact that a statue of her stood for centuries in a prominent spot by the steps ascending to the Topoi, near the baths of Arcadia.98 Possibly too, his whole family were long settled in the imperial capital and he was known as Zeno well before 465. His mother Lallis certainly lived there, while Zeno’s brother Longinus was married to Valeria who was most likely another local dignitary’s daughter.99 The family of Tarasis/Zeno may have been associated with the house of the patrician Zeno, the consul of 448, and his noble consort Paulina. He was part of a civilised provincial Isaurian elite which had emerged by the 5th cen­ tury.100 Changing his name to Zeno may well have resulted from such links of patronage. Indeed ‘Zeno’ was already the name of his own son by Arcadia and was to become that of his nephew as well.101 If Leo was responsible for the name change it can have occurred as early as 457, perhaps on first being enrolled in the protectores domestici, and before the marriage to Arcadia. This suggests that he was already called ‘Zeno’ by 465.102

96 Cf. Vernadsky (1941), 60 even assumes Zeno must be pagan so, along with changing his name, when he comes to Constantinople he has to be baptised. 97 Thompson (1946), 18–31; W. Burgess (1992), 874–7; Elton (1996), 396. 98 Patria Constantinopoleos 2.27 (Berger [2013], 66–7) with Janin (1964), 312 and PLRE 2, 130, ‘Arcadia 2’. 99 PLRE 2, 654 (‘Lallis’); 689–90 (‘Fl Longinus 6’); 1141 (‘Valeria’). 100 Lenski (1999), 450–2; Crawford (2019), 28–32. 101 PLRE 2, 1198 (‘Zenon 3’, nephew) and ‘Zenon 4’, son). Neither the position for Zeno, nor the local Constantinopolitan background, is accepted by Kosinski who prefers that Zeno suddenly arrived from Isauria in 465, assuming that ‘Tarasicodissa commanded his own attachment of Isaurian bucelarii, maintaining his influence in the mountain areas of Isauria’ (Kosinski [2010], 61). 102 Cf. Crawford (2019), 29–30. It is usually assumed that his change of name from Tarasis or Tara­ sikodissa can be ascribed to the witness of his fellow-Isaurian, Candidus. However, it is the words of the brief 9th century summary of Candidus by Photius which are being quoted (Candidus, frag. 1 Blockley [1982], 466–7). Photius summarises by saying that Candidus tells of how Zeno came to be promoted by Leo, then how he changed his name and made him his son-in-law. While Pho­ tius gives the impression that Candidus dated the name change to Zeno to the time of his marriage with Ariadne (468, accepted by, among others, Kosinski [2010], 66), this is far too late. His son Zeno was already growing up. It may be noted too that the contemporary Life of Daniel simply calls him Zeno from the start (465: v.Dan.Styl., 55).

72

DYNASTY AND ETHNICITY

(B) Zeno brought with him a large contingent of his Isaurian followers who checked the dominance of the Goths. It is frequently claimed that not only was Zeno summoned to Constantinople from Isauria but that he also arrived with an ethnic army at his back.103 Whether he was already at Constantinople, or whether he travelled from his base in Antioch with the evidence against Ardaburius, it is neither attested nor likely that Zeno was accompanied by an Isaurian contingent, let alone one that was large enough to challenge and over-balance the forces and allies of the magister militum praesentalis, Aspar. Underlying this claim is the assumption that Zeno was a sort of ethnic leader or comes foederatorum and that he was sought out and summoned to Constantinople naturally bringing with him the native Isaurian band whose chief he was. To repeat, there is simply no evidence at all for Zeno being either accom­ panied to Constantinople in 465 or having a large number of his fellow-Isaurians under command there. As already noted, Zeno was probably a court soldier of Leo, a protector domesticus. As such he was not responsible for leading large contingents, let alone Isaurians exclusively.104 As testimony to the introduction of large numbers of Isaurians into Constan­ tinople in the entourage of Zeno in 465/6 two comments by later historians are usually cited: (1) Procopius, Anekdota 24.17, where it is reported that from the time of the emperor Zeno the palace guards previously recruited among the Arme­ nians were open to ‘both cowards and wholly unwarlike men’ (kai. avna,ndroij kai. avpole,moij ou-si) and (2) Agathias, Histories 5.15.4, who claims that the scholarii were once recruited from honourable veterans but ‘Zeno the Isaurian seems to have been the first to introduce the present practice by enrolling in these regi­ ments, after his restoration, many of his fellow countrymen who, though they were men who had either not distinguished themselves on the field or had absolutely no military experience whatsoever, were nevertheless known to him in some other capacity and were his close friends’. Neither Procopius nor Agathias is really focussing on the period when Zeno first came to prominence at Constantinople in 465/6. Procopius in fact does not even claim that Zeno recruited Isaurians in par­ ticular, just unworthy candidates by previous standards. Agathias, it is true, does claim that Zeno recruited fellow-Isaurians who had not necessarily had military 103 Demandt (1970), 186: ‘Isaurierfürst . . . mit grossem Gefolge’; Lippold (1972), 154–5: ‘mit einem recht stattlichen Gefolge . . . um schon damit eine Art Gegengewicht zu dem ihn bedückenden Aspar zu schaffen’; Kaegi (1981), 27: ‘Leo in 466 had created a counterweight to Aspar by sum­ moning the Isaurian chieftain Tarasicodissa to become head of a crack native corps of Isaurian guards’; Averil Cameron (1993b), 30: ‘Leo tried to counteract the German influence by recruiting heavily for the army among the Isaurians’; Treadgold (1997), 152: ‘eager for allies against Aspar and his Germans, Leo summoned a company of Isaurians to Constantinople in 466’. 104 It is possible that something similar took place in the late 440s, but there is no need to assume comparable circumstances requiring comparable responses and consequences in 465/6, as pro­ posed by Thompson (1946), but this is too schematic an interpretation of Zeno’s role vis-à-vis the other generals of Theodosius II (noted by Zuckerman [1994], 175–6).

73

DYNASTY AND ETHNICITY

experience. Yet, he does not date this policy on Zeno’s part any earlier than ‘after his restoration to the throne’ (meta. th.n th/j basilei,aj a~na,kthsin) that is after the end of 476, while Procopius’ phrase ‘since the time Zeno succeeded to the throne’ (evx ou- de. Zh,nwn th.n basilei,an pare,labe) surely means no earlier than the beginning of his sole reign in November 474. When Zeno set out from Isauria to reclaim his throne in 476 he was totally dependent on a large force of local Isau­ rian recruits.105 That makes sense. These loyal troops of limited experience must soon have been absorbed into the imperial army, thereby contributing to the notion of undue dependence on Isaurians.106 It is this Isaurian presence in Constantinople after 476 that Procopius and Agathias lament, not that up to a decade earlier at the time of Zeno’s appointment as comes domesticorum. (C) Leo established a new imperial guard called ‘excubitores’ and stacked it with the Isaurian followers of Zeno. That Leo established a new corps of palace guards called excubitores, who were recruited from Isauria, depends on a brief reference by John the Lydian writing in the mid-sixth century, who says that ‘. . . Emperor Leo, who was the first to establish the so-called excubitores as guards of the side-exits of the palatium, put into service only three hundred to accord with ancient custom’.107 Actually, there had long been a unit of excubitores forming part of the palace guard. As John also says elsewhere, they were founded by the emperor Tiberius four cen­ turies before.108 So Leo did not so much establish the excubitores as give them a specific new role and command structure.109 It must have been an important contingent because its leader, now entitled comes excubitorum, ranked high in the imperial structure.110 Exactly what Leo intended by his new approach to the excubitores, how they were recruited, and who made up the 300 elite troops, are not known. That is, there is no extant testimony to this effect. That has not prevented the repetition of speculation and the automatic expectation that they must have been Isaurians. Further, the formation of the excubitores is directly linked to the arrival of Zeno at Constantinople and he is entrusted with respon­ sibility for them. The hesitation of Bury – ‘we may conjecture that [the unit of excubitores] was recruited from stalwart Isaurians . . . called upon to oppose the Germans’111 has

105 Theophanes A.M. 5969 (De Boor 112). 106 Anonymus Valesianus, pars posterior 9.40 (Festy [2020], 3): ‘in re publica [Zeno] omnino provi­ dentissimus, favens genti suae’; ps-Joshua the Stylite, Chron., 12, cf. Lenski (1999), 427–8. 107 On the Magistrates 1.16.3. For background: Mary Whitby (1987), 483–8. 108 On the Magistrates 1.12. 109 References in Fiebiger (1909), 1577; Grosse (1920), 270 with Croke (2005b) and Chapter 5, (134–52). 110 Jones (1964), 658, Mommsen (1889), 224–5; Frank (1969), 204–7; Haldon (1984), 136–41. 111 Bury (1923a), 318.

74

DYNASTY AND ETHNICITY

been dislodged by unqualified certainty.112 For instance, there is Baynes’ con­ tention that the new guards ‘were doubtless the followers of the Isaurian (komh,j domestikw/n)’,113 while Kaegi casts Zeno as ‘head of a crack native corps of Isaurian guards’114 and Frank claims that at about this time ‘a new corps of palace guards appears, the excubitores, recruited among the Isaurians, and their formation must surely be placed in connection with the appointment of Zeno’.115 He even charac­ terises them as replacing the ‘Arian Gothic’ scholares (204). Others are in similar vein.116 It needs to be repeated, however, that it is simply not known how the excubitores were recruited and who made up the 300 elite troops. Isaurians were possibly among the new excubitores, along with those of other backgrounds, but all being Roman soldiers primarily. John the Lydian is the sole testimony to Leo’s palace guard reformation and he makes no mention of Isaurians. The point is that we have no reason for assuming that the excubitores were necessarily recruited from Isauria, or even mainly from Isauria. Least of all can they be construed as a contingent under Zeno’s command. They were in fact under the authority of a new officer called the comes excubitorum, whereas Zeno was comes domesticorum. Nor do we have any indication of when Leo set about reforming the excubitores. Indeed, there is no reason for it not to have been in the late 450s or early 460s, that is, well before Zeno appeared on the scene. (D) Leo then made Zeno comes domesticorum and married him to his own daughter Ariadne. On the assumption that Leo sought out the Isaurian Zeno, summoned him to Constantinople along with his Isaurian war-band, then found that he happened to possess letters damaging to Ardaburius, the appointment of Zeno as comes domesticorum is seen as a sort of unlikely reward for outstanding special service. There is, however, a sounder way of approaching the appointment. If, as argued here, Zeno was a trusted member of the emperor’s personal forces (protectores domestici) then it will have been a small step to place him in charge of his own company. He may already have been one of the ten senior domestici (decemprimi), or even its leader (primicerius). Either position carried considerable status 112 Stein (1959), 358 (Isaurians, Thracians and Illyrians); Kaegi (1981), 27; Demandt (1970), 187; Williams/Friell (1999), 177: ‘Tarasicodissa was appointed to command a new corps of 300 Isau­ rian palace guards, the Excubitores’. 113 Baynes (1925), 399. 114 Kaegi (1981), 27. 115 Haldon (1984), 135. 116 Vernadsky (1941), 60: ‘der Kern einer weit grösseren isaurischen Einheit’; Jones (1964), 222: ‘With [Zeno’s] aid Isaurians were recruited in large numbers and stationed in the capital’; Haldon (1984), 127: ‘In the early period Isaurians predominated’; W. Burgess (1992), 875; Treadgold (1997), 152: ‘first the emperor made Zeno a commander of the imperial guard, and strength­ ened his hand by creating the new guard corps of the Excubitors, three hundred strong, composed largely of Isaurians’.

75

DYNASTY AND ETHNICITY

and reward including the title of clarissimus.117 As comes domesticorum Zeno received a promotion he might have expected one day. That he had successfully fulfilled a dangerous assignment for the emperor obviously enhanced his reputa­ tion and precipitated his promotion. However, it was not an unusual, unlikely or untimely advancement. Being promoted to comes domesticorum brought Zeno into the highest echelons of the imperial administration. It was the third ranking military post behind the magistri militum and involved the highest honorific title of illustris.118 It also followed earlier military appointments and held the likeli­ hood that he would be promoted to magister militum before long, as many of his predecessors had been. As comes Zeno was close to the emperor and had access to elite troops. That may go far towards explaining the enthusiasm of the Isaurian Candidus for the recognition suddenly achieved by a fellow Isaurian. To be entrusted with this position Leo must have felt that Zeno was qualified and trustworthy. Zeno is unlikely to have been appointed immediately after his evidence against Ardaburius, but he was probably made comes in late 465/early 466. The appointment was also linked to a betrothal to the emperor’s eldest daughter. There may have been some months between these two events. They are clearly separated in time in the vita of Daniel (cc.55, 65). While the appointment as comes may have been in late 465/early 466 the marriage appears to have taken place around mid-468. In any event, the marriage with Ariadne could not have taken place while Zeno’s first wife Arcadia was still alive, so she must have died around this time. There is no hint of enforced divorce. Having been born in c.425 Zeno was considerably older than his bride and was a very experienced soldier. As comes domesticorum and illustris Zeno was per­ fectly eligible to be betrothed to the emperor’s daughter. Indeed, he was no less eligible than Aspar’s son Patricius. Again, the marriage of Zeno and Ariadne was not necessarily an impossible or unlikely event which could only be explained by the emperor Leo’s anxiety to secure the support of a large band of Isaurians to counteract the influence of the ‘Gothic faction’ led by Aspar. Further, if Aspar had been holding out for a union between his son Patricius and Ariadne, his hopes were now finally dashed. Not only was she betrothed to someone else, but to the very man who had caused the downfall and disgrace of Ardaburius.119 A child of Ariadne and Zeno would likely lock Aspar’s family out of the succession once and for all. He would be back in the same position he was in with the birth of Leo’s son four years previously, in 463. Aspar is unlikely to have simply accepted the situation. After all, he was still the most senior general, senator and patrician. The emperor’s new son-in-law might soon be an equal or a more serious military rival. Leo was now his own man. Yet, Aspar’s tenure of office was not affected. Leo had used Zeno’s information to properly strip Ardaburius of office for a crime of high treason. Nothing more. As quickly as possible after the marriage of Zeno 117 Haldon (1984), 135. 118 Frank (1969), 65, 88–9 with Palme (1998), 98–116, esp. 108ff. 119 Brooks (1893), 212.

76

DYNASTY AND ETHNICITY

and Ariadne (468), in the late summer of 469, the first and only offspring of the marriage – Leo II – was born.120 The status of the infant son of Ariadne and Zeno, and the expectations he may have generated, are reflected in a contemporary image of gold and mosaic. It was located in a chapel constructed by Leo and Verina at Blachernai, to house a most important relic, the recently arrived robe (esqh, / j) of the Virgin Mary. The surviving tenth-century description of the image appears garbled, in a similar way to so many other later explanations of Byzantine monuments and works of art. The description says that the image depicted ‘Our Lady the immaculate Mother of God seated on a throne and on either side of her Leo and Veronica, the latter holding her own son, the young emperor Leo, as she falls before Our Lady the Mother of God, and also their daughter Ariadne’.121 The description goes on to explain that this young Leo succeeded to the throne on the death of his father Leo I. The writer of the description is confused. Leo II was the son of Ariadne not Verina (called ‘Veronica’).122 While it is impossible to date this event exactly it probably belongs in the late 460s or possibly later still.123 The traditions around the translation of the Virgin’s robe in its precious rel­ iquary ascribe the initiative to Leo and Verina.124 They clearly understood the relic’s significance and potential source of protection. Here for the first time the Virgin came to dwell in her city, Constantinople. On the very reliquary they had inscribed, ‘Having offered this honour to the Mother of God, they have secured the might of the Empire’ (th/j basilei,aj h~sfali,santo kra,toj).125 One part of the tradition also attributes the transport of the relic to two Byzantine aristocrats and generals named Galbinus and Candidus who are not otherwise known, at least if their names have been correctly preserved. Moreover, it is said that they were once close associates of Aspar and Ardaburius but had renounced their Arian beliefs 120 Malalas, Chron. 14. 47 (Thurn 299), with more detailed discussion in Croke (2003), with Chap­ ter 5 (134–52). 121 Cod.Par.Gr.1447, fols. 257–8 printed in Wenger (1952), 54–5. The translation is that of Mango (1986a), 35. 122 If the child was actually named as ‘Leo’ in the image the writer of this description, or the original observer, may have assumed on the basis of what he saw that the baby must be Leo II and that Leo II must therefore have been the son of Leo and Verina. If the child was not named then he may have assumed that it had to be Leo II. Another possibility, recently promoted (Lane Fox [1997], 189–90), is that the baby portrayed in this image was the otherwise unnamed male child of Leo and Verina and that the image was executed in the period just after his birth in April 463. Such an interpretation of the image is not self-evident, however. If it is meant to picture Leo’s family then Leontia who was born before the anonymous son is not mentioned, nor would the infant, who only lived for five months, be represented as ‘her own son the emperor Leo the Younger’. The son of Leo and Verina was never emperor in any sense, although he was obviously intended to be one day. Nor was his name necessarily Leo. 123 Theodore Anagnostes, HE Epitome 397 (111.7–12 Hansen); Cedrenus 383.4 (599 Tartaglia); Zonaras 14.31 (127.5–8 Büttner-Wobst). Wenger (1952), proposed 471–3. 124 Baynes (1955a), 240–7 and (1955b), 257–8. The text is that of the Oratio de S. Deipara with Latin translation in PL 115.560–6. For a more recent assessment: Shoemaker (2008), 53–74. 125 The text from Cod. Par. Gr 1447, fols 257–8 was published in Wenger (1952), trs. Mango (1986a), 35.

77

DYNASTY AND ETHNICITY

before setting off for the Holy Land where they came across the Virgin’s robe in the possession of a pious Jewess but tricked her out of it.126 Irrespective of the veracity of this tradition, it does highlight the extent to which Aspar and Arda­ burius could be isolated as Arians. Indeed, the Virgin’s robe provided a new source of power and patronage not open to Aspar at all. Leo seems to have understood and exploited this advantage. Much the same may be said for Leo’s relationship with Daniel the Stylite. The emperor was too busy in 459 for an audience with Sergius the bearer of the tunic of the late holy man Simeon the Stylite.127 At that time he preferred to petition Daniel through intermediaries (38, 41, 43), but before long he was prepared to journey directly to meet with him (44, 46, 48, 49). He even took Gubazes the king of the Lazi (51) and then all visiting dignitaries (54, 55, 57). By contrast, Eudoxia understood immediately the potential power of Daniel and unsuccessfully sought to appropriate it to herself. Daniel declined to be monopolised by her.128 Leo evi­ dently saw his association with Daniel as a source of special power and prestige, which he increasingly utilised for his own benefit.129 At one stage Leo feared that exposure to the elements would lead to the premature demise of the holy man. In resisting Leo’s proposed enclosure Daniel was reminded by the emperor: ‘. . . do not kill yourself outright, for God has given you to be fruitful on our behalf’.130 More importantly, Daniel was a source of power from which the Arian Aspar was excluded. Indeed, Leo’s assiduous support for the translation of relics gave rise to a concentration of spiritual power and protection within Constantinople. As their promoter, he doubtless envisaged special blessings for himself and his family. Daniel’s holiness became Leo’s chief bulwark. Yet, monopoly of the power deriv­ ing from his association with relics, especially the Virgin’s robe, and with holy men such as Daniel could not protect Leo from the more worldly machinations of Aspar.

Huns, Goths and imperial policy, 466–8 Relations between Leo and Aspar were further strained by the demands of man­ aging relations with the Goths and Huns. In the late 450s and 460s Leo and his regime had to maintain the peace with the constituent nations which Attila, king of the Huns, had temporarily formed into such a dangerous confederation in the 440s. The demise of Attila in the early 450s had not meant the end of instability in 126 PL 115, 565–6: ‘Dicebantur enim fuisse proximi genere Ardaburio et Aspari, qui illis temporibus in regia obtinuere tyrannidem’. If the brothers ‘Galbius’ and ‘Candidus’ (‘fratres genere’) were in fact generals (‘qui curam gerebant exercitus’), their relationship to Aspar might stem from them having married Aspar’s daughters. The names are not otherwise known unless one is a variant of Camundus, magister militum per Illyricum (PLRE 2, 256 [‘Camundus’]) 127 v.Dan.Styl. 22. 128 v.Dan.Styl. 35. 129 Miller (1970), 207–12; Trampedach (2013), 188–205. 130 v.Dan.Styl. 54.

78

DYNASTY AND ETHNICITY

the Danube region and the Balkans. As individual tribal groups emerged from the hegemony of the Huns in the 450s and 460s their interests soon confronted those of the Roman government. The so-called Amal Goths under their leader Valamer were settled in Pannonia, secured by the compulsory residence in Constantinople of a young Gothic prince Theodoric. Other Gothic groups remained restless, while yet others beyond the Danube pressed closer to their settlements. The sons of Attila, meanwhile, struggled to keep some semblance of unity among their differ­ ent war-bands. Occasionally open conflict erupted. Sometimes this involved the local Roman military forces. Marcian’s son-in-law Anthemius loyally served Leo by repelling both the Gothic king Valamer and the Hun Hormisdac.131 Immediately thereafter Leo had resolved the issue of any threat to his throne from Anthemius by appointing him as emperor in the West and sending him off to Rome. He will have been reminded of Anthemius’ exploits against the Goths by the City Prefect Diapharentius and the former City Prefect Dioscorus who both delivered panegy­ rics to the court on the reception of Anthemius’ imperial image in Constantinople in 467.132 In 466 hostilities erupted between the Sciri and the Goths, with neither having the capacity to prevail over the other. When they disengaged, both the Sciri and the Goths decided to seek support from the Romans. Accordingly, they each sent envoys to Constantinople. Priscus reports that ‘Aspar thought that they should ally with neither, but the Emperor Leo wished to help the Sciri. He sent letters to the general in Illyricum ordering him to send the appropriate help against the Goths’.133 Aspar was overruled. The attack of the Sciri must have been against the Pannonian Goths of Valamer if the magister militum per Illyricum was mobilised on their behalf. Further, it was surely during this encounter with the Sciri that the Gothic king Valamer was thrown from his horse, captured and killed by the Sciri.134 Leo evidently saw here an opportunity to push the Amal Goths out of Roman territory altogether and thought it worthwhile to lend support to the Sciri. Aspar preferred neutrality either hoping they would cancel each other out, or else deliver victory to the Goths because of his sympathy for them.135 Whatever the reason, there was now another cause of aggravation between Aspar and Leo, not long after Ardaburius’ disgrace. The Huns, now led by Dengizich, sent a legation to Leo seeking a treaty and a trading post. As with the Pannonian Goths, the Huns’ main concern was to secure 131 References in PLRE 2, 96–7 (‘Anthemius 3’). 132 Const.Porph., On the Ceremonies, 1.87 (Reiske 395.14–16, trs. Moffatt and Tall [2012], 395); Chron.Pasch. 597 (Dindorf). 133 Priscus, frag.45 (352/3 Blockley) = Exc. de Leg.Gent. 17. The date must be 466 because it occurs between Gobazes’ embassy to Constantinople in 464/5 (frag. 44 [352–3 Blockley] = Exc. de Leg.Gent 16; v.Dan.Styl. 51), and the Huns embassy in 467 (frag.46 (352–3 Blockley) = Exc. de Leg.Gent.18), contra PLRE 2.515 (Gobazes) misdating the fire to 465 not 464. Also Ensslin (1925), 1953. Others have dated this conflict to 468/9 (e.g. Wolfram [1988], 264 with 492 n.120). 134 Jordanes, Getica 276, Romana 347. 135 Scott (1976), 64.

79

DYNASTY AND ETHNICITY

sufficient land and income to feed their nation and to be able to defend its prop­ erty from the depredations of neighbours. Starvation was just as threatening as any hostile war-band. Leo had stood up to the Amal Goths not long before, so the Huns might expect the same treatment. When Dengizich learned that Leo did not believe the Huns should ever have access to Roman trade, in view of the damage they had earlier inflicted on Roman territory, he proposed attacking the empire as a consequence. His brother Ernach, distracted by his own local conflicts, counselled to the contrary.136 Yet Dengizich’s resolve eventually prevailed and he did threaten action. The Hun warlord kept to the bank of the Danube where he encountered a comes rei militaris Anagastes, son of Arnegisclus formerly magister militum of Thrace and fellow-general of Aspar in the 440s. Anagastes’ inquiries were rebuffed. The Huns preferred to deal with Leo directly. Again they journeyed to Constantinople, again they threatened destruction if Leo did not grant them land and money. By now the emperor was quite amenable, so we learn from Priscus, and prepared to agree to the Huns’ demands.137 Leo’s willingness to accommodate the Huns’ request on this occasion may have been due to more pressing priorities, especially his planning for the war on the Vandals. However, it may also reflect the pressure of Aspar, fearful of the Huns encroaching on the domain of the Thracian Goths with whom he was so closely allied through his wife’s family.138 Before long, in 467, the Romans had become involved somehow or other against both the Goths and the Huns. Perhaps the Goths had aligned themselves with the Huns, now experiencing drastic food shortages because of the Roman blockade. As with so many of the encounters in this period between the Romans and both the Goths and Huns, there is no record of what must have been a sig­ nificant military expedition, just its aftermath in the extract of Priscus preserved as a barbarian embassy.139 On this occasion, the Huns and Goths were opposed not only by the Roman generals Anagastes and Ostrys, both of Gothic origin but now leading Roman armies against the Goths, but also by the emperor’s brother­ in-law Basiliscus, then magister militum for Thrace. It was probably as part of this conflict that the recently disgraced Ardaburius vanquished the Gothic leader Bigelis.140 The Huns petitioned the Roman generals for a quick settlement because

136 Priscus, frag.46 (352/3 Blockley) = Exc. de Leg. Gent.18. 137 Priscus, frag.48 (354/5 Blockley) = Exc.de.Leg.Gent. 20. For Anagastes’ position at the time: PLRE 2, 75 (‘Anagastes’). 138 Argued by Maenchen-Helfen (1973), 166. 139 Priscus, frag.49 (356/7–358/9 Blockley) = Exc.de Leg.Gent. 21, with Wolfram (1988), 265 but dating to 469 rather than 467. On Ostrys’s position as a Roman general fighting against the Goths: PLRE 2, 814 (‘Ostrys’). 140 Bigelis probably led one of the Gothic war-bands previously attached to Attila (Heather [1991], 14; PLRE 2, 229 [‘Bigelis]). The victory over Bigelis is contained only in the statement of Jor­ danes that Ardaburius killed ‘Bigelis king of the Goths’ (Romana 338: ‘Bigelemque Getarum regem per Ardaburem Asparis filium interemit’ with Bury [1923a], 335–7; Vernadsky [1941], 63–7). There is no reason to doubt Jordanes’ information, nor to doubt the role of Ardaburius in the death of Bigelis. He had been recalled in disgrace a couple of years earlier and will not have

80

DYNASTY AND ETHNICITY

they were desperately short of food. Rather than commit the emperor to an emer­ gency agreement, the commanders decided to refer the request to Leo. Meanwhile they guaranteed to keep the Huns properly fed, provided they broke into smaller groups each guarded by a Roman general. The Huns agreed to comply and their envoys set off for Constantinople. One of the groups contained a mixture of Huns and Goths, but mainly Goths, and was now under the watchful eye of what Pris­ cus calls ‘Aspar’s men’ (oi~ :Asparoj). Exactly who these were is unknown. They were commanded by a certain Chelchal, a junior officer of Hun background, so would have been one of the units of the magister militum praesentalis deployed by Aspar as reinforcements for this campaign, or possibly even a unit of Aspar’s bucellarii or private retainers.141 Chelchal summoned the leaders of the Goths and explained that Leo would not be dealing with them independently and would not give land to them but only to the Huns. Since the Goths do not have a treaty with the Huns, Chelchal continued, they would lose out. Trusting in the integrity of the Hun Chelchal, the Gothic leaders were infuriated by his information and turned on the Huns amongst them. Word spread to the other groups under guard and similar massacres ensued. ‘Aspar’s men’ destroyed those in their supervision but some Huns, including Dengizich, managed to escape.142 The Priscus extract does not say what eventuated at Constantinople with the Huns’ envoys. Ardaburius was also involved in these campaigns and it was possibly in this period that an episode recounted by Candidus belongs. Candidus notes only that Ardaburius once sought to ‘attach the Isaurians’ to his cause (oivkeiopoih,sasqai tou.j ~Isau,rouj), but that his attendant Martin tipped off Zeno.143 Candidus does not say whether or not these Isaurians were soldiers although they probably were, nor that they were in the employ of Zeno. He does not date the episode either, but in Photius’s summary it falls between Zeno’s marriage to Ariadne (468) and the murder of Aspar and Ardaburius (471). So it was probably before 469 when both Ardaburius and Zeno were in Constantinople. By then, it is possible that Ardaburius actually had an Isaurian wife in Anthusa the daughter of Illus.144 Arda­ burius was trying to win over Isaurians not in order to undermine the Isaurian faction and promote the ‘Gothic faction’ but because they were potential allies to be enticed to his side in a straightforward power struggle.145 According to Candi­ dus, Ardaburius was planning on using the Isaurians not against Zeno but against

141 142 143 144 145

been in another senior command, he presumably was deployed to Thrace on behalf of his father Aspar, in a similar manner to the position held by Chelchal. It has been proposed that he was magister militum per Illyricum at the time which is not impossible but unlikely in the circum­ stances (Scharf [1993], 216–19). There is no certainty about the date of this episode. Jordanes places it between Anthemius’ coronation in Rome (March 467) and the Vandal expedition (sum­ mer 468), so 467 is most probable (Cf. Demandt [1970], 767 and [1986], 115). As suggested by Maenchen-Helfen (1973), 168. Priscus, frag. 49 [356–7 Blockley] = Exc.de Leg.Gent. 21 Candidus, frag. 1 (Blockley, 466–7)], with Demandt (1970), 765. As proposed by Scharf (1993), 221–3. Elton (1996), 397.

81

DYNASTY AND ETHNICITY

Leo (kata. basile,wj). In other words, whether Aspar was involved or not, his son Ardaburius was clearly plotting against the emperor. Following the Gothic and Roman attacks on the Huns in 467, Dengizich king of the Huns remained at large in Thrace. Eventually, he was run to ground and killed by the Roman general Anagastes. The defeat and death of the Hun king in battle was a considerable bonus for the Romans who had suffered humiliation at the hands of the Huns in previous years. For performing a feat which had so recently eluded a considerable part of the imperial army, Anagastes could look forward to acclaim and reward in the imperial capital as Dengizich’s head was paraded through the city on a pole amid scenes of enthusiastic rejoicing.146 All this took place in the latter months of 468, although it is normally dated to 469.147 By this time in Constantinople various new political forces had come into play arising from the uncertain outcome of the Vandal expedition which had sailed for Africa early in the summer.

Basiliscus and the ‘Fourth Punic War’, 467/8 The imperial expedition against the Vandals in North Africa in 468 proved to be a major turning point in the now uneasy relationship between the emperor Leo and his senior general Aspar. The new western emperor Anthemius’ departure for Rome in the spring of 467 was, according to Procopius, part of Leo’s planned strategy for dealing decisively with the Vandals.148 Since Procopius’ information probably originated with Priscus it should be deemed trustworthy. Consequently, Leo was engaged in detailed strategic planning for his mission against the Vandals as early as March/April 467. He would therefore have considered it essential in 467 to secure peace with the Goths and Huns along the Danube and in the Balkans. The appointment of Anthemius and the opening up of a more supportive relation­ ship with the western court brought the Vandal regime at Carthage more centrally into the policy orbit of the eastern emperor. Apart from holding hostage the daugh­ ter and granddaughters of Theodosius II since 455, the Vandals continued to raid Roman territory annually, not just the closest islands of Sicily and Sardinia but also the coasts of Italy, Greece and Egypt. Previous military challenges to the Vandals, 146 Marcellinus, Chron. s.a. 469, MGH AA XI, 90; Chron.Pasch. 598 (Dindorf). 147 It is dated to 469 because it is recorded in the chronicle of Marcellinus under that year. Mar­ cellinus, however, dates events by both indictions (beginning in September) and consulships (beginning in January) which sometimes creates confusion or uncertainty where the dating sys­ tems overlap (cf. Croke [1995] 54–5 and [2001], 173–5). On this occasion his source must have had the event in the 5th indiction (1 September 468/31 August 469) but assuming that it occurred in the latter part of the indiction he placed it in the consulship for 469, rather than 468. The dates are perfectly reconcilable by assigning Anagastes’ conquest of Dengizich to the earlier part of the indiction instead. That is, the slaying of Dengizich and the public display of his head took place in Constantinople between 1 September and 31 December 468, in the consulship of Anthemius, as recorded in the Chronicon Paschale 598 (Dindorf). 148 Wars 3.6.5, with Crawford (2019), 62–5.

82

DYNASTY AND ETHNICITY

such as that of the western emperor Majorian in 460, had not succeeded. Leo resolved to defeat the Vandals and expel them from Africa. As noted by Gautier, the eminent historian of North Africa, these campaigns could be construed as the ‘Fourth Punic War’.149 By now the imperial hostages had returned to Constanti­ nople, at least Eudoxia and her daughter Placidia had returned in 462. Eudocia remained in Africa as the daughter-in-law of the Vandal king, Gaiseric. Having married his son into the imperial Theodosian house, Gaiseric became keen to prosecute the imperial cause of Eudoxia’s other son-in law, Olybrius, then living at Constantinople. Indeed, demand for Leo’s support for Olybrius’ claim may have been part of the settlement leading to the release of Eudoxia and her daughter.150 Even if Leo tried to dissemble, his envoy Phylarchus was unsuccessful in deal­ ing with Gaiseric in 462,151 Tatian was not even accorded a hearing at Carthage in 464,152 while Phylarchus failed again in 467.153 The Vandals now preferred to fight. At about this point, if not earlier, Leo too had determined to abandon diplomacy and turn to an all-out expedition for the following campaigning season. Ships and troops were organised and assembled from far and wide. A total of 1,100 ships formed the imperial armada against Gaiseric. These vessels carried more than 100,000 soldiers, Roman and otherwise, who were now in the imperial pay. The cost of the expedition was comparably enormous.154 Whether or not some of these numbers were exaggerated by sixth-century writers keen to highlight the contrast with Belisarius’ more modest expedition in the 530s,155 it was still an immense fleet. Yet it was no larger than that in 440 actually led by Aspar to dislodge the Vandals just after they had occupied Carthage. Presumably all the ships and men gathered at the burgeoning docks of Constantinople and made preparations in the winter of 467/8, so they were ready to sail in the spring of 468. The large expedition had brought soldiers and sailors from everywhere, a most intense ethnic mix in what was already a polyglot capital. One such contingent may have been the bucellarii of the Gaul named Titus who deserted the service of the emperor for the company of the ascetic Daniel.156 Also among those sailing to Constantinople in 467 to join the African expedition was a contingent from Isau­ ria.157 They had lost several of their number en route at Rhodes. It was a bloody 149 Gautier (1935), 217–71. 150 Courtois (1955), 200. 151 Priscus frag.39 (Blockley, 342/3) = Exc. de leg. Gent. 14. Phylarchus’ embassy to Gaiseric in 462 is overlooked in PLRE 2, 884 (‘Phylarchus’). 152 Priscus, frag. 41.2 (Blockley 346/7) = Exc. de leg.Rom.11. On Tatian: PLRE 2, 1053–4 (‘Tatianus 1’), and Ensslin (1932), 2467–8 (‘Tatianus 4’). 153 Priscus, frag. 52 (Blockley 360/1) = Exc.de leg.Rom. 13. 154 As carefully computed by Treadgold (1995), 189–9. 155 As suggested by Courtois (1955), 202, cf. Gautier (1935), 255. 156 v.Dan.Styl. 60–1 with PLRE 2, 1122–3 (‘Titus 1’), cf. Kosinski (2010), 63. 157 John of Antioch, frag. 229 (Mariev 416–17). This fragment is frequently assigned to 469 but can be safely dated on internal evidence to 467/8. It comes between Anagastes’ murder of Ullibus

83

DYNASTY AND ETHNICITY

response to their aggressive behaviour towards the locals. Thereupon ‘they fled to their ships and came to Constantinople where they joined Zeno’. Even there the Isaurians were pelted with stones for harassing merchants in the market. The inci­ dent threatened to escalate but darkness brought calm. These Isaurians were tran­ sients who had brought their own ships to bolster the Roman fleet, part of the growing expeditionary force, not ‘the first introduction of an Isaurian garrison into Constantinople [which] was probably directed against Aspar’158). It was unruly behaviour like theirs which caused the emperor to legislate the following year to curb the actions of private retainers (bucellarii) and Isaurians among others.159 Although addressed to the Praetorian Prefect of the East, Nicostratus, the law applies to everyone in both town and country (‘omnibus per civitates et agros’). It was not designed to curb the influence of Isaurians or Goths in the imperial capital. The outcome of the Vandal expedition was probably still uncertain when this law was promulgated (28 August). Perhaps the reduction in armed forces in Constantinople and vicinity had caused these groups to suddenly flourish. To head up the complex and risky operation against the Vandals Leo chose his brother-in-law Basiliscus, appointing him as magister militum praesentalis.160 This promotion was bound to unnerve Aspar, irrespective of the outcome of the expedition. It would not be surprising if he actually opposed the appointment although there is no evidence to that effect. That Aspar declined to lead the expe­ dition because of his reservations about its likely outcome is also disputable.161 The foundation-stone of Aspar’s power since the early 450s had been his tenure as senior magister militum praesentalis. Until this point he had shared the office with Anthemius and he might have been relieved at Anthemius’ departure for the west, that is to say, if he had not been instrumental himself in the appoint­ ment to the western throne. In raising Basiliscus to the same position as Aspar in the course of transferring him from his post at Marcianople as magister militum per Thraciam, Leo was clearly signalling a change of policy towards Aspar and asserting his independent will. Even so, it appears that Ardaburius’ disgrace was set aside momentarily so that his military ability and experience could be called on for the expedition.162 The colossal imperial military force should have overwhelmed the Vandals. That it failed to do so was due to a range of factors. Above all, so it seems, the Vandals’ superior knowledge of the wind patterns and geography of North Africa

158 159 160 161 162

‘when Anthemius and Leo were emperors’ (frag.228 [Mariev 416–17], cf. Zachariah Mytilene, HE 3.12) and the revolt of Anagastes in 469 (frag.229 [Mariev 416–17]). It also appears to best fit circumstances in 467/8 (Demandt [1970], 469, 766, contra Lippold [1972], 155–6) rather than 469 (Brooks [1893], 470). Since they came to Constantinople specifically to Zeno they must be before he was absent, that is before mid-469. Brooks (1893]), 213. CJ. 9.12.10 (28 August 468) with Demandt (1970), 767; Scarcella (1997), 393–6. Demandt (1970), 777–8. Cf. Scott (1976), 65: ‘One can presume that Aspar could have had the command if he wanted it’. Damascius, vita Isidori (apud Photius, Bibl. 242) with Demandt (1986), 115.

84

DYNASTY AND ETHNICITY

gave them the upper hand. The Roman fleet parked on the west coast of Cape Bon, about 60km from Carthage, was unable to avoid the flaming Vandal fire-boats propelled towards them by the onshore winds. The instant loss of such a large part of the expedition emasculated the imperial project.163 The fleet retreated and its commander-in-chief Basiliscus returned to Constantinople in ignominy. He was forced to seek protection in Hagia Sophia until he was rescued by his sister Verina. Then he was redeployed to Herakleia for an unspecified period.164 He was still in Herakleia in 471. The contemporary historian Candidus described, according to the skeleton summary of Photius, both the successes and the setbacks experienced by Basiliscus.165 There is no mention of Aspar in Photius’ summary of Candidus. There is, however, in the account of Priscus (as preserved in Procopius) where Basiliscus’ motives for delaying the fight against Gaiseric and giving him the advantage are canvassed: ‘either as a favour to Aspar as promised, or because he had sold the opportunity for money, or because he thought it the best course’.166 There was obviously uncertainty at the time about what led to the destruction of most of Basiliscus’ fleet. Later writers found they could do little but report the variety of views. Theophanes, probably drawing on Priscus too, attributes the disaster to either incompetence or treachery on the part of the general. Zonaras in the 12th century and Nikephorus in the 14th effectively follow their predecessors except that Nikephorus reports it all as part of an elaborate conspiracy in which Basiliscus threw in his lot with Aspar and Ardaburius against Leo but failed.167 Hydatius, an exact contemporary, makes clear that Aspar and Ardaburius were under suspicion for supporting the interests of the Vandals.168 Given the confusion and contradictions in the extant documentation it is diffi­ cult to be certain of the relationship between Aspar and Basiliscus and how it was hindered or promoted by the war against the Vandals. What is likely, to judge from Basiliscus’ later behaviour, is that he did harbour his own ambitions for the throne. If Aspar had hoped still in 468 to secure a role for his son Patricius in the imperial succession, as appears to be the case, the ambition of Basiliscus had become an extra obstacle. Being now of comparable authority with Aspar made Basiliscus a positive threat. His failure against the Vandals can only have dented his standing and his prospects. It also removed him from the imperial presence. So it is not impossible that Aspar had been involved in events leading to exactly that result. He had a motive to undermine Basiliscus. On the other hand, Basiliscus may have been prepared to conspire with Aspar in order to secure Aspar’s backing for his own imperial aspirations. After all, it had been Aspar’s backing that ensured the 163 Gautier (1935), 255–7; Courtois (1955), 203. 164 Procopius, Wars 3.6.27; Nicephorus Callistus, HE 15.27 (PG 147.80C). 165 Candidus, frag. 1 (Blockley 466–7) = Photius, Bibl.79. According to Cedrenus 383.3 (599 Tarta­ glia) the war with the Vandals continued down to 473/4. 166 Procopius, Wars 3.6.16, cf. Theodore Anagnostes, HE.Epitome 399 [111.17–21 Hansen]. 167 Theophanes, AM 5961 (de Boor 116); Zonaras, 14.1.24–5 (Büttner-Wobst 126); Nicephorus Cal­ listus, HE.15.27 (PG 153.80). 168 Hydatius, Chron. 241, s.a.468 (R. Burgess [1993], 121), cf. Brooks (1893), 213 n.21.

85

DYNASTY AND ETHNICITY

elevation of both Marcian in 450 and Leo in 457. Perhaps Basiliscus had promised Aspar he would be a more accommodating emperor than his brother-in-law. At least, that seems to have been the version which made most sense to Nicephorus nearly a millennium later. While Aspar was not the only magister militum praesentalis he was still the only one at Constantinople. Leo was doubtless disappointed and frustrated at the failure of his Vandal expedition and its political and financial consequences. Aspar could say he told the emperor so.169 Before the Vandal campaign in 468 it was said to be likely that Aspar ‘would plot against the Emperor Leo, who had given him offence’ so that Aspar ‘was then fearful lest, if the Vandals were defeated, Leo should establish his power more securely’.170 That is why Aspar was accused of undermining Basiliscus, namely to get back at Leo, just as Ardaburius was culti­ vating local Isaurians to his side. Aspar’s remaining hopes for his son’s imperial marriage and succession were slipping away. Aspar himself seems to have felt the pressure first in the aftermath of the failed Vandal venture. Ambassadors who had been in Constantinople at the time reported back to the Suevi ‘that Aspar had been cashiered and his son executed after they had been discovered plotting with the Vandals against the Roman Empire’.171 While they heard wrongly about Aspar’s son the envoys were right to report that Aspar himself was under suspicion for his relations with the Vandals. In this tense situation Zeno remained at court as comes domesticorum and Leo possibly now regarded him as the imperial heir apparent, although only a promotion to Caesar would secure that. Meanwhile, as we have seen, the Huns’ king Dengizich was once again threatening Roman authority and property in the Balkans and would have to be contained.

Anagastes’ rebellion and Zeno’s flight, 469 Following the failed Vandal campaign, and Aspar’s evident role in its failure, rela­ tions between the emperor and his senior general were on a final collision course. Their mutual struggle for dominance was aggravated by the revolt of Anagastes in the Balkans the following year (469), that is the year after he had defeated the Hun king Dengizich. On the information brought forward by Zeno in 465, as we have seen, Aspar’s son Ardaburius had been convicted of treason and was stripped of all official power and titles. Subsequently, Ardaburius, and no doubt Aspar too, nursed his grudge against Zeno and his imperial father-in-law and cast about for an opportunity to exact revenge. This appears to explain what immediately followed in the aftermath of the failure of the Vandal expedition and Aspar’s exclusion of

169 Cf. Gautier (1935), 263; Stein (1959), 360, cf. Crawford (2019), 65–74. 170 Procopius, Wars 3.6.3–4 (from Priscus). 171 Hydatius, Chron. 241, s.a.468 (Burgess [1993], 121): ‘Asperem degradatum ad priuatam uitam, filium eius occisum, aduersum Roman imperium sicut detectique sunt, Vandalis con­ sulentes’; with Thompson (1982), 223–6.

86

DYNASTY AND ETHNICITY

Basiliscus from court. Anagastes, now magister militum for Thrace,172 did not receive the official thanks and glory he expected from his victory over Dengizich late in 468. In fact, he was an epileptic and the thought of an epileptic consul dis­ suaded the emperor from appointing him for the following year (470). What is more, the consulship was later announced as belonging to Jordanes, then magister militum in the East. In the appointment of Jordanes, Anagastes felt particularly slighted because his own father Arnegisclus had been responsible for the death of Jordanes’ father in 441. This is a sign that family feuds among the military nobility could be passed down from one generation to the next.173 From John of Antioch we discover that Anagastes then set about some sort of rebellion, and that Arda­ burius was instrumental in provoking Anagastes’ revolt just as he had attempted to do with the Persians in 465, and was accused of doing in assembling a local group of Isaurians. In fact, Anagastes eventually forwarded to Leo the incriminat­ ing correspondence from Ardaburius, just as Zeno had done in 465.174 What is not so clear, though, is precisely what Ardaburius hoped to gain from provoking Anagastes into rebellion. The most obvious reason is that in facing a new threat on another front closer to home Leo would be obliged to put another army into the field probably under his loyal son-in-law Zeno. With Zeno out of the way, and especially with Basiliscus now biding his time at Herakleia, Aspar and Ardaburius could more easily expect to re-assert their position at court. The fact that the rebellion of Anagastes took place in mid to late 469 after the announcement of Jordanes’ consulship for 470 is most instructive. That was the very year in which the current consul Zeno was sent against some enemy in Thrace as recounted in c.65 of the vita of Daniel the Stylite. What has never been suggested, so it would appear, is that in 469 Zeno was not sent against some Hun or Goth disturbance in Thrace, as normally supposed, but against the rebel­ lious general Anagastes. The account provided in the life of Daniel is a full one and deserves close consideration, but we must first focus attention on Theophanes’ version of the year 469. Theophanes, probably depending on Priscus, reports that 172 There has been an element of uncertainty about Anagastes’ position when he rebelled in 469. When he was part of the large-scale campaign against the Goths and Huns in 467 he held the posi­ tion of comes rei militaris while Basiliscus was magister militum per Thraciam (Priscus, frag.49 [Blockley 356/7] = Exc. De Leg. Gent. 21, with PLRE 2, 75 (‘Anagastes’), and 1292 (fasti).). When Basiliscus was transferred and promoted to magister militum praesentalis to lead the Van­ dal campaign it is not stated who replaced him in Thrace. On the mere assumption that Zeno succeeded Basiliscus in Thrace in 467/8, and that Anagastes was magister militum per Thraciam when he killed Dengizich, it is taken for granted that Anagastes succeeded a short tenure by Zeno, but on the mistaken assumption that Dengizich was killed in 469, not 468. The reality, so it would appear, is the exact reverse: Anagastes succeeded Basiliscus in 467 and Zeno was not appointed until he was consul in 469, as explained in the vita of Daniel (v.Dan.Styl. 65 with Seeck [1894]). 173 John of Antioch, frag. 229 (Mariev 416–17). Anagastes’ revolt is placed in 470/1 for no special reason, by Bury (1923a), 319; Stein (1959), 360. 174 John of Antioch, frag. 229 (Mariev 416–17). There is no evidence to link the rebellion with ‘the introduction of the Isaurians into Constantinople’ (Brooks [1893], 214).

87

DYNASTY AND ETHNICITY

‘In that year the emperor Leo sent Zeno ‘the general of the East’ and his son-in­ law to Thrace for some military purpose (evpi, tina crei,an polemikh.n) and ordered he be given a contingent of his own men for his assistance (strato.n evk tw/n ivdi,wn pro.j summaci,an)’, that is a contingent of the emperor’s personal military forces or protectores domestici under the command of the comes domesticorum. Theophanes goes on to explain that in accordance with the instruction of Aspar some of the emperor’s contingent who were supposed to be protecting Zeno were planning to take him captive. Zeno was tipped off, however, and escaped safely to Serdica. Thenceforth Aspar became an object of further suspicion to Leo, as explicitly stated by Theophanes.175 Strangely enough, Theophanes does not here indicate what enemy Zeno was sent to repel and it needs to be admitted that when Dengizich’s last stand is con­ signed to its proper place in late 468 we have no evidence of any other barbarian threat in Thrace precisely in 469.176 Nor is it clear why Zeno should be needed in Thrace if he was actually magister militum per Orientem at the time, as Theo­ phanes states. A Thracian threat was, as had been the case in the two previous years, the responsibility of the local duces and the Thracian magister militum. Theophanes, very often inaccurate in these matters of nomenclature, may have meant to say that Zeno was magister militum per Thraciam at the time.177 In that post he would not be out of bounds in Thrace. Theophanes’ evidence makes more sense when considered in conjunction with the more contemporary vita of Daniel the Stylite (c. 65). There we learn that Zeno became consul (469) and ‘shortly afterwards (met vou~ polu.n cro,non) when a barbarian disturbance broke out (tarach/j barbarikh/j genome,nhj) in Thrace Leo appointed him magister militum per Thraciam (strathla,thn th/j Qra|vkhj)’. It is clear from this passage that Zeno was appointed magister militum per Thraciam in the year of his consulship (469).178 Moreover, his precise task was to take charge of the threat emanating from some ‘barbarian disturbance’, tarach,, being the normal word for domestic 175 Theophanes AM 5962 [de Boor 116]). 176 The suggestion of Brooks (1893), 215 that Zeno was sent against the Ostrogothic king Theudimer is only a guess and an unlikely one since the Ostrogoths were then campaigning against the Suevi on the Danube (Jordanes, Getica 278–80). These events belong, however, to 473 (Heather [1991], 264–5). 177 cf. Demandt (1970), 767. 178 v.Dan.Styl. 65. There is much unnecessary doubt and confusion about Zeno’s positions in 469. Bury (1923a), 318 thought Zeno was acting in Thrace as magister militum praesentalis. Not yet having the vita Dan.Styl., Brooks (1893), 213 n.17 was forced to believe that Zeno had been magister militum per Orientem since about 467. More radically, Schwartz (1934), 183 n.3 places these events in Thrace in 471 and has Zeno relocated to Thrace from being magister militum per Orientem in 469/70. Kosinski (2010), 67 follows Theophanes. When the magister militum per Thraciam revolted in 469 the emperor immediately revoked his command and gave it to his son-in-law before despatching him to deal with Anagastes. This reconstruction has the added advantage of accounting for the odd fact that the newly appointed magister militum for Thrace should be forced to rely on a contingent recruited from the emperor’s own force instead of the Thracian army he was being sent to command. The army was still with Anagastes.

88

DYNASTY AND ETHNICITY

rebellion or tumult in fifth-century usage,179 while barbarikh,, suggests that Ana­ gastes had enlisted the support of local tribal groups or contingents. The Goth Ulli­ bos who rebelled at this time may have originally played such a role. He is linked with Anagastes by John of Antioch (probably from Priscus) who labels both of them ‘Scyths with a tendency to rebellion’ although they were opposed to each other on this occasion.180 It was after killing Ullibos that Anagastes’ revolt broke out in earnest. The vague description of the enemy given in both Theophanes and the vita, combined with the fact that Zeno was appointed general in Thrace in exactly the same year as Anagastes revolted, points to the fact that Zeno was sent to Thrace to deal with the uprising of Anagastes. This then raises the question of Anagastes’ precise status in 469. The exact timing of Zeno’s tenure as magister is important. It means that he was in Constantinople, not Marcianople, in 468 and into the campaigning season of 469. It was during this narrow but crucial period that Basiliscus was away on the Vandal expedition and was subsequently detained at Heracleia. Zeno’s presence will have reassured Leo. Further consolation and advice was able to be provided by Leo’s favourite holy man, Daniel the Stylite. The role of Daniel in events of 469 is that he was consulted by Leo and Zeno about the outcome of the expedition against Anagastes. He prophesied that a conspiracy would be formed against Zeno but that he would eventually escape unharmed.181 So the new magister militum per Thraciam Zeno set out for Thrace. At some stage the soldiers in the pay of Ardaburius (and Aspar) closed in but Zeno slipped away to Serdica.182 The vita goes on to say that Zeno eventually reached the ‘Long Wall’ and proceeded from there to Pylai, and some time later he reached Chalcedon. The vita does not say that he then proceeded to Constantinople, or his native Isauria, or anywhere else for that matter. Two items are of special significance here: (A) the role of Aspar and (B) Zeno’s escape route. A. The role of Aspar We have already observed that by 468 tension was building up between the fam­ ily and supporters of Leo and Aspar’s family and supporters. Zeno’s role in the dismissal of Ardaburius in 465 and his marriage to Ariadne in 468, together with Basiliscus’ appointment against the Vandals in 467, served to erode further the long-standing influence of Aspar and his family. Now Ardaburius’ complicity in the revolt of Anagastes could only create even greater antagonism and ten­ sion at Constantinople. The fact that Ardaburius, and presumably Aspar too, was behind Anagastes leaves them open to the suspicion that they engineered the 179 Gregory (1982), 10–11. 180 John of Antioch, frag. 228 (Mariev 416–17); Zachariah Mytilene, HE 3.12 with Norman (1953), 171–2; Sotiroudis (1989), 135–9. 181 v.Dan.Styl. 65. 182 v.Dan.Styl. 65; Theophanes AM 5962 (de Boor 116).

89

DYNASTY AND ETHNICITY

whole episode in order to extract Zeno from the court at Constantinople and thus place him in a vulnerable position by arranging his murder; while at the same time increasing their own access to, and influence over, Leo.183 It was similar to the tactic which saw Basiliscus removed from imperial proximity the previous year. If, as argued here, Ardaburius and Aspar encouraged Anagastes to revolt as a sure pretext for removing Zeno from Constantinople it is no less likely that it was Aspar who actually proposed Zeno’s appointment to Leo. When the carefully laid plans backfired, with Aspar and Ardaburius’ complicity revealed, Leo was powerless and isolated in the imperial capital. He could take no effective action against them. At the same time Zeno will hardly have considered it safe to return to Constantinople having narrowly avoided one attempt on the part of Aspar and Ardaburius to have him murdered. He would be best advised to keep well clear of the capital until events turned more favourably in his direction. B. Zeno’s escape route The position and whereabouts of Zeno between mid to late 469 and the murder of Aspar in 471 have never been clear, except that for much of this period he was magister militum per Orientem and therefore based in Antioch. By any account Zeno’s return route to Constantinople following the plot on his life was a round­ about one. On the surface, it appears that Zeno travelled from Serdica to the Long Wall, thence to Pylai and later reached Chalcedon. What is less certain is whether the ‘Long Wall’ he reached was the so-called Anastasian Long Wall 65km west of Constantinople, or that across the neck of the Chersonese. If the Anastasian wall was built de novo by Anastasius and therefore did not exist in 469 then it must have been the Chersonese Long Wall that was Zeno’s departure point. If the Anas­ tasian wall was built by 469, then it could be either wall.184 Irrespective of whether the Anastasian wall was already constructed in 469, it has been claimed that the Chersonese Long Wall cannot be the one mentioned in the vita because: (1) Zeno had no need to avoid Constantinople since there he enjoyed the protection of his Isaurians, (2) he was appointed magister militum per Orientem en route back from

183 It is possible that Ardaburius’ failure with the Isaurians prompted him to seek out Anagastes (Brooks [1893], 214). It should also be pointed out that there is no evidence for the romantic notion of Vernadsky (1941), 68 that Aspar now planned the establishment of his own independent Danubian homeland of Goths, Alans and Antae. 184 The case in favour of the Chersonese wall was advanced originally in Croke (1982) but resting on the presumption that the so-called Anastasian wall was not yet built. Strong arguments in favour of the Anastasian wall being built in the 440s, and therefore being the wall referred to in the vita Danielis were put by Michael Whitby (1985), 560–83. It now seems fairly clear the Anastasian Long Wall may have been begun earlier in the fifth century but can mainly be attributed to the initiative of the emperor Anastasius in the late fifth century (Crow [1995] and Crow and Ricci [1997], 239, 260). Incidentally, the Chersonese wall has been located recently but has not yet been the subject of significant research. It is explicitly mentioned, therefore as complete, just a few years after the 469 revolt (John of Antioch, fr.233 [Mariev 425]).

90

DYNASTY AND ETHNICITY

Thrace so he would naturally have taken the fastest route to the East, necessarily bypassing the capital, and (3) he therefore did not need to cross to Asia as far south as the Chersonese. None of these claims is beyond question.185 As for (1), Zeno had every reason to avoid Constantinople and the main Balkan highway leading there. The capacity to surround himself with an impregnable Isaurian retinue is severely over-estimated, as we have seen. Further, those who hoped to see him killed in the campaign against Anagastes, namely Ardaburius and Aspar, were in Constantinople with access to far more forces than Zeno. Hav­ ing reached Serdica, Zeno would have been ill-advised to take the route which brought him closest to the Goths of Aspar’s nephew Theodoric Strabo. As for (2) and (3) Zeno was not appointed to the East while on the run in Thrace. Rather, it is more likely that he was appointed while waiting at Chalcedon for an oppor­ tune moment to return to Constantinople. Then he set off for Antioch taking with him from Chalcedon the monk Peter the Fuller. His ultimate destination when he reached the Long Wall was Constantinople which is why he did not cross to Lampsacus and continue on from there but sailed the length of the Marmara to reach Pylai then proceeded overland to Chalcedon. Consequently, a fuller con­ sideration of the political context of the events surrounding the plot against Zeno in 469 suggests that his escape route must have taken him south from Serdica to link up with the Egnatian way east of Thessalonika. He then proceeded along it to the Long Wall at the Chersonese and then took a boat across to Pylai where the main highway through Asia Minor to Isauria and Antioch began. At Pylai he could decide whether to retreat to Isauria or return to Constantinople from the Asiatic side. In other words, Zeno carefully avoided the direct route from Serdica to Constantinople by way of Adrianople. All the same, if Zeno departed from the Anastasian Long Wall, and not the Chersonese Long Wall, the fact remains that he was avoiding the imperial capital for the time being. It is normally assumed that at this juncture Zeno actually returned to Constantinople and was then appointed magister militum per Orientem before departing at leisure for Antioch.186 Yet there is no testimony to this and it is unlikely in the circumstances. He was appointed while waiting at Chalcedon. There is, however, a curious turn of phrase in the chronicle of John Malalas which implies that Zeno actually travelled to Isauria at this time. In introducing Illus as an Isaurian, but friend and ally of Zeno, Malalas says that he ‘had escorted [Zeno] back with a large force on his second return from Isauria (th.n deute,ran auvtou/ evpa,nodon avpo. th/j ’Isauri,aj), after he had fled there from Constantinople while emperor’.187 Having been driven out of Constantinople by Basiliscus in January 185 Whitby (1985), 563–7. Zeno was not sent against the Hun king Dengizich (so Whitby [1985], 564) whom Anagastes had conquered the previous year but against Anagastes who had rebelled. 186 E.g. by Bury (1923a), 319; Lippold (1972), 157. 187 Malalas, Chron. 15.12 (Thurn 309), with translation from Jeffreys et al. (1986), 214. A similar turn of phrase is used by the author of the Parastaseis to describe the philosopher Galen’s predic­ tions about Verina and her son-in-law Zeno as emperor, but after his restoration in 476: meta. th.n

91

DYNASTY AND ETHNICITY

475, Zeno regrouped his family and allies in Isauria then began the campaign to return to Constantinople to reclaim his throne in 476. If that was his ‘second return’ from Isauria it begs the question of when was the ‘first return’. From what is known of the career and movements of Zeno it can only have been when he was turned out of Europe by the troops of Aspar in 469. Malalas, at least in his extant form, does not mention Zeno’s ‘first return from Isauria’ but it was doubtless his return after the murder of Aspar in 471. So, in effect, in the course of 469 Zeno entered a self-imposed exile journeying to Antioch by way of Isauria. If he were not appointed magister militum at Chalcedon then it may have been while he was in Isauria. What is certain is that in the same year he was involved as magister militum per Orientem in arranging for a rebellion in Isauria to be put down. As magister militum of the East he had the come Isauriae and the troops stationed in Isauria responsible to him.188 He was not using ‘Roman troops to consolidate his home in Isauria’.189 The legendary Isaurian speedster Indacus had been operating from his fortress on the hill of Papirius, named after his father, and ravaging the nearby countryside. Zeno managed to quell the raids of Indacus, perhaps being directly involved himself.190 His namesake Zeno, also as magister militum per Orientem, had done something very similar in the late 440s.191 At the time of Zeno’s appointment against Anagastes the magister militum for the East was Jordanes. He had been appointed consul for 470, however, and was presumably obliged to return to the capital to take up his consulship at some stage. Jordanes was definitely in Constantinople during his consulship in 470 because he fell under suspicion for taking a guided tour of the innermost part of the palace when the emperor was away, possibly at another palace during the summer months.192 It is evident that Zeno replaced Jordanes as magister militum per Orientem sometime late in 469. From Isauria Zeno moved onto the command headquarters in Antioch. Leo cannot have been pleased at having Zeno so far from the court especially since, according to Theophanes, it was as a result of this plot against Zeno that Aspar became suspect to the emperor Leo.193 Zeno was to remain at Antioch until the eve of Aspar’s assassination two years later, with his wife and infant son joining him there.

188 189 190

191 192 193

avna,kamyin Zh,nwnoj th.n avpo. ’Isauri,aj to. deute,ron (Parastaseis, 40 [ed. Preger [1901], vol. 1, 46]). Notitia Dignitatum, Or VII.56, ed. Seeck (1876). Shaw (1990), 252. John of Antioch, frag.229.2 (Mariev 419). On Indacus: PLRE 2, 590–1 (‘Indacus Cottunes’). Demandt (1970), 767–8 thinks that Jordanes was still magister militum per Orientem until the end of 469, so that for this foray Zeno must have been magister militum vacans, between two regional military posts (Thrace and the East). It is simpler, and better, to follow Lippold (1972), 156. Priscus, frag.16 (Blockley, 300–1). John of Antioch, frag. 31 (Mariev 421). Demandt (1970), 765 unnecessarily assumes that Jor­ danes was still magister militum per Orientem at the time and in fact lost his position as a result of this incident. Theophanes, AM 5962 (de Boor 116), with Crawford (2019), 79–80.

92

DYNASTY AND ETHNICITY

Patricius becomes Caesar, 470 By late 469 Aspar was now in the ascendancy once more. At Constantinople, sometime in 470, he arranged for Patricius to be finally named Caesar. This announcement had been his objective for well over a decade, just as it been Leo’s objective to foil Aspar’s request.194 Not only was Caesar a special title and posi­ tion, but it will have placed Patricius in the public position of heir-designate to Leo. One can only guess how Leo must have now felt after having so carefully deferred his original promise to Aspar back in 457 to elevate his son to Caesar, then evidently securing the succession through Ariadne’s marriage to Zeno in 468. Still, for whatever reason, he had not taken the final step of appointing Zeno as Caesar. Theodore Lector seems to have discussed the elevation of Patricius at length. Since his history is lost, however, we are obliged to make sense of the traces of it that survive in later writers. According to Evagrius, repeated exactly by Theophanes, it was ‘to gain the goodwill of Aspar’ that Leo made this appoint­ ment.195 In other words, the emperor might not like it, but he now had to accept the reality of Aspar’s position and influence. Zonaras reports that in the end Leo could resist Aspar’s demand no longer.196 The momentous appointment of Patricius as Caesar was clearly due to Aspar’s persistent coercion. That Leo was in no position to resist the demands of Aspar and Ardaburius is amply supported by episodes described in the vita of Marcellus, written in the sixth century on the basis of reliable testimony. In fact all power was now so con­ centrated in their hands that they were effectively emperors themselves.197 Much the same sentiment is echoed in the Syriac version of the Life of Simeon Stylites. It must have been around 470/471 that Leo made representations in Antioch, pre­ sumably to Zeno if he is the general mentioned, to have the remains of Simeon translated to Constantinople. At that time, according to Simeon’s hagiographer, Aspar and Ardaburius ‘were honoured like kings in the areas of their authority’.198 Leo was now powerless to resist. It was probably under these circumstances that Leo was so overwhelmed by his misfortune that he ‘not merely quitted the court – with phantoms haunting him as though he were an Orestes wanting in manhood – and dwelt elsewhere, but even came seriously to consider quitting the city itself . . .’. John the Lydian reports this anxious state of mind into which

194 Victor Tonnenensis, Chron s.a. 470 (ed. Hartmann, CCh 173a, 35 [12] = MGH.AA., XI, 188) with PLRE 2, 842 (‘Julius Patricius 15’). The testimony of Victor is frequently overlooked or unjus­ tifiably dismissed in dating this event to 468: Ensslin (1925), 1958; Stein (1959), 360; Demandt (1970), 772; Lane Fox (1997), 192. 195 Evagrius, HE 2.16: i[na th.n :Asparoj eu;noian kth,shtai; Theophanes, AM 5963 [de Boor 117.14]: i[na th.n :Asparoj eu;noian e;ch|. 196 Zonaras 14.3 (122. 8–10 Büttner-Wobst). 197 vita Marcelli, 32, ed. Dagron (1968), 314–16. 198 vita Sim.Styl. 125. It possibly refers to an earlier period when Ardaburius was magister militum per Orientem but the request for the transfer must have taken place sometime between the deaths of Simeon (469) and Leo (474) cf. Lane Fox (1997), 193–5.

93

DYNASTY AND ETHNICITY

Leo had fallen without locating it at a particular time, except that it was after the expensive failure against the Vandals in Africa.199 The vita of Marcellus, one of the ‘Sleepless Monks’ from their monastery across the Bosporus, provides two significant insights into the confidence of Aspar and his sons in the period from 469 to 471. The first is an incident concerning a cer­ tain John who had incurred the wrath of Ardaburius for some unspecified reason. The emperor was too powerless to protect him so he sought refuge in Marcellus’ monastery. Ardaburius hunted him down, and had a contingent of his own men besiege the monastery, repeating the vilest threats, but the holy man’s protection proved decisive in the end. The other incident originated in the popular resentment aroused at Constantinople by the enforced appointment of Patricius as Caesar and his marriage to Leontia in 470. As the vita of Marcellus notes, the Caesar was a ‘second emperor’ having all the trappings and apparel of an emperor except the crown. A Caesar became sole emperor on the death of the Augustus, no-one else daring to take precedence or to challenge him.200 Fearing that the death of the age­ ing and already infirm Leo would leave them with an Arian emperor, the orthodox populace, together with the patriarch Gennadius and Marcellus who left the mon­ astery on this occasion, marched from Hagia Sophia to the hippodrome in order to voice their protest. For hours they shouted their hopes for an orthodox emperor, that Patricius would not be made Caesar and would not marry the emperor’s daughter. In the end, Leo assured the crowd that Patricius had forsaken his Ari­ anism.201 The essential focus of this exchange is also reflected in Theophanes, probably relying on Priscus, who notes that Patricius was made Caesar only after he had agreed to abandon his Arianism and pledge his loyalty to Leo.202 More detail, but clearly derived from writers such as Priscus, is provided by Zonaras. He informs us that, when Leo announced that Patricius would be Caesar after all, the senate was disturbed and roused the people to revolt. Their real concern, accord­ ing to Zonaras, was the fear that when the imperial power passed to the family of Aspar the Arians would inflict greater atrocities on the orthodox than they had done earlier, in the reigns of Constantius II and Valens. So the clergy, the monks and the people, ‘whoever felt correctly about the faith’, implored the emperor to appoint a Caesar who was orthodox. ‘By both his words and his deeds’, concludes Zonaras, ‘the emperor quelled the insurrection’.203 Zonaras does not go any fur­ ther. He does not say that Patricius was in fact made Caesar, nor that he agreed to renounce his Arian adherence. 199 John Lydus, On the Magistrates. 3.44.3. 200 vita Marcelli 34 (Dagron [1968], 316–17): ~): ~O de. tou/to e;cwn to. avxi,wma deu,tero,j evstin basileu,j avlourgi,da kata,cruson kai. ta. avlla. tῆj basilei,aj para,shma evnduo,menoj di,ka mo,nou tou/ stefavnou & o[per evpi. toῦ Faraw. kai. tou/ ~Iwse.f avkhkoa,men) ~O kai/sar toi,nun kai. a;cri peri,estin o` basileu.j pa,nta koinh| pra,ttwn met vau/tou/( kai. teleuth,santoj monoj th.n basilei,an krateῖ mhdeno.j de. pro. au/tou/ mhde.. su. auvtw/| tolmw/ntoj evkei,nhn labei/n) 201 vita Marcelli 34 (ed. Dagron [1968], 316–18). 202 Theophanes AM 5961 (de Boor 116). 203 Zonaras 14.5–7 (122.11–123.8 Büttner-Wobst).

94

DYNASTY AND ETHNICITY

The doctrinal attachment of Aspar and his family was a serious block to their advancement. Progressively during Leo’s reign Arians and Arianism were out­ lawed. Unlike his predecessors, Leo was strongly hostile to Arians. He banned their churches and any congregation of Arians.204 That Patricius was an Alan or Goth, or the son of one, was never raised as a disqualification for the throne. Doctrinal allegiance was a far more significant obstacle. Aspar had challenged the doctrinal policy of the emperor and patriarch on at least one occasion,205 concern­ ing the banishment of Timothy ‘the Cat’ in 459, and is thought to be behind a law of 457 granting heretics a decent burial.206 He had also been prepared to argue with Leo in favour of the appointment of Arians. By 470 Leo was more firmly opposed to Arianism. Moreover, he had marshalled to his cause new and exclusive forms of spiritual authority in the shape of Daniel and the relics of Simeon the Stylite, as well as the Virgin’s robe in the chapel at Blachernae. If Aspar still wished his son to be Caesar then he would have no option but to declare himself orthodox. Now that the imperial succession was at least assured because Patricius was Caesar, there remained the final step of cementing the relationship with the family of Leo through marriage. Aspar had been planning for his son to marry a daughter of Leo’s since his coming to the throne in 457. As we have seen, there is evidence to suggest that Aspar’s will was continually thwarted not only by Leo but also public opinion, the clergy, monks and people of Constantinople.207 Thirteen years later, Aspar’s plan was to come to fruition. Patricius was to marry not Ariadne, espoused to Zeno since 468, but her younger sister Leontia who must have been no more than thirteen at the time. Now, Leontia is never actually named as the potential daughter-in-law of Aspar by any of the writers who mention this mar­ riage (vita Marcelli, Marcellinus, Jordanes, Malalas). Rather, it is inferred from the fact that since Ariadne was already married to Zeno by this time then it can only have been Leontia.208 Of Patricius as Caesar almost nothing is known. There are, for instance, no coins of Leo Augustus and Patricius Caesar which may well be a sign of reluctance on Leo’s part to permit this privilege.209 The only other

Malalas, Chron. 14.41 (Thurn 295); Chron.Pasch. 597 (Dindorf), with Wood (2011), 305–8. Theophanes, AM 5952 (de Boor 112). CJ 1.5.9, cf. Scott (1976), 68–9. This information is contained in the de insidiis extracts of Constantine Porphyrogenitus. and attributed to Malalas but may be from elsewhere. It is also possible that it does not so much refer to events over a long period of Leo’s reign, but just that episode relating to Patricius’ appointment to Caesar described by vita Marcelli and Zonaras. 208 The alternative – that Ariadne married Zeno in 466/7 but was divorced from him in order to marry Patricius in 470, then re-married to Zeno after Patricius’ demise in 471 – has its serious advocates beginning with Seeck (1920), 489 who persuaded himself that Leontia was simply too young to be anyone’s bride in 470, cf. Ensslin (1925), 1948; Lane Fox (1997), 192 but not Demandt (1970), 773–4. This is an unnecessary and unlikely scenario. 209 The claim of Grierson and Mays (1992), 162–3 that certain coins of Leo with an apparently redun­ dant ‘C’ in the reverse legend signify Patricius as Caesar cannot be sustained (cf. Kent [1994], 102, 109).

204 205 206 207

95

DYNASTY AND ETHNICITY

action of Patricius Caesar was a journey to Alexandria where he was received with all the pageantry accorded a Caesar.210 Meanwhile, Zeno and his family remained in Antioch. In journeying to Antioch, Zeno had been accompanied by a monk named Peter the Fuller with whose anti-Chalcedonian views he sympathised.211 After arriving there Peter began to identify and encourage the anti-Chalcedonians who would challenge the pro-Chalcedonian patriarch Martyrius. Eventually he invited monks into Antioch to promote his views more forcefully, especially his famous addition to the Trishagion, ‘who was crucified for us’, suggesting that God and not just Christ had suffered in the flesh. As Peter’s following grew, with Zeno’s blessing, the patriarch Martyrius sought the support of the emperor and travelled to Constan­ tinople where he was well received by both Leo and the patriarch Gennadius. In Martyrius’ absence Peter had himself consecrated patriarch of Antioch which led to the resignation of Martyrius on his return. He was alarmed at the civil unrest and violence and Zeno’s evident support for it.212 Shortly after, doubtless with fury against Peter, Leo invoked his exile from Antioch but he fled on receiving the news. On the same day that Peter was sent into exile, 1 June 471, the emperor sent a new law to Zeno forbidding monks to leave their monasteries and go to Antioch or elsewhere. Nor were they allowed to discuss doctrinal matters with the people, nor to incite to rebellion the ‘simple minds of the populace’.213 The emperor’s atti­ tude to Zeno who had condoned Peter’s behaviour is not stated but he is unlikely to have regarded him kindly over this development. From Antioch, in 470 or 471, Zeno was forced to confront an invasion of the Tzani into Roman Armenia,214 but we hear of no other military activity in the East at this time. Meanwhile, at Constantinople during the absence of Zeno, Aspar and Ardaburius were establishing a new stranglehold over Leo. They finally suc­ ceeded in pegging back the independence and advantage the emperor had previ­ ously acquired for the period 465–469.215 There was, for them, the additional advantage that not only was Zeno far away from court but so was Basiliscus. It was probably from his internal exile at Herakleia, as magister militum praesentalis, that he now led a new expedition against the Vandals which reached Sicily. An overland expedition under Heracleius and the Isaurian Marsus marched from Egypt to Tripoli to link up with Basiliscus’ forces. Before too long a settlement 210 Theophanes, AM 5961 (de Boor 116); Cedrenus 383.1 (Tartaglia 599). Theophanes probably obtained this story from his local Alexandrian document (Mango and Scott [1997], lxxviii–lxxx). The purpose of the journey is not specified. Schwartz (1934), 182 may be correct in suggesting that it was designed to show that the Vandal defeat had not diminished the empire’s strength. 211 Kosinski (2010), 72–5. 212 Theodore Anagnostes, HE Epit. 390–2 (109.19–110.16 Hansen); Cedrenus, 382.2 (Tartaglia 597–8); Nicephorus Callistus, HE 15.28 (PG 147 81–4) with Frend (1972), 166, 168, but claiming that Zeno was in Antioch as magister as early as 467. 213 CJ 1.3.29, with Scarcella (1997), 276–82. 214 John of Antioch, frag. 229.2 (Mariev 418–19). 215 Cf. Seeck (1896), 609.

96

DYNASTY AND ETHNICITY

was reached with the Vandal king, Gaiseric, probably in the summer of 471 with Roman forces close to Carthage.216 It had been precipitated by Leo’s desire to recall Basiliscus and his troops in order to strengthen his hand in dealing with the increased threat from Aspar and his son.217 The settlement with Gaiseric was proclaimed as a victory in both east and west. Basiliscus the magister militum praesentalis who was duly honoured in a statue set up for him in Philippopolis.218

The ‘butchery’ of 471 The creation of a Caesar for Leo was a vital and irrevocable step. For several months Patricius enjoyed his exalted position as the emperor’s son-in-law and designated heir. Doubtless his father Aspar was pleased to have fulfilled his cher­ ished goal. Yet, for both Aspar and Patricius, it came to a sudden and unexpected end the following year. The events leading up to Leo’s decision to finally despatch Aspar and his sons in a single action are very unclear. Speculation is unavoidable. Although absent from Constantinople, Basiliscus and Zeno, together or sepa­ rately, may have emboldened Leo to strike. They will each have had their own loyal allies at court and a common foe in Aspar. Beyond that, their interests may not have coincided and neither may have enjoyed Leo’s full support. Candidus, the writer closest to the event in question and quick to identify the fortunes of his fellow-Isaurians, explains how Ardaburius was discovered trying to lure the Isaurians over to his side which ultimately only aggravated the intense personal hostility between the emperor and Aspar’s family. It was sufficient to cause Leo to decide on murder.219 Aspar and Ardaburius were planning an usurpa­ tion, so John Malalas reports.220 Writing a little later in Constantinople, Procopius claims that Leo feared that Aspar and Ardaburius were plotting to kill him,221 obviously in order to expedite the promotion of Patricius to Augustus. The notion that Aspar was conspiring to kill Leo when neither Basiliscus nor Zeno was in Constantinople to defend the emperor, or foil the conspiracy, is further elaborated by Zonaras. Again, it needs to be recalled that Zonaras drew his information on

216 Following the account of Courtois (1955), 204. 217 Courtois (1955), 204. These events are clearly connected by both Theophanes, AM 5963 (de Boor 117), AM 5964 (de Boor 117), presumably originating with Priscus (cf. Blockley [1981], 76–7, 212 n.37), and Paul the Deacon, Hist. Rom. 15.2, possibly also originating in an eastern chronicle (cf. Goffart [1988], 352). The context is explained more fully in an important but unpublished paper by F. M. Clover (summary in Byzantine Studies Conference Abstracts 1976, 2–3 at www. byzconf.org/1976abstracts.html). I am grateful to the late Mike Clover for this and other assis­ tance with an early draft of this chapter. 218 PLRE 2, 212 (‘Basiliscus 1’), with Bersanetti [1943/4], 344–5). 219 Candidus, frag. 1 (Blockley, 466/7) = Photius, Bibl. 79 220 Malalas, Chron.14.40 (Thurn 294) with the additional sentence from the de insidiis extract (see Jeffreys et al. [1986] ~ Thurn 294); followed by the Chronicon Paschale, 296.17 (Dindorf): turanni,da meleth,santa. 221 Wars 3.6.27.

97

DYNASTY AND ETHNICITY

Leo’s reign from Priscus and others. ‘Some say’, so Zonaras reports, ‘that Leo feared Aspar and his sons would take advantage of his grandson’s age (that is, Leo II, then aged 2), so he killed them’.222 Earlier Zonaras had noted that, sometime after the installation of Patricius, Leo discovered that Aspar and Ardaburius were conspiring against his throne.223 Finally, there is Leo’s own admission in a letter to the western emperor Anthemius, intercepted en route, that: ‘I put to death Aspar and Ardaburius so that nobody should oppose my orders’.224 Leo’s motivation was intensely personal. He was not driven to action by ‘un mouvement germanophobe, entretenu sans doute par le cour’.225 Clearly, then, Leo felt threatened and desperate in the aftermath of the instal­ lation of his new Caesar. The initiative now lay once more with Aspar, not Leo, and the fear of a conspiracy to do away with the reigning emperor as a means of putting the succession beyond the slightest doubt is very probable.226 There may be a kernel of truth in the story told only by Nikephoros Kallistos Xanthopoulos: when Leo won a guarantee from Aspar and Ardaburius that they would renounce their heresy, and refrain from plotting against him, Ariadne (not Leontia) was married to Patricius. Since they did not hold back from plotting, one day in the hippodrome the people turned on them. Aspar and son fled along with their mili­ tary support to St Euphemia’s at Chalcedon. Leo sent the patriarch Gennadius to deal with them and they agreed to come out only if Leo guaranteed their safety. He did so and prepared to dine with them. However, Zeno was in the palace and decapitated them. This story is a blend of fact, fiction and speculation but it does raise the question of Zeno’s interlude in Chalcedon. For the months and weeks immediately prior to the assassination in 471 there is no information. All that is known is that Leo decided to take the initiative against his overlords. Given Aspar’s position at court, and the persistent plotting of Arda­ burius, Leo will have had to be extremely careful in devising their termination. Theophanes, again probably from Priscus, reports that Leo concluded peace with the Vandals in order to recall Basiliscus, Heracleius and Marsus whom he needed in the showdown with Aspar.227 It was perhaps now too that Zeno was recalled from Antioch. He was still there on 1 June when a law was addressed to him.228 At some stage thereafter Zeno travelled to Chalcedon presumably on the pretext of essential imperial business. At the time of the actual murder of Aspar and Arda­ burius, Zeno was waiting expectantly at Chalcedon while Basiliscus was also not far away at Herakleia.229

222 223 224 225 226 227 228 229

Zonaras, 14.29 (126.30–127.4 Büttner-Wobst). Zonaras, 14.8 (123.8–10 Büttner-Wobst). Malalas, Chron. 14.45 (Thurn 298). For the authenticity of this story: Bury (1886), 507–9. Stein (1959), 360. It is accepted, for instance, by Stein (1959), 361. Theophanes, AM 5963 (de Boor 117) and Lippold (1972), 157. CJ 1.3.29. v.Dan.Styl. 65 (Zeno).

98

DYNASTY AND ETHNICITY

The fullest account of these events is provided by John Malalas who writes as follows: During his reign he suspected that Aspar the patrician was planning a rebellion, being the leader of the senate, and so he put him to death inside the palace, together with his sons Ardabourios and [the Caesar] Patricius who were also senators, at a conventus and mutilated their bodies [throw­ ing them from the palace into panniers]. A riot began in Constantinople, for the victims had a band of Goths and comites and other followers, and a large number of supporters [whom he called foederati, from which are derived the annonae foederaticae]. Then a Goth who was one of Aspar’s associates, a comes named Ostrys, entered the palace with some other Goths, shooting with their bows. A battle broke out between the excubitores and Ostrys, and there were many casualties. He was surrounded and saw that he was beaten, so he fled, taking Aspar’s concubine, a beautiful [and rich] Gothic girl, who escaped with him on horseback to Thrace, where he plundered [many] estates. The Byzantines chanted an acclama­ tion about him, ‘the dead man has no friend – except Ostrys’.230 Malalas is here relying on a source or sources that can be traced to Leo’s own publicised account of events231 although the assassination of Aspar and sons was plainly well-planned. As part of the plan, Herminericus, the youngest son of Aspar, but in his 20s or 30s by 471, had been deliberately lured away to Chalcedon. Zeno could guarantee protection and a safe onward journey.232 Meanwhile, Aspar and Ardaburius were cut down inside the palace by eunuchs, that is, the cubicularii of the sacred bedchamber. Obviously Aspar and Ardaburius had no reason to sus­ pect what for them must have been a routine invitation. They were in and out of the palace all the time and had been for decades. On this occasion they were to attend a conventus or meeting of senators.233 The whole incident may have been just as much a surprise to Aspar’s friends and supporters too. The Arians Aspar and Ardaburius died there and then. The Caesar Patricius, who had publicly and necessarily forsaken his Arianism as a prelude to imperial promotion, a sure sign that public doctrinal allegiance mattered, was wounded but apparently escaped to safety although he is never heard of again.234 Perhaps this scenario, sparing the 230 Malalas, Chron 14.40 (Thurn 294–5), trs. Jeffreys et al. (1986), 204–5, but incorporating supple­ mentary material contained in the De insidiis version of Malalas (details in Jeffreys et al. [1986], 204–5 ~ Thurn 294–5). 231 Wood (2011), 298–314. 232 Theophanes, AM 5964 (de Boor 117–18) with Lippold (1972), 157, cf. Seeck (1920), 370; Stein (1959), 360. 233 Malalas, Chron 14.40 (Thurn 294), using the Latin term correctly as he normally does with such terms. On conventus: Jones (1964), 505. 234 Candidus, frag.1 (Blockley 467) = Photius, Bibl. 79 and Nic. Call., HE 15.27 say that he escaped. However, all other writers say he was killed. All that Photius says is that he survived and ‘lived

99

DYNASTY AND ETHNICITY

emperor’s Catholic son-in-law and heir, was also part of the plot. The immediate reaction in Constantinople was retribution. Aspar’s closest allies were a contin­ gent under a nearby Roman army leader, Ostrys. There was considerable unrest on the streets of Constantinople, as one might expect. Ostrys escaped on horseback through the long and crowded thoroughfare lead­ ing to the city walls. He headed into Thrace plundering ‘many estates’, probably the extensive imperial estates close to Constantinople. Ostrys was a comes rei militaris in Thrace in 467 fighting against the Huns for the Roman army and maybe still held the same position in 471. It is possible that he just happened to be in Constantinople at the time, or he might have been based there. He may also have been a relative of Aspar.235 Also roused were some of the Gothic foederati which must mean a contingent of the foederati under Aspar’s nephew, Theodoric Strabo, who had long been settled in Thrace.236 They were challenged not by Isaurians but by the troops of Armatus, magister militum for Thrace and nephew of the empress Verina and her brother Basiliscus, who had the Goths’ hands slashed off.237 In the final analysis, Leo does not seem to have been seriously threatened and the unrest was evidently short-lived. That is probably an indication that the troops and supporters of the emperor easily reclaimed the streets of Constantinople. If there was a large Gothic faction behind Aspar and his sons then it melted away at this crucial point. Ostrys was the only one prepared immediately to take up the cause of Aspar and breach the palace defences, if that is the point of the acclamation ‘the dead man has no friend – except Ostrys’. Leo was now desperate enough to support anything that promised liberation from the dominance of Aspar. However the operation was planned, it looks as if both Basiliscus and Zeno were part of it. Writing in the mid-550s, Jordanes certainly thought Zeno was implicated.238 Before long, both Basiliscus and Zeno reappeared. They had been placed in range and were soon in Constantinople once more. The vita of Daniel is very clear that Zeno did not return to Constantinople until after the deaths of Aspar and Arda­ burius, and this would appear to be confirmed by Theophanes.239 The aftermath of Leo’s ‘butchery’ was not necessarily smooth for the emperor but there was no concerted resistance from any ‘Gothic faction’. Theodoric Strabo

235 236 237 238 239

on’ (diesw,qh kai. die,zhsen). Perhaps he did live some time afterwards, say, months or a couple of years, but later died from the injuries sustained. If any store can be put on the words of Leo himself the following year in a letter to Ricimer, that is, ‘I put to death Aspar and Ardaburius’ (Malalas, Chron 14.45 (Thurn 298), then Patricius may still have been alive in 472. Wood (2011), 303 makes Ostrys the father of Theodoric Strabo which is erroneous. His father was Triarius, references in PLRE 2, 1073 (‘Theodoric Strabo 5’). Scharf (2001), 51–61. Malchus, Byzantiaka frag.9.4 (Blockley 416/7) = Suda A.3968 with PLRE 2.148, Armatus. Jordanes, Romana 338: ‘Zenonis generi sui instinctu’. v.Dan.Styl. 66; Theophanes A.M. 5964 (de Boor 117). Schwartz (1934), 184 considered that hatred of Zeno was ‘so intense and widespread at Constantinople that he did not re-enter the city until after his son Leo was made “co-emperor” in 473’. This may be so, in which case it will have occurred after he was made Caesar in 472 (for the date, Chapter 5 below [138–43]).

100

DYNASTY AND ETHNICITY

immediately claimed his uncle Aspar’s mantle but it was Leo who kept the upper hand. Immediately, he sent a court official (silentiarius) named Pelagius240 to The­ odoric and the Goths in Thrace. As reported by Malchus, it was all very formal and friendly: ‘The barbarians gladly received him and in turn sent envoys to the Emperor, wishing to be friends of the Roman’. There was no spontaneous Gothic revolt at Aspar’s death. Ostrys had fled to the Goths for safety, not to organise a revolt. Leo and the Goths commenced a normal diplomatic process with all its attendant courtesies and its protracted timetable. What is clear is that the process will have begun in the wake of Aspar’s death. Malchus explains why finalising the agreement with the Goths dragged on for eighteen months or so. Leo rejected the Goths’ initial proposals, whereupon Theodoric launched an attack and blockaded nearby Arcadiopolis and Philippi. Eventually the Goths realised that assuring their permanent food supply had to be their priority. So, by now in 473, they finalised an agreement with the Romans which included the appointment of Theodoric Strabo as magister militum praesentalis.241 Leo must have left the position vacant until he had secured the terms of Theodoric’s agreement. He was prepared to appoint Theodoric in 471. It was not Leo’s reluctance which delayed the appointment until 473. Being a close relative Theodoric had sought Aspar’s inheritance, land for his people in Thrace and Aspar’s post as magister militum praesentalis. Leo could not agree to the first two requests at the time but did accept immediately that Theodoric could replace Aspar as magister militum provided he agreed to become the emperor’s friend and not seek to undermine him. In other words, having done away with the power and proximity of Aspar, Leo was prepared to allow Aspar’s nephew, leader of the Gothic foederati, into exactly the same pow­ erful position at court. Such acquiescence highlights both the extent to which Leo’s antipathy to Aspar was personal, as implied by Candidus, and the fact that he was not seeking to eliminate Gothic influence or a ‘Gothic faction’.242 It was a personal power struggle and Leo had resorted to the desperate solution of eliminating his rival. His fears of a conspiracy may have been well-founded, but it seems that the sensitive issue of the doctrinal allegiance of Patricius, pos­ sibly Aspar too, was paramount. Several later writers sum up the ‘butchery’ as the elimination of dangerous Arian heretics rather than the removal of Goths or a Gothic threat.243 As for Aspar’s estate, Leo seems to have confiscated it for church building purposes.244

240 PLRE 2, 857 (‘Pelagius 2’), cf. 1058 (‘Telogius’), assuming with PLRE that Malchus meant ‘Pelagius’ for ‘Telogius’ (frag. 2 [Blockley 406–7] = Exc.de Leg.Gent. 2). 241 Malchus, frag. 2 (Blockley 406–7) = Exc.de Leg.Gent. 2, with Heather (1991), 264–71. 242 Candidus, frag.1 (Blockley 466–7) = Photius, Bibl., 79. 243 Marcellinus, Chron. s.a. 471 (MGH AA XI 90): ‘Aspar . . . Arrianus cum Arriana prole’; Cedrenus 379.1 (Tartaglia 593–4) w`j ~Arianou.j o;ntaj with Wood (2011), 305–8. 244 Anonymus Banduri (PG 122.1236C). In doing so he was possibly following the dictates of his own legislation (CJ 1.5.10).

101

DYNASTY AND ETHNICITY

The slaughter of Aspar and Ardaburius in 471 represented the culmination of a protracted struggle for succession which dominated and shaped imperial policy and behaviour for most of the time since Leo came to the throne, with Aspar’s con­ nivance, in February 457. Underlying Leo’s decision to liquidate two of the most powerful public figures at Constantinople was the fear that their growing influ­ ence threatened the secure transition from one reign to another within the family of Leo. In 471 Leo hoped that he would be succeeded by his infant grandson, if he lived long enough, otherwise the boy’s father, Zeno, then magister militum per Orientem in Antioch. Yet the cause of both young Leo and his father Zeno was currently bleak because the emperor already had an heir-apparent in the newly orthodox Patricius Caesar, son of Aspar. It was an appointment Leo had resisted unsuccessfully and doubtless resented. Growing increasingly infirm, he evidently feared that time was running out. Suspecting a conspiracy on the part of Aspar, probably his own demise and replacement as Augustus by Patricius, Leo was impelled to drastic pre-emptive action. The eclipse of Aspar paved the way for the return of Zeno, Ariadne and their son to Constantinople and the eventual elevation not of Zeno but of the emperor’s grandson as Leo II. It also brought Basiliscus back to court.245

Ethnicity: Goths and Isaurians Ethnic solidarity and a quest for ethnic supremacy have been taken to be the fac­ tors motivating personal and political action during the reign of Leo I, so that the emergence of Zeno in 465/6 has signified the deliberate creation of a new ethnic power-bloc led by Zeno (Isaurian) to balance that of Aspar (Gothic). Some responses to the original version of this chapter (2005) sought to restate the abso­ lute priority of the ethnicity of both Zeno and Aspar.246 Such an interpretation is belied, however, by a careful analysis of the key episodes considered here. Not only are ‘ethnicity’ and ‘identity’ modern concepts requiring their own interpre­ tation, but it is also instructive to heed what modern Roman historians have to say about the issue, things like ‘Greeks, Romans and even Jews imagined mul­ tiple lineages, intertwined ancestry and compound kinships. They did not, on the whole, trouble themselves about purity of the bloodline, and unlike moderns, they probably did not agonise much about ethnicity’.247 An underlying consideration, the structure and ethnic diversity of the Roman army at the time, also suggests that the factional interpretation of these years is misplaced. It is certainly risky to con­ sider ‘ethnic’ unit names as denoting a specific ethnic composition and identity, let alone loyalty.248 Isaurians were not foreigners or aliens but provincial Romans 245 Nicephorus Callistus, HE 15.27 (PG 147.80C) cf. Bury (1923a), 337: ‘he led a life of retirement at Herakleia on the Propontis, until he appeared on the scene of public life again after Leo’s death’. 246 Laniado (2015a), 49–53 and (2015b); Kaldellis (2018), 9–17. 247 Gruen (2013), 22, cf. Geary (2015), 5–15. 248 Dziurdzik (2017), 460–1.

102

DYNASTY AND ETHNICITY

in the military service of the empire, with their group identity a product of Roman stereotyping.249 They were scattered throughout the army but by the late fourth century the magister militum praesentalis at Constantinople had a specific unit of Isaurians.250 Moreover, being ‘Isaurian’ never constituted a particular linguistic or ethnic identity at all. Rather it was an indication of belonging to the Roman prov­ ince of Isauria or from the broader geographical region, an affiliation highlighted and then caricatured as a result of Zeno’s imperial rule.251 They actually called themselves ‘Cietae’ and spoke a Luwian language traceable back to Hittite times in the region.252 Isaurians had long had a reputation for being tough and separatist mountain fighters but, over time, being also Roman changed that. By the early fifth century we find an Isaurian advocatus at Constantinople,253 a grammaticus at Seleucia,254 then a distinguished Alexandrian sophist and mentor of Proclus.255 Zeno was part of the new educated and urban Isaurian provincial elite, which had become connected by marriage to the aristocracy of the imperial capital, just as Aspar’s family had done. By the 440s, when Aspar was already a senior Roman general, he had to deal with a powerful and well-connected Isaurian fellow-general named Zeno.256 Another Isaurian contemporary, Neon, was governor of Euphratesia whose administration drew high praise from a local bishop, Theodoret of Cyrrhus.257 So, the following decade, another Zeno who had married into the local aristocracy was simply fol­ lowing what had become a traditional career path for Isaurians, as was his brother Longinus who married a certain Valeria,258 and Athenodorus, senior senator (vir illustris) and son in law of a ‘patrician’.259 Isaurians were always keen for impe­ rial status because it considerably enhanced their prestige at home.260 It is no surprise to find Zeno among a unit of the emperor’s military staff, the protectores domestici, in 465. Nor is it surprising that he should be involved in military action as comes domesticorum in 467, as magister militum per Thraciam in 469 and as magister militum per Orientem in 469–71. Zeno did not have a large contingent of Isaurians, nor were the excubitores solely or mainly Isaurian. In fact, the only record of a contingent of Isaurians deliberately drafted into Constantinople would appear to be the emergency group put together by the general Zeno in 449 or so, 249 Wood (2009), 132. 250 Notitia Dignitatum, Or. 5.66, ed Seeck (1876). These were probably raised in the late fourth century by Theodosius I and were known as the ‘Felices Theodosiani Isauri’. 251 W. Burgess (1990), 109–121 and, for the impact of Zeno on the notion of an ‘Isaurian’, Elton (2000b), 293–307. 252 Shaw (1990) 261–3; Lenski (1999), 416–17; Kosinski (2010), 58–9; Crawford (2019), 18–19. 253 PLRE 2, 931 (‘Pylaemenes’). 254 PLRE 2, 68 (‘Alipius 2’). 255 PLRE 2, 666 (‘Leonas’). 256 Thompson (1946), 18–31. 257 PLRE 2, 774 (‘Neon 1’). 258 PLRE 2, 689 (‘Longinus 5’), with Crawford (2019), 31. 259 PLRE 2, 178 (‘Athenodorus 2). 260 Lenski (1999), 437.

103

DYNASTY AND ETHNICITY

to defend the city when its regular troops were elsewhere fighting against Attila’s Huns.261 Much later, Isaurian strong men who could afford it liked to assemble their own private guards.262 That ethnic solidarity always prevailed is patently untrue. There were occasions when Goths fought against Goths; for instance, Aspar and Ostrys against Goths in Thrace in 467, not to mention ‘Aspar’s men’ being under the control of a Hun, Chelcal. Likewise, Isaurians could be found in battle against Isaurians, as Zeno was against Indacus late in 469. In both cases it was a contingent of the emperor Leo’s army fighting against a particular enemy. On these occasions, ethnicity or provincial origin was plainly subsidiary to imperial service and policy. Nor do contemporary and later writers divide the historical actors into neat ethnic com­ partments. While Candidus ‘the Isaurian’ labels Zeno an Isaurian, Priscus (at least in the extant fragments) refers only to Zeno as Leo’s son-in-law, never as Isaurian. In the vita of Daniel he is referred to as Isaurian by birth but that is all. Theodore Lector, albeit only available in summary form, never mentions Zeno as Isaurian. Of later writers, the Chronicon Paschale never mentions Zeno during Leo’s reign but only introduces him subsequently as ‘Zeno the Isaurian’.263 Cedrenus makes no reference to Zeno being Isaurian, other than to introduce his origins,264 while in introducing his sole reign Zonaras says that ‘Zeno was characteristic of his nation’. In brief, the writers on whom we mainly depend for our knowledge of Zeno’s career in the reign of Leo do not indicate that his Isaurian background was the driving factor in his policy and behaviour, nor that his promotion and authority rested on his command of an exclusively Isaurian contingent. So too, the few incidents which have been taken as the signs of the ‘Isaurian faction’ asserting its weight have been significantly overrated. There is no testi­ mony during Zeno’s emergence at court, nor subsequently, that he led a large and threatening contingent of fellow-Isaurians. In fact, virtually all the known Isauri­ ans can be traced to three families.265 Not only is there scant evidence of an armed Isaurian faction in Leo’s reign, but hardly any high-ranking Isaurians are known from then. There is only the general Marsus who played a role in the campaign in Africa,266 and Illus who evidently held some office under Leo.267 Most of the snip­ pets of documentation cited to support the notion that Zeno initiated an Isaurian ascendancy at Constantinople from 465/6 apply to a later period. The role of Isau­ rians at the court of Zeno himself, as sole emperor from late 476, is quite another story altogether. In fact, much of the malice towards Zeno to be found in the later writers on whom we depend for our knowledge of the 460s is coloured by Zeno’s

261 262 263 264 265 266 267

Priscus, fr. 8 (Blockley 290–1). Lenski (1999), 451–3. Chron.Pasch., 599.12 (Dindorf). Cedrenus 384.1 (600 Tartaglia). W. Burgess (1990), 117. PLRE 2, 728–9 (‘Marsus 2’). Patria Constantinopoleos 3.33 (Berger [2013], 156–7).

104

DYNASTY AND ETHNICITY

later unpopularity as emperor. It is an attitude that lingered into the sixth century when it became clearer that ‘Isaurian behaviour was not determined by ethnicity, but by lack of pay, denial of benefits (real or hoped for) and political ambition’.268 Similarly, the ‘Gothic faction’ in the reign of Leo also dissolves on close inspec­ tion. Apart from the problematic nature of establishing precisely what a ‘Gothic identity’ entailed at this period,269 there are reports of Gothic contingents in action at different points, as there had been in the past. Most weight, however, is given to Malalas’ account of the events surrounding Aspar’s murder. What Malalas says is that Aspar had a military contingent of Goths (Gotqikh.n cei/ra), presumably one of the many such units in the Roman army and one of the many under his command as magister militum praesentalis. They were not a private army or a Gothic war-band maintained by Aspar to enforce the will of the ‘Gothic faction’. Damascius writing in the early sixth century about the cloud-seer Anthusa says that Aspar was a Gothic leader (to.n h‛gemo,na tw/n Go,tqwn).270 Damascius may not have troubled himself too much over such a description which is a stylised one required to interpret a cloud shaped like the silhouette of a Goth. Just as for Isau­ rian soldiers, the magister militum Aspar was bound to find himself occasionally with a unit of Goths or others. He had a range of such units under his regular com­ mand271 but so did Zeno. At some stage Aspar would have commanded Isaurians and Zeno would have commanded Goths but they were all Roman soldiers,272 and they were Roman generals. In fact, a few years later Zeno was commanding the forces of Herminericus, Aspar’s only surviving son.273 When news emerged from the palace of the slaying of Aspar and Ardaburius in June 471, Ostrys, so it would appear, had under his command a contingent of Theodoric’s foederati Goths plus many others. Again, this is insufficient to support the notion of a final showdown between two factions. It was a struggle between generals, as well as between individual generals and their emperor.274 The subsequent Gothic uproar in the city was quickly supressed.

Dynasty: emperors and contenders Leo I had owed his throne to the dominant position of Aspar. At his installation in 457 Aspar was the senior court general and his son Ardaburius held military authority throughout the eastern provinces. Their family and associates contin­ ued to enjoy the highest offices and honours for several years but gradually Leo 268 Elton (2000b), 301. 269 There is voluminous modern literature on the highly contested and ever-changing issue of Gothic identity, best approached through Gillett (2002); Pohl (2014). 270 Damascius, in Photius, Bibl., 242.69. 271 Scott (1976), 61 n.10. 272 For example, one of the palatine auxiliary units (the Felices Theodosiani Isauri) was under the command of the magister militum praesentalis (Notitia Dignitatum. Or. 5.66 [Seeck 14]). 273 John of Antioch, 237.4 (Mariev 432). 274 Kaegi (1981), 32.

105

DYNASTY AND ETHNICITY

began to assert his own independent authority as emperor. Open conflict with Aspar occurred over certain appointments to high office (the consulship for 466, the City Prefecture at some stage), over how to deal with Rome’s enemies (not to support the Sciri against the Goths in 466; not to mount a campaign against the Vandals), and over matters of religious policy. The most serious conflict, however, running right from 457 to 470 was over the promotion of Aspar’s son as Leo’s heir-designate. Leo had an advance understanding with Aspar that as emperor he would seal their relationship by marrying his daughter Ariadne to Aspar’s son Patricius, then would make Patricius his Caesar and heir. Before long, Leo obvi­ ously preferred to retain the prized imperial succession for his own family. The birth of his own son in 463 proved a short-lived solution, while the discredit­ ing of Aspar’s eldest son Ardaburius in 465 bought Leo more time. Gradually, as Ariadne grew closer to marriageable age, Aspar became more impatient. Her betrothal in 468 to Ardaburius’ nemesis, Zeno, combined with the appointment of the emperor’s brother-in-law Basiliscus as magister militum praesentalis in 467 must have threatened Aspar. He now had not only one of Leo’s family as an equal, at least in terms of military rank and authority, but had also lost the agreed opportunity for a marriage alliance with the emperor’s family. Aspar’s eventual response to both these impediments to his plans and expecta­ tions, either by careful planning on his part or accidents of circumstance which he shrewdly exploited, was to have his dynastic rivals banished. By mid-469 neither Basiliscus nor Zeno had direct access to the emperor and his court, the former operating as magister militum praesentalis from Herakleia in Thrace following the failure of the Vandal expedition, the latter being forced into fleeing to Antioch where he was magister militum per Orientem. In both cases it is Aspar and Arda­ burius who stand accused of engineering the events which led to the successive exiles of the senior generals in 469. The power vacuum created by the absence of Basiliscus and Zeno presented Aspar with the opportunity to reassert his authority and his control over the succession. So the isolated emperor Leo had no option but to agree in 470 to both the elevation of Patricius as Caesar and imperial heir, and to his marriage with the emperor’s daughter Leontia. The only concession extracted, and in response to intense public pressure at that, was the conversion of Patricius from Arianism to orthodoxy. Aspar’s interests had finally been secured. The struggle between Leo and Aspar which led to Aspar’s eclipse was not so much about competing ethnic groups or factions, presupposing ethnic identity as the main driver of personal behaviour and action. Rather it was more about the primal dynastic impulse of retaining power and securing succession, a tradi­ tional concern of imperial families and their rivals in every era.275 The struggle of the period from 457 to 471 abounds with the timeless motivating factors of selfinterest and self-preservation, not that of larger factional interests or policies on behalf of a particular ethnic or regional identity. Religion was a motivator, a binder

275 Cf. Geary (2015), 9–10.

106

DYNASTY AND ETHNICITY

and excluder. The emperor Leo’s strategy was to cultivate sources of authority, spiritual and political, from which Aspar was excluded and to promote his wife’s brother and other associates. Leo slowly built his own power-base but it was not exclusively dependent on ‘hardy Isaurian mountaineers’. For fourteen years Aspar and Leo manoeuvred persistently, but separately, towards the goal of securing and maintaining power and influence. In the end, it was a struggle of individuals, both for themselves and on behalf of their family’s interests. Only with the assassina­ tion of Aspar and Ardaburius in the latter half of 471 was a resolution achieved. Having used the power of his position to appoint both the current (457) and previ­ ous emperor (450), Aspar was denied appointing the next one. The house of Leo was now assured.

107

4 L E O I A N D T H E PA L A C E G U A R D *

The emperor Leo I (457–74) is usually credited with establishing the excubitores, as an imperial guard of elite soldiers who were predominantly Isaurian. His motive is considered to be the creation of a new base for challenging the overwhelming power of the general Aspar and his associates. This chapter proposes that the excu­ bitores were actually founded by the emperor Tiberius in the first century, and that Leo simply refocused their task as a guard for the imperial palace buildings, gave them a new command structure and confined their number to 300. Their Isaurian connection has been much exaggerated. The crucial testimony of John Lydus has tended to be ignored and references to excubitores before the time of Leo have been misconstrued as no more than general descriptions of guards, rather than a particular unit with defined functions. The imperial palace at Constantinople was a sprawling terraced complex of gardens, courtyards, and monumental buildings enveloped by the sea walls of the city and other protected precincts including the hippodrome. It was built and rebuilt over several centuries and occupied more or less continuously from 380 to 1204. The set of buildings constituting the imperial palace gradually expanded, its staff increased into the hundreds or possibly thousands, its ceremonial solidified, its security improved and its vista was enhanced. Each new emperor sought to leave his mark by new construction or distinctive decoration.1 The original palace built by Constantine was occupied at different points by Constantius II, Julian and Valens. With the advent of Theodosius I emperors came to spend consider­ able time there.2 It was expanded in the time of Theodosius II and extensively renovated and rebuilt by Justinian. From 395 to 565 the emperor hardly ever left the capital so the palace developed a permanent ritual, organisation and staff. As * This chapter originally appeared under the same title in Byzantion 75 (2005), 117–51 and is reprinted here with the permission of Peeters Publishers (Leuven). Repeated here are my thanks to Peter Bren­ nan for expert assistance with such an obtuse topic. 1 For orientation: Janin (1964), 106–21; Müller-Wiener (1977), 229–37; McCormick (2000), 136–42. Earlier, Bury (1912), 210–25. On the palace’s place in the public life of an imperial city: Her­ rin (1991), 213–30. On the development of Constantinople as an imperial capital: Dagron (1974); Mango (1986b) and (1990); Magdalino (2007). 2 Chapter 1, 25–61.

108

L E O I A N D T H E PA L A C E G U A R D

the palace grew in size, complexity and importance the task of guarding it became more intricate and extensive. Exactly who was responsible at any one time for the permanent guarding of the palace entrances, as well as the passageways connect­ ing the various buildings is not clear. The guard in the ninth-century Kletorologion of Philotheos and the tenth-century Book of Ceremonies for Constantine Porphy­ rogenitus is fairly well understood,3 but its evolution in the earlier centuries is less certain. Protecting the ‘sacred presence’ of the emperor and his household in the ‘sacred palace’ was the duty of a multiplicity of officers of varying rank and with various roles. Some guards were for show and decoration, promoting an aura of pomp and ceremony. Others were for effective and swift protection against any personal threat or other violence. The costumed colour of the grand palatial procession involved different units of guardsmen. Their appearance in the imperial mosaics at San Vitale in Ravenna, for instance, provides a conspicuous illustration of the guards’ ceremonial role. There was also the need for a physical guard twentyfour hours a day to ensure that no unauthorised intruders found their way into the sacred presence. The palace complex could be entered in several ways, formally and informally: through the main ceremonial entrance (the Chalke), through the kathisma of the hippodrome and down the spiral staircase, through different gates of the outer sea wall and other smaller entrances. Then there was need for protec­ tion of the sacred person of the emperor, wherever he was, and his family. Obvi­ ously the guards closest to the imperial presence needed to be the most trusted, the most experienced, and doubtless the best paid. Any foray into the organisation of the Roman palace guard from Augustus to Maurice is potentially hazardous, not least for the period from the 4th to the 6th century. A. H. M. Jones found part of it ‘an obscure and tangled problem’4 while Alan Cameron once noted that ‘few students of the Later Roman Empire could claim, if they were honest, that they fully understood all the nice distinctions between the various corps of palace guards’5 and Fergus Millar proposed that ‘it is best to admit, however, that we cannot in general do more than state that the impe­ rial presence always had a military aspect; the precise functions of the different units which accompanied him cannot be clearly stated’.6 Such blunt uncertainty arises because there are major gaps in the historical record, bridged by hypotheses of varying strength and durability.7 This chapter aims to advance a working hypothesis for one of those gaps, namely the emergence of the excubitores at the time of Leo I (457–74), in the hope that future research might clarify the issue more fully. It is focussed on the 3 4 5 6 7

Bury (1911); Haldon (1984). Jones (1964), 636. Cameron (1972), 136. Millar (1977), 63. A problem recognised long ago by Grosse (1920), 62 but highly illuminated by Bolognesi RecchiFranceschini (2008), 231–57.

109

L E O I A N D T H E PA L A C E G U A R D

key contextual elements of Leo’s approach to his palace guard, but is necessarily speculative in part. Recent excavations have rekindled interest in the imperial pal­ ace at Constantinople,8 particularly in the mosaics uncovered so far.9 They have also heralded the possibility of a more detailed knowledge of the various stages of construction, as well as the layout of the palace and its various buildings including the location of entrances and exits into, and around, the palace complex.10 New insights may eventually emerge on the excubitores and Leo I’s precise role in their development. Meanwhile, it is proposed here that Leo’s palace guard initia­ tive was both more targeted and more modest than has been assumed. The crucial testimony of John Lydus has tended to be ignored while references to excubitores before the time of Leo have been misconstrued.

John Lydus on the excubitores The emperor Leo I is normally credited with a significant and exclusive initia­ tive concerning the guarding of the imperial palace at Constantinople, namely the establishment of the so-called excubitores. Leo is held to have invented and named the excubitores as a palace guard de novo. Moreover, their invention is linked to a deliberate attempt to secure the support of an Isaurian imperial official named Zeno in a struggle with the powerful and dominant general Aspar to whom Leo owed the throne. The excubitores are therefore taken to be comprised entirely or mainly of Isaurians favoured and promoted by Zeno.11 While it is clear that there is nothing to support the notion of the excubitores being exclusively Isaurian, there is also less certainty than normally assumed about when and why they were established, their precise role, as well as their relationship to other imperial guard contingents. As John Haldon has observed, ‘evidence for the establishment of the excubitors [by Leo] is also lacking’.12 Attributing to Leo responsibility for establishing a corps of palace guards and labelling them excubitores depends on the solitary testimony of John the Lyd­ ian writing in the mid-sixth century. John knew the palace well. As an experi­ enced bureaucrat in the office of the Praetorian Prefect, he had frequently been in the sacred imperial presence. On one occasion in the early 530s he delivered a panegyric before the emperor Justinian and a Roman senatorial delegation and 8 On the various earlier excavations: Bardill (1999b), 216–30. See also Mango (1991), (1995) and (1997); Miranda (1983). The most recent discoveries (late 1990s) are noted in Dark and Özgumüs (2001); Bolognesi (2000). 9 Trilling (1989); Jobst and Vetters (1992); Jobst and Gurtner (1997). 10 Bolognesi Recchi-Franceschini and Featherstone (2002). 11 This general thesis can be traced from Mommsen (1889), 224–5 to (most recently) Lee (2000), 45–9. Specific versions of it are represented by Ensslin (1925) and (1958); Treadgold (1997), 14; Averil Cameron (1993b), 30; Stein (1949), 358; Kaegi (1981), 27; Demandt (1989), 187; Williams and Friell (1999), 177. 12 Haldon (1984), 137.

110

L E O I A N D T H E PA L A C E G U A R D

doubtless in Latin.13 John was certainly familiar with the palace’s main entrances and exits. So his contemporary testimony requires serious consideration. John offers two statements involving mention of the excubitores: one explaining their dress, the other explaining their limited number. 1. On the military dress of kings The earliest chapters of the de magistratibus (1.1–13) cover the period from Aeneas to the Roman republic, that is, the earliest institutions of Romulus and the kings. John justifies starting with Aeneas (1.1), then moves to Romulus who is labelled a ‘tyrant’, having explained the distinction between a king, a tyrant and an emperor (1.3–6). There follows a discussion of the various emblems and apparatus of the kingship (1.7–8) before essentially explaining the origin of cer­ tain contemporary terms or labels which can be dated from that period. Among these (1.9–12) are some military terms, in particular the difference between the various types of soldiers and their distinctive shields (scuta, clipea and parmae). Following this, John describes the original military dress as worn by Aeneas: bra­ zen helmet, ringed breastplate, broad-sword on the left thigh, two javelins at right, woven black greaves, shoes called garbola/crepidae. He then goes on to note that over time Roman military dress changed considerably, although there are still some soldiers dressed in the manner of Aeneas. These are the excubitores which Tiberius formed as palace guards. John Lydus, On the Magistrates 1.12.614 Tw/n de. nῦn stratiwtw/n barba,rouj zhlwsa,ntwn( evkei,nwn de. auvtou,j para, mo,noij tou/ palate,ou fu,laxin $le,gontai de. para. ~Rwmai,oij evkskoubi,torej( oi=oneiv fu,lakej a;gripnoi(ou]j prw/toj meta. ~Rwmu,lon Tiberi,oj Kai,sar ‛ j e;fhn( evx Ai`nei,ou e,xeu/ren% to. toiou/t/ o parape,meine sch/ma avpo. ~Rwmu,lou w th.n avrch.n e;con) ‘But since the present-day soldiers, however, have emulated the bar­ barians, and the latter the former, such form of dress as had its origin from Romulus, as I said, derived from Aeneas, remained only with the guards of the palatium (they are called by the Romans excubitores, that is to say, ‘vigilant guard’, whom Tiberius Caesar was the first after Romu­ lus to have devised)’. John is here explaining that the soldiers of his day (550s at time of writing) have taken to dressing in barbarian fashion while barbarians now dress like Romans. There is only one group, however, who still wear a form of the traditional martial 13 On the Magistrates, 3.28. The occasion is not specified but it was possibly during negotiations which formed part of the prelude to Justinian’s war against the Goths. Given the audience, an ora­ tion in John’s native Greek would appear unlikely. 14 On the Magistrates, 1.12.

111

L E O I A N D T H E PA L A C E G U A R D

dress dating back to the time of Aeneas. They are the excubitores. Romulus first utilised the dress now worn by the excubitores, although it was Tiberius who first founded the excubitores and dressed them accordingly. These excubitores were palace guards otherwise known as ‘vigilant guards’, that is, ‘night guards’. John may here be implying some connection between the excubitores and the vigiles of the time of Tiberius. John’s point about ancient Roman dress surviving among the excubitores is reinforced by the summary of this chapter given in the list of chap­ ters which says plainly ‘that the Roman army received from Aeneas the method of equipping themselves in such manner as the so-called excubitores still do even to this day’ (o[ti avpo. Aivnei,ou pare,laben o‛ ~Rwmaiko,j strato.j ou[tw ste,llesqai* w/j e;ti kai. nu/n oi] kaloume,noi evxkoubi,torej).15 John is not saying here that Leo founded the excubitores at all or was responsible for their title. In fact, Leo is never even mentioned. The establishment of the excubitores is clearly attributed to Tiberius. 2. On the patricians and their number Following his discussion of the excubitores and their traditional dress John moves to treat in clear and numbered sequence each of the major institutions of Roman government down to the reign of Titus (cf. 1.49). He begins with the cavalry commander (magister equitum) and its replacement by the praetorian pre­ fect (1.14–15), then comes the patricians (1.16–23) and the quaestors as regal offices (1.24–9), followed by the consuls (1.30–3), the decemviri (1.34–5), dicta­ tors (1.36–8), censors (1.39–42), tribunes (1.44–7), praetors (1.48) and finally an explanation of why he stopped at Titus. Next comes a section on the night watchmen (vigiles, 1.50) and a conclusion noting that all these magistracies and functions were absorbed into the sole authority of the emperor from the time of Augustus (1.51). In the course of explaining the second of the magistracies or powers, the patri­ cians, John again mentions the excubitores. The context for this reference is a discussion of how Romulus established the patricians and why they were set at 300. This numerical pattern became traditional in the organisation of Roman insti­ tutions including the legionary reorganisation of Marius. John then specifically links this traditional practice of having units of 300 to Leo’s involvement with the excubitores. John Lydus, On the Magistrates 1.16.316 kai. Le,wn de. o‛ basileu.j prῶtoj tou.j legome,nouj e‛ kskoubi,twraj tw/n parexo,dwn tou/ palati,ou fu,lakasj prosthsa,menoj triakosi,ouj mo,nouj e‛ stra,teuse kata. th.n avrcaio,thta

15 Keph. 5 (trs. Bandy [1983], 5). 16 On the magistrates, 1.16.3 (trs. Bandy [1983], 31).

112

L E O I A N D T H E PA L A C E G U A R D

‘and even Emperor Leo, who was the first to establish the so-called excu­ bitores as guards of the side-exits of the palatium, put into service only three hundred to accord with ancient custom’ What John says is not so much that Leo established the excubitores outright but that he first set them up (prosthsa,menoj) as guards to the side-entrances of the palace (tw/n parexo,dwn tou/ palati,ou fu,lakaj). In other words, here is a specific designated function for them which Leo was first to formulate or codify. The very point of John’s explanation here, however, is their number. Leo restricted them to 300, explains John, and there was a strong traditional reason for that number, namely the example of Romulus and later. Since John is the sole testimony linking Leo and the excubitores the notion that the fifth-century emperor actually founded them de novo depends on inter­ preting John in exactly that fashion, as most scholars have done. Further, to link directly both passages of John the Lydian, on the interpretation that Leo actually founded the excubitores sometime after he came to the throne in 457, implies that the emperor deliberately identified and resurrected an archaic republican form of dress for his new guard, just as confining their number to 300 was a conscious classicising and archaising decision.17 Yet, John suggests that the excubitores’ uni­ form was not the product of revival but of survival, that is to say, it had originated in the days of Aeneas, was used by Romulus and by the mid-sixth century had all but disappeared except for the 300 excubitores. Whether their uniform was a deliberate revival of an obsolete form or a living relic of long tradition, the fact remains that Leo is credited with inventing the excubitores as a new unit with the specific function of guarding the palace entrances. Yet, if Leo did in fact create them as a new guard, it has to be asked who had previously been responsible for guarding the imperial palace at Constantinople from its foundation in 330 up to the 460s. In modern accounts John’s explicit attribution of the excubitores as a palace guard to Tiberius is simply overlooked. From John the Lydian’s passages mentioning the excubitores it would appear that two main conclusions emerge: (1) Tiberius actually founded the excubitores as an imperial palace guard and with a uniform which could be traced back to Aeneas and Romulus; and (2) Leo’s initiative involved defining their responsi­ bilities as guarding the side-entrances of the palace, and reorganising the unit as 300 picked men, a number deliberately reminiscent of Romulus and Marius. Assuming, for the moment, that John accurately reflects the development of the excubitores with whom he and his audience were personally familiar, it is neces­ sary to (1) identify and explore testimony to the excubitores between Tiberius and Leo, then (2) examine Leo’s involvement with them more closely since he cannot have simply introduced them for the first time.

17 Cf. Maas (1992), 45.

113

L E O I A N D T H E PA L A C E G U A R D

Excubitores from Tiberius to Leo Whatever Leo did he was not responsible for inventing a palace guard contingent designated excubitores. They had long existed, for centuries in fact, by the time of his accession in 457. Exactly how they had developed, and how they functioned and were organised by the mid-fifth century, is difficult to assess.18 Many of the scattered literary references to them are problematic and ambiguous. Even so, it would appear that a guard contingent known as excubitores had already been formed as early as the reign of Tiberius, just as stated by John the Lydian. Earlier references than Leo have never been comprehensively collected and analysed. Rather they have tended to be dismissed as merely generalised descriptions of guards or sentries without denoting any particular unit or function.19 A reconsid­ eration of the contemporary references to excubitores from the first century to the fifth suggests that, from the time of Tiberius as implied by John the Lydian, they may well have been used in the technical sense of guards with a specific imperial role. It is, at least, an hypothesis worth pursuing. In his first-hand account of the siege of Alesia in 52 BC Julius Caesar describes the final state of Roman preparations. Having noted that the fortified Roman camp included 23 towers with sentries stationed between them to deal with any sudden attack, he goes on to say that they were guarded in the night by excubitores and a strong garrison.20 These excubitores appear to have been the night guards of the fort itself rather than Caesar’s personal bodyguard which were the praetorians now mainly asleep. He also surrounded himself with a contingent of hand-picked Germans. It was these personal forces of republican generals which in 27 BC Augustus turned into a permanent force for the new princeps stationing them in and beyond Rome.21 The praetorian guard was soon established as a separate corps with its own command structure and barracks at Rome, originally for the three praetorian cohorts stationed there. The German bodyguard of horsemen was turned into the emperor’s own personal bodyguard, the Germani corporis Augusti custodes otherwise known as ‘Batavians’22 who now accompanied the emperor wherever he moved.23 The establishment of a permanent household or palace at Rome for the princeps also created the need for a palace guard. A praetorian cohort with its tribune was

18 References in Fiebiger (1909), 1577; Bury (1911), 57; Jones (1964), 318. 19 E.g. Alan Cameron (1972), 137 and (1976a), 182 n.8; but not the Thesaurus Linguae Latinae vol., 5.2, col. 1288 (‘excubitor’) which ascribes a specialist meaning to excubitor as used by Suetonius, Lactantius and Ammianus. 20 De Bello Gallico 7.69.7: ‘Castra opportunis locis erant posita ibique castella viginti tria facta, quibus in castellis interdiu stationes ponebantur, ne qua subito eruptio fieret: haec eadem noctu excubitoribus ac firmis praesidiis tenebantur’. 21 Keppie (1996). 22 Durry (1938), 22–3; Bellen (1981); Speidel (1994), 15–18. 23 Tacitus, Annales 15.55; Suetonius, Tiberius 60.2.

114

L E O I A N D T H E PA L A C E G U A R D

permanently on guard for the emperor in the palace,24 but only the tribune was permitted to enter the imperial bedchamber.25 Each of the three cohorts took turns for eight hours of palatine guard duty26 and they wore a toga rather than regular military dress.27 Precisely how many soldiers were deployed in guarding the pal­ ace is not known except that part of this cohort had a special function and separate designation, namely the speculatores. These troops numbering 300 (as the excubitores were under Leo) were essentially praetorians promoted to special duties.28 One of their number killed Otho.29 Given the proximity of the praetorians to the emperor it was inevitable that their prefect would become a powerful figure at court. Sejanus was evidently the first to capitalise on this power from around AD 17. Thereafter, despite occasional variations in size and distribution they remained influential especially in the transition from one emperor to the next.30 It was their tribune who received the secret daily watchword, or password.31 One of the components of the emperors’ palace guard was evidently the excubitores. It is not clear when and how the excubitores were formed although John the Lydian’s ascription to Tiberius is perfectly feasible. Tiberius was certainly responsible for creating at Rome a permanent camp for the praetorian cohorts stationed in the city,32 for forming a dedicated palace guard from the praetorians,33 and for constructing a new imperial palace (domus Tiberiana) on the north-west half of the Palatine.34 Not long after, in the reign of Claudius, is found a ‘tribunus excubitor’, which may well denote the existence of a unit known technically as excubitores whether founded by Tiberius or not.35 The excubitores, possibly a 24 Tacitus, Annales 12.69, Historiae 1.29; Suetonius, Otho 6, with Turcan (1987), 76–80; Winterling (1999), 119fff. 25 As explained by the Prefect Plautianus to the tribune Saturninus in devising a plot against Severus in 205 (Herodian, 3.11.6). The guards of the imperial bedchamber (3.12.1) were distinct from the emperor’s personal bodyguard (swmatofu,lakej) which was also close at hand (3.12.9). On one occasion in Augustus’ palace an Illyrian was apprehended near the emperor’s bedroom having avoided the notice of those tending the entrance (‘ianitores’: Suetonius, Augustus 19), while Nero panicked on waking up at midnight and discovering that his guard (‘statio militum’) had deserted him (Suetonius, Nero 47.1). 26 Durry (1938), 275, deduced from Martial 10.48.1–2. There was evidently a regular pattern in the guard rostering as Otho discovered in 69 when planning to ambush Galba at dinner in the palace. He had to abandon his plans when he discovered that the cohort was the same one as the one which had been on duty when both Caligula and Nero had been killed in 41 and 68 respectively (Sueto­ nius, Otho 6, cf. Tacitus, Historiae 1.29). 27 Tacitus, Historiae 1.38; Martial 6.76. 28 Durry (1938), 108–9; Speidel (1994), 33–4. 29 Tacitus, Historiae 1.35. 30 Kennedy (1978), 275–301 (including discussion on rates of replacement). 31 Scriptores Historiae Augustae, Ant.12.6; =Marc. 7.3; Pertinax 5.7, 6.3; Severus Alex. 23.3. 32 Tacitus, Annales 4.2; Suetonius, Tiberius 37; Dio 57.19.6. 33 Suetonius, Tiberius 24 34 Turcan (1987), 25–7. 35 Suetonius, Claudius 42: ‘quotiens quidem hostem uel insidiatorem ultus esset, excubitori tribuno signum de more poscenti non temere aliud dedit . . .’

115

L E O I A N D T H E PA L A C E G U A R D

whole separate cohort, and their tribune are also in evidence subsequently in the reign of Nero.36 In fact, following the death of Claudius in 54 young Nero had presented himself to the guards (excubitores) who endorsed his election and car­ ried him to the praetorian camp between the 6th and 7th hour.37 When the emperor went to an entertainment at the house of another person, the excubitores appear to have accompanied him, and to have kept guard as in his own palace.38 So it was remarked that Trajan once dined with Sura after dismissing his guard, presumably for privacy.39 Other references to the excubitores, especially in Greek texts, are harder to pinpoint. Cassius Dio, for example, appears to denote the emperor’s personal German guard as simply ‘bodyguards’ (swmatofu,lakej)40 with the prae­ torians being designated dorufo,roi41 It is possible that excubitores was the name given to the single praetorian cohort which guarded the palace at night, just as they did for Julius Caesar at Alesia although at Nero’s accession they were clearly on duty during the day as well.42 John the Lydian implies that the excubitores were related to the vigiles.43 After the disastrous fire at Rome in AD 6 Augustus sought to improve the safety and security of the city, including the control of fires, by an administrative reorgan­ isation. The city was divided into 14 separate regions and at the same time he established a new force called vigiles, a unit of 7000 freedman divided into 7 separate cohorts of 1000 men each, under the command of a praefectus vigilum. Each cohort was entrusted with responsibility for two regions and they were based in a statio vigilum in one of the two regions. Part of each cohort was dedicated to 36 Tacitus, Annales 12.69: ‘ comitante Burro Nero egreditur ad cohortem, quae more militiae excu­ biis adest; Suetonius, Nero 9 : ‘primo etiam imperii die signum excubanti tribuno dedit “optimam matrem” . . . cf. Otho 4: ‘aureos excubanti cohorti viritim dividebat . . .’; TaCITuS, Historiae 1.24: ‘cohorti excubias agenti’. 37 Suetonius, Nero 8: ‘Septemdecim natus annos, ut de Claudio palam factum est, inter horam sextam septimamque processit ad excubitores, cum ob totius diei diritatem non aliud auspicandi tempus accommodatius uideretur; proque Palati gradibus imperator consalutatus lectica in castra et inde raptim appellatis militibus in curiam delatus est discessitque iam uesperi, ex immensis, quibus cumulabatur, honoribus tantum patris patriae nomine recusato propter aetatem’. 38 Suetonius, Otho 4. 39 Dio 68.15.5. 40 Dio 58.9.2, 61.9.1, 64.17.2 (cf. Suetonius, Gaius 58); 63.27.3; 64.17.2; 74.9.2–4; 76.14.2. In the late sixth century they were equated with the excubitores of the time by John of Epiphaneia (fr. 5 [FHG 4, 275]: ko,mhta de. tou/ton [Tiberius, in this case] evxkoubito,rwn ~Rwmai/oi kalou/sin), and likewise in the early seventh century by Theophylact Simocatta (Historiae 3.11.4, cf. 7.15.7: h‛ gemw.n d ou-toj evtu,gcanen ὢn tw/n tou/ basile,wj swmatofula,kwn kai. u‛ paspistw/n ὂn ko,mhta evxkoubito,rwn th/| sunh,qei fwnh/| e;qoj ~Rwmai,oij kalei/n). Dio may have had a similar connection in mind too. 41 Dio 58.9.5, 61.8.5, 62.14.3, 62.24.1 [tribune], 63.27.3 (cf. John of Antioch, fr.91 [FHG 4, p. 576]). He also notes the palace’s doormen (qurwroi,), presumably ianitores: Dio 77.4.2 [205]; There was a minimum guard at the palace when the emperor was absent for long stretches, such as Domitian who preferred to reside in the Gardens of Sallust (Dio 65.10.5). 42 Cf. Grosse (1920), 270. 43 John Lydus, On the Magistrates 1.12.

116

L E O I A N D T H E PA L A C E G U A R D

guard duty and divided between both regions. In each region was a guardhouse (excubitorium) where they were stationed.44 The excubitorium of the seventh cohort was discovered at Trastevere in the 1860s and is a clear testimony to the working lives of the guardsmen. It was possibly this sub-group of the vigiles which was known as the excubitores, that is, those responsible for guarding the build­ ings within a particular city region. By extension, for the excubitores in Region 10 the main area and building requiring protection was the domus Augustana et Tiberiana, the imperial palace. Greater certainty and precision appear impossible. The excubitores remained in their role of palace guard. By the time of Domi­ tian/Trajan, however, the palace had been considerably enlarged and modified, and a further imperial guard corps had been formed – the equites singulares Augusti commanded by the praetorian prefect.45 Like the Germani custodes they replaced, they were a mounted bodyguard for the emperor rather than a palace guard and comprised 500 troops, later 1000, chiefly Pannonians and Germans. Their camp was near the Lateran. The equites remained active until about the midthird century before being disbanded. By then, possibly replacing the equites, or a combination of equites and praetorians, there emerged a new group of guards known as the protectores. These protectores were assigned to particular officers with those assigned to the palace called protectores domestici or domestici.46 They were under the command of the comes domesticorum. None of these had anything to do with the excubitores. By this time emperors were hardly ever at Rome. The mobile court now required a mobile guard, thereby increasing the importance of the mobile emperor‘s per­ sonal bodyguards. So too, at other imperial palaces way beyond Rome the same guards were required. Lactantius recounts a plot which must have unfolded at the imperial palace at Marseilles in 310. Maximian, father-in-law of Constan­ tine, sought to induce his daughter Fausta to betray her husband by leaving their bedroom door ajar to permit the nocturnal entry of an imperial assassin. Without divulging anything to Maximian, a lowly eunuch was substituted for the emperor. Maximian found his way to the imperial bedchamber, skewered the occupant of the imperial bed and was then surprised by Constantine who emerged from another entrance with a large group of scholae. What is clear from Lactantius’ account of this incident is that the imperial bedchamber was shut and guarded at night, as it had been when Galba alarmed the guards outside his door by falling out of bed one night.47 Maximian’s entrance was facilitated by the reduction in the guard which Lactantius denotes as excubitores.48 44 45 46 47 48

Details in Sablayrolles (1996), but never making any connection with excubitores. Speidel (1978) and the inventory of records in Speidel (1994). Frank (1969), 33–43. Dio 63.7.2. De mortibus persecutorum 30. 4 : ‘. . . Surgit ille nocte intempesta, videt omnia insidiis opportuna. Rari excubitores erant et ii quidem longius; quibus tamen dicit vidisse somnium quod filio suo narrare vellet. Ingreditur armatus et spadone obtruncato prosilit gloriabundus ac profitetur quid admiserit. 5 Repente se ex altera parte Constantinus ostendit cum globo armatorum . . .’

117

L E O I A N D T H E PA L A C E G U A R D

Meanwhile, the praetorian cohorts including the palatine cohort remained at Rome, functioning mainly as a city garrison. In 312 they were prominent once more, along with some equites singulares Augusti, in helping Maxentius defend Rome against the forces of Constantine but following their defeat at the Milvian bridge Constantine moved quickly to dissolve both guard units and dismantle their respective camps.49 By about 318 he had formed an entirely new imperial guard from different cavalry units and together they constituted the scholae palatinae. Essentially these scholae replaced the equites singulares.50 The tribunes of the prefectural cohorts (tribuni praetoriani), presumably including the tribune of the cohort which had once guarded the palace (possibly known as excubitores), were preserved as a group of court functionaries. The urban cohorts (cohortes urbanae) were likewise maintained.51 With the administrative reorganisation lead­ ing to the formation of several prefectures and prefects, responsibility for the scholae and their tribuni was removed from the praetorian prefect and given to the newly instituted magister officiorum.52 The precise stages and chronology under­ pinning these developments are unclear.53 In general, it appears that early in the fourth century there was a schola armaturarum, a unit of the drill-squad,54 and already in Constantine’s time there were the schola scutariorum and the schola scutariorum clibanariorum. They prob­ ably formed part of the guard at the palace at Nicaea. When Constantine enter­ tained the bishops assembled there for the ecumenical council in 325 they found ‘bodyguards and armed men’ (dorufo,roi me.n ga.r kai. o‛ pli/tai) posted around the entrance vestibule.55 By mid-century was added another unit recruited from foreign troops (schola gentilium). All these scholae were organised as cavalry units and heavily recruited from Germans and other barbarians, like the equites singulares and the custodes Germani for centuries before them.56 At the same time, the protectores continued in their role as officers of the emperor and court responsible to the magister officiorum, while the scholae formed the garrison for the imperial household and its properties with the protectores responsible to the comes domesticorum.57 Despite this major reorganisation of Roman civil and mili­ tary administration affecting the various imperial guard units, exactly which units

49 Aurelius Victor, Caesares 40.25; Zosimus 2.17.2, with Speidel (1986), 253. It was proposed by Speidel (1995a), 83–7 that a certain Valerius Victorinus served in a schola palatina belonging to the emperor Licinius (308–24), but it is more likely to have been a vexillatio palatina of Constan­ tine (Woods [1997c], 85–93). See also, with some differences, Speidel (1995b), 131–6. 50 Speidel (1997), 75. 51 Symmachus, Relatio 42. 52 Boak (1924), 60–1; Clauss (1980). 53 For details: Hoffmann (1969), 279ff. 54 Frank (1969), 53–4. 55 Eusebius, Vita Constantini 3.15.2. 56 Frank (1969), 59–72. 57 Frank (1969), 87.

118

L E O I A N D T H E PA L A C E G U A R D

guarded the imperial palace at Constantinople is not clear either. What is clear, however, is that the excubitores remained. When Constantinople was founded as an imperial capital by Constantine a new imperial palace was built. By the end of the fifth century there was a part of the palace, just inside the main ceremonial entrance, called the exkoubita which presumably was the quarters of the excubitores. According to Byzantine tradi­ tion, which may well be authentic, the exkoubita was an integral part of Constan­ tine’s original palace.58 The next time we encounter excubitores is in the course of Ammianus Marcellinus’ account of the proclamation of Julian as emperor at Paris in 360. Ammianus was himself part of the imperial guard structure, at least he was one of the protectores assigned to particular senior generals, in this case the gen­ eral Ursicinus.59 Ammianus is a particularly important witness on this topic. As a protector himself, he knew the role and that of the various other imperial guard units. The extant books of his history provide extensive testimony to the various units of the scholae palatinae and their commanders.60 There is no reason to doubt Ammianus’ understanding of the contemporary structure of the imperial guard units and their nomenclature.61 Ammianus describes how, acting on a false rumour of Julian’s immediate assassination, a contingent of nearby troops immediately descended on the Parisian palace (regia). The resultant commotion in the palace caused consternation among the imperial guards (excubitores), the tribunes and the comes domesticorum named Excubitor. They refused to leave until they had actually sighted the new emperor in person in the imperial consistorium: hocque conperto milites, quos ignota pari sollicitudine movebant et nota, pars crispantes missilia, alii minitantes nudatis gladiis, diverso vagoque, ut in repentino solet, excursu occupavere volucriter regiam, strepituque inmani excubitores perculsi et tribuni et domesticorum comes Excubitor nomine veritique versabilis perfidiam militis, evanuere metu mortis subi­ tae dispalati . . . non antea discesserunt, quam adsciti in consistorium, fulgentem eum augusto habitu conspexissent. (20.4.21–2)62 58 Patria Constantinopoleos 1.59 (ed. and trs. Berger [2013], 34–5) with Berger (1988), 215–17; Kostenec (1998), 290. Similarly, the ‘scubitum’ recorded at Ravenna must reflect the excubitorium of the imperial palace there (McCormick [1986], 140). 59 Matthews (1989), 74–80. 60 The attempt by Woods (1997a), 269–91, to construct a list of commanders of five scholae between 353 and 364 reads too much into Ammianus and needs to be used with caution, as demonstrated by Brennan and Barlow (2001). 61 Likewise, a date around the time of Ammianus is argued for the vision of Dorotheus because of the way the guards of the heavenly palace are described in the vision: Bremmer (1998), 82–8. 62 Szidat (1977), 162–3; den Boeft et al. (1987), 107–8; ‘Excubitor’ as a name has long been an object of suspicion about likely textual contamination, but is generally accepted: Szidat (1977), 163, cf. Szidat (1975), 493–4; den Boeft et al. (1987), 108: ‘The most likely explanation is that at some point in the tradition of the text the name of the comes was misread and adapted to the preceding word excubitores the original name may have had a similar beginning e.g. Exsuper(ant)ius’.

119

L E O I A N D T H E PA L A C E G U A R D

Here we see the excubitores guarding the palace where the campaigning emperor was currently residing and they were alarmed. The tribunes and the comes domesticorum left in fright. When Ammianus uses the phrase excubitores he presumably does so consciously. Indeed, he appears to designate the excubitores as a separate unit with palatial guard responsibilities. As emperor, Julian had deliberately sought to reduce the influence of the impe­ rial guard but after his unexpected death on campaign in 363 his protectores domestici proclaimed Jovian as emperor.63 Jovian was himself a tribune (primicerius) of the guard.64 Jovian’s successor, Valentinian was a scholarian tribune, while his father had been tribune of the schola secunda scutariorum and his impe­ rial brother Valens was protector domesticus.65 In August 378 Valens lost the cata­ strophic battle at Adrianople because his troops were undisciplined and untrained. His field marshal Sebastian knew this and tried to get the troops in shape through small forays. Eunapius’ and Ammianus’ reports of the events – hitherto wrongly thought incompatible – combine to show that the troops being thus trained were the scholae palatinae.66 The inventory of palace guard units from this period and slightly later is reflected in the Notitia Dignitatum although it does not mention the excubitores. The Notitia is a complex and highly contestable document.67 It lists the units of the scholae palatinae under the magister officiorum and the domestici under the respective comites domesticorum. In the East there are seven scholae. Each of them included 500 scholares, thus 3,500 in all. In addition, under the comes domesticorum peditum and the comes domesticorum equitum there were two units each. From 395 most scholares and domestici were regularly outside the palace on imperial errands or assignments. Although the extant documentation for the period after the Notitia is scanty, there is no further mention of the excubitores until the mid to late fifth century, with the possible exception of a letter of Nilus of Ancyra writing in the earlier part of the fifth century. One of his letters is addressed to a certain Isidore who is titled an ‘excubitor’ (VIsidw,rw| evxkoubi,tori)68 This letter is usu­ ally thought to be anachronistic and therefore bogus on the assumption that, as a designated guard unit, the excubitores did not exist before the later fifth century.69 Yet, this need not be so. The letters of Nilus are a mixture of the genuine and the 63 Cf. Woods (1997b), 335–67 arguing that the passion of Sergius and Bacchus which describes events in the reign of Maximian is actually based on a lost account of two Christian members of the schola gentilium put to death at Antioch in 363 under the emperor Julian’s orders. 64 Lenski (2000), 492–515. 65 Ammianus Marcellinus, res gestae 26.1.5; 30.7.4. 66 Speidel (1996), 434–7. 67 The date of the Notitia and its various sections is a complex issue because it was a dynamic document and various parts show revisions at various times (for 394 [except for Illyricum] see Hoffmann (1969), 25–53). For its character and context, as well as a proposal to identify Macrobius as its author, see Brennan (1996); Woods (1996); Kulikowski (2000). 68 Ep 2.372 = PG 79, col. 357. 69 Vasiliev (1950), 64 n.45; Alan Cameron (1976b), 182.

120

L E O I A N D T H E PA L A C E G U A R D

dubious. This letter to Isidore could just as easily be located among the genuine ones. The designation excubitor need not necessarily disqualify it after all. The foregoing analysis strongly suggests that for the four centuries from Tiberius to Theodosius II the excubitores were an imperial guard contingent with specific responsibilities on behalf of the emperor. Such a position diverges from the usual claim that in texts from Tacitus to Ammianus excubitores is used as noth­ ing more than a purely literary term for ‘guards’ or ‘sentries’ without denoting any particular function or locale. Rather it could apply generally to any unit.70 More­ over, the excubitores maintained their appellation and their protective function throughout this period, even though most of the remaining guard units were trans­ formed and reinvented over the years. From Tiberius’ time the excubitores may have been part or all of a single praetorian cohort (or a subsection of the vigiles). With the abolition of the praetorians by Constantine they possibly became part of the newly formed scholae under the magister officiorum, or part of the domestic military staff under the comes domesticorum. This prompts closer scrutiny of how the excubitores related to the other guards. The excubitores are not the only specialist unit missing from the Notitia. The reason for the absence of the candidati is that they were an elite sub-group which have been taken to be the personal bodyguard of the emperor.71 A similar explana­ tion is probably required for the decani, another candidate for palace guard,72 as well as for the excubitores. That is, they were a small specialist guard unit, either a section of one of the scholae (or part section of each of them) or one of the domes­ tic units, probably one of the domestici peditum. On the other hand, if they were a cohort of the praetorians, like the speculatores of the early empire, then they survived Constantine’s dissolution of the praetorians. In that case they would not have been part of the newly formed scholae under the magister officiorum.73 As palace troops they would have been under the direct control of the comes domesticorum.74 However they were organised, in the fourth century the excubitores were still doing what they were doing in the palace of Claudius – guarding the emperor and his property particularly at night. Writing in the early fifth century and commenting on a Virgilian phrase (vigilum excubiis Aeneid, 9.157) the grammarian Servius explained that the excubiae was a day guard and the vigiliae a night guard (3.66). John the Lydian over a century later likewise linked the excubitores and the vigiles as if they were simi­ lar in function and structure except they operated at different hours. In the early seventh century, Isidore of Seville in the section of his Etymologiae concerned 70 E.g. Alan Cameron (1976c), 182 n.8; Szidat (1977), 162. 71 Frank (1969), 127–38; Haldon (1984), 129–30. 72 Argued by Frank (1969), 107–8 but disputed by Delmaire (1995), 80ff. On the other hand the deputati responsible to the comes domesticorum (Notitia, Or. 15.8; Occ. 13.8.7) were appointed for a specific task (John Lydus, On the Magistrates, 1.46; 3.10) 73 Accordingly, Mommsen (1889), 233 n.7 thought they were not recruited from the scholae because they appear to have been different from them, but he was not certain. 74 Frank (1969), 88–9, contra Clauss (1980), 41.

121

L E O I A N D T H E PA L A C E G U A R D

with ‘De Regnis militiaeque vocabulis’ (9.3) refers to excubitors in similar terms: ‘Excubitores dicuntur, pro eo quod excubias semper agunt. Sunt enim ex numero militum et in porticibus excubant propter regalem custodiam. Excubiae autem diurnae sunt, vigiliae nocturnae’. In other words, the excubitores were a military unit and they guarded by day the imperial presence along the palace porticos. With Isidore, as with John the Lydian to a certain extent, it is never clear how much of his explanation is based on contemporary usage and how much is merely antiquarian. In this case it may be a bit of both. Certainly, it cannot be maintained that Isidore was unaware that the term excubitores was a contemporary one. What has happened though is a considerable transformation in meaning in the centuries between Caesar and Isidore. The etymological complementarity of excubitores and vigiles as officers whose duties were discharged at night has dissolved. By now their functions have diversified but their titulature remained. Exactly how the excubitores and vigiles were related is uncertain. There is no doubt that originally a single praetorian cohort, under its tribune, provided a 24-hour imperial guard spread over three 8-hour shifts, with or without some sup­ port from the vigiles. Alternatively, it may be that the day guard was a specialist sub-group of the vigiles and at night it was replaced by the regular vigiles. It is possible that after Constantine the responsibility was split between a day guard and a night-guard. By John the Lydian’s day the vigiles had virtually disappeared. Their fire-fighting functions had been transferred to the collegiati in the time of Valentinian I (364–75) but their security functions may have remained.75 The pre­ fect of the vigiles or ‘night prefect’ (u[parcoj tw/n nuktw/n in Greek) was only abolished by Justinian. In doing so he commented that ‘we don’t know how’ the title of ‘night prefect’ came to be obsolete and confusing in that its responsibilities were not confined to night-time hours at all. So the title was changed to ‘Praetor of the People’.76 Most, if not all, of the references to excubitores between the time of Tiberius and that of Leo at least imply a guard of the imperial household, exactly as John the Lydian explains. The imperial palace always provides the context. Even Colu­ mella, in the 60s, may have had an imperial guardsman in mind in suggesting that you’ll not find a more alert excubitor than a good dog.77 So too, Seneca’s Neronian audience would surely have recalled the emperor’s bedroom bodyguard on hearing Thyestes proclaim ‘I need no guard (excubitor) outside my door while I sleep’.78 Excubitores is not just a general term for guards or sentries used casu­ ally without any designated functional meaning. Rather, they are the first or final cordon for the imperial presence, the personal guard within the palace and outside the imperial bedchamber. Although their role developed significantly over the intervening centuries, from the time of Tiberius excubitores has a precise technical 75 76 77 78

Sablayrolles (1996), 61–5. Justinian, Novel 13 (535) praef. Columella, De re rustica 7.12. Seneca, Thyestes, line 457.

122

L E O I A N D T H E PA L A C E G U A R D

meaning and is used by later writers accordingly. The absence of any epigraphic record of an excubitor as a palace guardsman does not deny their existence in that role.

Leo I If, as proposed above, excubitores had formed part of the imperial guard from the time of Tiberius, then Leo’s role in their development requires fresh appraisal. In attributing their invention to Leo the explanation provided is that by the 450s the traditional palace guards (the scholae palatinae) had outlived their usefulness. The 3,500-strong corps of elite soldiers, including the 1,000 members of the two scholae stationed in Constantinople, so it is argued, had become merely decorative and were no longer capable of ensuring the protection of the emperor and his pal­ ace. Hence a new unit of hardened military operatives was formed by Leo to take over the role the scholae had traditionally filled.79 There are several deficiencies in this line of explanation. In the first place the scholae were not replaced. They continued to function and continued to be recruited for two centuries afterwards.80 It is unlikely they spent all their time dressing up and standing around. Further, there is no clear testimony to the fact that they had become dysfunctional by the mid-fifth century, nor to any crisis or concern about the declining security of the imperial court at that time, nor that any of them were suddenly sidelined to make way for the newly-formed excubitores. Rather, the political and military importance of the scholae was increasing at precisely this time.81 Evidence cited for the decline in quality of the scholae is somewhat later, in the reign of Zeno, and derives from Procopius and Agathias looking back from almost a century afterwards. Procopius reports that ‘since the time Zeno succeeded to the throne’ (evx ou- de. Zh,nwn th.n basilei,an pare,labe) that is, no earlier than the beginning of his sole reign in November 474, the palace guards were recruited from ‘both cowards and wholly unwarlike men’ (kai. avna,ndroij kai. avpole,moij ou=si).82 He does not claim in fact the functional decline of the scholae by the 470s, just the beginning of such a process through the recruitment of unworthy candidates by previous standards. Agathias claims that the scholares were once recruited from honourable veterans but ‘Zeno the Isaurian seems to have been the first to introduce the present practice by enrolling in these regiments, after his restoration to the throne (meta. th.n th/j basilei,aj avna,kthsin – that is, after 476), many of his fellow countrymen who, though they were men who had either not distinguished themselves on the field or had absolutely no military experience 79 Hoffmann (1969), 301. Typical is Jones (1964), 658: ‘The scholae having become an ornamental body, Leo enrolled a small corps, the excubitores, 300 strong, to do the real work of guarding the palace. The original members of the corps were certainly genuine soldiers’. 80 Haldon (1984), 126–7; Bolognesi Recchi-Franceschini (2008), 233–4. 81 Cf. Whitby (2000b), 471. 82 Secret History 24.17.

123

L E O I A N D T H E PA L A C E G U A R D

whatsoever, were nevertheless known to him in some other capacity and were his close friends’.83 He simply contends that Zeno recruited fellow-Isaurians who had not necessarily had extensive military experience, not that the 3500 scholares had become merely ornamental and ineffective. The testimony of Procopius and Agathias cannot be taken to imply a change in role and status for the scholae by the time of Leo, nor a necessary connection with the excubitores. Neither Pro­ copius nor Agathias is really focussing on the reign of Leo when Zeno first came to prominence at Constantinople in 465/6, but a decade later at the earliest. Any change in function, status or extent of the scholae follows, rather than precedes, Leo’s promotion of the excubitores. Even so, there is no cause for proposing that it was the role of the excubitores from the time of Leo which turned the scholares into ‘superfluous relics’.84 The scholae had never been solely responsible for the functions now enjoyed by the excubitores. If the 300 excubitores replaced any­ one as palace guards at Constantinople it will have been part of the protectores domestici.85 John the Lydian’s passing comment on Leo and the excubitores is the singu­ lar basis for the argument that Leo founded the excubitores. Again, John says simply that ‘and even Emperor Leo, who was the first to establish the so-called excubitores as guards of the side-exits of the palatium, put into service only three hundred to accord with ancient custom’.86 On the foundation of this solitary sen­ tence a whole superstructure has been erected. Vasiliev, for instance, claimed that John ‘explicitly ascribes the formation of the excubitors to Leo I’,87 and that Leo ‘formed a new body of palace guards with the title of excubitors as a coun­ terbalance to the excessive influence of the Germans. The new corps was to be recruited from the residents of various regions of the empire, provided that they were stalwart and brave’.88 Much the same sentiment still prevails.89 In particular the establishment of the excubitores has been linked to the rise of the Isaurian gen­ eral Zeno. Baynes contended that the new guards ‘were doubtless the followers of the Isaurian ko,mhj domestikw/n’,90 while Kaegi casts Zeno as ‘head of a crack 83 Histories 5.15.4 (trs. Frendo). 84 Grosse (1920), 93. 85 Cf. Hoffman (1970), 303 who was inclined to consider the excubitores as having taken over func­ tions of the scholae. There is no evidence either that the excubitores took over the guard duties of the 40 candidati, as suggested by Whitby (2000a), 291, cf. Haldon (1984), 129–30 and 139. The protectores domestici disappeared in the reign of Justinian (Haldon [1984], 134). The candidati seem to have grown out of the scholarii (Bolognesi Recchi-Franceschini [2008], 237). 86 On the Magistrates 1.16. 87 Vasiliev (1950), 65 n.46 88 Vasiliev (1950), 64–5. This view is evident in Brooks (1893), 212–14; followed by Seeck (1920), 369–71; Ensslin (1925), cols. 1947–62; Stein (1959), 361. Bury was rather exceptional in interpret­ ing John more cautiously (1923a), 318 and (1911), 57 (‘organised probably by Leo I’). 89 For example: Demandt (1989), 185–7; Averil Cameron (1993b), 30; Treadgold (1995), 13 and (1997) 150–6; Williams and Friell (1999), 177 and, but far more nuanced, Lee (2000), 45–9. A sober dissenting voice was Jones (1964), 224. 90 Baynes (1925), 40 and (1925), 399.

124

L E O I A N D T H E PA L A C E G U A R D

native corps of Isaurian guards’91 and Frank claims that at about this time ‘a new corps of palace guards appears, the excubitores, recruited among the Isaurians, and their formation must surely be placed in connection with the appointment of Zeno’.92 He even characterises them as replacing the ‘Arian Gothic’ scholares.93 Others are in similar vein.94 Most of these claims are pure conjecture. There is no evidence that in the time of Leo the excubitores were exclusively, predominantly, or even partially Isau­ rian. That is not to say that there were no Isaurians, nor several, among the 300 guardsmen. Procopius comments on Zeno’s recruitment of Isaurians, it is true, but not before the period after 476 when he was emperor, and not for the excubitores either, but the scholae.95 Nor can it be upheld that Leo’s dealing with the excubitores was part of a larger and more deliberate ‘anti-German’ policy designed to counter the influence of the Arian general Aspar and his family.96 Nor, indeed, can the re-assignment or re-establishment of the excubitores be necessarily connected with Zeno. All that John the Lydian says is that Leo was the first to set up the excubitores as guards of the ‘side entrances’ of the palace and that there were 300 of them in total. He makes no comment on how they were composed, why they were entrusted with their particular responsibility and the contemporary political or military context. Nor does John say at what point during his reign Leo enacted his decision on the excubitores. Usually it is dated to 466 but only because it is related to the appointment of Zeno as comes domesticorum and Leo’s alleged policy of delib­ erately creating an ethnic counterweight to Aspar and allies.97 Leo’s excubitores were certainly in place by 471 when they protected the palace from the attack on it by the supporters of Aspar after his murder in the palace.98 It is possible that excubitores took on their new function as early as 457 at the beginning of Leo’s reign. If, as John the Lydian asserts, they existed as a unit of the imperial guard since the time of Tiberius then they had always been a small elite group and still effectively functioned as such in the 450s at Constantinople. If so, the action of 91 92 93 94

95 96 97 98

Kaegi (1981), 27; Elton (1996), 101. Frank (1969), 46. Frank (1969), 9, 204, but see Alan Cameron (1972), 137. Boak (1924), 61; Vernadsky (1941), 60: ‘der Kern einer weit grösseren isaurischen Einheit’; Stein (1959), 358 (Isaurians, Thracians and Illyrians); Jones (1964): ‘With [Zeno’s] aid Isaurians were recruited in large numbers and stationed in the capital’; Kaegi (1981), 27; Demandt (1970), 187; W. Burgess (1992), 875; Treadgold (1997), 152: ‘first the emperor made Zeno a commander of the imperial guard, and strengthened his hand by creating the new guard corps of the Excubitors, three hundred strong, composed largely of Isaurians’; Williams and Friell (1999), 177: ‘Tarasicodissa was appointed to command a new corps of 300 Isaurian palace guards, the Excubitores’. Secret History 24.17 Explained in detail in Chapter 3 (110–11). Frank (1969), 205; Clauss (1980), 43. It could have been from a later period when Leo was under more serious pressure from Aspar, that is, 468/9 (Ensslin [1925], 1958). Malalas,Chronicle, 14.40 (Thurn 294–5 = de insid. 31, de Boor 160.25–161.14) with Stein (1959), 358; Frank (1969), 204; Clauss (1980), 43; Kaegi (1981), 27.

125

L E O I A N D T H E PA L A C E G U A R D

Leo was to upgrade them and focus their precise function on guarding the palace entrances. They continued to be recruited from the most competent soldiers, as they always had. They were not merely a ceremonial guard but were designed to be able to respond effectively to any threat or intrusion. Their role in 471 and later demonstrates that the excubitores were an elite, skilled and efficient fighting force. Then there is the question of their commander. It has been regularly assumed that in the time of Leo the excubitores were under the command of the comes domesticorum,99 again probably because of their foundation being linked to the appointment of Zeno as comes. Since, however, there is no necessary connection between Zeno and the excubitores during the reign of Leo, let alone at precisely the time he was comes domesticorum (465/6), there is no need to assume that they must then have been under the command of the comes domesticorum although the comes domesticorum continued to exist.100 There were still separate comites domesticorum towards the end of Anastasius’ reign to judge from the example of Paulus, great-nephew of the emperor,101 while Philoxenus (cos.525) was comes under Justin.102 Neither of them can have had responsibility for the excubitores, however, because there was by now a comes excubitorum and there is no reason why the position of comes excubitorum had not existed from the time of Leo even though for the next 25 years no such comes is recorded. The first known comes is the emperor Justin who appears to have been in the position by 515 or so.103 The excubitores must have been an important contingent because their comes ranked high in the imperial structure.104 If the comes excubitorum was established by Leo 99 Fiebiger (1909), col. 1577; Grosse (1920), 270; Frank (1969), 206–12; Haldon (1984), 136–9; Treadgold (1997), 177. 100 According to the Notitia Dignitatum which reflects the situation in the early fifth century there were by then in both east and west two equal comites domesticorum, one each for cavalry and infantry, with their respective staffs of domestici (Notitia Dignitatum Or. 15; Occ. 13). There was still such a division in the East in the 450s. Flavius Sporacius (cos. 452) was comes domesticorum peditum in 450/1 (PLRE 2, 1026 [‘Fl. Sporacius 3’]), at about the same time as Aetius (cos. 454) was comes domesticorum equitum (PLRE 2, 29–30 [‘Fl. Aetius 8’]). After 452 the next comes for whom we have any trace is Zeno. Following him there was Aedoingus in 478 (PLRE 2, 11: ‘Aedoingus’). By this stage too there are records in Egyptian papyri of several comites domesticorum who can only be honorary holders of the title (Palme [1998], 98–116). The proliferation of honorary titles began around this time, for which see Mathisen (1991), 191–221. This fact makes it more difficult to determine the precise status, honorary or actual, of those subsequently described as comites. 101 PLRE 2, 83 (‘Fl. Anastasius Paulus Probus Sabinianus Pompeius Anastasius 17’). 102 The consul of 525 Philoxenus is called simply ‘com.domest.’ on his four extant diptychs. Again this is thought to be an honorary position (PLRE 2, 879: [‘Fl. Theodorus Philoxenus Soterichus Philoxenus 8’). Since Philoxenus was recalled from exile by the emperor Justin immediately on his accession in July 518, he is likely to have been given an active position (as Justin did to other exiles such as Vitalian and Appion). Philoxenus had previously been magister militum for Thrace and was therefore well-fitted for the post of comes domesticorum. There is no other objection to him being comes domesticorum from 518 and being still comes in 525. 103 PLRE 2, 649 (‘Iustinus 4’); Vasiliev (1950), 68. 104 Jones (1964), 658; Mommsen (1889), 233; Frank (1969), 204–7; Haldon (1984), 136.

126

L E O I A N D T H E PA L A C E G U A R D

to command the full complement of 300 excubitors then it is unclear who was the normal commander of the unit or units on duty at any particular time. By the reign of Justinian we hear of scribones who appear to have fulfilled some such function.105 It is possible that the institution of the scribones was also part of Leo’s restructuring of the excubitores. Given the patchiness of the surviving records for the period, it should not surprise that there is no earlier trace of them although they were possibly ‘the leading excubitors’ (tou.j tw/n evxkoubito,rwn) whom Justin summoned on the death of Anastasius in July 518.106 It would make sense for Leo to have created not only a senior commander (comes excubitorum) but also the subsidiary commanders (scribones) at one and the same time. Certainly in much later times the scribones act as unit commanders of the excubitores under the general command of the dome,stikoj tw/n evxkoubitw/n.107 Leo did not so much establish the excubitores as give them a specific new role and command structure. That is essentially what John the Lydian attributed to Leo, namely, a specific function (guarding the palatial entrances) and a fixed number (300).108 John gives emphasis to the number because in his antiquarian approach to describing the Roman magistracies he links Leo’s decision with the traditional number for special contingents employed by Romulus and Marius. The excubitores which had existed from the time of Tiberius, as John explains elsewhere, may have evolved over the previous four and a half centuries into a unit smaller or larger than 300, possibly smaller. What Leo did in assigning them control of the palace entrances was to narrow, but upgrade, their role. As an elite group, entry to the ranks of the excubitores was selective in terms of physique, skill and experi­ ence. As a physical guard, they were well armed but as a guard whose very display exemplified strength and safety they continued to be dressed in the time-honoured way. It was a feature of the excubitores which John the Lydian noted in passing for his readers. Leo’s reconstitution of the excubitores as a palace guard also involved detaching them from their traditional responsibility and entrusting them to their own comes excubitorum with the same high status. When all this took place, and what precipitated it, are simply unknown. Certainly they were in their new role and under their new command structure by 471. To repeat, Leo could well have moved on the excubitores early in his reign in the late 450s.

Excubitores after Leo I Autonomus was an Italian bishop who had been forced to seek refuge in Bithynia around 300. He settled in Soreoi on the Gulf of Nicomedia and eventually con­ verted sufficient locals to establish a large congregation there and built a church dedicated to Archangel Michael. He also preached elsewhere and soon attracted 105 106 107 108

Haldon (1984), 137–8. Const. Porph., On the Ceremonies 1.93 (Reiske 426.14–15, trs. Moffatt and Tall [2012], 426). Bury (1911), 59. On the Magistrates, 1.16.3 (trs. Bandy [1983], 5).

127

L E O I A N D T H E PA L A C E G U A R D

an arrest order from the imperial court, during the persecution of Diocletian, whereupon he withdrew to Claudiopolis on the Black Sea. Later he returned to Soreoi, departed again and returned again, this time to nearby Limnae. There some of his zealous followers attracted the ire of the pagans whose shrine they had demolished. The enraged pagans set upon Autonomus during the divine liturgy and stoned him to death. Autonomus was buried nearby and in Constantine’s reign a church was built over his tomb. In about 430 the church had to be pulled down and, according to the account of Autonomus’ miracles, some 60 years later his relics were found fully preserved on the site of the former church at Limnae. The person who was led to this discovery c.490 was an excubitor named John.109 It was at the same time, in 490, that the excubitores throttled the patrician Pelagius in the palace at Constantinople on the orders of Zeno.110 Whether John was an excubitor at the time, or earlier or later, is not apparent. If so, then he may not have been on duty at court at the time but on a special imperial mission in Bithynia. In any event, John may have known another contemporary court soldier, the Illyrian Justin. Justin, along with some of his friends, had left his Balkan homeland around 470 and, according to Procopius, came to Constantino­ ple and enrolled ‘in the palace-guard’ (evj tou/ palati,ou th.n fulakh,n).111 It seems Justin possessed the physique and skill to secure a position in the excubitores but it is unlikely that he was enrolled immediately. They were, like earlier imperial guards such as the equites singulares, a picked guard of experienced soldiers. Entry to the restricted corps of excubitores was only achieved through promotion of those outstanding individual soldiers who had proven themselves in other units. So by ‘palace guard’, Procopius probably means that Justin commenced his new career among the 3,500 scholae palatinae rather than be directly enrolled, as a raw recruit, in the 300 excubitores.112 In that he would not be unusual. What would be unusual would be an immediate enrolment in the elite ranks of the excubitores without any prior military experience. All the other excubitores of the sixth cen­ tury, men such as Tiberius, Maurice and Philippicus were senior military men before they appear in the record as an excubitor. Justin held subsidiary commands (comes militaris) during the Isaurian war of the 490s and for the Persian war in the following decade. He was not among the excubitores who in 498 advanced from the imperial box in the hippodrome to 109 Acta Sanctorum, September, vol. 4, 18 (= PG 115, col. 696). This John should be inscribed in the list of persons missing from PLRE 2. For the context of Autonomus (identified as a Novatian) and for his church at Soreoi: Foss (1987), 187–98. 110 References in PLRE 2, 857 (‘Pelagius 2’). 111 Procopius, Anekdota 6.2–3. 112 It is usually assumed, but is highly unlikely, that Justin entered the ranks of the excubitores imme­ diately on arrival in Constantinople (Vasiliev [1950], 67; PLRE 2, 649 [‘Iustinus 4’], cf. Kaldellis (2018), 9–11. Such an immediate appointment may have been possible in the much later case (770s) of the monk Ioanikos who had spent many years in the army, but it should be noted that his two hagiographers are divided on the point. References and discussion in www.pbe.kcl.ac.uk/ person/p4060.

128

L E O I A N D T H E PA L A C E G U A R D

fatally attack a Moor who had thrown a stone at the emperor Anastasius.113 It was possibly after the Persian war (c.506/7) that Justin had sufficiently distinguished himself to be enrolled among the very unit responsible for guarding the entrances to the palace at Constantinople. Given Justin’s lengthy experience and age, now in his mid-50s, he may well have become comes excubitorum at this time. If so, then it was as comes that he was called on to repel the seaborne attack on Con­ stantinople by Vitalian in 515.114 By the time of the emperor Anastasius’ death in July 518 Justin was definitely commanding officer of the guards. As comes excubitorum he was one of the most senior dignitaries of the state. He and his 300 troops could ensure no-one left the palace on the emperor’s death, nor gained entry. Even though Anastasius was an elderly man and his death was not unex­ pected there appears to have been little planning for the eventuality, or at least the planning misfired. As a result, there was in effect a contest between the scholae and the excubitores, a sign that both contingents had simultaneous responsibili­ ties inside the palace at night. The excubitores prevailed and their comes Justin became emperor.115 The excubitores guarded the palace of Justin and his nephew Justinian from 518 to 565. Their daily task of standing guard naturally goes unno­ ticed. They only come to light when they are deployed outside the palace to help restore order or prevent a serious civil disturbance (usually caused by the Blues and/or Greens) from threatening the palace and the emperor. The excubitores were prominent, for example, during the Nika riots in January 532,116 as well as in deal­ ing with dangerous situations in 547, 561 and 563.117 Sometimes they provided an armed escort for those summoned to the imperial presence such as the feisty monk Z’ura who had been lodged in a monastery for Monophysite monks at Sykai.118 Certain individual comites excubitorum also come to light on occasion: Priscus who was banished at the request of the empress Theodora in 529, Theodorus mur­ dered while on special assignment in Carthage in 535, Marcellus (541–52) who was close to Justinian, and Marinus in 561/2.119 At the death of Justinian in November 565 the excubitors were still at their post. In his poetic description of the accession of Justinian’s successor, Corippus pictures the newly designated Justin II arriving at the imperial palace where he is to be officially inaugurated by the assembled senators. The first imperial officials to greet and acclaim Justin, at the very entrance to the palace, are the excubitores. Corippus explains that their role is to guard the entrances and exits leading into the 113 Malalas, Chronicle, 16.4 (Thurn 321–2 = de insid., 38 (de Boor, 168.11–25) and 39 (168.26–34). 114 Vasiliev (1950), 67–8 (but not seeking to date Justin’s appointment as comes). 115 Anonymus Valesianus, Pars posterior, 76; Constantine Porphyrogenitus, On the Ceremonies. 1.93 (Reiske 426.14–15, trs. Moffatt and Tall [2012], 426), with Vasiliev (1950), 68–82. 116 Chron.Pasch., 626 (Dindorf). 117 547: Malalas, Chronicle, 18.99 (409, Thurn); 561: 18.135 (423–4 Thurn); 563: 18.151 (431–2 Thurn) =de insid. 514 (175.29–176.17 de Boor). 118 John of Ephesus, Lives of the Eastern Saints 2 (ed./trs. E. W. Brooks, PO 17.1, 28). 119 Priscus: PLRE 3, 1051 (‘Priscus 3’); Theodorus: PLRE 3, 1248 (‘Theodorus 9’); Marcellus: PLRE 3, 814–16 (‘Marcellus 3’); Marinus: PLRE 3, 831 (‘Marinus 2’).

129

L E O I A N D T H E PA L A C E G U A R D

palace, as Leo had assigned them according to John the Lydian, and to ensure that no one with hostile intent can gain access to the palace. They were fully armed: Excubiae primum, quae summa palatia servant, Imperium felix dominis intrantibus optant, Et cunctos aditus armato milite vallant, Ne quis in Augustam contrarie audeat aulam Infensum conferre pedem.120 Subsequently Corippus describes in detail the scene inside the palace a week later as the various dignitaries arrive to attend the reception of the embassy of the Avars. Corippus itemises the various groups of palace officials under the authority of the magister officiorum, placed by due rank and dressed in their variously coloured uni­ forms. Then in an elaborate and striking account he singles out the excubitores as the guard lining the porticoes from the entrance to the palace (Chalke). It was a deploy­ ment designed to impress and intimidate the Avar legation which will have been conducted along the passage between the rows of excubitores, like a river flowing between two walls of giant overhanging oak trees on either bank. To quote Corippus: The great excubitors who guard the sacred palace (165) were gathered close together in the long porticoes from the very gate and protected right hand and left like a wall, linking their golden shields with their upright javelins. Their sides girded with swords, their legs gripped by boots, they stood of equal height and glittering equally, their wide shoulders and strong arms towering: as leafy oaks amid sacred rivers, through which the calm water flows with a sharp whispering, make a dense shade with their leafy strength, and shut out the daylight with their boughs, and hap­ pily lift their uncut heads and hit the high stars with their tops. On the left and the right you could see lines of soldiers standing and glittering in the glancing light of their axes terrible and matched in age. The imperial palace with its officials is like Olympus.121 Next, in sketching the scene in the great hall (consistorium) of the palace at the installation of Justin as consul on 1 January 566, Corippus describes the new consul providing a consular gift to each of the palace guards as they were called forward strictly according to unit and seniority. With proceedings completed in 120 Corippus, In laudem Iustini Augusti minoris, 1.202–6 (ed. and trs., Averil Cameron [1976d], 42). 121 Ingens excubitus divina palatia servans porticibus longis porta condensus ab ipsa murorum in morem laevam dextramque tegebat, scuta sub erectis coniungens aurea pilis. (Corippus, In laudem Iustini Augusti minoris, 3.165ff, ed. Averil Cameron [1976d], 65)

130

L E O I A N D T H E PA L A C E G U A R D

the consistorium the imperial entourage processed into the more open part of the palace (the delphax) where an enthusiastic audience was eagerly awaiting the spectacle. As Justin processed, raised high in his consular chair, he was surrounded by a protective wall of imperial guards, including the excubitors at his back: Incedunt densae mixto lictore cohortes. Hinc armata manus dextram laevamque tuetur Caeserei lateris; clipeis pia terga tegebat Ingens excubitus, protectorumque phalanges Fulgebant rutilo pilis splendentibus auro.122 These passages from Corippus, describing a contemporary scene in contempo­ rary language albeit poetic, are fairly clear. There are still different categories of guards, both excubitores and protectores.123 What marks out the excubitores is that they guarded the entrances to the palace. They had the bearing, the physique and the arms to ensure that no unwelcome intruder dared pass. On these occasions, perhaps always, they stood side-by-side armed with their shields and javelins held upright, and with their swords at their side. This exactly fits the description of John the Lydian. Corippus also appears to describe them as brandishing battle-axes (‘bipennes’) although there is some uncertainty about that.124 At any rate, it is not clear exactly where such axes would be worn to make them so visible behind the large shield, or how they might have been otherwise carried. In addition, Corippus mentions the excubitores’ high-laced soldiers boots called ‘cothurni’ (3.169). John the Lydian also refers to their boots as being a distinc­ tive part of their dress. He describes the antique uniform of Aeneas as involving black greaved boots named ‘crepidae’ (mag. 1.12) and implies that they are still worn by the excubitores. Another contemporary description of an imperial proces­ sion, namely that of Paul the Silentiary for Justinian rushing to the newly dam­ aged church of Hagia Sophia in 558, makes similar reference to the black boots (melagkrh,pida) of the excubitors as part of a traditional costume which carried the dignity of an archaic garb.125 The boots were obviously a marked feature of the

122 Corippus, In laudem Iustini Augusti minoris, 4.238–242 (ed. Averil Cameron [1976d], 80). 123 Alan Cameron once suggested (CR 22 [1972], 137] that a phrase in the vita of Daniel (75: scola,rioi oi‛ ta. evxkou,bita poiou/ntej) means that by the sixth century the excubitores had become part of the scholae, but it is more likely the hagiographer simply means to reinforce the guard function of the scholae rather than distinguish two separate units, cf. Haldon (1984), 398 n.172. 124 Noted by Averil Cameron, In laudem Iustini, 187, in commenting on this passage. There is some textual uncertainty here but the resolution of the difficulty seems to lie with attributing the glitter­ ing ‘bipennes’ to the excubitores. Otherwise, this line may refer to the scholae who did carry axes (Frank [1969], 150 n.11). Perhaps, given the intricate description on this occasion, ‘bipennes’ does not mean ‘axes’ specifically but more generally ‘weapons’. It may also be a reference to either the excubitores’ double-edged swords or the double-edged tips of their javelins. 125 Paulus Silentiarius, Ekphrasis of Hagia Sophia 261. For background and explanation of Paul’s use of such a recondite poetic vocabulary: Mary Whitby (1987), 486–8.

131

L E O I A N D T H E PA L A C E G U A R D

guards’ costume. Moreover, it was the antique boots in particular which evoked the aura of hallowed Roman tradition. Such boots were worn by the guard of Romulus, a select group of soldiers known as celeres,126 and in the early empire by the elite contingent of the praetorian guards known as the speculatores.127 There was another connection between these republican and imperial guards in that they were confined to a total of 300. Some deliberate continuity is suggested by the fact that the excubitores also were 300 (at least from the time of Leo), and that they wore similar boots. Possibly the excubitores took over the dress of the speculatores when they were disbanded. It may even be the case that the speculatores were never disbanded at all. Rather they were, for a period, simply an alternative name for the excubitores established by Tiberius. A more coherent explanation eludes. During and after Justin II’s time excubitores continued to provide an imposing and striking guard for the palace entrances, and for the emperor himself on occa­ sion. Frequently the excubitores were called upon to intervene in some threatening turn of events in and around the imperial capital,128 such as helping the emperor Maurice defend the Thracian Long Wall against the invading Avars in 600.129 They were still housed in the exkoubita not far from the main palace entrance they guarded, namely the Chalke, just as they were in Constantine’s original palace construction.130 Their commander remained one of the most senior and powerful imperial officials. Justin II’s immediate successor Tiberius Constantine was comes excubitorum when elevated to the throne.131 Tiberius’ successor, the emperor Maurice, was also comes before being given military command against the Persians on the eastern frontier,132 and was followed as comes by the no less powerful Philippicus (comes from 582/4 to 603), Priscus (603–11/12), and Nicetas (612–3) through the successive reigns of Maurice, Phocas and into that of Hera­ clius.133 A century later emperor Leo III (reigned 717–741) downgraded the com­ mander of the excubitores but the position had been elevated again by the time of Constantine V (reigned 741–775).134 By then, the excubitores had become more 126 Livy 1.15.8; 2.20.5. 127 Suetonius, Gaius. 52; Tertullian, de corona 1. 128 Alan Cameron (1976b), 116–17; Haldon (1984), 136–9. They also were involved in forceful interventions against doctrinal dissidents (John of Ephesus, HE 3.1.16, 2.37, 2.38). 129 Theophylact Simocatta, Historiae 7.15.7; Theophanes, Chronicle AM 6092 (279 de Boor). 130 Mango (1959), 73; Bolognesi Recchi-Franceschini and Featherstone (2002), 39. 131 Corippus, In laudem Iustini 1.202–6; Theophanes, Chronicle, AM 6066 (247 de Boor); John of Epiphaneia, fr.5 (FHG 4, 275); Theophylact Simocatta, Historiae 3.11.4 with further references in PLRE 3, 1324 (‘Tiberius Constantinus 1’). 132 Gregory, Ep. 3.61; John of Ephesus, HE 3.6.14, 6.27 (replacing Tiberius in 574); Theophylact Simocatta, Historiae, 3.15.10, with further references in PLRE 3, 856 (‘Mauricius 4’). 133 PLRE 3, 1022 (‘Philippicus 3’); PLRE 3, 1056–7 (‘Priscus 6’); PLRE 3, 941 (‘Niketas 7’). 134 The known excubitores (mainly from lead seals) from the 7th and 8th century can be found in PBE 1 (at www.pbe.kcl.ac.uk/data/off/index.htm): Georgios 272 (mid/late 7th century); Ioannikios 2 (mid-8th–mid-9th); Konstantinos 90 (6th/7th century); Konstantinos 122 (7th/8th); Niketas 47 (7th).

132

L E O I A N D T H E PA L A C E G U A R D

of a ritual guard with a defined role and rank in Byzantine court ceremonial. In 949 there were 700 excubitors divided into 18 banda, each headed by a scribo.135 When Iohannikos became an excubitor in 771 (or 781) he was enrolled in the 18th bandon.136 The scholae still continued as well. They were divided into 30 banda (each headed by a comes scholon), although it is not certain how many scholarioi there were in total. The domestici were by now under the authority of the ‘domes­ tic of the scholae’ which suggests they had been merged with the scholarioi.137 To sum up, the excubitores had survived in name, and to a considerable degree also in function, for 700 years since their establishment by Tiberius at Rome in the first century. Leo I did not establish the excubitores in the fifth century. Rather, he redefined their role as a guard for entrances to the imperial palace at Constan­ tinople, confined their number to 300 and provided them with a new command structure involving scribones and a more senior commander, the comes excubitorum. Leo did not so much invent the excubitores as reinvent them.

135 Bury (1911), 57. 136 References in www.pbe.kcl.ac.uk/person/p4060 (‘Ioannikos 2’). 137 Bury (1911), 52–5.

133

5 THE IMPERIAL REIGNS OF LEO II*

The boy-emperor Leo II has bequeathed us more than his share of chronological puzzles. He was only five years old when he died in 474 but he had possessed imperial authority for more than half his short life. There is uncertainty about the date of his birth, and most of his career as emperor with his grandfather Leo I. If we still had the chronicle of Nestorianos which actually terminated at the death of Leo II we would probably be fully and clearly informed on these matters.1 Instead, several of the relevant documents have been consistently misconstrued, that is, they have been made to fit a predetermined interpretation or order of events. Moreover, one important text has been ignored altogether. The chronology of the boy-emperor’s career depends on Otto Seeck’s Regesten2 which has inevitably shaped the coverage in the standard accounts of J. B. Bury and Ernst Stein.3 The authority of Seeck also underpins the entry in the Prosopography of the Later Roman Empire which sets out the following chronology of Leo’s brief life and imperial career:4 467 – birth; 473 – October, elevated to Caesar by Leo I; 474 – before 18 January, elevated to Augustus by Leo I; 474 – death, November. This investigation is designed to show that the multiple phases in the imperial rule of Leo II have never been properly disentangled. To do so not only brings greater precision to the sequence of events, and to Leo’s short career, but also casts light on the wider context of these turbulent few years.

* This chapter originally appeared under the same title in Byzantinische Zeitschrift 96 (2003), 559–76 and is reproduced here with the permission of De Gruyter (Berlin). Account has been taken in this version of the subsequent corrections proposed by Feissel (2006); Kosinski (2008). 1 Feissel (2006), 189. 2 Seeck (1920), 417–21. 3 Bury (1923a), 323, 389; Stein (1959), 361–2. 4 PLRE 2, 664–5 (‘Leo 7’).

134

THE IMPERIAL REIGNS OF LEO II

Family Leo II was the son of Ariadne, eldest daughter of Leo I, and Zeno, a military man of Isaurian lineage. Ariadne was born well before Leo I ascended the throne in 457 while Zeno had probably been a member of the emperor’s personal military staff, the protectores. He came to prominence in Constantinople in 465 when he brought to court incriminating letters of the magister militum of the East, Arda­ burius. The emperor Leo I took immediate action dismissing Ardaburius from his military post and stripping him of the prestigious title of ‘patricius’. Ardaburius had been one of the leading military figures in the empire and the son of Aspar, the most senior general and senator to whom Leo I owed his placement on the throne. Aspar and his son, the two most powerful men in the empire at the time, were clearly stung by the rebuke of Leo and nursed their resentment. Aspar remained as magister militum praesentalis and was able to continue to exert his power and influence over Leo. Gradually, Leo asserted his own independent authority by securing office and power for his own appointees. Chief among them was his brother-in-law Basiliscus and Zeno who became his son-in-law. As an Isaurian, Zeno was never particularly popular in the capital it would seem. In 466 probably, Leo appointed him as head of the emperor’s military staff, comes domesticorum, and later (468) betrothed him to Ariadne.5 The struggle to secure the imperial succession for the next generation continued in the late 460s/early 470s between Aspar and his supporters and Leo I and his supporters, including Zeno. While it is usually cast as a struggle of antagonistic ethnic warbands, the Goths of Aspar against the Isaurians of Zeno, it was rather a conventional dynastic rivalry among extended aristocratic families. In 469 Zeno was appointed as magister militum per Thraciam to replace Anagastes who had rebelled. However, due to the machinations of Aspar and Ardaburius, Zeno was prevented from completing the expedition against Anagastes and was forced to flee to the East, whereupon he was soon appointed as magister militum per Orientem at Antioch. He remained there for two years, with his wife Ariadne and his son Leo who had joined him, before returning to Constantinople in the second half of 471 after the murder of both Aspar and Ardaburius. Young Leo II soon became the vital link in establishing the dynastic continuity which had been the policy objective of his imperial grandfather for so long.6

Birth The record closest to the birth of Leo II, the anonymous Life of Daniel the Stylite, places the birth during the consulship of his father Zeno (469). Daniel sat on his 5 For all these events and the sources which underpin them: Bury (1923a), 314–23; Stein (1959), 354–60; Jones (1964), 221–4; Demandt (1989), 185–7; Lee (2000), 45–9. 6 For the present chronology, especially of Zeno’s career, and interpretation of Leo I’s dynastic rivalry with Aspar: Croke (2005a) and Chapter 3 (51–107).

135

THE IMPERIAL REIGNS OF LEO II

column at Anaplus just up the Bosporus from Constantinople and was regularly accessible to the emperor Leo and his courtiers. The Life of Daniel describes the birth of Leo as follows: ‘now while the patrician Zeno was still absent at the war (e;ti o;ntoj auvtou/ evn tw/| pole,mw|) a male child was born to him by the Emperor’s daughter and received the name of Leo’.7 The author of Daniel’s life was a disciple of the holy man and when he came to writing up the life some decades later he relied heavily on his own memory and the recollections of others. Even though this dependence on personal recollection and oral tradition gives rise to much useful and accurate information, the chronology of the life is not always clear.8 There are two successive chapters of the vita in question here. The first chapter (c.65) dealt with the plot against Zeno which Aspar had initiated when Zeno was appointed as local magister militum to take on the rebellion of Anagastes in Thrace after he had become consul in 469. The second chapter (c.66) records that while Zeno was still absent from Constantinople on the campaign his son was born. To say, as the Life of Daniel does, that Leo II was born while his father was ‘still at the war’ probably means in the late summer of 469 if ‘still’ ( e;ti) is meant to carry any force at all. The best we can say then, is that Leo II was born in the late summer (August/September) in 469, or possibly a little later in the year.9 The sixth-century Antiochene John Malalas says, and it is confirmed by a simi­ lar statement in the Syriac Chronicle of 1234, where the author explicitly cites the sixth-century Ecclesiastical History of John of Ephesus,10 that when Leo II died in November 474 he was seven years old. If Leo was seven in November 474 then he must have been born no later than November 467 (counting in whole months), that is, if he had only celebrated his seventh birthday in the same month in which he died. Malalas, however, struggled with reconciling the chronology of Leo and his grandson’s reign so it is very possible that he has erred here and that prefer­ ence must be given to the more contemporary Life of Daniel. In that case, Leo II could have been born mid- to late-469, and therefore been conceived in late 468. So, assuming he was conceived in wedlock, the marriage of his parents Zeno and Ariadne must belong somewhere in the period 467/8. Further, the politics of the period suggest that the union of Ariadne and Zeno would be unlikely to be as early as 466. Although married into the Roman aris­ tocracy, Zeno had only come to notice in Constantinople in 465, according to the scanty extant historical information for his career. Even though he made an instant impact at court by having the general and patrician Ardaburius stripped of his office and titles, he will still have needed time to further establish his credibility 7 v.Dan.Styl. 66 (64.26–65.1 Delehaye) with the translation of Dawes and Baynes (1977), 47. 8 On the vita’s sources of information, and its chronology: Delehaye (1913), 225–7 and (1923), XLIII–V, LIV–LVII. 9 Accepting here the argument put by Kosinski (2008), 209–10 against the earlier version of this chapter (Croke [2003]). 10 Malalas, Chron 14. 47 (Thurn 300); Chronicon ad annum 1234 pertinens, 48 (CSCO Scr.Syr., 56, 146). Since John of Ephesus, in turn, made extensive use of Malalas’ Chronicle it is possible that this statement is one such example.

136

THE IMPERIAL REIGNS OF LEO II

with the emperor Leo I and his entourage. In any event, there was the fact that Zeno was already married at the time.11 The possibility of a union with the emper­ or’s daughter can only have emerged after his first wife’s death which may have been shortly after the disgrace of Ardaburius. The marriage to Ariadne may have occurred at the same time as Zeno was appointed to head the military staff of the imperial household as comes domesticorum, but not necessarily. Whenever the union was solemnified, it would not be surprising in the current political circum­ stances, with Leo I so anxious about the imperial succession, to find that Ariadne and Zeno had been under pressure to produce a potential emperor as soon as possible. They appear to have obliged. The most likely date for the marriage of Ariadne and Zeno is therefore mid-late 468 with Leo being born sometime during the Roman campaigning season, that is between March and September, 469. If Leo II was born in 469 then he can have been no more than five years old when he died in 474. For his birth, and therefore age, other dates have been advanced, however, and need to be considered. Baynes, for instance, construed Malalas’ w;n e;niautw/n x , as saying that when he died Leo II was in his ‘seventh year’, that is he was six years old at the time and was therefore born in 468.12 Seeck evidently interpreted Malalas the same way.13 Now, it is one of the consistent aspects of Malalas’ narrative that he concludes each reign with the age of the emperor at his death. He claims to have taken this information from the chronicle of Nestorianos which concluded in 474.14 In most cases it is very difficult to assess the accuracy of Malalas’ figure because we have no independent testimony to an emperor’s date of birth. Whether Malalas’ participial formula for imperial ages (w;n e;niautw/n [number]) refers to the current year or completed years is also unclear. On balance, it would appear that the figure he quotes for an emperor’s age means completed years. This supposition is reinforced by the fact that in some cases he even adds months. The age of the emperor Anastasius, for example, is put at 90 years and five months.15 That Malalas’ imperial ages at death are for completed years, or the number of birthdays, means that the 468 date proposed by Seeck and Baynes for Leo II’s birth can be discounted. By Malalas’ reckoning Leo II was already seven years old when he died, although he is clearly mistaken here. A more recent and more literal interpretation of the vita of Daniel proposes, however, that Leo II was born in 471 and was only three years old, not five, when he died.16 The argument proceeds thus: Ariadne was married to Aspar’s 11 His first wife was Arcadia (PLRE 2, 130 [‘Arcadia 2’]). Their son, named Zeno after his father, died young (PLRE 2, 1198 [‘Zenon 4’]). 12 Note to translation of the vita S. Danielis Stylitae in Dawes and Baynes (1977), 81. Earlier Baynes (1925), 400 had opted for 469. 13 Seeck (1920), 415. 14 Malalas, Chron. 13.14 (Thurn 249) cf. 14.47 (Thurn 376). For Malalas’ formulaic treatment of emperors, but not discussing their ages in particular, see Jeffreys (1990a), 138–43. 15 Malalas Chron 16.22 (Thurn 335). Other examples are the emperors Carus, ‘60 and one-half years’ (12.34 [Thurn 233]), and Constantine I, ‘60 years and 3 months’ (13.14 [249 Thurn]). 16 Lane Fox (1997), 191–2.

137

THE IMPERIAL REIGNS OF LEO II

son Patricius when he became Caesar in 468, but in the following year they were divorced so that Ariadne ‘was taken from the Caesar Patricius and married to the new favourite’ (Zeno) while their offspring Leo II was born in 470/1 and died three years later. This concatenation of events is unsustainable for several reasons: (1) it is wrong to date Patricius’ elevation to Caesar in 468, rather than the reliably attested 470;17 (2) the vita (c.67) would appear to imply that the three years is the length of Leo II’s reign, not the length of his life, thereby vitiating the contention that ‘alone of all our sources, it knows the age of the unfortunate Leo II’;18 (3) it ignores the clearly documented separate stages of his reign as Caesar and Augustus; (4) it ignores the clear evidence for the connection between Zeno’s rise to prominence at Constantinople in 465/6 and his later marriage (468) to Ari­ adne; (6) it requires the unlikely event of an imperial divorce before, not after, the deaths of Aspar and Ardaburius and the attack on Patricius who was still Caesar in later 471.19 Further, if Leo was only three years old when he died in November 474 then he cannot have been born any earlier than December 470. Yet, the vita of Daniel says that Leo was born while Zeno was ‘still absent at the war’ (c.66). As noted above, the war in question, on this chronology, must be that against Anagastes in the summer of 469. By late 469 Zeno was in Antioch as magister militum per Orientem despatching troops to battle Indacus in Isauria,20 while in 470 he was still at Antioch as magister.21 In any event, there is a plain contradiction here: if Leo II was born while his father Zeno was ‘still at the war’, that is before the end of the campaigning season of 469, then he will have been five, not three, years old when he died. In brief, this impossible interpretation appears to be based on a misguided quest to prove the literal chronological accuracy of every statement in Daniel’s vita at any cost and in spite of any other testimony.

Caesar Leo II’s imperial rule can be divided into distinct phases. The first two occurred in the lifetime of Leo I when the boy-emperor was successively Caesar and Augustus. Both these stages are clearly differentiated by Nikephoros Kallistos Xanthopoulos, writ­ ing at Constantinople in the fourteenth century According to Xanthopoulos, first Leo was made Caesar, and later Augustus – ~O de. basileu.j Le,wn o‛ me,gaj ) ) ) to.n mikro.n Le,onta Kai,sara pro,teron ei;ta kai, basile,a ~Rwmaiw/n a,nhgo,reue.22 At this point in his work he is probably depending on the lost Ecclesiastical

17 Victor Tonnennesis, Chron., s.a. 470 (ed. Hartmann, CCh 173a, 35 [12] = Mommsen, MGH AA, XI 188) with PLRE 2, 842 (‘Julius Patricius 15’). 18 Lane Fox (1997), 192. 19 Following Seeck (1920), 489. 20 John of Antioch, fr.229 (Mariev 416–18). 21 References in PLRE 2, 1201 (‘Zenon 7’). 22 HE 15. 29 (PG 147 84).

138

THE IMPERIAL REIGNS OF LEO II

History of Theodore Anagnostes written in the 520s. Theodore’s history does not survive but there is an extant summary of it, and even in this summary we can discern that Theodore distinguished between Leo’s proclamations as Caesar and as Augustus (Le,wn o‛ basileu.j proeba,leto Kai,sara Leo,nta to.n mikro.n ) ) ) Le,wn o‛ mikro,j ) ) ) avnhgoreu,qh basileu.j)23 Each phase needs to be considered carefully in turn. Other authors besides Xanthopoulos and Theodore refer to Leo as having been Caesar.24 Above all, the contemporary account preserved by Peter the Patrician is unequivocal because he describes young Leo as Caesar.25 So, at some point the emperor Leo I elevated his young grandson to a subordinate share in the impe­ rial power. When and why he did so remain obscure. If there was an overriding impulse to action on Leo’s part it was evidently the escalating severity of his illness.26 He knew his time was now limited. The position of Caesar was a sign of emperor-designate and had generally been reserved for children of reigning emperors.27 In Leo’s case there had been a very significant recent precedent in that Aspar’s son Patricius had been created Caesar in 470 under strong pressure from Aspar and with considerable public opposition in the streets of Constantinople.28 They were reminded by the portrayal of Patricius on the latest gold coins of Leo I,29 and their opposition was based on the expectation that Patricius would soon replace the ageing Leo. Accordingly, Leo’s motives in bestowing the title of Caesar on his infant grandson are fairly transparent. Having struggled for so long to secure his own succession, creating a Caesar from within his own family was a guarantee of imperial continuity, not to mention religious orthodoxy. Exactly when Leo became Caesar is not evident. It cannot have been before the murder of Aspar and Ardaburius in 471. Allowing for the resettlement of Zeno and his family from Antioch to Constantinople in the aftermath of Aspar’s demise, as well as Leo’s consolidation of power, it would not have been before 472. On the other hand, given the urgency of Leo’s preoccupation with guaranteeing his suc­ cession it is likely to have been sooner rather than later. Up to the demise of Aspar there had been a Caesar, albeit not Leo’s preference. A new Caesar was probably a priority for Leo, so a coronation sometime in 472 is probable. Further, it might 23 HE Epitome 398 (111.13 Hansen) and 400 (112. 11 Hansen). 24 Victor Tonnennesis, Chron. s.a. 473.1 (ed. Hartmann, CCh 173a, 37 [12] = Mommsen, MGH AA XI 188); Chronicon ad annum 1234 pertinens, 48 (ed. J. Chabot [1937], 146) – neither cited by PLRE. 2 on Leo II as Caesar. 25 On the Ceremonies, 1.94 (431–2, Reiske), trs. Moffatt and Tall (2012), vol. 1, 431. 26 On the Ceremonies, 1.94 (431–2, Reiske), trs. Moffatt and Tall (2012), vol. 1, 431; Theodore Anagnostes, HE Epitome 400 (112.11–12 Hansen). 27 Cf. Guilland (1967a), 25–43. 28 Patricius as Caesar: vita Marcelli 34 (316–18 Dagron) cf. Zonaras 14.4–7 (122. 11–123.5 BüttnerWobst). For the date: Victor Tonnennesis, Chron. s.a. 470 (ed. Hartmann, CCh 173a, 35 [12] = Mommsen, MGH AA XI 188). 29 Grierson and Mays (1992), 162–3, although there are doubts about their interpretation of these coins expressed by Kent (1994), 109–10. Generally, a Caesar does not appear on imperial coinage after the mid-fifth century (cf. Rösch [1978], 37).

139

THE IMPERIAL REIGNS OF LEO II

have been expected that his son-in-law Zeno would be the preferred candidate. According to the contemporary Isaurian writer Candidus, Leo wanted Zeno to be emperor and manoeuvred accordingly but the populace of Constantinople actively spurned his choice.30 They were not convinced of Zeno’s orthodoxy,31 while Leo too came to doubt whether Zeno possessed the necessary physical and mental qualities of an emperor.32 In addition, in the early 470s the vociferous Byzantine public was also not keen on the idea of an Isaurian emperor. At just this time (September/October 472) many Isaurians were killed when set upon in the hippo­ drome. This deadly outburst may have been a protest against a proposed elevation for Zeno.33 In any event, time was running out for Leo. By October 472, but pos­ sibly months earlier, the emperor’s physical condition was worsening so he now resolved to elevate his three-year-old grandson Leo II to the position of Caesar.34 The Ecclesiastical History of Theodore Lector, would appear to suggest that the conferral of the title Caesar took place before November 472. Theodore implies that Leo II was made Caesar in the same year that the massive eruption of Mt Vesuvius deposited ash as far away as Constantinople: Le,wn o‛ basileu.j proeba,leto Kai,sara Le,onta ton. Mikro.n to.n u‛io.n vAria,dnhj th/j e‛autou/ qugatro.j kai. Zh,nwnoj tou/ gambrou/ auvtou/) h‛ ko,nij tou,tw| tw|/ cro,nw| kath/lqen mhni. Noembri,w| puraktou,ntwn nefw/n fase,ntwn evn ouvranw| pro,teron( w‛ j pa,ntaj pisteu/sai o[tipu/r h]n to. me,llon katafe,resqai35 Several centuries later, in his Synopsis Historion, Cedrenus adds a significant detail, namely that Leo was already Caesar at the time the volcanic dust from the Vesuvian eruption started falling on Constantinople (6 November 472). In fact he was exercising his imperial authority by presiding at games in the hippodrome when the ash began to descend at the sixth hour (Anagoreuqe,ntoj toi.nun tou/ mikrou/ Le,ontoj( ‛ippikou/ avgome,nou w[ra j , th/j hvme,raj sko,toj baqu. evkaluye th/n po,lin).36 For this part of his history, Cedrenus’ information was taken from

30 Candidus fr.1 (466–7 Blockley) = Photius, Bibl.Cod.79; Lippold (1972), 157. 31 Zonaras 14.1.28 (126.14–15 Büttner-Wobst); Nicephorus Callistus, HE 15.29 (PG 153.84) cf. Theodore Anagnostes, HE.Epitome 390 (109.19–110.6 Hansen). 32 Zonaras 14.2.2 (128.6–7 Büttner-Wobst), possibly a reference to Zeno’s alleged epilepsy although it has been proposed that the epilepsy was merely a Chalcedonian slur with no factual basis (Con­ rad [2000], 61–81). 33 Marcellinus, Chron. s.a. 473.2 (ed. Mommsen, MGH AA XI 90). Marcellinus dates this fracas to the consulship of Leo (473) and the 11th indiction (1 September 472 to 31 August 473) which makes possible a date in October 472. For events in Constantinople, the indictional date of Marcel­ linus is generally the correct one, cf. Croke (2001), 173–5. 34 On the Ceremonies, 1.94 (431–2, Reiske, trs. Moffatt and Tall [2012], 431); Theodore Anagnostes, HE Epitome 400 (112.11–12 Hansen). 35 HE Epitome 398 (111.13–16 Hansen). 36 Cedrenus 383.5 (Tartaglia, 599).

140

THE IMPERIAL REIGNS OF LEO II

the Chronicle of pseudo-Symeon the Logothete but probably derives ultimately from Malalas or from Theodore.37 Starting with this summary passage from Theodore’s history, the date usually given for Leo II’s elevation to Caesar is October 473, not 472.38 There are two components to fixing this particular date: (A) its link to the eruption of Vesuvius; and, (B) its determination as the beginning point for calculating the length of Leo II’s reign. Neither of these components is as secure as the traditional interpretation presumes. (A) Link to Vesuvian eruption Both PLRE and others have relied for the 473 date on the arguments of Seeck39 who followed Theodore and Cedrenus in linking Leo II’s elevation to Caesar with the eruption of Vesuvius. Unfortunately, Seeck misdated by a year the eruption of Vesuvius and therefore Leo’s elevation to Caesar. It is clear from the entry in the sixth-century chronicle of Marcellinus that the eruption was a catastrophic and memorable event which definitely took place in November 472. Volcanologists have confirmed that it was the greatest eruption of Mt Vesuvius since that of 79AD which destroyed the neighbouring cities of Pompeii and Herculaeum.40 Two other later chronicles date the eruption to different years: the Chronicon Paschale to 46941 and Theophanes to 474.42 Neither can be preferred to Marcellinus for an event so close to his own experience. Given Marcellinus’ double dating system it sometimes happens that events are primarily dated to the indiction rather than the consulship.43 If the Vesuvian eruption is not to be dated to 472 then the only other possibility is 471, that is, if Marcellinus’ date falls in the 10th indiction (1 September 471–31 August 472) rather than the consulship of Marcian and Fes­ tus (1 January to 31 December 472). Still, there is no reason not to accept 472. Seeck explains, somewhat perversely, that he has put this event in 473 rather than 472 because in his view Marcellinus’s dates cannot be trusted. Much the same assumption is made by Kosinski in denying the possibility of Leo II being made

37 38 39 40

Hunger (1978), 393ff; Jeffreys (1990b), 264–5. Seeck (1920), 419 followed by Bury (1923a), 323; Ensslin (1925), 1961 and PLRE 2, 665 (‘Leo 7’). Seeck (1920), 419 with supplementary discussion at 425–6. Marcellinus, Chron. 472.1 (ed. Mommsen, MGH AA XI 90); with Croke (1995), 99 and the carbondating of debris in Rosi and Santacroce (1983), 249–71. Note their statement (250) that ‘the 472 eruption was extremely disruptive and must be considered the most violent and fatal [at Vesu­ vius] during the last nineteen centuries. A similar eruption occurring today, considering the present urbanization of the area, would have apocalyptic consequences’. See also Mastrolorenzo et al. (2002), and for the growing recent scientific literature on the 472 eruption and its impact: De Sim­ one and Russell (2019). Kosinski (2008), 213 downplays the date of the eruption and Marcellinus’ entry on it. 41 Chron.Pasch. 598.10–14 (Dindorf). 42 Theophanes, AM 5966 (119 de Boor). 43 Explained in Croke (2001), 173–5.

141

THE IMPERIAL REIGNS OF LEO II

Caesar in 472.44 Furthermore, argues Seeck, since John Malalas’ date for the death of Leo II (November 474) is correct, then so must his statement of the duration of Leo’s reign (1 year, 23 days). So, Seeck counted back from November 474 to locate the beginning of his reign (elevation to Caesar) in October 473. To these points he adds that it was customary for a new emperor to be honoured with the consulship the following year. Since Leo II was consul in 474 he must have been made Caesar in October 473, according to Seeck. However, if he was proclaimed in the latter part of the year in 472 and at short notice it was probably already too late for him to take on the consulship for the following year. It had already been announced as being the consulship of Leo I. The young Leo would have to wait until the next year for his consulship. (B) Calculating reign-length Malalas’ statement of the duration of Leo’s reign, ‘1 year and 23 days’, cannot be used so directly and partially as Seeck does. What Malalas says is more complex: ‘after the reign of Leo the Elder, Leo the Younger ruled for 1 year and 23 days’ (meta. de. th.n basilei,an Le,ontoj tou/ mega,lou evbasi,leuse Le,wn o‛ mikro.j etoj ; a, kai. h‛ me,raj eivkositrei/j).45 Now, if Malalas meant to count from the death of Leo I (18 January 474) he was way out; if he meant to count from the time when Leo I was no longer sole Augustus, that is, from November 473 (as we shall see), then he is rather closer. What is clear, however, is that Malalas could not be calculating from the time Leo II became Caesar. Seeck is misleading in using Malalas’ count in isolation from his starting point – ‘after the reign of Leo the Elder’. Seeck’s use of Malalas does not outweigh his more serious shortcoming at this point in dating the eruption of Vesuvius to 473, rather than to 472. If anything, it only compounds the difficulty of sustaining his October 473 date for Leo II’s accession as Caesar. On any reckoning, therefore, the combined testimony of Theodore and Cedrenus means that Leo II was already Caesar by November 472, when he was 3 years old. The Chronicle of Victor of Tunnuna would also appear to confirm 472 as the date although the text of the chronicle is rather corrupt at this point.46 Which month and day in 472 saw the imperial installation simply cannot 44 Kosinski (2008), 212–14. 45 Malalas, Chron. 14. 47 (299 Thurn). 46 Victor Tonnenensis, Chron. 473.1 (ed. Hartmann, CCh 173a, 37 [12] = Mommsen, MGH.AA. XI 188). Victor’s chronicle, at least as transmitted, omits altogether the consuls of 472 (Marcian and Festus). Instead, it lists a series of events under the consuls ‘Leo VI et Probino’ which is a fiction arising from a confusion of the consulships of 471 (Leo IV et Probianus) and 473 (Leo V). The first event recorded (Hartmann, CCh 173a, 37 [12] = s.a. 473.1) is the elevation of Leo II as Caesar, then follows what looks like an insertion covering the succession of bishops in each of the sees of Rome (s.a. Hartmann, CCh 173a, 38 [13] = 473.2), Constantinople (s.a. 473.3), Antioch (s.a. 473.4), and Jerusalem (s.a. 473.5). This apparent gloss is itself followed by the arrival of Olybrius in Rome and the expulsion of Anthemius (Hartmann, CCh 173a, 39 [13] = s.a. 473.6), which is definitely dated to April/May 472 (references in PLRE 2,797 [‘Anicius Olybrius 6’]). This entry

142

THE IMPERIAL REIGNS OF LEO II

be determined, but it could have been quite early in the year. The most we can say is that it was no later than October, because it predates the fallout at Constanti­ nople from the Vesuvian eruption in early November. From what we can surmise, a date earlier, rather than later, in 472 would make better political sense. Coins from the period when Leo I was Augustus and Leo II Caesar were produced, but they do not help narrow the proclamation date.47

Augustus Leo Caesar was proclaimed Augustus at some stage and then ruled as a consti­ tutional equal with his grandfather. There are surviving coins which demonstrate this status clearly. On them we see two emperors enthroned and both are crowned, signifying that both are Augusti.48 Not only does the contemporary coinage show clearly that Leo II was an Augustus during the reign of Leo I, but his consular nomenclature demonstrates the same. In none of the extant documents noting Leo II’s consulship in 474, some contemporary and some much later, is Leo noted as Caesar. Instead, he is always ‘Leo Iunior Augustus’49 and must have been so from the commencement of the consular year on 1 January, if not from the time his consulship was proclaimed and broadcast in 473. Unlike Leo II’s elevation as Caesar, we are rather better informed about his ele­ vation to Augustus although it is usually assumed to have taken place by default, that is, only on the death of Leo I in January 474.50 Regrettably, however, a key document describing the elevation of young Leo as Augustus is never cited by Seeck nor in the PLRE entry on his career That document is Constantine Por­ phyrogenitus’ tenth century Book of Ceremonies which incorporates sections of a work on ceremonial attributed to Peter the Patrician in the sixth century. The chap­ ters included in the Byzantine ceremonial book which derive from Peter cover the imperial coronations of Leo I, Anastasius, Justin and Justinian.51 Also included from Peter in the Book of Ceremonies is an account of the elevation of Leo II. It

47 48 49 50 51

is introduced by ‘his consulibus’. The next entry (Hartmann, CCh 173a, 39 [13] = s.a. 473.7) is a completely jumbled sequence in which a certain ‘Herculanus’ (presumably Glycerius, cf. PLRE 2, 544 [‘Herculanus’]) is said to be the son of Orestes and both were killed before Nepos succeeded Herculanus. A likely explanation for this rather chaotic sequence of events is that Victor originally had the entries for Leo II as Caesar (s.a. 473.1) and Olybrius (s.a. 473.6) under the consuls ‘Mar­ cian and Festus’ (472) with a notice of the last western emperors Glycerius, Nepos and Romulus (the son of Orestes) under the consulship of ‘Leo V’ (473). In the course of transmission these few entries, along with other parts of Victor’s chronicle were miscopied and confused. See further: Placanica (1997), 80–1. For the coinage: Grierson and Mays (1992), 163. Kent (1994), 103. CLRE, 482–3, noting that, Leo was ‘apparently proclaimed Augustus between 1.i and 18.i’ (483), that is before becoming sole emperor on 18 January. There is no reason, however, why Leo II was not proclaimed Augustus before his consulship, as argued here, perhaps even in anticipation of it. Seeck (1920), 378; but not Stein (1949), 361; Crawford (2019), 100–1. Bury (1907), 212–13; Antonopoulos (1990), 196–221.

143

THE IMPERIAL REIGNS OF LEO II

constitutes a template for the ceremonial involved when someone who is already a Caesar is elevated to Augustus.52 Seeck offers no reason for ignoring this account of the elevation of Leo II as Augustus by Leo I, but in the PLRE it is discarded because of the alleged inaccuracy of two circumstantial details: (A) Eusebius as magister officiorum, an office it is claimed he did not hold before 474,53 and (B) the coronation is dated in the Book of Ceremonies to November in the ‘consulship of Leo the Younger’ (474), a patently incorrect date.54 Before resolving whether to endorse or impugn the authenticity of the description of Leo II’s coronation in the Book of Ceremonies, these two alleged dating defects require more detailed examination. (A) Eusebius as magister officiorum While Eusebius was magister officiorum at some stage during the joint reign of Leo II and Zeno Augusti, that is, between 29 January and November 474 so also was the patrician Hilarianus.55 Undated laws from these months issued as ‘Impp. Leo et Zeno AA’ are addressed to both Eusebius56 and Hilarianus.57 Since Hilari­ anus was definitely magister officiorum on 27 March 47058 it is assumed that he must have held the office continually until well into 474 during the reigns of Leo II and Zeno, before being succeeded by Eusebius. That is the sequence assumed in PLRE 2.59 It would mean, because the magister officiorum presided at imperial coronations, that Hilarianus played a key role in no less than three coronations: Leo II as successively Caesar and Augustus, as well as for the coronation of Zeno as imperial colleague of Leo II in January 474. Yet, Hilarianus is also later attested as magister officiorum under Zeno, that is, after November 47460 which is unnecessarily explained away by the PLRE as being ‘probably also [a law] of Leo and Zeno’.61 There is no ambiguity about the inscription to this law, so it can only mean a subsequent tenure of the same office for Hilarianus. The PLRE is also inclined to add to Hilarianus’ career another law addressed ‘Imp. Zeno A. Illyriciano magistro officiorum’.62 Assuming ‘Illyriciano’ to be a corruption of ‘Hilariano’, if that is what PLRE intends, it is quite possible that Hilarianus is the magister addressed here. The attribution of this law 52 On the Ceremonies, 1.94 (431–2, Reiske, trs. Moffatt and Tall [2012], 431). 53 PLRE 2, 431 (‘Eusebius 18’); Seeck (1907), 1370. 54 There is no citation or discussion of the On the Ceremonies (De caerimoniis) account in the PLRE entry on Leo II (PLRE 2, 664–5 [‘Leo 7’]). 55 PLRE 2, 561–2 (‘Hilarianus 2’); Seeck (1913), 1599; Clauss (1980), 161. Cf. Begass (2018), 146. 56 CJ 12.29.2. 57 CJ 12.25.4. 58 CJ 1.23.6. 59 PLRE 2, 431 and 561, cf. Clauss (1980), 154, 161. 60 CJ 12.7.2. 61 PLRE 2, 561, likewise by Clauss (1980), 161 n.97. 62 CJ 12.40.11.

144

THE IMPERIAL REIGNS OF LEO II

to Hilarianus only reinforces his tenure under the sole reign of Zeno despite the untenable claim of PLRE that this law too should be dated earlier to the joint reign of Leo II and Zeno.63 In other words, Hilarianus was magister officiorum not only under Leo I but also at some point under Zeno. Further, since Hilarianus’ last recorded tenure of the office before that of Eusebius is 470, the only other laws addressed to him by Leo being undated,64 it is very likely that Eusebius had suc­ ceeded him as magister by November 473, possibly as early as 471. Eusebius was still in that post in February 474 at the commencement of the joint reign of Leo II and Zeno. It was Eusebius therefore, not Hilarianus, who was magister officiorum at the coronations of Leo II as Augustus in 473, and possibly as Caesar in 472, and for the coronation of Zeno in 474. At least this detail of the Book of Ceremonies coronation account is vindicated. While Eusebius was magister officiorum in 473 the previous magister Hilari­ anus was leading a Roman army in Illyricum against the Pannonian Goths.65 There is no subsequent trace of him except for the law of Zeno definitely addressed to him as magister officiorum. Since he was already in that position again before the death of Leo II in November 474 he will have replaced Eusebius sometime before­ hand. Moreover, Hilarianus remained in the post when Zeno assumed sole author­ ity after the death of young Leo II. By 9 January 475 Basiliscus had claimed the throne and Zeno was in exile.66 That Basiliscus’ magister was a different person, namely Theoctistus,67 means that Hilarianus was out of office once more. Perhaps he had fled with Zeno. It also means that Zeno’s laws addressed to Hilarianus (assuming that the inscription of one of them once read ‘Hilariano’ not ‘Illyriciano’) can be dated more narrowly to the period between November 474 and early January 475. In summary, the fasti of PLRE.2 for the magistri officiorum therefore need to be amended to accommodate the following sequence: 1 2 3 4

Leo I (457–c.471/3): Hilarianus (Cod.Just. 1.23.6)/Eusebius (?); Leo I/Leo II (November 473–February 474): Eusebius (Const.Porph., de caer. 94); Leo II/Zeno (February–November 474): Eusebius (Cod.Just. 12.29.1)/Hilarianus (Cod.Just. 12.25.4); Zeno (November 474–January 475): Hilarianus (Cod.Just. 12.7.2; 12.40.11). (B) The ‘consulship of Leo’

Seeking to establish when the ceremony in which Leo II was crowned Augustus took place ought to be straightforward because there is a date given in the 63 64 65 66 67

PLRE 2, 561–2, likewise by Clauss (1980), 161 n.97. CJ 12.19.10; 12.59.9. Jordanes, Getica 286–7 with Heather (1991), 264–5. References in PLRE 2, 213 (‘Fl Basiliscus 2’). PLRE 2, 1066 (‘Theoctistus 3’).

145

THE IMPERIAL REIGNS OF LEO II

description – th|/ ou[n pro. dekape,nte Kalandw/n Dekembriw/n evn u‛ pati,a| Le,ontoj tou/ mikrou/). The day and month (17 November), are probably correct. The year, however, cannot be right because it is given as 474, the sole consulship of Leo II. To reject the whole document, with all its circumstantial detail, on the basis of this discrepancy is too extreme a reaction. At least some attempt should be made at explaining the dating discrepancy. The only solution seriously proposed is that the date given refers not to Leo II’s elevation, the very point of the document, but to his death, that is to the ceremonial surrounding his death and burial in Novem­ ber 474.68 To adopt this view requires an assumption that only the date is correct, while the detailed context described in the Book of Ceremonies can be ignored altogether as fictitious and irrelevant. Yet, the description of the coronation is accurate in its ceremonial detail. It is the date that is the problem. Besides, it is more likely to be a textual problem than anything else, that is to say, since the con­ sulship is palpably erroneous it requires emendation in some way. The simplest solution, more often than not the best, is the substitution of mega,lou for mikrou/. This does minimum violence to the text, and recognises the frequent confusion of senior and junior emperors of the same, or similar, name in later Greek and Latin manuscripts. Accepting this emendation would have the effect of redating the cer­ emony from the ‘consulship of Leo the Younger’ (mikrou/) in 474 which makes no sense, to the ‘consulship of Leo the Elder’ (mega,lou) in 473, which makes perfect sense. If so, then we may conclude that Leo II became Augustus on 17 November 473, having been Caesar for over a year.69 Not to accept this proposed emendation or something similar, leaves the date as nonsense. This analysis suggests that the dating details in the Book of Ceremonies’ account of the coronation of Leo II as Augustus which have been found to be problematic can be explained otherwise. We can therefore be more confident that this descrip­ tion is an authentic record of an actual ceremonial event which took place in Constantinople. It begins with the fact that Leo I’s increasing infirmity embold­ ened him to elevate his young grandson to full imperial partnership. Hence, on 17 November in (as proposed above) ‘the consulship of Leo the Elder’ (473) and under the guidance of the magister officiorum Eusebius who had responsibility for such a ceremony, there assembled in the hippodrome the people of Constantinople and the visiting ambassadors ‘for there happened to be many there from various nations’, along with the city’s military contingents bearing their colourful stan­ dards. The soldiers were stationed at the Stama, a covered portico at a level below the imperial enclosure (kathisma) where soldiers traditionally assembled.70 The people chanted in Greek for Leo I to show himself, the soldiers did likewise but in Latin. The emperor duly appeared preceded by the senate. The young Caesar meanwhile remained in the triclinium or inner court of the imperial palace where the emperor usually received senators. With him was the patriarch Acacius. When 68 This curious interpretation appears to originate with PLRE 2, 431 followed by Kent (1994), 109. 69 Accepted without question or explanation by Stein (1949), 361, cf. Feissel (2006), 193 n.38. 70 Janin (1964), 24; Guilland (1969), 451–7.

146

THE IMPERIAL REIGNS OF LEO II

the emperor entered the imperial enclosure of the hippodrome, he stood before his seat and began to greet the soldiers and the people. Evidently appreciating the difficulty the emperor had standing they all chanted ‘sit down, we beg you’ (parakalou/men i[na kaqesqh/|j). Whereupon the ill and ageing emperor Leo thanked his considerate audience and continued with the ceremony while seated. The people then chanted the ‘Augustus’ for Leo I and there were many chants for the crowning of the new boy-emperor. Leo then agreed to proceed with the coronation. Next there was chanting for the magister Eusebius and the patricians to fetch the young Caesar, which the emperor ritually promised, duly despatch­ ing Eusebius and certain patricians to the triclinium. When they arrived back the young Caesar stood on the emperor’s left and the patriarch Acacius on his right. The patriarch offered a prayer to which the audience replied ‘amen’. Then the praepositus, the emperor’s chamberlain, who must have been Urbicius,71 presented him with the crown. Leo stood and placed it on the head of the young emperor saying three times: ‘good luck’ (euvtucw/j / feliciter). The patriarch then departed and the emperor sat down again. The new Augustus then greeted the people and they all chanted in unison ‘Augustus’. Following the acclamations, the Prefect of the City, on the left, and the senate offered the new Augustus the customary gold crown (modi,oloj). Finally, the young (four-year-old) emperor addressed the soldiers himself, promising them each the customary amount of five copper coins (nummi) and one pound of silver as a coronation gift (augustaticum). From this day Leo II held the title and authority of Augustus with his grandfa­ ther Leo I. It was possibly when Leo I was seriously struck by disease in 473, the very point at which the historian Malchus begins his Byzantiaka,72 that he resolved on the elevation of Leo as Augustus. Following the death of Leo I on 18 January 474, although there is uncertainty about this date too,73 the four-year-old Leo II was now sole Augustus. While old enough to possess a distinctive personality, obviously his power was mainly managed and mediated by his parents Zeno and Ariadne Augusta. As Leo’s II’s mother, and the daughter of Leo I, she made her son’s position secure. His grandmother, Aelia Verina Augusta, was clearly keen to retain imperial power and authority for herself, within her own family. Her brother Basiliscus was the senior court general, magister militum praesentalis, in the same position that Aspar had been on the death of the previous emperor Marcian. The 71 The name of the praepositus is not given in the De caerimoniis, but Urbicius was praepositus at the time (PLRE 2, 1188–90 [‘Vrbicius 1’]). He was a distinguished and long-serving chamberlain. Details in: Honigmann (1949/1950), 47–50, 212–13 (‘Notes additionelles’). 72 Malchus, Testimonia 1 (402–3 Blockley) = Photius, Bibl. 78. 73 There are several different dates on offer among the extant writers. Malalas (14.46 [Thurn 300]) says ‘3 February’ (adopted by Bury (1923a), 322) but the correct date must be 18 January (cf. Seeck [1920], 421) as in the Auctarium Havniense Prosperi. ordo posterior (ed. Mommsen, MGH AA, IX, 307) while Theophanes (AM 5966 [120 de Boor]) says merely ‘January’. As corroboration for 18 January, Cyril of Scythopolis notes that Euthymius died on 20 January 473 (vita Euthymii 40 [60.1–4 Schwartz]) and that the emperor Leo I died ‘at the end of the first year after the death of the great Euthymius’ (vita Euthymii 43 [62.5–6 Schwartz]) – not cited in PLRE 2.664.

147

THE IMPERIAL REIGNS OF LEO II

boy-emperor will have been vulnerable against any serious rival, His father would have to be his main protector and Zeno was never popular at Constantinople, neither at court nor on the streets of the city. Be that as it may, giving Zeno equal power with his son was obviously a priority following the death of Leo I. There are different accounts of how Zeno came to be crowned Augustus by his son. The more likely version is that it was a properly planned public ceremonial as had been the case for both Leo I and Leo II. That is, the coronation took place on the advice of the senate (Candidus, v.Dan.Styl.) and was staged in the kathisma in the hippodrome with full ceremonial (Theodore, Theophanes) supervised by the magister officiorum Eusebius.74 There is also the version of an accidental crown­ ing. That is, it is reported that the young Leo was more or less tricked into tak­ ing the crown off his own head and putting it on that of his father Zeno. This is the version recorded by Malalas and those who depend on his account, at least for the Chronicon Paschale75 and the Syriac chronicle tradition.76 It may also be what Victor of Tunnuna meant by saying that the coronation was unusual (contra consuetudinem).77 As for the date, there is yet further uncertainty. The seventh century Continuator of Prosper offers ‘29 January’,78 while Malalas has ‘9 Febru­ ary’ (followed by Theophanes’ simple ‘February’).79 Since the Continuator’s date for Leo I’s death is definitely to be preferred to that of Malalas there is no reason not to do the same for the elevation of Zeno. Hence, 29 January 474 for the coro­ nation of Zeno. In both cases, Malalas’ later dates (3 and 9 February) may well represent not the date of the actual event but of its subsequent announcement in Antioch.80 As sole emperor for the period from 18 to 29 January 474 Leo II’s coinage fol­ lows the pattern of the later coinage of Leo I. Even for such a brief period, there are two patterns for the reverses: one shows the boy-emperor Leo holding the cruciate globe and standing on a low platform; the other, probably later, shows him standing without the platform.81 For the period as joint emperor with Zeno from 29 January the coinage shows the relationship clearly. Leo and Zeno are both Augusti but Leo is depicted as the senior of the two, despite his age.82 These coins 74 v.Dan.Styl. 67 (65.9–19 Delehaye); Theodore Anagnostes, HE Epitome, 400 (112.13 Hansen); Candidus, fr. 1 (466–7 Blockley); Theophanes AM 5966 (120 de Boor). 75 Malalas, Chron. 14.47 (Thurn 299–300); Chron.Pasch. 599.8–13 (Dindorf). 76 For example in Chronicon ad annum 1234 pertinens, 48 (ed. J. Chabot [1937], 146) deriving from the relevant part of John of Ephesus’ Church History, cf. Nau [1897], 460). 77 Victor Tonnenensis, Chron. s.a. 474.2 (ed. Hartmann, CCh 73a, 41 [13] = Mommsen, MGH.AA XI, 188). 78 Auctarium Havniense Prosperi. ordo posterior, s.a. 474.2 (ed. Mommsen, MGH AA, IX, 307). 79 Malalas, Chron. 14.47 (Thurn 300); Theophanes, AM 5966 (120 de Boor). 80 As proposed by Schwartz (1934), 185 n.1. The 9 February date for Zeno’s accession is the one usu­ ally accepted: Seeck (1920), 421, 424 (followed by PLRE 2, 1202 [‘Zenon 7’]); Bury (1923a), 389; Stein (1949), 362 and all works dependent on them for such details, reargued by Feissel (2006), 193. 81 Kent (1994), 109. 82 Kent (1994), 109, summarising some of the disputed points about this coinage.

148

THE IMPERIAL REIGNS OF LEO II

were minted until the death of Leo II towards the end of the year. The same pat­ tern applies to the imperial laws of this period where Leo II appears as the prior emperor.83

Death Leo II was still alive on 10 October 474 the latest date on which a law was issued under his name.84 The usual date given for his death is November 474 and is based on the testimony of John Malalas: The emperor Zeno Kodissos the Isaurian reigned with his son Leo for a short time. The most sacred Leo the Younger became consul in the year 522 according to the era of Antioch of the 12th indiction. But in the 11th month of his consulship the most sacred Leo the Younger fell ill and died, in the month of November of the 13th indiction, in the year 523 according to the era of Antioch, at the age of seven (w]n evniautw/n x ,), as was written by the most learned Nestorianos whose chronicle ended with Leo the Younger (kaqw.j sunegra,yato Nestoriano.j o‛ sofw,tatoj cronogra,foj e[wj Le,ontoj tou/ mikrou/).85 Except for Leo II’s age (5 not 7), there is no need to take issue with Malalas’ infor­ mation here. He is emphatic and clear, confirming his Julian date by both indiction and era of Antioch. Leo II had taken ill and died aged five years sometime in November 474, having reached the 11th month of his consulship. Moreover, Malalas takes the opportunity to cite his source of information for this topic – the lost chronicle of Nestorianos. Theodore Anagnostes writes that Leo ruled with Zeno for 10 months, that is from January to November and this was repeated by Theophanes.86 The circumstances surrounding Leo II’s death also appear uncomplicated. All later writers agree that he died from natural causes.87 Given the tense and complex politics in which he was immersed and over which as a five-year-old child he had little control, it might be expected that he would be vulnerable to ambitious rivals. A likely suspect for any foul play would be his great-uncle Basiliscus. Even so, we do well to accept the plain statements of Malalas that Leo II died from an ill­ ness, probably rather suddenly, which he contracted in November and succumbed 83 CJ 1.14.11 (‘Impp. Leo et Zeno AA’); 2.7.16 (‘Impp. Leo iunior et Zeno AA’); 10.15.1 (‘Impp. Leo et Zeno AA’). 84 CJ 10.15.1. 85 Malalas, Chron.14.47 (Thurn 300). Chron. Pasch. 599.13–16 (Dindorf). The translation is from Jeffreys et al. (1986), 208. 86 Theodore Anagnostes, HE.Epitome. 400 (112.13–14 Hansen); Theophanes, AM 5967 (120 de Boor). 87 v.Dan.Styl. 67 (65.18 Delehaye); Evagrius, HE.2.17; Theophanes, AM 5966 (120 de Boor); Cedre­ nus, 383.7 (Tartaglia 600); Nicephorus Callistus, HE 15.29 (PG 147.85).

149

THE IMPERIAL REIGNS OF LEO II

to on 9 December 474.88 His burial is not recorded but it is assumed that he was interred beside his imperial forerunners in the Church of the Holy Apostles at Constantinople, possibly in the green Thessalian marble tomb of his grandfather.89

Duration How long Leo II’s reign lasted depends on the starting point of one’s reckoning. Later authors as different as the Chronicon Paschale and the Anonymus Valesianus took the purely practical view that he reigned alone for almost a year so ‘one year’ summed up his reign.90 Victor of Tunnuna, closer to the event, says that Leo II was crowned Caesar and then reigned for two years which is about right, although not all that time as Caesar.91 John Malalas appears to have something more precise in mind with his 1 year and 23 days ‘after the reign of Leo the Elder’ (meta. de. th.n basilei,an Le,ontoj tou/ mega,lou). If he means from October 472 when Leo I began to share power with his grandson as Caesar then he is way out; if he means after Leo’s death on 18 January 474 then he is also way out. If, however, if he means after the sole reign of Leo Augustus, that is after 17 November 473 when Leo II became his imperial partner, he is spot on.92 This suggests that Malalas’ figure is exactly right. An additional curiosity, however, usually just ignored as nonsense, is the state­ ment in the Life of Daniel the Stylite that Leo II died after the passing of a ‘three year period’: As time went on it befell that the pious Emperor Leo the Great fell sick and died; he made a good end and left as successor to the throne his own grandson Leo, son of the patrician Zeno. Then the senate convoked a meeting because the emperor was an infant and unable to sign docu­ ments; and they determined that his father Zeno should hold the sceptre of the empire. And thus he was crowned and became Emperor. After three years had passed (cro,nou de. dielqw,ntoj trietou/j) the Lord took the infant, the pious Emperor Leo, into his eternal kingdom; and he went to the land of his fathers, and left the Empire to his father.93 As noted earlier, some of the chronology in the Life of Daniel is muddled but most of it is sound. The vita is based on the direct recollection of Daniel’s 88 Following the calculations of Feissel (2006), 193. 89 Cedrenus 383.7 (Tartaglia 600), with Grierson (1962), 44 n.60. The persistent rumour that Leo did not die but was secretly spirited away to a monastery by his mother Ariadne and a substitute killed by Zeno has been explained elsewhere (Croke [1983], 81–91, cf. Crawford [2019], 106–7). 90 Chron.Pasch. 599.6 (Dindorf) – one year, after the death of Leo I; Anon. Val. 9.39. 91 Victor Tonnennesis, Chron. s.a. 473.1 (ed. Hartmann, CCh 173a, 37 [12] = Mommsen, MGH AA XI 188). 92 As correctly noted by Feissel (2006), 193. 93 v.Dan.Styl. 67 (65.9–19 Delehaye; trs. Dawes and Baynes (1977), 47).

150

THE IMPERIAL REIGNS OF LEO II

contemporaries and disciples, and it is written within a generation of his death.94 The intervals between events recorded in the vita are otherwise correct and there is no reason to believe that the hagiographer is being careless or whimsical in recording, merely in passing, that Leo died in the third year of his reign. Norman Baynes, an astute student of Byzantine authors, was suspicious about this state­ ment and he felt rather uncomfortable in having to dismiss it: ‘The words cro,nou de. dielqw,ntoj trietou/j (185.16) I do not understand. Are they simply a mistake, or do they loosely refer the death of Leo II to the third year of his reign, and was he really crowned in 472 and not, as is usually thought, in 473?’.95 Festugière con­ cluded that it must be a manuscript error, proposing ‘one year’ for ‘three years’.96 The foregoing investigation suggests that Baynes was right to be suspicious. It can be shown that Leo II’s reign did begin in 472, not 473, after all. The vita of Daniel makes perfect sense. If, as proposed here, Leo II’s reign began with his elevation as Caesar in 472, at some time up to October, then his regnal years were as follows:97 I October 472–September 473, II October 473–September 474, III October 474–December 474. Leo II’s death in November 474 therefore occurred in his third regnal year, exactly as the vita of Daniel noted. The suggestion that the vita’s three-year period refers to Leo’s life rather than the length of his reign is unsupportable.98

Finale Leo II enjoyed imperial power for less than a triennium, from just three to five years of age. In that short period the boy-emperor exercised his authority in a variety of modes: (1) as subordinate emperor (Caesar) with his grandfather Leo I; (2) as equal emperor (Augustus) with Leo I; (3) as Augustus for a total of one year and twenty-three days precisely as Malalas recorded it; and, (4) with his father Zeno. Differentiating between each one in this rapid succession of regnal phases must have been difficult enough for contemporaries, let alone for the later writers on whom we are now mainly dependent. The loss of the chronicle of Nestorianos is to be particularly regretted for our purposes here. This study has attempted to make up for that loss in a small way. So, it has focussed on distinguishing between each phase of Leo’s reign and on sharpening our understanding of the sequence of regnal phases. 94 95 96 97 98

Delehaye (1913), 225–7. Baynes (1925), 401 n.1. Festugière (1961), 140 n.134. Cf. Schwartz (1934), 184 n.4 but without detailed explanation. Lane Fox (1997), 193.

151

THE IMPERIAL REIGNS OF LEO II

The story of Leo II’s reigns can now be reinforced, modified and enhanced as follows: 1 2 3 4 5 6

Leo II was most likely born in August/September 469, not 467; he was proclaimed Caesar in 472 (but no later than October), not 473, with the presiding magister officiorum possibly being Eusebius, not Hilarianus; he became Augustus on 17 November 473, not in 474, with Eusebius, not Hilarianus, as the presiding magister officiorum; Leo II became sole emperor (Augustus) when Leo I died on 18 January 474; he became joint-emperor (Augustus) with his father Zeno on 29 January (not 9 February) 474, under the ceremonial supervision of Eusebius once more; he died at five years of age on 9 December 474.

152

6 A R I A D N E A U G U S TA Shaping the identity of the early Byzantine empress*

Between the mid-fifth and the late sixth centuries not a single male born to a Roman emperor at the imperial capital of Constantinople survived into adulthood. Yet it remained the case that only a male could hold imperial authority.1 So it was during this period that the mothers, wives, daughters, and sisters of emperors rose to prominence as the real imperial power-brokers.2 The most significant of them all, it will be argued here, was Ariadne Augusta (ca. 450–515) and it was during her lifetime that the role and identity of the early Byzantine empress was essentially shaped. Apart from being the spouse of two emperors, Zeno (474–491) and Anastasius (491–518), Ariadne was the daughter of one emperor (Leo I, 457– 474), and the mother of another (Leo II, 471–473), not to mention the niece of yet another (Basiliscus, 475–476). In addition, of the two junior emperors or Caesars one (Marcus, 475) was her cousin, the other (Basiliscus/Leo, 476) was the son of another cousin. Another Caesar, Patricius (470–471), was at one stage her brother­ in-law. An Augusta was now a defined position within the imperial structure and Ariadne became the most prominent symbol yet of the power and status that could be achieved by an early Byzantine empress.3 At Constantinople, Ariadne was born into a world which had long boasted a plethora of palaces and places named after imperial women (Helena, Arcadia, Marina, Pulcheria, Placidia).4 If Ariadne’s own mother, the empress Verina, can be considered a ‘lost empress’ whose importance is totally underrated and understudied,5 then all the more so for Ariadne herself. To date she has mainly played a bit part in most histories,6 * This chapter first appeared in G. Dunn and W. Mayer (eds). Christians Shaping Identity from the Roman Empire to Byzantium (Leiden 2015), 293–320. It is reprinted with permission of Brill (Leiden). 1 Holum (1982), 1–5; Laiou (1999), 161–7; McCormick (2001), 146–8; James (2002), 126–9. 2 McCormick (1997), 244–7; Smith (2000), 536; Herrin (2013), 174–5; Croke (2014), 103 ~ Chap­ ter 2, 34 (above). 3 Lucchelli and Rohr Vio (2012), 503. 4 Angelova (2015), 147–60. 5 James (2002), 133. Cf. Laiou (1999), 84; Leszka (1998), 128. More recently: Vallejo Girvés (2017). 6 E.g., Bury (1923a). More recently: Lee (2012), 98–9; 159–60; Mitchell (2014), 123, 126. Ariadne takes up less than a page of PLRE, 2: 140–41 (‘Aelia Ariadne’).

153

A R I A D N E A U G U S TA

while only recently has there appeared the first, but quite inadequate, monograph dedicated to her.7 Meanwhile, there has been a good deal of research on Byzantine empresses generally but without devoting any special attention to Ariadne,8 even with the efflorescence of interest in the reigns of her husbands Zeno and Anasta­ sius.9 Arguably the most significant studies have been by art historians focussed on illuminating the representational dimension of her long imperial life which point to her heightened status and authority as an empress.10 Despite all this work, much of it quite recent, Ariadne’s fundamental significance remains shadowy, not least because of the patchy records for the period of her lifetime. Her role in defining the identity of an independent and influential imperial consort during her nearly sixty years in the imperial palace has never been properly formulated and credited to her. Instead, Ariadne has generally been overlooked as attention has focussed more on her successors: Theodora, wife of the emperor Justinian (527– 565).11 and Sophia, wife of Justin II (565–578),12 whereas she actually paved the way for them.

The family of Leo Ariadne was born about 450,13 probably at Selymbria (Silivri) west of Constanti­ nople where her father Leo, who was a senior military officer, was then stationed. Her mother Verina was in her late twenties but otherwise unknown.14 Being merely the daughter of a soldier Ariadne enjoyed no imperial or aristocratic con­ nection although Leo was a close associate of the powerful senior court general Aspar and possibly of Marcian, another fellow soldier who succeeded Theodosius II as emperor in the same year Ariadne was born (450). When, strongly backed by Aspar, Leo himself was chosen as emperor on the death of Marcian (457) he 7 Magliaro (2013), preceded by the first serious scholarly article on the empress by Meier (2010), 277–91. 8 Most notably by Judith Herrin whose several sophisticated studies are now collected in Herrin (2013). Note also Garland (1999); Herrin (2004). 9 There are several minor studies plus four major monographs: Kosinski (2010); Crawford (2019) on Zeno; Haarer (2006); Meier (2009) on Anastasius. 10 James (2001); Angelova (2004), 1–15; McClanan (1996), 50–72 and (2002); see also Connor (2004), 48–9, in the wider context of artistic patronage by imperial women. 11 Among the welter of literature on Theodora where she is generally treated as unique and excep­ tional some treatments stand out for contextualising her in terms of bias in surviving records and the significantly developed role of the empress by her predecessors: Foss (2002), 141–76; Leppin (2002), 438–41. See also Allen (1992), 93–103. 12 A role first highlighted by Averil Cameron (1975a), 5–21, rp. Cameron (1981); Garland (1999), 40–57. 13 Siebigs (2010), 235 n.181, as does Kosinski (2008), 211 n.14 before him, unnecessarily places Ariadne’s birth in 455 on the strength of Marcellinus’ ‘60 years’ (Chron. 515.6 [ed. Mommsen, MGHAA 11 [1894], 99]), but the chronicler is referring to the duration of her period in the palace (in palatio) not her whole life. 14 Verina has been taken to be a woman of barbarian background but the argument is too tenuous: Siebigs (2010), 750–68.

154

A R I A D N E A U G U S TA

moved into the imperial palace at Constantinople with his wife and young daugh­ ter. Either that same year, or not long after, another daughter named Leontia was born in the palace.15 In 463 a younger brother, and potential emperor, was also born but lived only a few months.16 Had the young boy grown to manhood he would have become emperor in succession to Leo, and Ariadne might never have come to enjoy the power and status of an emperor’s consort.17 For most of the 460s Verina and her young daughters enjoyed the privileged court life with its routine of ritual, processions, banquets, and liturgies. On one occasion she met the local holy nun Matrona and sought her blessing on herself, her husband, and her young daughters.18 The girls’ education in Greek and Latin but also in the Scriptures took place inside the palace where they shared a common tutor.19 Otherwise, the differences between the sisters in terms of age, dispositions, and predilections remain a mystery.20 At the same time, Leo was preoccupied with the politics and administration of the imperial court. In particular, he had to deal with Aspar’s expectations about connecting his own family more permanently with that of the emperor through his daughters. Evidently some sort of pact between the two was settled which involved Leo promising that in due time he would marry Ariadne to Aspar’s son Patricius thereby making him a likely imperial successor to Leo. As events unfolded, rela­ tions between Aspar and Leo progressively deteriorated and Leo sought to make himself less dependent on Aspar by promoting others such as his brother-in-law Basiliscus and an Isaurian named Tarasicodissa, then Zeno, who both became senior generals. As they reached marriageable age Ariadne and Leontia also had to face the challenging uncertainty of identifying a suitable husband. Ariadne will have known, for instance, the young Gothic prince Theodoric who was growing up with her in the palace21 but by the time he returned to his people in 469 she had married Zeno. At last in 470 Leo was forced to concede his original promise to Aspar, so Patricius became Caesar and was married not to Ariadne as originally proposed but to her younger sister Leontia. The tension between Aspar and Leo was finally released in June 471 when Leo arranged for the execution of Aspar and his sons. Patricius was allowed to escape but in doing so forfeited his marriage to Leontia. Leo now reclaimed authority for himself and his family.22 Central to Leo’s method in protecting Ariadne and Leontia, while dealing with the personal and political dominance of Aspar in the 460s, was developing sources of power and influence beyond the reach of Aspar whose adherence to the Arian affiliation put him outside the mainstream orthodoxy of the court at a time when 15 16 17 18 19 20 21 22

PLRE 2, 667 (‘Leontia 1’). PLRE 2, 664 (‘Leo 6’). Magliaro (2013), 58. Vita Prima Matronae 32. English translation is in Featherstone (1996), 13–46. Magliaro (2013), 51. Magliaro (2013), 67. PLRE 2, 1078 (‘Theodericus 7’). For detailed background and interpretation see Chapter 3.

155

A R I A D N E A U G U S TA

doctrinal preference increasingly mattered. First, Leo tried but failed to have the body of Simeon Stylites relocated to Constantinople as a source of protection,23 but how he and his family came to channel and control this distinctive power is highlighted by the progressive impact of Simeon’s imitator, the Syriac-speaking holy man Daniel who was set on his pillar at St Mamas (Bešiktaš) up the Bospo­ rus. Empress Verina and her daughters will have at least seen and been aware of Daniel and may have accompanied Leo on visits or separately.24 This identity of the pious imperial woman had been cultivated especially by Pulcheria, daughter of Theodosius II and wife of Leo’s predecessor Marcian. Leo is said to have paid special honour to Pulcheria by having her likeness placed on her tomb and by duly observing the annual liturgical commemoration of her life. In addition, he would draw attention to her picture in the imperial palace as the model of a blessed life.25 Whether she liked it or not, Verina was a new Pulcheria. Another source of the new power of the holy at Constantinople which Ariadne experienced was the introduction into the city of relics of the Virgin Mary follow­ ing her newly defined status as Mother of God (Theotokos) which Pulcheria had promoted in particular. Verina was responsible for the reception into Constanti­ nople of the first Marian relic, the veil or mantle of the Virgin. She had it deposited in a newly constructed shrine of the Virgin at Blachernai, just outside the city walls along the Golden Horn. The reception of the relics at Blachernai would have given rise to a busy liturgical scene rather like that depicted on the famous Trier Ivory.26 Likewise, Verina takes credit for the Virgin Church of the Chalkoprateia.27 Found­ ing, or refounding, a church in a God-dominated society such as Constantinople was a demonstration of singular power.28 Such power is evident from the confused tenth century description of an image in the Blachernai church which was set in gold and coloured mosaic. It is usually taken to be a genuine dedicatory image from the 470s depicting, as recorded, ‘Our Lady the immaculate Mother of God seated on a throne and on either side of her Leo and Veronica [read ‘Verina’], the latter holding her own son, the young emperor Leo and she falls before Our Lady the Mother of God and also their daughter Ariadne’.29 If authentic, this would be the earliest visual testimony to Ariadne. 23 Syriac Life of Simeon 128 (Doran [1992], 194). 24 v.Dan.Styl. 38, 46, 51, 55, 63, 65, and 92 (relics brought by Leo from Babylon of Ananias, Azarias, and Misael), (ed. Delehaye (1923), 35, 44, 49, 54, 62, 64, 86 [relics brought by Leo from Babylon of Ananias, Azarias and Misael]). 25 Parastaseis 45 (Preger [1901], 1.52) with English translation in Averil Cameron and Herrin (1984), cf. Holum (1982), 227–8. 26 For example, see Holum and Vikan (1979), 113–33. There are several uncertainties around the Marian relics at Blachernaae (Shoemaker [2008], 53–74). 27 Justinian, Novel 3.1. In both cases, Verina may well be the original builder despite later ascription to Pulcheria (James [2011/12], 65–8). 28 Laiou (1999), 165. 29 Text in Cod.Par.Gr.1447, fols. 257–8 with translation of Mango (1986a), 35. The problem of identification concerns the young boy said to be Ariadne’s son Leo II that may well be accurate. Often it is claimed that the young boy is the anonymous short-lived son of Leo and Verina who

156

A R I A D N E A U G U S TA

While the date of Ariadne and Zeno’s marriage is disputed, it must have occurred in 468,30 by which time Leo would have made Zeno co-emperor were it not for local opposition.31 The offspring of the marriage was born the following year (469), a young boy named Leo after his imperial grandfather.32 They had no children after Leo, that is, discounting the claim for another son33 and the subsequent tales in several languages of the two daughters of whom one secretly left Constantinople and ended up disguised in a male monastery in Egyptian Scetis but was summoned at one stage to the capital as the one most likely to cure her younger sister.34 Ariadne’s sister Leontia, once freed from her shortlived marriage to Patricius, following the massacre of his family in June 471, was soon united with Marcian, son of Emperor Anthemius (467–472) and grandson of Emperor Marcian.35 Marcian may well have been Leo’s first choice of husband for his daughter all along, while his consulship in 472 would have reinforced his new status. Yet Leo I was the origin of the imperial authority inherited by his wife and daughters. Moreover, he was responsible for making Verina, then Ariadne, the first empresses to be financially independent, with all that flowed from that. It was during this latter part of Leo’s reign that the imperial treasury, the res priuata, which was continually replenished and expanded through rents on imperial estates and various other taxes, was divided into two: one part for the emperor and his household, another for ‘our wife, the most serene Augusta’ and her household.36 This was a most significant development which enabled Ariadne to accumulate, over the coming decades, unprecedented authority and independence for an empress.

30 31 32 33

34 35 36

was perhaps called ‘Leo’. If so, it is odd that Ariadne’s younger sister Leontia is absent because she was born before their brother. See further: Lane Fox (1997), 189–90. For critical background: Wortley (2005), 177–83 (unduly sceptical); Shoemaker (2008), 53–74. The case for 468 is made by Kosinski (2008), 210–11. Cf. Magliaro (2013), 66, superseding the case for an earlier date argued in Croke (2003), 561–3 (see Chapter 5, 135–38, above). Candidus, frag. 1 (Blockley, 466–7). PLRE 2, 664–65 (‘Leo 7’); and Croke (2003), 559–75 with Kosinski (2008), 223. Rather than Ariadne, as sometimes claimed (e.g., by Treadgold (1997), 158, 164, 927 n.6), the soli­ tary reference to another son, also called Zeno (Malchus, frag. 8 [Blockley, 2.414–15]), suggests that he was born of Zeno’s previous wife Arcadia (cf. PLRE 2, 1198 [‘Zeno 4’]). A potential co­ emperor, and designated imperial heir, could be expected to have left more traces in the historical record. Details in Drescher (1947), 69–82. A daughter of Zeno also features in the legends related to St Menas (Budge [1909], 44–58). Another possibility is that these were actually the daughters of Ariadne’s sister Leontia and her husband Marcian (Malalas, Chron. 14.46 [Thurn 299.10]). PLRE 2, 717–18 (‘Fl. Marcianus 17’). By 472, there were separate palace officials (decani) and couriers for the empress (CJ 12.59.10.3), with separate parts of the imperial treasury for the emperor and, referring to Ariadne, ‘our wife the most serene Augusta’ (CJ 10.32.64 [485–486]). Cf. ‘most pious Augusta’ (CJ. 10.32.66 [497–498]); details in Jones (1964), 424–5, cf. Bolognesi Recchi-Franceschini (2008), 244 and for comparable resources for early Roman empresses such as Livia: Angelova (2015), 144.

157

A R I A D N E A U G U S TA

Imperial power of Verina and Ariadne, 465–476 By 474 Ariadne was a capable young woman who had already spent most of her life close to power and influence in the palace. With her father’s health fading her mother Verina was keen to ensure dynastic continuity and arranged for the elevation to the throne of Ariadne’s young son and her grandson as Leo II, first as Caesar in October 472 then Augustus on 17 November 473.37 Grandfather and grandson ruled together for several months. Leo then died (18 January 474) whereupon Verina was no lon­ ger an emperor’s wife but his grandmother, while the imperial daughter Ariadne now became an imperial mother and her sister Leontia an imperial aunt. This sudden but subtle shift in status and power between mother and daughters who potentially shared equal power derived from Leo I was to have significant ramifications. If not before, Ariadne now became an Augusta and acquired all the concomitant privileges of that title.38 Young Leo II was not himself a healthy child and not necessarily destined for a long reign. This potential threat to the power of Verina and Ariadne was clearly perceived and the optimal solution of also making Ariadne’s husband Zeno emperor soon presented itself. Leo II therefore crowned his father Zeno on 29 January 474. When young Leo II died in November 474 Ariadne’s position changed once more, from imperial mother and wife to just imperial wife, while Verina was said to be the instigator of Zeno’s enthronement but evidently soon regretted her action. Leontia’s husband Marcian, meanwhile, remained a more than eligible emperor. Verina, so it would seem, was keen to elevate her new lover, the magister officiorum Patricius, as the replacement of Zeno39 but others manoeuvred in favour of her brother Basiliscus.40 In any event, relations between Verina and Ariadne clearly became strained at this point. Although Ariadne and Zeno still occupied the palace, as did Verina, he was faced with an untenable position and, while absent from the city on an imperial expedition to Chalcedon across the Bosporus, Verina took the initiative. Thereupon, Zeno fled from Chalcedon to his native Isauria. Whether Ariadne accompanied him in flight or followed later is a moot question.41 Fleeing with them was a number of Isaurians at Constantinople while substantial treasury assets were also relocated. Then, acting on the authority of an Augusta and as widow of Leo I, Verina had Basiliscus crowned emperor on 9 January 475 at the Hebdomon outside the city walls. The crowd acclaimed: ‘Long life to Verina the orthodox Helena’. Verina was not just a new Pulcheria but also a new Helena, identified with the mother of Constantine as a woman of public piety and orthodoxy.42 Ariadne and her mother were clearly now estranged. 37 38 39 40 41

For the dates: Croke (2003), 563–72 (Caesar in 472); Kosinski (2008), 212–14 (Caesar in 473). Crawford (2019), 104–5. PLRE 2, 838–39 (‘Patricius 8’). PLRE 2, 212–14 (‘Fl. Basiliscus 2’). Marcellinus, Chron. 475.1 (ed. Mommsen, MGH.AA, XI, 91); v.Dan.Styl. 69 (Delehaye [1923], 80); and Theophanes, Chron. AM 5967 (de Boor 120). 42 Parastaseis 29 (Preger [1901], 1.37). On the enduring significance of Helena as an imperial example see Brubaker (1997), 52–75.

158

A R I A D N E A U G U S TA

Basiliscus, meanwhile, fearing that Verina would opt for her lover Patricius at the first opportunity, had him killed, which instantly alienated Basiliscus from Verina and probably their nephew Armatus. Turning a blind eye to the local violence at Constantinople against Isaurians alienated the now powerful Isaurian courtier Illus, while the emperor’s anti-Chalcedonian encyclical aroused the ire and action of the local people.43 Within a year most of the same alliance which had conspired to force Zeno and Ariadne out of Constantinople were united in seeking to secure their rapid return. By the end of August 476 Zeno and Ariadne formally entered Constantinople once more and reoccupied the palace where Ariadne Augusta was to spend the next thirty-eight years of her life. Again she resumed the duties and position of an imperial wife. She will have supported Zeno in the proclamation as Caesar in the hippodrome of Armatus’ son Basiliscus, also known as Leo. Zeno soon grew suspicious of Armatus and had him murdered in the Kochlias, the narrow staircase leading from the palace up to the hippodrome. Ariadne, however, capitalising on her own authority, protected the young Basiliscus by getting him ordained and smuggled out of the city. The former Caesar lived well into the reign of Justinian and the story of his background as a relative of Ariadne became confused with that of her son Leo II.44 Even now Zeno and Ariadne were not fully secure on the throne. In 479 another rebellion was afoot. This time it involved not Ariadne’s mother but her sister Leontia. Ariadne might be the current empress but her sister considered her claim superior because she was ‘born in the purple’, that is while her father Leo I was actually emperor. In the raging battle at Constantinople Marcian almost succeeded in replacing Zeno who had been driven back to the safety of the palace where he stood firm, doubtless reinforced by Ariadne as Theodora was to do for Justinian during the Nika riots in 532. Outside, Marcian was opposed by Illus and his forces arriving from Chalcedon and his coup failed at the last minute whereupon he took to the high ground and the sanctuary of the Church of the Holy Apostles. He was later removed from there, forcibly ordained a priest by Patriarch Acacius and exiled to Cappadocia, later to Isauria.45 By then Leontia had joined him.46 Verina too was in exile in Isauria, having incurred the wrath of Illus the previous year (478), and she evidently backed and encouraged Leontia’s claim.47 In using his wife so blatantly in the contest for power Marcian highlights the very significance of female imperial power whose limits were now being tested,48 but enjoyed alone at Constantinople by Ariadne Augusta.

43 44 45 46 47 48

Redies (1997), 211–21. Croke (1983), 81–91. Theodore Lector, HE 419–20 (Hansen 116.10–19). John of Antioch, frag. 234.4 (Mariev 430–1). John of Antioch, frag. 237.2 (Mariev 434–6). Dagron (2003), 43.

159

A R I A D N E A U G U S TA

Ariadne as empress, 476–491 By now, Illus had put himself in an unassailable position because he held, as hostages in Isauria, Ariadne’s mother (Verina), sister (Leontia), and brothers-in­ law (Marcian and Longinus). Yet, Ariadne was not dissuaded from standing up to Illus. She had recent first-hand knowledge of the life of an exile in the mountain­ ous cold of Isauria so sometime in 480, on receipt of a begging letter from her mother which doubtless complained about the trials of her Isaurian exile, Ari­ adne pressured Zeno to have Verina restored to Constantinople. ‘Ask the patrician Illus about her’, was the emperor’s reply. Summoning Illus to her presence, as an empress could do, she made her tearful petition. The enigmatic reply of Illus, ‘You are seeking to make another emperor instead of your husband’, only infuriated Ariadne further. Whether or not Illus was insinuating that Ariadne had a preferred alternative to Zeno is unclear. In any event, it underscores the power of an Augusta by 480 to make and unmake an emperor even if she was not otherwise actively involved in imperial decisions and deliberations. Ariadne remained a risky suspect to Illus who denounced her to his close friend the emperor Zeno whose response was to authorise his wife’s assassination in the separate palace quarters where she lived. Relations between Ariadne and her husband had never been good but this was clearly a point of no return. Still, the empress retained the independence of her own financial resources, household, and loyal staff and so now avoided her fate by substituting the chambermaid for herself in her bed then fled to Patriarch Acacius for refuge. Zeno, presuming he was now a widower, went into mourning only to be interrupted by the patriarch seeking security for the empress’ safety. Ariadne sought immediate vengeance on Illus with an ultimatum: ‘Either Illus stays in the palace or I do’, to which Zeno replied, ‘I want you. If you can do anything to him do it’.49 So, Ariadne now had no compunction in resolving on cold-blooded murder. Through the bedchamberlain Urbicius she organised for the assassination of Illus, now conscious that he had also planned to do away with her if he could.50 In the confused and bungled attempt on Illus’ life, as he was processing through the Kochlias, the same narrow staircase where Armatus was cut down in 476, he escaped with just a severed ear. Shortly after, Illus excused himself from the imperial capital, was made magister militum per Orientem and relocated with imperial sanction to Antioch. From there he had easier access to his native Isauria and the Cilician coast. Through this whole episode Ariadne had shown herself an independent woman of strength and steely calmness, feared by both her husband Zeno and the most powerful court officials. She had also demonstrated a far more ruthless streak than her husband51 and had achieved what Verina could not, 49 Malalas, Chron. 15.13 (Thurn.311): καὶ λοιπὸν ἡ Αριάδνη εἶπε τῷ Ζήνωνι· ἢ Ἰλλοῦς ἐστιν εἰς τὸ παλάτιον, ἢ ἐγώ. ὁ δὲ Ζήνων λέγει ὅτι·εἴ τι δύνῃ, πρᾶξον. ἐγὼ σὲ θέλω.’, and Jordanes, Romana, 349–51 (Mommsen, MGH.AA, 5/1, 45). Cf. PLRE 2, 586–90 (‘Illus 1’). 50 Kosinski (2010), 126–7. 51 Cf. Magliaro (2013), 117–18.

160

A R I A D N E A U G U S TA

namely, the banishment of Illus from Constantinople and his powerbase at the imperial court. Zeno himself clearly distrusted Illus by now, doubtless fuelled by Ariadne, and when Illus refused to release Zeno’s brother Longinus the emperor stripped him of his position and installed a new general at Antioch, John the Scythian.52 In 484, knowing Calandion the bishop of Antioch stood behind him and against Zeno because of his unpopular Henotikon decree,53 Illus rebelled. Zeno was then forced to send the magister militum per Thraciam Leontius with an army to con­ front Illus but Leontius was persuaded to change sides and agreed to be nomi­ nated as emperor himself. The imperial coronation was only made possible by the authority of Verina who was escorted from her exile in Cherris to Tarsus where, however unwillingly, she proclaimed the new emperor on 19 July 484. In her rescript to the Antiochenes and other provincial capitals authenticating her choice of emperor she wrote: The Augusta Verina to our governors and Christ-loving people, greet­ ings. You know that the empire is ours and that after the death of my husband Leo, we appointed as emperor Trasakalissaios, subsequently called Zeno, so that our dominion would be improved. But now see­ ing that the state is being carried backwards as a result of his insatiate desire, we have decided that it was necessary to crown for you a Chris­ tian emperor embellished by piety and justice so that he may save the affairs of the state and that wars be stilled. We have crowned the most pious Leontius as emperor of the Romans who will reward you all with his providence.54 These were potent claims, eclipsing those of Pulcheria in appointing Marcian in 450, and they had been reinforced by her earlier proclamation of her brother Basiliscus (475) and her backing for the usurpation of her son-in-law Marcian (479).55 That Ariadne retained and augmented her power in the face of clear provocation and political insecurity on all these occasions can be attributed, at least partly, to her own personal resolution and capability. Ariadne knew her way 52 PLRE 2, 602 (‘Ioannes Scytha 34’). 53 Evagrius, HE 3.16. 54 Malalas, Chron. 15.13 (Thurn 314): ‘Αἰλία Βηρίνα ἡ ἀεὶ Αὐγούστα Ἀντιοχεῦσι πολίταις ἡμετέροις. ἴστε, ὅτι τὸ βασίλειον μετὰ τὴν ἀποβίωσιν Λέοντος τοῦ τῆς θείας λήξεως ἡμετερόν ἐστιν. προεχειρισάμεθα δὲ βασιλέα Στρακωδίσσεον τὸν μετὰ ταῦτα κληθέντα Ζήνωνα, ὥστε τὸ ὑπήκοον βελτιωθῆναι καὶ πάντα τὰ στρατιωτικὰ τάγματα. ὁρῶσι νῦν τὴν πολιτείαν ἅμα τῷ ὑπηκόῳ κατόπιν φερομένην ἑκ τῆς αὐτοῦ ἀπληστίας ἀναγκαῖον ἡγησάμεθα βασιλέα ὑμῖν στέψαι εὐσεβῆ διακαιοσύνῃ κεκοσμημένον, ἵνα τὰ τῆς ̔Ρωμαϊκῆς πολιτείας περισώσῃ πράγματα καὶ τὸ πολέμιον ἥσυχον ἄξει, τοὺς δὲ ὑπηκόους ἅπαντας μετὰ τῶν νόμων διαφυλάξͅῃ· ἐστεψαμεν Λεόντιον τὸν εὐσεβέστατον, ὅς πάντας ὑμᾶς προνοίας ἀξιώσει.’ Slightly fuller in Theophanes, Chron. AM 5974 (de Boor 129) with translation of Mango and Scott (1997), 198. 55 Laiou (1999), 165.

161

A R I A D N E A U G U S TA

around the corridors of power and her relations with court and army sustained her. If Zeno now wavered, as in 479, she was not for turning. In the end, Leontius’ usurpation had all the hallmarks of a narrow and hastily assembled rebellion rather than a calculated plan on the part of Illus which had been years in the making.56 Defeated by Zeno’s army near Antioch in September 484 they retreated once more into Isauria to the fort of Cherris-Papyrion, then blockaded by the Roman army for four years before being captured and behead­ ed.57 Ariadne perhaps gazed upon the grisly sight of the heads of Leontius and Illus which were paraded around the hippodrome on poles as a victory procession before been taken across to Sykai for public spectacle.58 Meanwhile Ariadne had not seen her mother for years and was never to see her again. However, she knew from her example and position that an imperial wife and widow was now a key repository of political power and influence. Verina had died in exile in 484 and her body was transported to Constantinople only in 488 where Ariadne will have played a key role in her burial in the Church of the Holy Apostles.59 Since 484, the mantle of Verina, the supreme authority of an Augusta, now lay on Ariadne’s shoulders.

Ariadne and Anastasius, 491–515 When Zeno died on 9 April 491 there was a clear expectation that, as Augusta, Ariadne would properly take charge of proceedings, as Verina had done in 484. On the evening of 10 April she summoned to the palace the leading imperial officials and senators along with the city’s patriarch Euphemius. Meanwhile the soldiers and the people of Constantinople, with the circus factions all in their appropriate places, had gathered in the hippodrome to await news of a successor but also to guarantee an outcome.60 As part of this reception Ariadne assured them that ‘even before your requests we gave a command to the most glorious office-holders and the sacred senate with the common consent of the most noble, to choose a man who is Christian, Roman, and endowed with every imperial virtue’.61 They replied 56 The subtle but ultimately unlikely case made by Kiel-Freytag (2010), 291–301. 57 PLRE 2, 670–71 (‘Leontius 17’). The details of the campaign against Leontius and Illus in 484 followed by their retreat to Papyrion are uniquely found in the local history of Joshua the Stylite (pseudo-Joshua, Chronicle 13–17 (trs. Trombley and Watt [2000], 12–16). 58 Malalas, Chron. 15.14 (Thurn 315.64–6). 59 John of Antioch, frag. 237.12 (Mariev 442–3). 60 For the accession of Anastasius: Lilie (1995), 3–12; Dagron (2003), 65–8; Haarer (2006), 1–6; Meier (2009), 63–75. The author of the Life of Daniel reports hearing Daniel prophesy that after Zeno’s death Ariadne would reign over the empire in conjunction with Anastasius (v.Dan.Styl. 91 [Delehaye (1923)], 86). 61 Const. Porph., On the Ceremonies 92 (Reiske 418–19, trs. Moffatt and Tall [2012], 418–19): ὅτι καὶ tῶν ὑμετέρων αἰτήσεων ἐκελεύσαμεν τοῖς ἐνδοξοτάτοις ἄρχουσι καὶ τῇ ἱερᾷ συγκλήτῳ μετὰ κοινῆς τῶν γενναιοτάτων δοκιμασίας ἄνδρα ἐπιλέξασθαι Χριστιανὸν Ῥωμαῖον καὶ πάσης γέμοντα βασιλικῆς ἀρετῆj. The detailed record of these events was set down by Peter the Patrician in the sixth-century and copied in the tenth-century De ceremoniis for Emperor Constantine VII

162

A R I A D N E A U G U S TA

by insisting the choice of emperor belonged to Ariadne. Her chosen emperor Anastasius was required to swear an oath that he would not use his new position to harbour grudges against those with whom he had previous dealings and that he would rule conscientiously. It was presumably at this stage of the process that Ariadne insisted with Patriarch Euphemius that an oath of adherence to Chalcedo­ nian orthodoxy also be solicited from Anastasius and deposited in the patriarchal archives.62 Anastasius’ relatively lowly official position made him an unexpected choice of emperor. He was the head of the imperial ‘silentiaries’, those who serve the court as officials responsible for regulating access to the imperial presence and pre­ serving its dignified silence. Obviously, contemporaries had to explain Ariadne’s choice for themselves. The one person expected to succeed Zeno was Longinus his brother and the brother-in-law of Ariadne.63 She will have been sensitive to the mood of the city and court that another Isaurian emperor, especially the brother of Zeno could not be countenanced so Longinus was quickly despatched, along with his family.64 This was no less than the bold act of a powerful empress. Other senior generals, as Marcian had been in 450, as well as current and former impe­ rial officials and senators, might also have seen themselves as the next emperor. Rumours therefore swirled including that Anastasius had long been Ariadne’s secret lover,65 perhaps as Patricius is said to have been for Ariadne’s mother Verina. The fact remains that Anastasius proved to be more than a competent emperor despite periodic opposition and revolts generated by his ecclesiastical and economic actions. Both Ariadne and the imperial bed-chamberlain Urbicius had already spent decades in the imperial palace and clearly understood that Anas­ tasius would make a good emperor, a deliberate and adroit choice. Indeed, the imperial power and influence overtly wielded by Ariadne (Zeno in 474, Anasta­ sius in 491) and later Sophia (Tiberius in 578), as the maker of emperors following the earlier example of Pulcheria in 450 and Verina in 475 and 484, arguably raises them above Theodora. It was as an imperial widow and Augusta that Ariadne selected Anastasius to be emperor and participated in his installation on 11 April 491. Only a month later did she marry him. While Ariadne was around forty years of age at the time Anastasius was in his early sixties. If she had a precedent for her actions it was the example of Pulcheria in 450 who selected Marcian as emperor before they celebrated their nuptials. Ariadne may have been very conscious of this herself

62 63 64 65

Porphyrogenitus. Note that for ‘esteemed/highly esteemed archons’ the words ‘glorious/most glo­ rious office-holders’ have been substituted to reflect more accurately the technical titulature of the original eudoxos/eudoxotatos or gloriosus/gloriosissimus, while sixth-century use of ‘archon’ is translatable as ‘office-holder’. Theodore Lector, HE 447 (Hansen 125.26–126.15); Evagrius HE 3.32; and Victor Tonnenensis, Chron .s.v. 491.1 (MGHAA 11.191–92), with Kosinski (2012), 66–9. Evagrius, HE. 3.29. Haarer (2006), 24 n.72. Pseudo-Zachariah of Mytilene, HE 7.1 (Brooks, 2.17–20), with Haarer (2006), 4.

163

A R I A D N E A U G U S TA

if that is the explanation for the commemorative gold coinage struck to celebrate the occasion. ‘FELICITER NUBTIIS’ was the legend and its iconographical pat­ tern was clearly modelled on the similar one which commemorated the union of Pulcheria and Marcian.66 On this coin the bride and groom, dressed in full impe­ rial regalia, stand either side of Christ who blesses and symbolises the sanctity of their union. Yet, it is only Ariadne who wears the full imperial crown with hanging prepondoulia which is thought to represent the fact that in effecting this union she is the more powerful one because the bridegroom is her choice.67 The marriage of Ariadne and Anastasius will also have produced a marriage ring to be passed out as gifts to courtiers and aristocrats. None survives unless the one discovered at Trebizond (now in Dumbarton Oaks) in the same treasure as the marriage solidus represents Ariadne and Anastasius.68 In that case the difference in imperial cos­ tume between Ariadne (crowned) and Anastasius (not crowned) is even starker.69 The coin, combined with the ring, produced in 491 for the new imperial marriage were part of the propaganda immediately promoting the stability and continuity of imperial power enabled through Ariadne. Following her key role in the accession of Emperor Anastasius and her subse­ quent marriage to him, Ariadne largely disappears from view for over a decade. While the emperor and his court, including a relative of Ariadne,70 were focussed on a protracted war with the rebels in Isauria (492–498), then a more extensive engagement against the Persians on the imperial frontier (502–505), Ariadne remained actively occupied in the life of the court and the city. The daily routine and ritual of an imperial life make less impact on the subsequent record than the wars and riots of the reign. It is not necessarily true that, in contrast to her role with Zeno, she was now deliberately shunning the limelight.71 Certainly she was associated with the emperor as a target of hostile opposition in Constantinople in 493 when, according to an eye-witness, ‘statues of the emperor and the empress were bound with ropes and dragged through the city’ by the violent crowd.72 It was also somewhere in this period that she proposed to her husband that the patri­ cian Anthemius be made praetorian prefect, an offer steadfastly rejected by the emperor on the dubious grounds of insufficient education,73 just as Theodora was 66 67 68 69 70

Bellinger (1992), 4; Hahn (1984), 101–6. Walker (2010), 852–5. Cf. Brubaker and Tobler (2000), 851–2. Ross et al. (2005), 6–57 (item number 66). Walker (2010), 857. Malalas, Chron. 16.3 (Thurn 320), mentions in passing that Anastasius’ general Diogenianus was a ‘relative of the Augusta’. Precisely how they were related is very unclear. That he first appears in 492 and was still a senior general thirty years later (PLRE 2, 362 [‘Diogenianus 4’]) suggests he was of the next generation to Ariadne. If so, he was perhaps the husband of one of her nieces, daughters of Leontia and Marcian. 71 As suggested by Meier (2009), 311. 72 Marcellinus, Chron. 493.1 (ed. Mommsen, MGH.AA 11.94). Perhaps one of them, later replaced where it had originally stood, was the full-length statue of Ariadne that was still located near the palace entrance in the eighth century (Parastaseis 32 and 80 (Preger, 1.38.5 and 1.70.13). 73 John Lydus, On the Magistrates 3.50, with Meier (2009), 120.

164

A R I A D N E A U G U S TA

rebuffed by Justinian in objecting to his restoration of John the Cappadocian in 532. Anthemius was the son of an emperor himself and his brother Marcian had been married to Ariadne’s sister Leontia. Even though he was part of Marcian’s unsuccessful revolt against Zeno in 479 after which he fled to Rome, Anthemius remained loyal to Ariadne and Anastasius, becoming consul at an advanced age in the year of Ariadne’s death (515).74 Indeed, her image appears on his now lost consular diptych.75 Anastasius and Ariadne were also responsible for the construc­ tion or reconstruction of a Church of St Euphemia in Petra at Constantinople that does not survive although one can imagine the dedication and its inscriptional record.76 Other churches attributed to Anastasius and Ariadne as joint dedica­ tions are those of St Michael the Archangel in the Nea,77 and the Forty Martyrs near the Baths of Constantine, near the bronze tetrapylon,78 plus the church of Elias attributed to Ariadne and Zeno.79 For Ariadne, as for Pulcheria and Verina before her and Theodora afterwards, church building symbolised not only her piety but also her political power and patronage consolidated by her association with Anastasius.80 Of particular interest is Ariadne’s involvement in ecclesiastical matters. Through the theologically turbulent period of the reign of her husband Zeno, and despite his failed attempt at unity through his Henotikon decree in 482, which sim­ ply avoided mention of Chalcedon, she remained a firm Chalcedonian. Even the challenge of Anastasius who was increasingly seen as having anti-Chalcedonian sympathies, if not an outright Monophysite agenda, did not cause her to waver. The oath which she and Euphemius made Anastasius take on his inauguration was constantly under pressure in the fraught relations between emperor and patriarch culminating in the exile of Euphemius in 496 on a trumped up charge of collud­ ing with the Isaurian enemy.81 Ariadne must have been pained by this but would remain loyal to her husband, just as she was on Friday 22 July 511 when we catch a glimpse of her at the Hebdomon where the imperial couple were then in summer residence. On this occasion, probably at the Church of John the Baptist, Euphemius’ patriarchal successor Macedonius had evidently aroused the concerns of the empress sufficiently enough for her to join Anastasius in refusing to take communion from the patriarch’s hand.82 The estrangement between Macedonius and Anastasius was now irreconcilable and the patriarch was exiled one night the following week (7 August 511). 74 75 76 77 78 79 80 81 82

Assuming that PLRE 2, 98 (‘Anthemius 5’) and 99 (‘Procopius Anthemius’ 9) are the same person. Delbrueck (2009), 217–19. Details in Janin (1969), 126–7. Patria 3.181 (Berger [2013], 212–13). Patria 3.55 (Berger [2013], 168–9). Patria 3.66 (Berger [2013], 172–3), with Janin (1969), 137–8. James (2011/12), 66. Dijkstra and Greatrex (2009), 227–30; Kosinski (2012), 72–8. Pseudo-Zachariah of Mytilene, HE 7.8 (CSCO Scr.Syr. 39, 41–8); with Dijkstra and Greatrex (2009), 230–64 (quoting contemporary letters of Severus); Greatrex (2010), 125–9.

165

A R I A D N E A U G U S TA

The replacement of Patriarch Macedonius by Timothy was not well received at Constantinople where religious tension was regularly provoked by the orthodox monks agitating against the policies of Anastasius and his increasing support for the enemies of Chalcedon in the city and elsewhere. A decree requiring the contro­ versial clause ‘who was crucified for us’ to be added to the Trishagion prayer in all churches so infuriated the people that they rioted over five days (4–8 November 512). The porticoes from the palace to the Forum of Constantine were destroyed and the people were eventually pacified by the hippodrome appearance of the emperor in a gesture of humility and defeat.83 When Anastasius’ imperial tenure was threatened he sought refuge in the Blachernai church. Ariadne considered this act degrading and abused her husband for causing such harm to the orthodox Christians.84 On this occasion, politically secure and independent, she was clearly on the side of the emperor’s opponents. Ariadne died some time in 515 and was buried in a casket of Aquitanian marble in the imperial mausoleum at the Church of the Holy Apostles with Anastasius laid to rest beside her when he died aged 90 in August 518.85 Marcellinus, who then lived in the city, noted that she had spent ‘sixty years in the palace’ (58 to be exact) while a later church history recorded that she ‘administered the empire’ for forty years (41 to be exact) as the wife of Zeno then Anastasius.86 The so-called Oracle of Baalbek produced around 503/4 predicted her own ‘power and dynasty’ would endure for 52 years (to 526 to be exact).87 Only a couple of years before Ariadne’s death a local grammarian Priscian delivered a panegyric on behalf of the emperor Anastasius in which he said of Ariadne that, ‘She has achieved more than her sex allowed her to do’.88 This was another clear signal to Priscian’s Con­ stantinopolitan audience, and to posterity, that Ariadne had broken new ground for a Roman empress.

Ariadne as Augusta Except for Verina briefly, Ariadne was effectively the first empress capable of oper­ ating independently with her own finances, staff and imperial quarters which she consolidated over a forty-year period as Augusta. Her brittle relationship with her older husband Zeno, which gave rise to the rumour of her having him killed,89 and 83 Meier (2009), 269–88. 84 Theodore Lector, HE 508 (Hansen. 3.144.24–145.19); and Theophanes, Chron. AM 6005 (de Boor, 159). 85 Grierson (1962), 45. 86 Pseudo-Zacharias of Mytilene, HE 7.13 (39.57). 87 Alexander (1967), 18 (text: line 148), 27 (trans.), 42–3 (date), and 82–4. Cf. 140 (discussion). 88 Priscian, Pan. Anast. 307: ‘plus fecit quam quod sexus concesserat illi’. 89 What appears legendary is the later story of how Ariadne arranged for her husband to be buried alive in the coffin designated for him, then ignored his cries for help (Zonaras, Epit. 14.2.23–3.5 [3.132–33]; and George Kedrenos, Hist. 388 [605–6 Tartaglia]). This tale originated in later Chal­ cedonian polemic against Zeno, as explained by Conrad (2000), 61–81.

166

A R I A D N E A U G U S TA

apparent distance from the even older Anastasius contributed further to her inde­ pendence as an empress and to the institutionalisation of supportive structures and protocol, as did the chasm between Ariadne and her husbands in terms of religious belief and practice. The Antiochene patriarch Severus who had known her while in Constantinople in recent years explained to his congregation that former empresses had usually concentrated on themselves and their passions. Ariadne, however, was different: she was actually ruling jointly with Anastasius in a collaboration bestowed by God.90 Yet, she remained a firm Chalcedonian at a time of intense pressure to conform to an imperial policy which favoured compromise through Zeno’s Henotikon and Anastasius’ overt support for anti-Chalcedonian bishops such as Severus, thereby arousing popular animosity and opposition at Constantinople. We see this resolute independence of Ariadne clearly in the visit of the Palestinian holy man Sabas to the imperial court in 511 including to the separate quarters of the empress. She assured him of adherence to the decisions of the Council of Chalcedon (and asserted her orthodoxy). Sabas then blessed her and ‘exhorted her to hold firmly onto the faith of her father the great emperor the sainted Leo’, whereupon she replied to him, ‘You speak well, venerable father, as there is One who hears us’.91 For decades Ariadne was the core element ensuring continuity at court of orthodox belief while Zeno and Anastasius were kept at the doctrinal margins. So, the much vaunted doctrinal difference between Justinian (Chalcedonian) and his empress Theodora (anti-Chalcedonian) in the next generation had a notable recent precedent. Although the literary records for Ariadne’s life are relatively scant and frag­ mentary, the independent role of the empress as co-ruler is represented in vari­ ous ways in the iconography of this period. A large number of portrait statues of Ariadne survive which may be explained by her authority and longevity,92 while coins show that she extended the female basileia.93 She was the first empress not to have a crowning hand of God on her gold solidi,94 for instance, but she was the last empress to be depicted on them altogether.95 Theodora never appears on Justinian’s coinage. For the first time, however, an empress is portrayed with the emperor on consular diptychs such as that of Clementinus in 513 where the sepa­ rate but equal clipeate images of Ariadne and Anastasius point to her role in the imperial hierarchy as a co-ruler, as emphasised by Severus in 513, with a status above that of a consul.96 Not only did she confer legitimacy on Anastasius but 90 Severus, Homily 24 (16 May 513, Brière and Graffi [1975], PO 37.1, 143). 91 V.Sab. 53 (Schwartz, 145.1–5): Οὕτως ἐξελθὼν ἀπὸ τοῦ βασιλέως ὁ πρεσβύτης εἰσῆλθεν πρὸς τὴν αὐγούσταν Ἀρεάδνην καὶ εὺλογήσας αὐτὴν παρεκάλει ἀντιλαβέσθαι τῆς τοῦ ἐν ἁγίοις πατρὸς αὐτῆς Λέοντος τοῦ μεγάλου βασιλέως πίστεως. καὶ λέγει αὐτῶι ἐκείνη καλῶς λέγεις, καλόγηρε, ἐὰν ἔστιν ὁ ἀκούων. trs. Price (1991), 154. 92 Itemised in Alföldi-Rosenbaum (1968), 25–7. 93 McClanan (2002), 92; Grierson and Mays (1992), 176, 410. 94 Kent (1991), 35. 95 James (2001), 109. 96 McClanan (2002), 71, 185; Angelova (2004), 8–10; James (2002), 135–6, but doubted by Rubery (2013), 109.

167

A R I A D N E A U G U S TA

also on the annual consuls. Indeed, the indivisible imperial power deriving from Ariadne continued after her death in 515 to the end of Anastasius’ reign three years later. Representations of Ariadne on diptychs for 517, for example, are a posthu­ mous sign of her legitimation of both the emperor and the consuls.97 Ariadne may also be the empress depicted alone in a palatial setting on two renowned extant ivories.98 In each of these we see an empress wearing the chlamys over the right shoulder which is fastened with a jewelled fibula. She has a bejewelled diadem on her head and holds a cross and sceptre. The necklaces and pendentives are distinctly imperial.99 While these two ivories may not in fact have been for Ariadne there would probably have been similar ones with the emperor on the other leaf of the diptych, again displaying their co-rulership.100 The name of the empress does not appear on any of these images because none was required anymore. By now, certainly for a contemporary audience, the identity of the empress was stylistically self-evident.101 In summary, through the fifth century women became more powerful and prominent at the imperial court, first as the wives, mothers, sisters and daughters of emperors then in their own right. Ariadne embodied and extended this transi­ tion. Indeed, none of the early Byzantine empresses can compare to Ariadne. She lived most of her life ‘in the purple’, inhabiting the palace at Constantinople for nearly sixty years from the accession in February 457 of her father Leo I until her death in 515 by which time she could be regularly depicted as co-emperor. Ari­ adne’s role in shaping the identity of the early Byzantine empress in the context of the court culture and ideology that surrounded her was fundamental. Her fortyyear occupation of the palace as Augusta in which she ‘achieved more than her sex allowed her to do’ (Priscian) facilitated the development of the independent household, ritual, and iconography of an empress. Moreover, it was she, and not her husbands, who provided the surety of orthodox Christian belief and practice for which the populace of Constantinople regularly clamoured. Through her influ­ ence, authority, and patronage, and the many formal ways they were represented, Ariadne set a new imperial mould into which were fitted the empresses Euphemia (518–523), Theodora (527–548) and Sophia (565–582) after her.

97 McClanan (2002), 81–2; Olovsdotter (2011), 112. 98 One in Florence (Bargello), one in Vienna (Kunsthistorisches Museum) with details in James (2002), 136–45 arguing that both ivories may not represent the same empress and that neither may be Ariadne (but cf. James [2002], 130). In favour of Ariadne: Magliaro (2013), 73; Rubery (2013), 99–114 (same empress but different phases of her public life). 99 Angelova (2004), 4. 100 Angelova (2004), 10. 101 McClanan (2002), 65.

168

7 JUSTINIAN UNDER JUSTIN Reconfiguring a reign*

On ascending the throne in July 518 the emperor Justin I ‘already had one foot in the grave and was utterly without learning when it came to letters’, according to the sharp invective of Procopius of Caesarea’s Secret History. He goes on to say of Justin that he was ‘at the outer reaches of old age and very senile now, caus­ ing laughter in his subjects Everyone had contempt for him because he was not aware of anything that was going on and they paid him no attention. Justinian, on the other hand, they courted in great fear . . .’.1 Already by the 550s, Procopius is able to collapse the imperial reigns of both Justin and his nephew Justinian (from 518 to 565). In the Secret History he calculates the reign of Justinian as having begun with the coronation of Justin as Augustus in 518, not when he himself became Augustus in 527. Procopius charges specifically that Justinian ‘used to administer the entire government’ even during the reign of his uncle.2 For cen­ turies, these passages of the Secret History have promoted the notion of Justin as a bumbling, boorish and feeble-minded old emperor dominated for all nine years of his reign (July 518–August 527) by Justinian. While scholars continue to accord appropriate allowance to the Secret History’s portrayal of the emperor Justinian and his consort Theodora, as well as to that of the general Belisarius and his wife Antonina,3 virtually no attention has been paid to Procopius’ characterisa­ tion of Justin as emperor. In most modern accounts the nature and veracity of his treatment of Justin is taken literally. Although rightly valued as a literary tour de force, the Secret History remains a vehicle of vituperation.4 Its picture of the reign * This chapter originally appeared in Byzantinische Zeitschrift 100 (2007), 13–56 and is reproduced with permission of De Gruyter (Berlin). As duly noted here, parts of it have been subject to intel­ ligent critique by Leppin (2011); Pazdernik (2015). 1 Procopius, Secret History. 6.11: tumboge,rwn me.n gegonw,j h;dh( avma,qhtoj de. gramma,twn avpa,ntwn kai. to. dh. lego,menon avnalfa,bhtoj; 9.50 o` de. h`qlia,zwn te kai. komidh/ evscatoge,rwn geno,menoj pro.j tw/n avrcome,nwn ge,lwta w=flhn( ovligwri,a| te pollh/| evs auvto.n evco,menoi a[pantej a]te tw/n prassome,nwn ouv xunie,toj u`perew,rwn(VIouistiniano.n de. xu.n de,ei pollw|/ evqera,peuon (trs. Kaldellis [2010]). 2 Procopius, Secret History. 6.19: νε,oj w' e;ti diww|kei/to th/n avrch/n xu,mpasan 3 Fisher (1984), 287–313; McClanan (2002), 107–20. 4 Averil Cameron (1985), 49–66; Tinnefeld (1971), 33–5.

169

JUSTINIAN UNDER JUSTIN

of Justin and the role of Justinian during those years is internally inconsistent and tendentious. In other contemporary accounts of Justin’s reign, Justinian is far from being the Procopian caricature. In particular, those authors closer to events, such as the chroniclers Marcellinus and Malalas writing in the early 530s and who had lived through both reigns as adults, portrayed the reigns of both Justin and Justin­ ian as discrete. For both of them, at their time of writing, Justinian had so far only been emperor for six or seven years, which was not yet as long as Justin’s reign had been. Nor did they portray Justin as an inactive stooge plagued by ill-health and controlled by his nephew. So far as we can tell, lost works covering Justin’s reign did not treat it merely as an early phase of the reign of Justinian either. In particular, Hesychius wrote a separate account of Justin’s reign and that of Justinian,5 while the Church History of Basil the Cilician also covered the reign of Justin only.6 Under the spell of the Secret History, modern scholars continue to portray Jus­ tin as a mere puppet emperor with Justinian pulling the strings from the very beginning of his reign in July 518. Justin is then cast as ineffectual and largely incompetent, if not feeble-minded. By contrast, Justinian is taken to be the source of inspiration for all Justin’s policies and the driving force behind their imple­ mentation. His power and influence are considered all but absolute, as if he were emperor himself. Typically, Stein cast Justinian in the decisive role and claimed that once Justin became emperor he ‘rapidly slid into a state of senility’.7 Although Vasiliev devoted a large monograph to Justin, it is predicated on the assumption that Justinian’s ‘predominant power behind the throne dated . . . from the first day of his uncle’s rule’.8 Later, despite the thoroughness of his research and the breadth of his approach, Rubin considered that ‘both reigns must be treated as one’.9 The master student of Procopius could not escape the pull of the Secret History.10 The formulation of Stein, Vasiliev and Rubin has usually dominated subsequent scholarship,11 and recurs in even the most recent works by Evans, Meier, and others.12 Occasional doubts have been expressed,13 but they have never been systematically pursued. The weight of the interpretive tradition, based on the magnetic attraction of Procopius’ Secret History, is difficult to shift.

5 PLRE 2: 555 (‘Hesychius Illustrius 14’), with Kaldellis (2005), 383. 6 Photius, Bibliotheca 42 (Henry vol. 1, 26–7). 7 Stein (1949), 222: ‘Il semble bien, en effet, que Justin, une fois empereur, glissa rapidement dans une [223] espèce d’hébètement sénile’, cf. Stein (1919), 1328. 8 Vasiliev (1950), 102, cf. 93. 9 Rubin (1960), 78. 10 Rubin (1960), 67. 11 Barker (1966), 23. 12 Evans (1966a), 97; Gauthier (1998), 55; Maraval (1999), 22; Meier (2003b), 124, 185; (2004), 30; Tate (2004), 80. 13 For instance by Honoré (1978), 15–16 and most recently by Moorhead (1994), 2; Rosen (1999), 777; Averil Cameron (2000), 63. Most notably now, the underpinning of Leppin (2011), 29–81, a long chapter entitled ‘The Nephew (518–527)’.

170

JUSTINIAN UNDER JUSTIN

The reign of Justin thereby becomes merely the first phase of a longer-term supremacy of Justinian. It is not considered a legitimate independent entity in its own right. Rather, it is characterised as Justinian’s ‘apprenticeship to statecraft’,14 in which he plays the role of ‘emperor-in-waiting’.15 An abundance of telling syn­ onyms are employed: ‘introduction’,16 ‘prelude’,17 ‘preface’,18 ‘prologue’.19 Such a metaphorical construct allows no separate authority or credibility to the regime of Justin. The years from July 518 to August 527 are simply absorbed into a larger whole and suppressed, exactly as Procopius did in his Secret History, thereby leading to a putative reign of Justinian from 518 to 565, rather than from 527 to 565. Consequently, the precise nature of Justinian’s authority during Justin’s reign, deriving from the successive positions he held, has never been thoroughly examined. A reconsideration of the substantial contemporary documentation for the activities and relative roles of Justin and Justinian during these years suggests, however, that modern interpretation of Justin’s reign has been much distorted by the prejudicial hindsight of Procopius. Instead, Justin remained a strong and pur­ poseful emperor from July 518 almost to his final months in 527 while at every step he had to be reluctantly persuaded to increase Justinian’s rank and power. Justin’s reign needs reconfiguring.

Justinian candidatus and Justin’s accession, 518 The first issue to consider is how Justin himself came to the throne in July 518 and the position and role of Justinian at that time. It was not an inevitable transi­ tion to power because great uncertainty followed the death of Anastasius. There was no empress (Augusta), no junior emperor (Caesar), no superior family mem­ ber (nobilissimus), no designated heir at all. There were eligible candidates within Anastasius’ own family not to mention among others, but no obvious and imme­ diate choice was to hand. Hence, protracted discussion, subtle lobbying and an accusation of vote buying. Of Anastasius’s three nephews, the highly experienced general Hypatius was absent in the east. As magister militum per Orientem he was stationed in Antioch, while his brother Pompeius was in Marcianople as magister militum per Thraciam. Their cousin Probus was evidently in Constantinople at the time but his strong support for the anti-Chalcedonian cause was a significant handicap. Also in the capital was Olybrius, husband of Anastasius’ niece Irene and son of Areobindus and Anicia Juliana, the living legacy of the Theodosian imperial family. Olybrius had been destined to succeed, according to an oracle cir­ culating in the time of Anastasius.20 As noted by Evagrius in the next generation, 14 15 16 17 18 19 20

Bury (1923b), 23. Evans (2002), 10. Vasiliev (1950), title. Maraval (1999), 22. Evans (1996a), 13. Stein (1949), 273. Alexander (1967), 126 n.15.

171

JUSTINIAN UNDER JUSTIN

Justin’s election was rather unexpected and there were different explanations for it.21 Peter the Patrician’s contemporary record portrays the slow and roundabout process which resulted in agreement on the selection of Justin, who was then commanding officer of the excubitores, the very unit responsible for guarding the entrances to the palace.22 He could ensure no-one left, let alone gained entry, until a new emperor was proclaimed. Peter’s account is an important document worth closer scrutiny. As Peter explains, at the death of Anastasius the imperial court’s two vital officials were summoned, namely Celer the magister officiorum (leader of the emperor’s private retinue, the scholae palatinae) and Justin, the comes excubitorum (leader of the palace guard). Both were extremely experienced soldiers and courtiers. Despite a considerable age difference, they had fought together in the war against Persia some fifteen years ago, had subsequently continued to serve Anastasius loyally and had become his closest, and most trusted, military officials. Celer, it was said, was Anastasius’ most intimate friend.23 Celer now explained the situation to his troops, including the emperor’s personal guards (candidati) among whom was numbered Justin’s nephew, Justinian. Justin likewise explained the situation to his troops and to other military officials. Both conveyed the same message to their audience: ‘Our mortal lord has passed away. We must all con­ fer together and choose a new emperor pleasing to God and beneficial for the empire’. Co-ordinating proceedings was Amantius, the eunuch chief of the emper­ or’s bedchamber (praepositus sacri cubiculi).24 At dawn on 10 July 518 the lead­ ing senators arrived at the palace, dressed in clothes befitting their state of grief. Meanwhile, word of Anastasius’ passing had leaked out and the local populace soon assembled in the hippodrome chanting for the senate: ‘Long live the senate! Roman senate, you are victorious! A God-given emperor for the army! A Godgiven emperor for the world!’, and many similar slogans. The foremost senators along with the patriarch of Constantinople, John, conferred inconclusively for some time in the Hall of the Nineteen Couches, the large niched imperial space used for formal and ceremonial occasions. Celer addressed them all, urging a quick resolution lest the people and the army impose on them an undesirable choice. He may have spoken too late. Justin’s excubitors now in the hippodrome opted for a tribune named John, which only antagonised the partisans of the Blue faction. They started pelting stones and sev­ eral were killed in the ensuing turmoil as the excubitors struck back. Next, Celer’s scholares nominated Patricius, then the empire’s most senior general (magister militum praesentalis), and had him summoned. This incited the excubitores to 21 Evagrius, HE 4.1–2. 22 Preserved in Constantine Porphyrogenitus, On the Ceremonies, 1.93 (Reiske 426.1–430.21, trs. Moffatt and Tall [2012], 426–30). Detail in Boak (1919), 39–40; Dagron (2003), 68. 23 John Lydus., On the Magistrates 3.17.3. Celer was himself a more eligible choice as emperor than Justin but was perhaps considered too old (cf. Rubin [1960], 55). 24 PLRE 2, 67–8 (‘Amantius 4’) with Vasiliev (1950), 76.

172

JUSTINIAN UNDER JUSTIN

immediate anger and they were about to set upon Patricius when Justinian, one of the candidati, intervened. Patricius passed to safety among the excubitors. Justin­ ian, however, now found himself the subject of nomination, and was summoned to the guardroom (exkoubita), but he hastily spurned the offer. At last the whole of the senate agreed on Justin, a choice endorsed by the army and the people. He was then conducted to the imperial box in the hippodrome where he was duly crowned by the patriarch John, followed by the chanting of ‘Justin Augustus, you are victorious! Long live the new Constantine!’ and so on.25 Peter’s account, unlike other subsequent accounts, does not even hint that Justin may have actively solicited support for his cause. The white-haired Justin, aged 68 at his accession, has often been taken to be not only an unexpected choice as emperor but also an unusual and unlikely one, a misfit for the position. Jones, for example, echoing Procopius, asserted that the senate and officials had elected a ‘boorish old comes excubitorum’ who rep­ resented a compromise choice by being ‘an obscure outsider’.26 This judgment is questionable on both counts: Justin was no incompetent, as we shall see, and he was definitely no outsider. Rather, as comes excubitorum he was one of the empire’s most senior military dignitaries, one of those closest to the emperor.27 That someone of such power and influence might himself become emperor was no surprise. He was simply the first of several comites excubitorum turned emperor in the ensuing century. Not only has Justin been underestimated by being dismissed as an unsuitable compromise choice as emperor. He has also been dismissed as incompetent and illiterate. The charge of illiteracy stems from the information that his use of a stencil to produce his elaborate and stylised imperial signature meant that he could not even write his own name. All it meant is that, unlike Theodosius II for example, he was no calligrapher. Having been a Roman general implies that he must have been at least sufficiently literate to undertake the considerable paper­ work involved in that position. Certainly, he may not have possessed the literary education or culture of leading civic officials but he was far from illiterate. Using a stencil to trace any emperor’s intricate monogram was not a marker of literacy.28

25 Date of 10 July: PLRE 2, 650 (‘Iustinus 4)’; Stein (1949), 217 and (1919), 1315 relying on the clear testimony of Cyril Scyth., v.Sab. 60 (ed. Schwartz 161.14, 162.4) and Ps.-Zachariah Mytilene, HE 8.1. That Justinian may have influenced the protocol is highly unlikely. 26 Jones (1964), 328; Vasiliev (1950), 74–6, contra Stein (1919), 1327 and (1949), 221. 27 An anecdote preserved in a sixth century Italian document known now as the Anonymus Valesianus. Pars Posterior nicely illustrates his proximity. Anastasius once dreamed that the first person to greet him the following morning would be his successor. When he awoke, Justin as chief of the palace guard was the first one through the entrance to the imperial bedchamber, a clear indication of his closeness. Then, on another occasion when Justin was changing sides while the emperor was in some procession or other he happened to tread on the imperial cloak. ‘Why are you rushing?’ Anastasius was knowingly moved to comment (Anon.Val., 13.74–8). 28 Vasiliev (1950), 82–4; Baldwin (1989), 124f. The accusation of illiteracy carried implicit cultural discrimination, cf. Kraus (2000). Nonetheless, as noted by Meier (2003b), 128, the social implica­ tions of his inferior birth and education cannot be overestimated.

173

JUSTINIAN UNDER JUSTIN

One of the most enticing of Justin’s qualifications to be the new emperor was his religious orthodoxy. Anastasius’ support for the opponents of the Christologi­ cal position formulated at the Council of Chalcedon in 451 had long antagonised a large portion of the people of Constantinople as well as senior imperial officials, especially the most powerful military men who tended to favour doctrinal ortho­ doxy. The revolt against Anastasius in 514–6 by his disgruntled general Vitalian had been on behalf of the Chalcedonian supporters. Besides Vitalian, the Chal­ cedonian view was strongly promoted by the mainly Illyrian generals and other military commanders. Consequently, with the orthodox Illyrians Celer and Justin playing the instrumental roles, it was certain that Anastasius’ successor would be someone who would reverse his anti-Chalcedonian policies. This fear prompted the attempt of some officials to effect the election of another anti-Chalcedonian emperor. The praepositus Amantius was clearly the leader of this group. He intended the throne for his own domestic named Theocritus.29 Other palace cham­ berlains were part of the plan. This surely explains their continued refusal to pass out the imperial robes, as requested on the successive nominations of John and Patricius. They agreed on Justin, however, the most senior Chalcedonian adherent alongside Celer.30 Some have accused Justin and friends of deliberately orchestrat­ ing the entire process in advance, and of cunningly deceiving Amantius by accept­ ing his money to secure the nomination of Theocritus but actually using it to win support for himself. However, the evidence does not bear out this contention,31 because it is usually founded on the unnecessary assumption that Justinian rather than Amantius was the ambitious power behind the throne.32 Similarly, the pub­ licly proclaimed ‘new Constantine’ is unlikely to have been known as merely a Chalcedonian of convenience, rather than conviction, as recently proposed.33 Justin had come to Constantinople in the 470s (in his 20s) from his hometown of Bederiana (modern Badar, near Skopje) in Thrace in search of a worthwhile career in the army. How and where he was originally deployed in the reigns of Leo and Zeno is not known, but by the 490s (in his 40s) he was a commander dur­ ing the war in Isauria, while the following decade (in his 50s) in the conflict with the Persians he held a similar position (comes rei militaris) under the magister militum Celer. He was later (in his 60s) stationed in Constantinople as part of the local palace guard (excubitores) and was instrumental in the defeat of Vitalian in 515. About the same time he was made comes excubitorum.34 By this stage, 29 PLRE 2, 1065 (‘Theocritus’). 30 An important point in his favour, as highlighted by Greatrex (1996), 136. For more detailed eluci­ dation of this aspect: Solari (1948), 339–49. 31 Vasiliev (1950), 81–2. 32 Bury (1923b), 18; Rubin (1960), 68; Browning (1987), 23; Honoré (1978), 7. 33 Menze (2008), 20–2. 34 PLRE 2, 648–51 (‘Iustinus 4’); Stein (1919), 1314; Vasiliev (1950), 66–8. It seems unlikely that the emperor Leo personally subscribed Justin in the excubitores immediately in the 470s, just because he was ‘tall and brawny’, as proposed by Kaldellis (2018), 14. In Procopius’ tendentious account (Secret History 6.2–3) he merges in time a distinction between (1) Justin’s recruitment into the army and (2) his subsequent promotion to the emperor’s personal palace guard. The fact

174

JUSTINIAN UNDER JUSTIN

possibly from the early sixth century when Justin had become well-established as a senior military commander, his sister’s family had joined him in the imperial capital. When her son Justinian was coming of age he may have wanted to follow the successful path of his uncle in the Roman army. Perhaps around the same time Justin’s brother’s family also moved to Constantinople where his sons Germanus and Boraïdes likewise found their niche in the army. The tale of how Justin rose to the highest positions of power in the Roman world is not so unusual at all. In any event it was depicted in a public tableau series by Marinus.35 In essentials it fol­ lows closely the pattern evident in the family of Anastasius, another Illyrian from coastal Dyrrachium, who also found his fortune at Constantinople.36 Justin was childless. At some point therefore he formally adopted his sister’s son as his own, according to the established procedure of Roman law. There are two clear indications of Justinian’s adoption: (1) his name, Justinian, which preserves the usual nomenclature of the adoption process, the addition of ‘-ianus’ to the name of the adoptive father (as ‘Octavius’ became ‘Octavianus’ when adopted by Julius Caesar). His original name, to judge from his consular diptych in 521, was Petrus Sabbatius; (2) references to Justin as his father, the earliest in an epigram carved on a church at Constantinople which describes him as being the ‘son of the great emperor’ [Justin],37 others in laws of the 530s and 540s in which he refers to Justin as ‘our father’.38 Justin had other nephews besides Justinian, namely his brother’s sons Germanus and Boraïdes. Precisely why he singled out Justinian as the one for adoption can only be a matter of speculation. Possibly, he showed par­ ticular talent and Justin was keen to take advantage of the opportunity to promote him. That he also promoted Germanus just as successfully suggests more complex motives. Nor is it clear when Justinian’s adoption took place. Whether the adop­ tion predated Justin’s rise to the throne, and when, can only be guesswork too:39

35 36 37 38

39

remains that Justin fought in Anastasius’ war against the Isaurians in the 490s and the Persians in the early sixth century. It was probably during this period that he had his foot pierced by an arrow, a wound which years later led to his death. It was perhaps around 505 or so that he was promoted by emperor Anastasius to the excubitors, the palace guard at Constantinople maintained as an elite and hand-picked corps of highly experienced soldiers (cf. Rosen [1999], 765). Despite Procopius’ usage (cf. Kaldellis [2018], 11) he may mean that Justin’s military career began in one of the seven scholae palatinae, each of which was a much larger unit (500 soldiers) than the total excubitores (300). Ps.-Zachariah Mytilene, HE 8.1. Capizzi (1969), 43–4. Anth.Pal., 1.97: kai. ui`e,i? pambasilh/oj with Alan Cameron (1976b), 282. CJ 2.7.29 (9); Justinian, Novel 28.4 (535); 97.2 (539); 109 pr. (541); 124.4 (545), cf. Justinian, Institutes. 2.7.3 (‘divus Iustinus pater noster’); 2.12.5 (‘patris mei’). In his relatio to the seventh session of the Council of Constantinople (26 May 553), the quaestor Constantine refers to Justin as Justinian’s ‘father’ (ACO 4.1: 186.38, 39, 41). Cf. Evans (1996a), 96. Others, such as Stein (1949), 240, cf. (1919), 1326; Diehl (1901), 6, have put it unjustifiably late, just before Justinian became Augustus in 527; still others (e.g. Rosen [1999], 765) unnecessarily early, c.505. The earliest extant contemporary testimony to the name ‘Justinian’ is his letter to Pope Hormisdas, dated September 518 (Ep. 147 = CA 35, 592). The Collectio Avellana is an important sixth-century collection of papal and imperial documents put

175

JUSTINIAN UNDER JUSTIN

perhaps as early as 505 when he may have arrived back in Constantinople from the East, perhaps only a decade later when he was comes excubitorum and stationed in the imperial palace itself, or perhaps only at the point when he acquired the throne in July 518.40 Indeed, it may have been a response to those who had been prepared to nominate him as emperor ahead of his uncle in July 518, although he was still clearly ‘Petrus Sabbatius’ on his consular diptych. By 518 Justinian was one of the imperial candidati serving the emperor Anas­ tasius.41 He was in his late 30s at the time. The candidati were an elite unit of the scholae, carefully selected for their physical prowess and bearing. They formed a corps of forty soldiers who acted as the emperor’s permanent personal bodyguard. In view of their position, they enjoyed special ration privileges and took their name from their white ceremonial uniform.42 Justinian’s military appointments over the previous 15 to 20 years are unknown but it is quite possible that he had held junior positions in the Roman army in the east before being promoted to the scholae palatinae. A period of service in the East may well be the explanation for the otherwise late and implausible story that he met his future wife Theodora there.43 To be a candidatus at the court of Anastasius means that he must already have established a military reputation corresponding to that of Justin in earlier years. Indeed he may have carved out a path similar to that of his adoptive father and his cousin Germanus, as well as other Illyrian generals in previous times. Given his background and military experience, combined with his connection to Justin, Justinian was a likely candidatus in 518. He was also definitely someone who could already be offered the throne, as was proposed during the protracted decision to appoint a successor to Anastasius. The notion that Justinian followed an extensive legal and theological education, with a legal or administrative career at Constantinople throughout these years, though often repeated,44 is unfounded and unlikely. That is not to deny, however, his intellectual acumen, the diversity

40

41 42 43 44

together in Rome most probably, but which has only recently been studied extensively for the first time (Lizzi Testa and Marconi 2019). The preference of Vasiliev (1950), 93. While Peter the Patrician refers to Justinian by that name in recounting the accession of Justin on 9/10 July 518 (On the Ceremonies, 1.93 [Reiske 428.3–4] trs. Moffatt and Tall [2012], 428) it looks most like a retrospective application of the name. It does not necessarily mean that he must have been called Justinian by then. Victor Tonnenensis, 101: s.a. 518 (ed. Hartmann, CCh 173a, 33 = Mommsen, MGH.AA XI, 196); Petrus Patricius in Const.Porph., On the Ceremonies, 1.93 (Reiske 428.4, trs. Moffatt and Tall [2012], 428). Haldon (1984), 129–30. Michael the Syrian, Chronicle 9.20; Chron. 1234, 40 (CSCO Scr.Syr. 81/36. 192). For the Mono­ physite Syrian tradition on Theodora: Harvey (2001), 1–31, cf. Pazdernik (1994), 271–81; Evans (2002), 18; Foss (2002) 141–9. Vasiliev (1950), 85, who has Justinian moving straight from his studies to the post of candidatus, cf. 93 (‘an extensive and excellent education’) and 135 (‘an excellent theological education’), cf. Capizzi (1994) 26; Tate (2004), 77. Stein conceded ([1919], 1329) that Justin knew the value of the education he did not himself enjoy and ensured it for his nephew, cf. Rubin (1960), 83–4, 90, contra Honoré (1978), 26.

176

JUSTINIAN UNDER JUSTIN

of his interests, his predilection for theological questions and literature, and the extent of his acquired knowledge. When Justin became emperor his nephew Jus­ tinian already knew the innermost secrets of the imperial palace and how it func­ tioned. He was also in a position to know all the key figures in the family of Anastasius and the other courtiers and senators. He was himself no non-entity, nor was Justin. Given their family connection it was inevitable that they would work closely together now that Justin was emperor. But was Justinian already the dominant power behind the throne, as Procopius would have us believe?

Justinian comes: replacing Anastasius’ regime, 518–519 What needs to be considered next is how Justin consolidated his hold on the throne and what his accession meant for Justinian. From his vantage point in the midsixth century Procopius attributed all Justin’s actions to the influence of Justinian, beginning with forcing aside ‘all the kinsmen of Anastasius, although they were numerous and also very distinguished’.45 John the Lydian writing at Constanti­ nople around the same time, but who had lived through these events at close view, was no less blunt in widening the circle of exclusion: ‘those who received their advancement at the hands of Anastasius were removed’.46 Exactly how Justin and his allies progressively marginalised the claims of Anastasius’ family and support­ ers, or removed them from major positions altogether as alleged by Procopius and John the Lydian, remains to be explained. It is true that Anastasius’ men were soon displaced, beginning with the most senior and influential, namely the generals, the magister officiorum, the Praeto­ rian Prefect of the East and the comes domesticorum, while Justin needed to be replaced himself in the powerful position of comes excubitorum. He wasted no time in filling all these key positions. First of all, he recalled from exile certain powerful individuals who had earlier fallen out with Anastasius: Apion, Dioge­ nianus, Philoxenus and Vitalian. Of these, Apion immediately became Praetorian Prefect of the East, stationed at Constantinople. He replaced the barrister Sergius who had been hand-picked by Anastasius for the post just a few years earlier.47 Diogenianus became magister militum for the East replacing Anastasius’ nephew, the long-serving Hypatius.48 Philoxenus replaced Anastasius’ great-nephew 45 Procopius, Wars 1.11.1. The same perspective appears a half-century later in Evagrius HE 4.1: ‘there were many prominent members of Anastasius’ family who had achieved great prosperity and wielded all the power needed to invest themselves with such a great office’, trs. Whitby (2000), 200 with n.3, but inadvertently substituting Patricius (PLRE 2.840 [‘Patricius 14’]) for Pompeius (PLRE 2, 898–9 [‘Pompeius 2’]) as a nephew of Anastasius. Evagrius is probably copying Pro­ copius at this point, cf. Allen (1981), 173. 46 John Lydus, On the Magistrates, 3.51.5. As an imperial functionary, John witnessed first-hand the transition from the reign of Anastasius to that of Justin, then to that of Justinian, cf. Maas (1992) 28–9. 47 PLRE 2, 994–5 (‘Sergius 7’); 111–12 (‘Apion 2’). 48 PLRE 2, 362 (‘Diogenianus 4’); 879–80 (‘Fl. Theodorus Philoxenus Soterichus Philoxenus 8’).

177

JUSTINIAN UNDER JUSTIN

Pompeius Anastasius (cos.517) as comes domesticorum.49 Vitalian became magister militum praesentalis replacing the loyal Patricius, the scholares’ preferred emperor,50 while the Chalcedonian veteran Romanus was possibly appointed at this time as the other magister militum praesentalis.51 One of the most important appointments, namely Proculus as quaestor seems to have come from outside this circle, although Justin may have known his father as an accomplished bureaucrat at Constantinople.52 Within a week of Justin’s accession Vitalian was back in Constantinople to take on the position of senior court general, although Justin had to travel across to Chal­ cedon to meet him. A contemporary explained Vitalian’s immediate appointment as a dutiful act on the part of Justin53 who made the unreliable Vitalian swear an oath of allegiance to the new emperor, fittingly enough in the church of St Euphe­ mia, site of the disputed council of Chalcedon in 451.54 Since Justin had played an important role in the contingent which defeated Vitalian just three years earlier the need for an oath was an understandable precaution. There is no evidence for the presumption that Justin was accompanied to Chalcedon by Justinian, nor that Justinian too swore an oath with Vitalian.55 Once more, the fiction derives from the mere assumption that Justin could do nothing without the sanction of Justinian. Justin’s nephew Germanus took the place of Anastasius’ nephew Pompeius as magister militum for Thrace,56 while Celer was soon swapped for Symmachus as magister officiorum.57 Although Celer was replaced almost immediately he later participated in negotiations with Pope Hormisdas in 519, a sign that he remained in Constantinople and remained in favour. Others were not so lucky. The gen­ eral Patricius, having managed to keep a foot in both anti- and pro-Chalcedonian camps under Anastasius, not only lost his post as the most senior general but was proscribed and expelled from Constantinople.58 The consul for Justin’s accession year (518) was Magnus, the son of Anastasius’ anti-Chalcedonian nephew Probus. Magnus was sent into exile along with his whole family.59 The tribune John, who 49 PLRE 2, 82–3 (‘Anastasius 17’), with Alan Cameron (1978), 260–1. The position of comes is assumed to be honorary in PLRE but more likely to be genuine for an imperial kinsman (cf. Stein [1919], 1316). 50 PLRE 2, 841–2 (‘Patricius 14’). 51 PLRE 2, 948 (‘Romanus 8’). 52 Pazdernik (2015), 227. 53 Marcellinus, Chron. s.a. 519.3 (ed. Mommsen, MGH.AA. XI, 102): ‘principis pietate’. 54 Ps.-Zachariah Mytilene, HE 8.2; Marcellinus, Chron. s.a. 519.3 (ed. Mommsen, MGH.AA. XI, 102): ‘data acceptaque fide’ (not part of Marcellinus’ original text but the version of a later scribe drawing on an unknown source). 55 The assumption of Vasiliev (1950), 109–10; Tate (2004), 82. 56 PLRE 2, 505–7 (‘Germanus 4’). Not long after he defeated the invading Slavs (Procopius, Wars 8.40.7, with Vasiliev [1950], 310–11). 57 PLRE 2, 1043 (‘Symmachus 4’). Before long Symmachus was replaced by Tatian who was defi­ nitely magister militum the following year (PLRE, 2.1054–5 [‘Tatianus 3’]). 58 Greatrex (1996), 125–6. 59 John of Ephesus, HE Part 3, 2.12 (trs. Brooks CSCO 55, 53). Although John does not date this episode precisely, the most likely time is the accession of Justin when Magnus was at the peak of

178

JUSTINIAN UNDER JUSTIN

had been nominated as emperor before Justin during the process of determining Anastasius’ successor on 9/10 July 518 was spared but neutralised by later being ordained bishop of Heraclea in Thrace.60 In summary, Justin immediately demonstrated his intentions by summoning certain former colleagues and supporters from exile and placing them in key posi­ tions. Most of these new appointees, and potential alternative emperors, had also served the court of Anastasius at some stage. Their very appointment suggests, it has been justifiably remarked, that Justinian had little say in their selection.61 They were all more experienced and more senior than Justinian. So, where did that leave the emperor’s nephew? At this stage he was a candidatus, under the command of the magister officiorum Celer, then under Symmachus. Exactly what happened next is unclear. In a letter which Justinian wrote to Pope Hormisdas on 22 April 519 he styled himself ‘comes’.62 This could be nothing more than a purely honorary title designed to create senatorial rank for the emperor’s adopted son. It is usually considered, however, as the signal that he must have been made comes domesticorum, thereby replacing Anastasius’ grand-nephew Pompeius Anastasius.63 There are some problems with this attribution. The letter does not specify ‘comes domesticorum’ when it might have done so, just ‘comes’. Further, it is clear that sometime later Philoxenus was definitely comes domesticorum. Since Philoxenus was one of those summoned from exile by Justin immediately on his accession, it is surely unlikely that he was recalled merely to a life of courtly idleness. Philoxenus had previously been magister militum for Thrace. He was a perfect candidate for the post of comes domesticorum, head of the troops at the emperor’s direct service. It makes sense to see him as filling that position imme­ diately in 518 and being still in it in 525. As for Justinian, there is only one important court position remaining for which we do not know the immediate incumbent, that is, Justin’s previous position of comes excubitorum. This may be the explanation for the title ‘comes’ which Jus­ tinian held by April 519.64 As comes excubitorum he would be well positioned to be a magister militum in due course. Yet, the near contemporary testimony of Victor of Tunnuna is very clear on one point: Justinian was still a candidatus when he was promoted to general (magister militum) in 520.65 There is no good reason to doubt Victor on this point. So, rather than becoming the all-powerful factotum directing the emperor Justin from his very accession, Justinian remained what he had been under Anastasius – just one of the forty imperial bodyguards. His title of

60 61 62 63 64 65

his prestige and influence. He cannot therefore be the ‘Probus’ acclaimed by the pro-Chalcedonian council at Tyre in September 518 (ACO 3, 102.31, 103.13) as proposed by Greatrex (1996), 131–2. PLRE 2, 609 (‘Ioannes 65’). Greatrex (1996), 139. CA 162 (614). Schwartz (1934), 259; Stein (1949), 222; Vasiliev (1950), 93–4; Tate (2004), 80; but not Meier (2004), 29. Cf. Holmes (1912), 1.304 n.4. Victor Tonnenensis, Chron. 103 (Hartmann, CCh 173a, 33 = Mommsen, MGH.AA XI, 196).

179

JUSTINIAN UNDER JUSTIN

comes in 519 was therefore honorary, as was that of illustris which he evidently acquired around the same time and which enabled him to be ranked among the emperor’s senatorial advisers.66 Obviously, his connection by blood and law to the emperor will have enhanced his influence and status from July 518 but no room was found for Justinian in the cleanout of the imperial administration that accompanied Justin’s accession. He was simply following a conventional military career – soldier, candidatus, magister militum. Further, both Justinian and his uncle were seasoned courtiers, not novice outsiders who suddenly found them­ selves saddled with supreme power. There was nothing compromising about them or their careers, even if Justin could cast himself in the role of reluctant emperor. That much is clear from the expeditious way the new emperor dealt with his rivals, his favourites and his predecessor’s appointees. Justinian now had to secure and expand his own position and influence in an environment of antagonism and resentment provoked by Justin’s arrival on the throne.

Justin, Justinian and reconciliation with Rome, 519–520 The most urgent task for Justin was ending the schism with the bishop of Rome and restoring religious unity and harmony between Rome and New Rome. The populace of Constantinople soon demonstrated their expectation. In the passion­ ate encounters in the imperial church of Hagia Sophia on 15 and 16 July 518, that is, less than a week after Justin’s accession and at the same time as he was travelling to Chalcedon and back to negotiate with Vitalian, they chanted accla­ mations for Justin and his wife Euphemia (‘the new Constantine and the new Helena’). They also chanted for the recognition of the decrees of Chalcedon, the removal of the powerful eunuch Amantius, and the deposition of the patriarch of Antioch, the anti-Chalcedonian Severus. The congregation eventually secured all its requests.67 Amantius and his anti-Chalcedonian allies were killed or exiled, although in his private satire forty years later Procopius blamed Justinian for their fate.68 However, in accounts closer to the time such as the chronicles of Marcel­ linus and Malalas, as well as in the detailed eyewitness record of the rowdy events of 15 and 16 July, there is no mention of Justinian. On the persistent assumption of Justinian’s domination of Justin, his responsibility for Amantius’ murder has been justified and the silence of the records explained by the fact that, as yet, ‘he had not had the opportunity to manifest himself as a leading figure’.69 Instead of such suppositions, it is more helpful to examine closely the considerable contemporary documentation and evaluate what it implies about the relative status and roles of Justin and Justinian during these years, beginning with the restoration of relations with the bishop of Rome. 66 67 68 69

CA, 154 (letter to Justinian from Pope Hormisdas, late 518/early 519). ACO 3.27 (71.30–76.25) with extensive translation in Vasiliev (1950), 136–44. Secret History 6.26, with Vasiliev (1950), 102ff. Vasiliev (1950), 105, 144, cf.108, 135.

180

JUSTINIAN UNDER JUSTIN

A synod of those bishops currently present at Constantinople was urgently assembled and held on 20 July 518. It considered the petitions of 15/16 July from Hagia Sophia as well as a submission signed by the city’s monastic leaders. The bishops endorsed the main requests and communicated their synodal decrees to the emperor and patriarch.70 Subsequently, the patriarch advised his fellow met­ ropolitan bishops and sought their approval. So they convened similar synods at Jerusalem on 6 August and at Tyre on 16 September, as well as others at Apamea in Syria Secunda and elsewhere.71 In the extant records of the synod in Tyre it is notable that, in addition to the acclamations for the emperor and empress, on sev­ eral separate occasions there were also shouts of ‘Long live the patrician Vitalian! Long live the orthodox [or general] Vitalian’/ Bitalianou/ patrikiou/( polla. ta e;th) Bitalianou/ ovrqodo,xou $strathla,tou% polla. ta. e;th)72 At Apamea they voiced ‘Long live Vitalian, worthy of the emperor’/ a;xioj tou/ basile,wj.73 By September 518 it was clear to the bishops, clerics and monks of the eastern provinces that Vitalian was the next most important figure to Justin and Euphemia, because of his championing of orthodoxy. Throughout these detailed accounts the name of Justinian never appears. By the time Vitalian’s name was being popularly invoked at Tyre, Apamea and in other places Justin had also made contact with Pope Hor­ misdas at Rome and commenced the process of formulating terms for the settle­ ment of the so-called ‘Acacian schism’ between Rome and Constantinople. The surviving correspondence of the years 518 to 521 between pope and emperor, plus various others in Constantinople, provides an important insight into their mutual understanding of people and events. This documentation has not always been fully exploited, but it requires close and careful attention. Justin first wrote to Pope Hormisdas on 1 August 518, three weeks after his installation. By then he had already publicly established his clear support for the overturning of Anastasius’ religious policy and full-scale reconciliation with Rome. Only the details remained to be resolved. In his initial letter Justin simply advised the pope that he had been chosen emperor against his will but with the support of the court dignitaries, the senate and the army. He asked the pope to pray that the beginning of his reign would be authoritative.74 A month later on 7 September 518 Justinian sent his first letter to Hormisdas in which he reinforced Justin’s desire for unity and suggested that the pope might like to come to Con­ stantinople himself to complete the process.75 While Justinian was able to write directly to the pope this letter does not necessarily mean that he was already in 70 ACO 3 25, 62–6. 71 Jerusalem: ACO 3 28–30 (76–80); Tyre 31–2 (80–90); Apamea: 33–7 (91–100) with Vasiliev (1950), 145–60; Speigl (1996), 8–20. 72 Vitalian as ‘patrician’: ACO 3.85.26; 86.21; as ‘orthodox’: 85.27; as ‘general’: 86.22; 103.2, 16. 73 Vitalian as ‘worthy of the emperor’: ACO 3.103.16. 74 CA, 141 (586). It is possible that Justin, as well as Euphemia, relied on the eloquence of the impe­ rial quaestor to express their sentiments in a Latin acceptable to the papacy (suggested by Honoré [1975], 107). That is not to diminish the authenticity of the views of the emperor and empress. 75 CA 147.2 (592–3).

181

JUSTINIAN UNDER JUSTIN

complete charge of imperial affairs himself.76 Rather, his proximity to the impe­ rial court meant he could take advantage of the courier Gratus, not least because he was apparently a like-minded friend.77 In the ensuing months there was further correspondence between the pope at Rome and Justin, Justinian and patriarch John at Constantinople.78 Among the letters written in mid to late January 519 and carried by the papal envoys who arrived in Constantinople in March 519 to formalise the settlement were ones for both Justin and Justinian. To Justin the pope explained at length the issues at stake in supporting the Council of Chalcedon and the tome of Pope Leo,79 while to Justinian he noted the significance of Justin’s family being on the throne.80 In another letter shortly after, replying to one of Justinian which has not been preserved, the pope refers to Justinian’s letters as being ‘full of the holy faith’ and he was now encouraged to help bring unity to fruition.81 Hormisdas did not single out Justinian for special treatment. He also wrote to others in similar vein. To the empress Euphemia he noted that ‘since God elected you to the empire’ the peace of the churches is already being resolved by you and you can play a role in increasing ‘your husband’s piety for completing the ecclesiastical peace’.82 To the distinguished ladies Anastasia, wife of Anastasius’ nephew Pompeius, and Palmatia (lineage unknown) he wrote in similar terms.83 Hormisdas also sent a single combined letter to the veteran generals Celer and Patricius. Celer was evi­ dently still in favour with the emperor to judge from the letter, which urges him to assist in the cause of bringing about peace between the churches. Patricius, meanwhile, had been banished from Constantinople by the time the pope’s letter reached there.84 76 As supposed by Vasiliev (1950), 163, cf. Rosen (1999), 766–7; Meier (2003b), 124 n.129. 77 CA, 147.3 (593): ‘Gratum uirum sublimem, unanimum mihi amicum’. 78 Advice from Hormisdas to Justin: CA, 142.4 (587): ‘ubi deus recte colitur, adversitas non habebit effectum’ (October/November 518); CA, 144 (588–9): ‘tenete itaque hanc piae sollicitudinis curam et pro catholicorum pace, sicut coepistis, insistite’ (144.2 [588]; and from Hormisdas to John: CA, 145.2 (589): ‘quia maior ei de hac pace quam de quibuslibet proeliis triumphus acquiritur’. 79 CA 149 (595–8). The terms were already set out in the libellus of Hormisdas which the pope had failed to persuade Justin’s predecessor Anastasius to sign, as explained in detail and set in context in Haacke (1939), 77–91. 80 CA 148 (593–4): ‘pro bonae uoluntatis initio genus uestrum meruit culmen imperii’ (148.2 [594]). 81 CA 154 (601–2). 82 CA 156.1–3 (603–4): ‘Ecclesiarum pax iam caelesti ordinatione componitur, cum uos ad imperium deus elegit, apud quos esse integrum semper religionis suae cognouit affectum . . . ut per uos ad perficiendam ecclesiae pacem mariti uestri pietas amplius incitetur’. 83 CA 157 (604–5). 84 CA 152.2 (600): ‘ut pro ecclesiastica pace allegationes eorum, qui directi sunt, apud animos sereni­ ssimi principis adiuuetis’. The PLRE distinguishes this Patricius (PLRE 2, 839, ‘Patricius 11’) from the general of Anastasius (PLRE 2, 840–2 [‘Patricius 14’]). There is no evidence to support this distinction and no need to assume it (cf. Greatrex [1996], 126). Perhaps Patricius was still sup­ ported in exile by the court since in November 519 he was responsible for the violent expulsion of the bishop Paul from Edessa: Chronicon Edessenum 88 [Guidi 9]; Ps.Dionysius of Tel-Mahre, II.24–5 (trs. Witakowski (1996b), 25–6). When the papal envoys were staying at Scampae on their

182

JUSTINIAN UNDER JUSTIN

When the papal envoys finally approached the imperial city on Monday 25 March 519 they were met at the tenth milestone by Vitalian, Pompeius and Justini­ an.85 That was probably the welcoming party’s due order of rank and status. As we have seen, at this stage Vitalian was widely regarded as the next most significant individual after the emperor and empress.86 Pompeius was one of Anastasius’ fam­ ily who lost his position after Justin came to the throne. Since he did not share his uncle’s anti-Chalcedonian sympathies, however, he was still in favour. Alongside these two, Justinian was the junior partner in terms of status and experience. The legates were escorted into the city in a vast aduentus procession of the faithful carrying lighted candles, then taken to their lodgings. A couple of days later the papal envoys had their first meeting with Justin in the presence of the whole sen­ ate. Justin’s direct and active involvement in discussions is made clear by their account. Negotiations proceeded and on Thursday of Easter week agreement was reached and the pope’s libellus signed by John.87 Justinian is not mentioned at all, although he will have been present. Similarly, there is no mention of his par­ ticipation in the many conferences on the new doctrinal proposition of the Scyth­ ian monks. Justin and Vitalian were principally involved88 and it was Justin who effected a reconciliation between Paternus, bishop of Tomi, and Vitalian. It was also Justin who took personal responsibility for the appointment of Paul as patri­ arch of Antioch in 519.89 The papal legates subsequently reported to Pope Hormisdas on the zealous faith of Justinian and that his sincerity for the Catholic religion was known to all. This was by way of justification for a request put on behalf of Justinian that the pope might send some precious relics to him for his new church of Sts Peter and Paul which was being constructed in his palace of Hormisdas.90 Through the

85

86 87 88

89 90

way to the capital, they were informed that Patricius had been exiled but were not told why (CA 213 [671–3] – a copy of the suggestio of the envoys). Note, however, the reservations of Begass (2018), 205–6. CA 223.1 (683): ‘quod tantum in ipsis, qui dignitate funguntur, inuenimus religionis ardorem, ut Uitalianus, Pompeius et Iustinianus nobis occurrerent in decem milibus et de aduentu nostro cum uestra gratiarum actione gloriari non arbitrarentur indignum . . .’. According to the Liber Pontificalis, they were officially greeted at the so-called Castellum Rotundum by the emperor himself and Vitalian, then in the city by the emperor again (Liber Pontificalis 54, trs. Davis [1989], 48). Cf. Evagrius, HE 4.4. CA 167.9–15 (620–1) with Vasiliev (1950), 175–6. For the libellus itself: CA, 159 (607–10), with Haacke (1939); Menze (2008), 58–105. CA 217.6 (678): ‘coacti piissimi principis et domni Uitaliani magistri militum iussione frequenter ad audientiam causae conuenimus . . . (7) clementissimus imperator in conuentu publico, ubi et nos interesse iussit, Paternum praedictum [bishop of Tomi] et magnificum uirum Uitalianum reduxit ad gratiam [i.e. conciliated]’. CA 217.4 (677): ‘piissimus imperator sua auctoritate Paulum nomine presbyterum de ecclesia Constantinopolitana elegit episcopum fieri in ecclesia Antiochena’, with Menze (2008), 48–57. CA 218.1 (679): ‘Filius uester magnificus uir Iustinianus res conuenientes fidei suae faciens basili­ cam sanctorum apostolorum, in qua desiderat et beati Laurentii martyris reliquias esse, constituit: sperat per paruitatem nostrum ut praedictorum sanctorum reliquias celeriter concedatis’ (2) ‘quia talis feruor est fidei eius . .’ (3) (680): ‘unde si et beatitudini uestrae uidetur, sanctuaria beatorum

183

JUSTINIAN UNDER JUSTIN

same courier Justinian himself also sent the pope a note warning him of the iniq­ uitous doctrine of the Scythian monks then on their way to Rome.91 This clearly implies Justinian’s active engagement with the doctrinal issues and the need for unity, even if he had not been a direct party to the emperor’s discussions with the monks. He also took the opportunity to reinforce the request for relics for his newly constructed church.92 Earlier he had written to the pope offering prayers for the continued support of the papal cause and for Justin and his realm.93 Again, Jus­ tinian was not being singled out. Other letters to Hormisdas were from Pompeius begging prayers for the welfare of the emperor,94 from Anicia Juliana express­ ing hope for strengthening unity95 and from Anastasia with similar hopes for the emperor’s safety and prosperity.96 Later (22 July), Hormisdas expressed gratitude to Justinian for his support for the restoration of unity with the see of Rome and noted that he should expect a heavenly reward for such assistance to the practice of the ‘good emperor’.97 Much the same thankful sentiment was offered to both Pom­ peius98 and to an unnamed high official, possibly Vitalian.99 On the same day Pope Hormisdas wrote to Anicia Juliana not only acknowledging her support for unity with Rome but expressing the hope that ‘just as imperial blood ennobles your person, so may your conscience shine forth with the light of good deeds’.100 He also wrote to Anastasia, wife of Pompeius, offering to pray that God may continue to inspire with religious zeal Anastasia and her family, along with the emperor, as he has already done at the beginning of his reign.101 Yet again, Justinian does not appear as the exclusive and dominant power behind the throne but rather as one of a group of distinguished papal supporters.

91

92

93

94 95 96 97 98 99 100 101

Petri et Pauli secundum morem ei largiri praecipite . . . petit et de catenis sanctorum apostolorum, si possible est, et de craticula beati Laurentii martyris’. CA 187.1–4 (644/5). Justinian later wrote a special letter to Hormisdas attempting to withdraw some of the strong reaction to the monks that he had previously expressed. It seems that he had been persuaded to change his mind by their patron, ‘our most glorious brother Vitalian’ (CA 191.1 [648]). CA 187.5 (645): ‘praesumentes autem de beatitudinis uestrae beniuolentia paternam dilectionem nimium petimus, quatenus reliquiis sanctorum apostolorum tam nos quam basilicam eorum hic in domo nostra sub nomine praedictorum uenerabilium constructam illustrare et illuminare large dignemini, cognoscentes, quod nullum nobis maius nec munus nec beneficium praestare potestis, domine beatissime pater, quam si hanc nostrum petitionem adimpleueretis’. CA 162. 2 (614): ‘petimus uti pro sanctissimo Augusto nostro, totius fidei fautore, proque eius re publica, pro nobis quoque mandatorum uestrorum custodibus aeterno regi consueti impetra­ biles preces offerre dignemini nosque uobis fideliter supplices uestris salutiferis rescriptionibus uisitare’. CA 163.2 (615). CA 164 (615). CA 165.2 (616). CA 176.2 (633). CA 174 1–2 (630–1). CA 177 (634). CA 179.1–2 (635). CA 180.1–2 (635–6).

184

JUSTINIAN UNDER JUSTIN

Two issues remained unresolved between Hormisdas and Justin by the end of 519: the reinstatement of certain eastern bishops deposed in the reign of Anastasius, and the doctrine of the Scythian monks. In both cases there was protracted cor­ respondence between Rome and Constantinople; in both cases Justin and Justinian were involved. On behalf of the deposed bishops Hormisdas wrote separately to Justin102 and his wife Euphemia,103 as well as to both Justinian and Germanus.104 He further observed that their just cause was being undermined by ‘some malevo­ lent people around the emperor’.105 In relation to the Scythian monks, discussion at Constantinople involved the pope’s envoy Dioscorus106 as well as Justin and Vitalian,107 with Justinian evidently playing a lesser role.108 The papal legation which had been in Constantinople since March 519 finally set out on the return journey to Rome around 9 July 520, exactly two years into Justin’s reign. They were bearing with them letters for Hormisdas from all the pope’s key supporters in the imperial capital. Celer hoped for continued resto­ ration of unity, while Anicia Juliana promised to ‘support you against the mad dogs’.109 The new bishop of Constantinople Epiphanius notified Hormisdas of the imperial support for his election and seeking papal prayers not only for himself but also for the continued success of Justin and Euphemia in fortifying the church.110 Euphemia herself repeated the request for prayers on behalf of the protection of the empire,111 as did Justin.112 He also took the opportunity to propose to Hormis­ das that the papal position on unity required a little more subtlety.113 Justin’s letter is clear and concerted, as is his continued engagement with the issues. Around 102 CA 202 (661). Justin replied to Hormisdas on 7 June 520: CA, 193.1–4 (650–1): ‘monentes prae­ cipue, ut pro incolomi statu rei publicae numen supernum uestris exoretur precibus’, 4 (651). 103 CA 203 (662). 104 CA 207 (666): Justinian; CA, 211 (669–70): Germanus. In writing to the bishops themselves Hormisdas advised that he had sent pleading letters on their behalf to the ‘distinguished and magnificent men, our sons Justinian and Germanus’ CA, 210.2 (669). 105 CA 175 (632). Precisely who they were is not known and not easily inferred. 106 CA 188 (645–6); CA, 227.5 (693). 107 CA 224 (685–7). 108 CA 188 (645–6) cf. 206.1–5 (665–6). 109 Celer: CA 197.1–4 (657); Juliana: CA 198.1–2 (657–8) – an expression considered as being ‘femminilmente affettuose oltre che riverenti’ by Capizzi (1973), 83. 110 CA 195.1 (652): ‘et electione Christianissimi et iustissimi principis nostri Iustini et piissimae reginae, quae ei ad omne stadium communicat diuinum’ 8 (654): ‘pro serenissimo principe nostro et pro Christianissima Augusta, quia eorum salus communis sanctarum ubique est ecclesiarum profuturum firmamentum’. The papal envoys reported a month after his election in February that, although they had not yet met Epiphanius the recently agreed peace between the churches was not put at any risk by the new appointment. 520 (CA 222.1–4 [682–3]). 111 CA 194.1 (652): ‘ut orationibus uestris numquam excedat nomen meum ac praecipue serenissimi coniugis nostri sed tam utrique nostrum quam rei publicae supernum precibus uestris placetur praesidium’. 112 CA 192.5 (650): ‘. . . ut suis orationibus pro nostro utatur imperio proque incolomi statu rei publicae’. 113 CA 192.1–5 (649–50).

185

JUSTINIAN UNDER JUSTIN

the same time, he also demonstrated a decisive approach in a new law requiring all soldiers to be orthodox Christians (as testified by three witnesses in the pres­ ence of the unit’s commander).114 Among the letters carried by the envoys to the pope was one from Justinian. The letter provides an early demonstration of his genuine interest in theological issues, as he quotes from Augustine’s treatises de trinitate and de fide. Justinian’s message to Hormisdas does no more than rein­ force that of Justin, namely that the time has come to forget about condemning all the contemporaries of Acacius.115 Much has been read into this letter’s failure to mention Justin,116 yet such failure means nothing here. Of greater moment is the fact that the letters of both Justin and Justinian converge around a common theme. Justinian’s letter simply bolsters and complements the attitude of Justin. It implies collaboration of emperor and nephew rather than domination by Justinian over Justin. Justinian was not running his own policy. However, he and Vitalian were seen as interested and influential figures in resolving the remaining issues with Rome.117 By now Vitalian was not only the most powerful man at court after the emperor, but he had immediately succeeded Justin as consul for 520. Doubt­ less, the papal envoys in Constantinople had been honoured guests at Vitalian’s consular inauguration celebrated on 1 January. From the crucial testimony of the papal correspondence it is evident that in the two years from Justin’s accession in July 518 to July 520 the emperor remained in full command. More than anyone else, it was Justin himself who sought and achieved reconciliation with the papacy. He was responsible for his dealings with the pope and other senior figures and was recognised by them as the authoritative ruler.118 Justinian, by contrast, did not display any more power and influence than might be expected from someone in his very senior position and a close personal relation to the sovereign.

Justinian as magister militum and consul, 520–522 Although Justinian was now a court general, presumably replacing Romanus as magister militum praesentalis,119 he was still less senior, less experienced and 114 Noted by Jacob of Edessa, Chron. (ed. Brooks [1905], 240). The law is partly preserved in CJ 1.4.20 as explained by Vasiliev (1950), 242–3. 115 CA 196.1–7 (655–6): 1 (655): ‘Domino nostro Iesu Christo fauente regnat in saecula, qui sacra religione suum fundat imperium’ 4 (655): ‘ut redimatis plebem sanguine, quam deus noster reg­ endam commisit’ 6 (656): ‘quemadmodum sanctus Augustinus ait’. 116 Vasiliev (1950), 199. 117 As indicated in a letter to Hormisdas from the African bishop Possessor (received at Rome on 18 July 520) who hopes Hormisdas is aware of the problems at Constantinople and notes that both Vitalian and Justinian are keen to resolve this issue: CA, 230.4 (696). 118 Cf. Frend (1972), 238; Anastos (1985), 127–39. 119 CA 230.4 (696): ‘magistri militum Uitalianus et Iustinianus . . .’. He had been made general some time in 520 according to Victor Tonnenensis (Chron., 103 [Hartmann, CCh 173a, 33] = Mommsen, MGH.AA XI, 196), elevated from candidatus. At Vitalian’s death he was described as his ‘colleague’ as general (Ps.-Zachariah Mytilene, HE 8.2). For the likelihood of his replac­ ing Romanus: PLRE 2, 948 (‘Romanus 8’). Sometimes it is stated, incorrectly, that Justinian

186

JUSTINIAN UNDER JUSTIN

less powerful than the consul Vitalian. However, it was not long after the papal legation had left Constantinople in July 520 that Vitalian was suddenly and unex­ pectedly murdered in the delphax of the imperial palace, presumably by the excubitores. The circumstances remain unclear and Justinian remains implicat­ ed.120 It must be supposed, nonetheless, that a plot to kill the consul and the second most powerful man at court was known to the emperor Justin, that is, if he was not himself the instigator of it. Some decades later, safely beyond the end of Jus­ tinian’s reign, but only just, Victor of Tunnuna attributed Vitalian’s murder to the ‘factio’ of Justinian.121 He stops short of blaming Justinian directly. Procopius is less reticent, but in the case he assembled in the Secret History in the late 550s it suited his design to saddle Justinian with responsibility for Vitalian’s death.122 Another contemporary, Jordanes, blames no-one but notes that the murder arose from Justin’s suspicion that Vitalian might do to him what he had attempted to do to his imperial predecessor, Anastasius.123 Those even closer to the event are more circumspect. Marcellinus, himself part of the court at exactly this time as a cancellarius to Justinian but writing in 534, chose to record the event without imputing blame to anyone, least of all Justinian.124 Around the same time, John Malalas, in a badly preserved account, linked the murder to a destructive rebel­ lion by Vitalian and centuries later the Constantinian excerptor saw that Malalas’ account of Vitalian’s demise should be included in the collection of extracts on rebellions (de insidiis).125 While it is generally presumed that Malalas is referring to Vitalian’s rebellion under Anastasius some years previously, this need not be the case. He certainly does not say so. Instead, he could well be referring to a rebellion by Vitalian in mid-520 that is not otherwise recorded. According to John of Nikiu, probably tak­ ing his information from the original complete chronicle of Malalas, Vitalian was actively plotting to overthrow Justin, and so the emperor was obliged to have him liquidated. In fact, John of Nikiu pointedly distinguishes this rebellion from that under Anastasius.126 A usurpation against Justin is a likely and easily explicable

120 121 122 123 124 125 126

actually replaced Vitalian in the same position as magister militum and therefore may have initi­ ated his murder in 520 to achieve precisely this outcome (Vasiliev [1950], 113; Barker [1966], 66; Moorhead [1994], 17). The notion evidently originated in the 12th century with Zonaras, 14.5.34 [150.8–9] who was heavily influenced by Procopius in his account of Justin and Justinian. Justin­ ian may have replaced Romanus in May or early June (cf. Begass [2018], 224). Ps.-Zachariah Mytilene, HE 8.2. For the timing: Vasiliev (1950), 200. Victor Tonnenensis, Chron. 107 (Hartmann, CCh 173a, 35 = Mommsen, MGH.AA XI, 197). Procopius, Secret History 6.28 Justinian is taken to be the culprit by Stein (1949), 230. Jordanes, Romana 36: ‘quem rursus in suspicionem habens facti prioris’. Marcellinus, Chron. s.a. 520 (ed. Mommsen, MGH AA, XI 101). Malalas, Chron. 17.8 (Thurn 339): w`j turannh,saj ~Rwmai,ouj kai. polla.j po,leij kai. cw,raj th/j ~Rwmani,aj praideu,saj with Excerpta de Insidiis, 170.23–171.5 de Boor (= Thurn 339). John of Nikiu. 90.11–12 (trs Charles): ‘Now Vitalian’s death was brought about by his plotting, after his appointment by the emperor Justin, to raise a revolt (against him) as he had done against the emperor before him. [12] And thereupon (Justin) gave orders for his execution’. Theophanes A.M. 6012 (166.19–21 de Boor) may well have used the same version of Malalas as John of Nikiu

187

JUSTINIAN UNDER JUSTIN

scenario. Justin’s whole relationship with Vitalian was based on a sworn agree­ ment that the general would never attempt to exceed his power and authority, as he had done on three occasions under Anastasius. A rebellious Vitalian was exactly what Justin had feared and sought to avert. He would obviously respond force­ fully to the slightest hint of disloyalty from his consular general. That Vitalian was killed for having planned or initiated rebellion is supported by the way his death was subsequently treated with the full force of the damnatio memoriae reserved for a usurper (tyrannus). In traditional fashion his memory was systematically and officially erased. Even the record of his consulship was expunged.127 Justinian was now the sole court general. As such, he was in a powerful position but he was still a relatively inexperienced general. It does not necessarily follow that he now automatically became the dominant and unchecked influence over Justin,128 or that Justin was now more constrained and less able to function as emperor. The project of settling doctrinal harmony continued, but with a less aggres­ sive tone than that propagated by Vitalian.129 Meanwhile, on 7 August 520 Justin had sent a special request to Hypatius who was by now reinstated as magister militum per Orientem. In a fascinating document delivered to Antioch by the emperor’s agent Thomas,130 Justin provides a rare insight into imperial business at the Roman court. He begins by reporting to Hypatius that the recent records of the defensor civitatis at Antioch had been read out to him in the palace at Constantinople. Among the events described was the report of a procession in Cyrrhus organised by the bishop Sergius to honour the image and memory of Theodoret, a former bishop of the city whose theology was now under suspicion. Justin goes on to explain that he subsequently read the records of the defensor of Cyrrhus itself in the copies available at Constantinople, as well as other records of Sergius and the other bishops of the province but could find nothing less than full support for the four councils and rejection of Nestorius. He chides Hypatius for not paying proper attention to the records from Cyrrhus, a signal that he must

127

128

129 130

but describes the murder as revenge by the Byzantines for the deaths Vitalian incurred during his rebellions against Anastasius. Begass (2018), 260–1 considers revenge for the behaviour of previ­ ous years, rather than a fresh rebellion, is the only explanation for Vitalian’s murder. Vitalian’s consulship was obliterated from the record, as well as being retrospectively deleted from an imperial law of 28 May (CJ 7.63.4): CLRE 575 and Alan Cameron (1982b), 93–4. There are extant coins from the time of Justin which have been altered to read ‘DNVITALIANVS’ (Wroth [1908], 10, cf. xiii n.1). This may well represent a crude attempt to produce coinage in recognition of an imperial usurpation by Vitalian that was quickly suppressed. Stein (1919), 1317: ‘Seither übt Iustinianus einen unbeschränkten Einfluss auf die Regierung aus’; Bury (1923b), 21: ‘during the remaining seven years of the reign we may, without hesita­ tion, regard him as the directing power of the Empire’. The circumstances of Vitalian’s murder and associated iconographic representation of the Archangel Michael as a new symbol of imperial triumph over usurpers are outlined in detail in Croke (forthcoming a). Vasiliev (1950), 222–3; Frend (1972), 241 – attributing the change of style to Justinian but there is no explicit evidence for such a claim. The process of subscribing to the libellus of Pope Hor­ misdas continued to drive a wedge between Chalcedonian and non-Chalcedonian. Thomas should be added to the lists of ‘missing persons’ in PLRE.

188

JUSTINIAN UNDER JUSTIN

have had routine access to them at Antioch, or at least for not following up what he must have realised was obviously questionable behaviour by Sergius. Justin then orders Hypatius to summon Sergius and the soldiers named in the Antiochene records and undertake a thorough inquiry. If the soldiers are found to be lying they should be expelled from their unit and tortured. A full report back to the emperor is required. Justin concludes by adding that just in case the requisite documents are no longer available to Hypatius he has also included copies of the relevant records of events in Antioch and Cyrrhus but they must be returned to Constantinople after use.131 Justin’s request to Hypatius clearly evokes a picture of the industri­ ous emperor having records of provincial cities read to him and reading others for himself, then discussing and acting on their contents. This episode reinforces other contemporary documentation in highlighting Justin’s complete command of imperial business in August 520. Likewise, the continuing correspondence of Jus­ tin and Justinian with pope Hormisdas demonstrates emperor and nephew work­ ing in concert,132 not Justinian dictating policy, as has been presumed.133 It was around this time that the City Prefect Theodorus built a chapel in Con­ stantinople which he dedicated to both Justin and Justinian, another indication of the latter’s importance as the nephew and adopted son of the emperor but not yet designated as Justin’s heir.134 By now it was known that the consul for the follow­ ing year would be Justinian. He was now finally the equal of what Vitalian had been – senior general and consul. Plans will already have been well underway to ensure the spectacular show that inaugurated his consulship in the hippodrome at Constantinople on 1 January 521. It was considered a more elaborate and more expensive inauguration than anyone could remember, perhaps deliberately eclips­ ing the most recent one for Vitalian.135 In the early summer of Justinian’s consul­ ship the long-awaited papal replies to the letters from the emperor, especially 131 The document is to be found in the acts of the seventh session of the Council of Constantinople, on 26 May 553 (ACO 4.1, 199.22–200.27), trs. Coleman-Norton (1966), 981–3. 132 CA 199.1–3 (658–9): Justin; 200.1–4 (659–60): Justinian, cf. CA, 243.1 (743): ‘Diligenter apos­ tolatus uester cognoscit, quanto fidei calore filius uester serenissimus imperator nosque fuimus ab initio: numquam cessauimus agere, quae pertinebant ad firmamentum religionis diuinae’. 133 E.g. by Meier (2003b), 124 n.129, claiming Justinian is palpably blunt in speaking to Hormis­ das of res publica nostra ‘in the sense of an emperor’. On the contrary, Justinian was merely employing conventional usage just as in other letters he referred to the emperor Justin as ‘noster inuictissimus imperator . . . noster serenissimus princeps’ (CA 147 [591]: 7 September 518). There is nothing uniquely imperial about such descriptors. It is also found, for example, in contemporary letters of Pompeius (‘domini nostri principis’: CA 163 [615]), Anastasia (‘Domini nostri Augusti’: CA 165) and patriarch Epiphanius (‘principis nostri’: CA 195 [652], cf. 233 [707]). So too, Justin­ ian’s one-time employee, the chronicler Marcellinus could speak routinely of ‘princeps noster’ (s.a. 532, 533) and of ‘ductores nostri’ (s.a. 503) and ‘ecclesia nostra’ (s.a. 380, 399, 431 without suggesting any special sense of ownership. 134 Anthologia Graeca, 1.96 97, see also 9.696, 7 with Cameron (1976b), 282. 135 Marcellinus, Chron. s.a. 521 (ed. Mommsen, MGH. AA, XI, 101–2) with Croke (1995) 122; Cut­ ler (1984) 102–4. Incidentally, as argued by Croke (forthcoming a), Justinian’s consulship and the victory over Vitalian may have provided the occasion, or at least the inspiration, for one of the most renowned of late antique ivories, that of the Archangel Michael in the British Museum

189

JUSTINIAN UNDER JUSTIN

those originally despatched on 9 September 520, reached Constantinople. They were dated 26 March 521.136 Hormisdas exalts even more forcefully the reputa­ tion of Justin as the champion of religious unity: ‘For the peace which [Christ] gave the disciples, the world has found through you’. Yet the sentiment is direct and personal. Justin is addressed as ‘the slayer of schisms and arrogance, and the restorer of the old form of worship’. The pope also confesses that he was becom­ ing totally discouraged about ever achieving unity when Justin’s arrival on the throne rekindled his hope. The emperor’s letters, says the pope, gave him ‘new peace of mind’.137 Justin’s purposeful religious policy, and his active involvement in its promo­ tion, not only attempted to unite east and west. It even extended beyond the bound­ aries of Roman territory, as evident in his dealings with monarchs at opposite ends of the empire, in Lazica and Ethiopia respectively. Fearing that the growing influence of the Persian king Kavadh might result in the imposition of Zoroastrian practices on the Christian population of Lazica, their new king Tzath immedi­ ately travelled to Constantinople and sought formal installation from the Roman emperor in 521/2. His new royal tunic included a segment with an image of Justin embroidered on it, a clear sign of his new status as a Roman vassal. Tzath also sought baptism or rebaptism at Constantinople.138 John Malalas, then an official in the office of the comes Orientis at Antioch through which this matter may well have been conducted, is well informed about this episode and is the solitary source of all later versions. He even quotes from a letter of Justin which he had probably seen.139 The Persian king Kavadh regarded Tzath’s alliance with Justin as a dec­ laration of war. Before long both the Romans and Persians were seeking support from the neighbours of the Lazi, the Sabir Huns and their king Zilgibis. Roman negotiations with Zilgibis resulted in him accepting from Justin a sum of money to attack the Persians, but he then proceeded to ally himself with the Persian king. When Justin informed the Persian king of Zilgibis’ double dealing Kavadh had the Huns’ king executed and attacked his forces. Again, Malalas is our main source

136

137

138 139

(cf. Cutler [1984], 112, contra Vasiliev [1950], 418–26 who associated it with the settlement with the Papacy in 519). Curiously, in all five letters dated to 26 March 521 the consulship of Justinian is omitted. Hor­ misdas includes only the western consul Valerius. Surprising as it might be, one can only assume that this omission is explained by his ignorance at that stage of the year of the eastern consul for 521. It could hardly be a deliberate snub. CA 238.1–16 (734–8): 5 (735) ‘nam pacem quam ille discipulis dedit, per te mundus inuenit’; (6) ‘tu quoque scismatum et superbia dissipater et cultus ueteris restitutor’; (8) tu me, uenerabilis imperator, ecclesiae uulnera tam longa maerentem ad spem redendae salutis animasti: tu me post tam continuas turbidines iam paene desperatione cessantem ad nouam tranquillitatem directis ultro piis litteris excitasti’. Details in Vasiliev (1950), 258–64; Engelhardt (1974), 80–4. Malalas, Chron., 17.9 (340–1 Thurn); Chron.Pasch., 613.3–615.4 (Dindorf); Theophanes (168.14–169.12 de Boor); John of Nikiu 90.35–41; Cedrenos, 398.4 (Tartaglia 619–20), with Vasiliev (1950), 260–2; Greatrex (1998), 132–3. For Malalas’ access to ambassadors’ reports and archival documents: Croke (1990), 9–11.

190

JUSTINIAN UNDER JUSTIN

of information and he possibly had access to the reports of envoys between Justin and Kavadh.140 Meanwhile, events in Ethiopia and Yemen had now begun to occupy the emperor’s time too. Over recent decades the fortunes of the two kingdoms both separated and united by the Red Sea had become complicated by strengthening religious affiliations. By Justin’s time the ruler of the Axumites of Ethiopia (Ella Asbeha) was Christian, albeit anti-Chalcedonian, and that of the Himyarites of Yemen (Dhu Nuwas) was Jewish. Tensions exploded in November 523 with a massacre of Christians by Dhu Nuwas in the Himyarite town of Najran. The details of the massacre were widely circulated. In February 524 the massacre overshadowed a meeting in Ramla to which Justin had sent his envoy Abramius to finalise a treaty with Persia’s vassal Arabs and to ransom two Roman generals they had previously captured. It was presumably in reporting back to Justin that Abra­ mius provided more details of the massacre at Najran and furthered the emperor’s resolve to support the overthrow of Dhu Nuwas. Organising the requisite ships, men and arms was to take some time but in 525 a successful invasion of Himyar was undertaken.141 As a result, the anti-Chalcedonian cause was firmly established in south Arabia and, paradoxically, the persecutor of the anti-Chalcedonians, Jus­ tin, became immortalised in Ethiopian history.142 In the various detailed accounts of all these events Justinian is never sighted. The emperor Justin, however, always appears in confident and complete control.

Justinian as patricius and nobilissimus, 523–524 The magister militum Justinian had not only been preoccupied with religious policy in the early years of Justin’s reign. He also found time for supporting his favoured Blue circus faction. At Constantinople and in cities throughout the East the violence engendered by the rival Blue and Green factions was becoming pro­ gressively worse by the early 520s. As the court general, Justinian’s partisanship was backed by access to force. In 523 persistent rioting by the Blues created seri­ ous disorder in several eastern cities, including Constantinople. So serious did the unrest become that the emperor Justin expressly encouraged the City Prefect Theodotus, who was responsible for ensuring peace on the streets, to be as strict as he needed to be in quelling the violence. He was to show favour to no-one. The prefect took the emperor at his word. As a result he arrested and executed

140 Malalas, Chron. 17.10 (Thurn 341–2); Chron. Pasch., 615.5–616.8 (Dindorf); Theophanes, A.M. 6013 (de Boor 167); John of Nikiu 90.42–6. This episode probably belongs to 522, the date of the Chronicon Paschale (explained by Mango and Scott [1997], 254–5 n.1 but without reference to Wirth (1964), 376–83 where the same point is argued more fully). 141 Vasiliev (1950), 274–83; Pigulewskaja (1969), 225ff; Engelhardt (1974), 135–47; Shahid (1995), 40–6; Greatrex (1998), 227–31; and, on the complex and disputed chronology, Beaucamp et al. (1999/2000), 15–43. 142 Vasiliev (1950), 297–302; Shahid (1995), 728–32.

191

JUSTINIAN UNDER JUSTIN

Theodosius, a senior dignitary with a name evoking imperial connections.143 That might have sufficed to alarm the emperor. Next, the emperor’s own nephew Jus­ tinian found himself implicated. Not only was he a well-known supporter of the Blues who were fomenting trouble in the imperial capital but he was now accused by Theodotus of starting some of it himself. The conscientious prefect was mak­ ing arrangements for Justinian’s arrest when the general was rescued by a sudden illness.144 Indeed, he was not expected to recover. Justinian was now under a cloud, his career possibly terminated. Even his life was under threat. By the time that Justinian’s illness unexpectedly lifted, however, Justin had been persuaded to have his adopted son spared.145 Instead, it was the Prefect who was now under threat. Justinian mounted a campaign against him and before long, in the late summer or autumn of 523, the emperor removed Theodotus from office. It was not an easy decision. Justin’s chief legal advisor, the quaestor sacri palatii Proculus, sided with Theodotus. In terms of intellect and integrity Proculus was considered the outstanding person of those times and Justin surely took a risk in ignoring his advice.146 Not long after being convicted Theodotus left the city. The following year he took refuge in a church in Jerusalem where he remained until he died.147 The chronicler Marcellinus, at that very time acting as cancellarius to Justinian, does not suggest any involvement by his employer but he does express a view in favour of Theodotus’ measures, without mentioning him.148 Justin remained in command throughout this episode, except that Justinian was able to persuade him to ignore the advice of Proculus. It was probably in the aftermath of this period of great uncertainty for Justin­ ian that he was given the most distinguished title possible for an imperial official, that of ‘patrician’.149 As patricius he was now of equal status with all his potential 143 References in PLRE 2, 1102 (‘Theodosius qui et Zticcas 19’). 144 John of Nikiu 90.16–23, here (as elsewhere) more fully preserving the original text of John Mala­ las. For Justinian’s support of the Blues: Whitby (1999), 242–4. 145 According to Procopius, Justinian recovered ‘suddenly and unexpectedly’ (Secret History 9.39). If there is any truth in the later and garbled account of Justinian’s cure by Sampson (Acta Sanctorum, vol.5, 267.5ff), then this may have been the occasion. 146 Procopius, Wars 1.11.11, cf. Secret History. 9.41; John Lydus., On the Magistrates, 3.20, with Pazdernik (2015), 241–59. 147 References in PLRE 2, 114–15 (‘Theodotus qui et Colocynthius 11’). While the details of Justin­ ian’s involvement and arraignment appear only in John of Nikiu it is clear that he is here drawing on the complete version of the chronicle of John Malalas otherwise preserved only in abbrevi­ ated form (see Jeffreys et al. [1986], 235–6). It cannot be dismissed as merely an ‘unreliable record’ (Vasiliev [1950], 117) or as an ‘extrapolation from Justinian’s involvement with the Blues’ (Mango and Scott [1997], 253 n.4). Confirming the severity of Justinian’s illness: Procopius, Secret History 9.35. 148 Marcellinus, Chron. s.a. 523 (ed. Mommsen, MGH.AA. XI, 102) with Croke (1995), 123. 149 Victor Tonnenensis, Chron. 107: s.a. 523 (ed. Hartmann, CCh 173a, 35 = Mommsen, MGH.AA XI 197) refers to Justinian as patricius in 523, the earliest reference, but in the course of describ­ ing the death of Vitalian which he misdates to that year. Even so, the title may be correct for 523 but not for 520 when Vitalian was killed. Further, there is no indication that he became patrician immediately following the death of Vitalian, as claimed by Guilland (1967b), 136.

192

JUSTINIAN UNDER JUSTIN

political rivals, particularly Olybrius and the nephews of Anastasius. Justinian subsequently appears as patricius on different occasions: in assisting the reloca­ tion of the anti-Chalcedonian bishop Mare from one place of exile to another around 524/5,150 and at the time of the devastating flood in Edessa in 525.151 The relocation of Mare stemmed from one of Mare’s monastic allies Stephen who had arrived in Constantinople to petition Justinian’s new consort Theodora about Mare’s plight. John of Ephesus, who is the source of this information and had first-hand knowledge, mentions that not only was Justinian patricius at the time but Theodora was also a patrician (patricia). From Procopius we learn that it was Justinian who had her advanced to the honour of patricia, and he implies that this occurred before they were married.152 If so, it would appear that Theo­ dora achieved this honour not long after Justinian himself, that is, in 523/4. Moreover, as the patron of holy men and propagator of monastic life she was now following a well-established pattern for female patricians.153 On the model of aristocratic women, so visible in the settlement with Pope Hormisdas in 518/20, Theodora also naturally involved herself in doctrinal politics. While much has been made of Theodora’s religious patronage, such engagement was not at all unusual for a patricia. Despite her extraordinary and elevated status Theodora was still not married to the magister militum Justinian. The marriage would have occurred sooner but for two reasons: the objection to the union by the empress Euphemia, and the need for special legislation making possible Justinian’s union with a former actress. Although not yet husband and wife, it is plain that Justinian and Theodora were already living in the palace of Hormisdas where Justinian had recently built his church of Peter and Paul. Euphemia’s refusal to allow Justinian to marry Theodora as long as she lived is a striking fact. Its legitimacy presumably derived from her position as spouse of Justinian’s adopted father. That Justinian was obliged to accept the situation is surely another sign that he was not the dominant influence at court. By the end of 524 it appears that Euphemia had died. One impediment to marriage was removed. Another remained. Procopius accuses Justinian of coerc­ ing Justin into making special legislation purely to enable him to marry Theodora. Justin’s marriage law was addressed to Demosthenes as Prefect but cannot be dated any more precisely than between 520 and 524. It was promulgated by 19 November 524 so Justinian was probably married before then or very shortly 150 John of Ephesus, Lives of the Eastern Saints 13 [PO 17.2, 189]) – not cited in PLRE 2.647, (‘Jus­ tinian 7’). 151 John of Nikiu 90.16–18.; Chronicon Edessenum, s.a. 836 [10]), and Cyr.Scyth., vita Sabae 68 [170.10ff]. 152 Secret History 9.30. 153 Women such as Anastasia (PLRE 2, 76–7 [‘Anastasia 3’]), Caesaria (PLRE 2, 248–9 [‘Caesaria 3’]), and Georgia (PLRE 2, 501 [‘Georgia’]) – all connected to the family of the emperor Anas­ tasius. An earlier example is Herais (PLRE 2, 543 [‘Herais’]) wife of Anthemius (PLRE 2, 98 [‘Anthemius 5]’) and connected with the households of the emperors Anthemius (through her husband) and Zeno (through her son).

193

JUSTINIAN UNDER JUSTIN

thereafter.154 Despite Procopius’ accusation, typical of the Secret History, it is not certain that the law was initiated because of Theodora’s former status. Nor was such a law required at all to enable Justinian to marry if, as patricia, Theodora had already transcended any legal status impediment to marriage.155 Be that as it may, it is safe to assume that by the end of 524 Justinian and Theodora were husband and wife.156 The union of the patricius Justinian and the patricia Theodora may well have prompted Justin to confer on the middle-aged bridegroom the rarer rank of nobilissimus. As the adopted son of the emperor it was an honour to which Justinian was entitled since it was reserved exclusively for the sons of emperors. Being nobilissimus did not constitute any right of imperial succession, but in terms of rank and status it will have helped differentiate Justinian further from his con­ temporaries. In particular, he finally stood on equal terms with Anicia Juliana, the empire’s only nobilissima. Marcellinus refers to Justinian in 527 as having been designated nobilissimus by Justin for some time without specifying the exact year.157 Interestingly, it was not something which Justin had sought. Rather, the senate prevailed upon him to award this title to his nephew.158 In fact, they had earlier petitioned him to make Justinian a co-emperor but Justin had pointedly gathered up his imperial cloak and replied to them ‘You should pray that no young man puts on this cloak’.159 Clearly, Justin felt that Justinian (early 40s) was not, or not yet, worthy of more power and honour. It was only a year previously that his life was in danger, as he battled the charge of the City Prefect Theodotus of being involved in provoking civil strife at Constantinople. Perhaps Justin implies that supreme power was the special prerogative of age and experience. The emperors he had known, Zeno and Anastasius, had certainly met that criterion. In pressing such a nomination upon Justin the senate probably wanted to ensure a smooth transition to the next emperor, unlike the situation that confronted it in July 518. As a title, nobilissimus complemented that of patricius and could be kept until becoming superseded by his elevation to the rank of full emperor (Augustus) whenever, if ever, that should come to pass. The ceremonial for bestowing the title of nobilissimus took place inside the imperial palace, mainly in the Hall of the 19 Couches, replete with acclamations. Constantine Porphyrogenitus’ Book of Ceremonies describes how this particular ritual was conducted centuries later but it had probably not changed all that much from the 520s.160 There is no indication of when the bestowal of the title took place but a suitable occasion would have

154 Honoré (1978), 10; Garland (1999) 14. 155 Argued by Vasiliev (1950), 395 but rejected by Daube (1967), 380–99. 156 The marriage need not have been as late as 525, as proposed by Browning (1987), 40; Evans (1996a), 20. It could well have occurred even early that year. 157 Marcelinus, Chron. s.a. 527 (MGH.AA, XI, 102). 158 Zonaras, 14.5.37 (150.14–17 Büttner-Wobst) with Vasiliev (1950), 94. 159 Zonaras, 14.5.35 (150.9–13 Büttner-Wobst). 160 Const.Porph., On the Ceremonies I. 44(53) (Reiske, 225–9, trs. Moffatt and Tall [2012], 225–9).

194

JUSTINIAN UNDER JUSTIN

been the annual celebrations to commemorate the day of Justin’s inauguration, that is, 10 July 524.

Justinian Caesar, 525 Further promotion lay ahead for Justinian the nobilissimus. In 525 (not 522, as often thought) the settlement of relations between Justin and the Persian king Kavadh resulted in a special proposal to the Roman emperor.161 Caring closely for his own succession and wanting to ensure the preservation of his favourite son Chosroes, the Persian king asked Justin to adopt Chosroes formally. As an adopted son, Chosroes would presumably have the same rights and legal status as Justin’s other adopted son, Justinian. Initially, the emperor regarded the proposal with some favour. Then he sought advice from the senate. They were opposed to the idea, persuaded by the advice of the emperor’s chief legal official, Proculus, that it introduced complications and was probably a Persian ruse.162 Proculus pointed out that if Chosroes became Justin’s adopted son then he would be eligible to inherit the throne on the emperor’s death. That could lead to the unification of the thrones of both Rome and Persia. Proculus considered such an outcome too dangerous to contemplate. Instead, it was agreed to propose to Kavadh an alternative form of adoption with less enduring potential, namely ‘adoption through arms’.163 At the Persian monarch’s request, senior Roman envoys were sent to Persia to negotiate the proposition. Justin sent Hypatius, still at Antioch as magister militum, along with Rufinus and the venerable Pharesmanes. The negotiations subsequently broke down with Kavadh offended by the emperor’s rejection of his adoption proposal. On this occasion, unlike that in 523 when Justinian’s future hung in the balance, Justin heeded Proculus’ advice. On this occasion at least, it would appear that Proculus, rather than Justinian, was the power behind the throne.164 In the course of discussing the proposal at Constantinople earlier in the year the quaestor Proculus turned to Justinian and remarked that the adoption of Chos­ roes would prejudice his own chances of succeeding Justin. Proculus urged Jus­ tinian to support rejection of the proposal ‘lest you prove a stumbling block to yourself as regards coming to the throne’.165 Procopius’ report at least indicates 161 Often dated to 522 (Vasiliev [1950], 269) but must be later. Either 525 (Greatrex [1998], 137) or 525/6 because the Roman envoy Rufinus was sent to discuss the adoption further with Kavadh in the fourth indiction, i.e. 1 September 525 to 31 August 526 (depending on Ps.-Zachariah Mytilene, HE 9.7, cf. Mango and Scott [1997], 255 n.5). For a critical analysis of this episode: Greatrex (1998), 134–8; for the legal issues involved in adoption in Roman and Sasanian law: Pieler (1972), 399–432. 162 Theophanes, Chron. A.M. 6013 (de Boor 167–8), probably deriving from the original version of the chronicle of John Malalas but not otherwise preserved. 163 Explained in Greatrex (1998), 135 n.43. 164 Pazdernik (2015), 244–7. 165 Procopius, Wars 1.11.16. There is no need to doubt the authenticity of Procopius’ information here, as does Solari (1948), 355–6. See Pazdernik (2015).

195

JUSTINIAN UNDER JUSTIN

that there was by now an expectation that Justinian would be the next emperor. There was no indication how soon this might be, nor could there be any guarantee about it. Being the adopted son of the emperor did not by itself ensure succession. Indeed, trepidation over this very outcome may well explain the reported attempt by Hypatius to undermine Chosroes’ offer which Procopius attributed to malice towards Justin.166 Although the events have never been connected before, as far as can be ascertained, it is likely that the senate’s opposition to the Persian king’s adoption proposal focussed their attention on the succession to Justin. This pro­ vides a compelling context for the promotion of Justinian to the rank of Caesar at precisely this time, thereby putting the succession beyond doubt. Under the year 525 the chronicler Victor of Tunnuna reports that ‘At the behest of the senate, Justin Augustus unwillingly makes Caesar his nephew Justinian’ (Iustinus Augustus Iustinianum nepotem suum ad senatorum supplicationem inuitus Cesarem facit).167 As with the title of nobilissimus previously, the senate pre­ sumably petitioned Justin once again to make Justinian a co-emperor. Once more Justin gave them something less than they desired. His reluctance may indicate that he was still not convinced Justinian was the appropriate person to succeed him. That Justinian was Caesar between 525 and April 527 when he was made co-Augustus with Justin is normally overlooked,168 yet there is no good reason to doubt it.169 It is arguably confirmed by Victor’s contemporary Peter the Patrician, a trusted emissary of Justinian. In his record of imperial coronations preserved in the 10th century Book of Ceremonies compiled for Constantine Porphyro­ genitus, Peter records the coronations in order: Leo I, Anastasius and Justin as Augustus. Then follow two others: (1) Leo II which is introduced as illustrating ‘how an emperor is proclaimed by an emperor’(o[pwj kai. basileu.j u`po. Basile,wj gi,netai).170 that is, how a living emperor creates a new Augustus, and (2) the brief description of the coronation of Justinian as Augustus in April 527. The brevity of this last account is explained by the omission of most of the ceremonial because it paralleled the previous coronation of Leo II except that it took place inside the palace (at the delphax) rather than in the hippodrome.171 By juxtaposing and 166 Procopius, Wars 1.11.31. 167 Chron. 109 (s.a.525) Hartmann, CCh 173a, 35 = MGH AA XI. 197. As for the date, there can be no certainty. It is worth remarking, however, that Justinian was still patricius when he was petition­ ing Justin to restore Paul to the see of Edessa. This must have occurred in the latter part of 525 following the bishop Asclepius’ death on 27 June. Paul was reinstated in March 526 (Chronicon Edessenum [ed. Guidi 10.4–10, CSCO Scr.Syr. 2]). 168 It is more often just ignored (as in PLRE 2, 647) rather than considered and dismissed, as in Bury (1923b), 21 n.6; Stein (1949), 240 n.3 (‘certainly false’). 169 Victor otherwise records Justinian’s positions accurately: Victor Tonnenensis, Chron. 101 (Hart­ mann CCh 173a, 33 = 518.2 [Mommsen, MGH AA. XI, 196]; 103 (Hartmann CCh 173a, 33 = 520.2 [XI, 196]). cf. Vasiliev (1950), 94–5 n.70. 170 Const.Porph., On the Ceremonies 1.94 (Reiske 431.1–2 trs. Moffatt and Tall [2012], 431). 171 Const.Porph., On the Ceremonies 1.96 (Reiske 433.8–9 trs. Moffatt and Tall [2012], 433): kai. pa,nta kata. to. o[moion sch/ma evge,neto ou- me,ntoi evn tw/| i`ppikw/| a;nw( avlla. evn tw/| de,lfaki)

196

JUSTINIAN UNDER JUSTIN

linking the ceremonial promotion of Leo Caesar to Augustus with that of Justin­ ian, Peter implies that the same ceremonial applied equally to Justinian in 527. In other words, like Leo II, when crowned Augustus Justinian was already Caesar. In addition, there is at least one other clear subsequent reference to Justinian as Caesar,172 while in a law of 535 concerning the proconsul of Cappadocia Justinian may also be alluding to having once held the title of ‘Caesar’ himself by linking it to the provincial capital of Caesarea.173 There had not been a Caesar for some decades, but after Justinian more Caesars were to follow. Justinian himself was later to be pressured to make his nephew Justin a Caesar but, perhaps with his own uncle’s example still in mind, he resisted.174 The installation ceremonial for a Caesar is described in detail in the Book of Ceremonies, but much of it was already standardised by the 520s. There are clear similarities between the record of the coronation of Leo II as Caesar in 472 and the ritual set out in the Book of Ceremonies, for instance the shout of euvtucwj( euvtucwj( euvtucwj.175 To all intents and purposes, the Caesar was another emperor, having many of the imperial privileges and wearing the imperial garb, except for the full crown. The title and position did constitute, however, the assurance of succession.176 To be inaugurated as Caesar Justinian will have been considered the obvious heir, now possessing sufficient authority and presence to be the next Augustus. Becoming Caesar entailed a distinctive inaugural ritual involving the senior emperor and all the court officials. Further, it immediately created a separate court ceremonial with associated dignitaries and began with the ritual process of send­ ing inauguration gifts to foreign leaders. As soon as Tiberius became Caesar in 574, for instance, we see him sending such gifts to the Persian king Chosroes even though the Roman and Persian armies were currently engaging each other in Mesopotamia.177 A Caesar could also preside at races in the hippodrome from the imperial box, as we see Leo II doing in 472. So too, he could receive acclamations, 172 Const.Porph, De thematibus 12.34 (ed. Pertusi 52), 76.29–31: o` me,gaj evkei/noj kai. peribo,htoj basileu.j VIoustiniano,j( o[te tw/n skh,ptrwn evkra,tei th/j ~Rwmaikh/j basilei,aj o` tou,tou Qei/oj VIousti/noj( monostra,thgoj tw/n ~Rwmaikw/n tagma,twn Kai/sar w;n evcrhm,aze. The title Caesar is also recorded in the Syriac tradition but fused with Justinian’s elevation to Augustus in 527 : Ps.Zachariah Mytilene, HE 9.1; Michael the Syrian 2.189; Bar Hebraeus 2.81 (trs. Wallis Budge [1932], 73). 173 ‘. . . et urbem praebet magnam amantissimi Caesaris nobis cognominem, qui dedit principium bonum quae nostra est monarchiae, per quem in omnibus terrae gentibus nominatissimum est Caesaris nomen et quo nos pro alio quodam imperialium signorum nobilitamur’ (Justinian, Novel 30. praef, 18 March 535). 174 John of Ephesus, HE 3.2.10. 175 Leo II: On the Ceremonies 1.94 (Reiske 432.11, trs. Moffatt and Tall [2012], 432). 176 Guilland (1967a), 25–43; Rösch (1978), 36–7. The Caesar’s role is neatly summarised in a passage not cited by Jones, Guilland or Rösch, namely vita Marcelli 34 (ed. G. Dagron, 316– „ 17): ~O kai/sar toi,nun kai. a;cri peri,estin o` basileu.j pa,nta koinh| pra,ttwn met vau/tou/(kai. teleuth,santoj mo,noj th.n basilei,an krateῖ mhdeno.j de. pro. au/tou/ mhde.n su.n auvtw/| tolmw/ntoj evkei,nhn labei/n) 177 John of Ephesus, HE 3.6.22 (trs. Brooks, CSCO 55, 243).

197

JUSTINIAN UNDER JUSTIN

and raise levies but he did not have the right to mint coinage, nor did his name appear in imperial laws.178 To express his elevated status, a Caesar needed his own palace for himself and family, his own staff and resources, his own ceremonial. Justinian Caesar and Theodora held court in their residence, the palace of Hor­ misdas where Justinian had lived since 518. They also undertook construction of a rather grandiose new church on their property, namely that dedicated to St Sergius which stands to this day. For the foreseeable future it was to be the church of the Caesar and his consort.179 Meanwhile, it was not long after Justinian became Caesar in 525 that a bishop of Rome first set out for New Rome. The settlement with Pope Hormisdas in 519 and afterwards had not entirely secured a permanent harmony between pope and emperor. Justinian’s invitation to Hormisdas in 518 to come to Constantinople was never taken up but his successor Pope John arrived in the capital with great ceremony and approbation in the early part of 526. The papal party was first met at the 15th milestone from the city. It was presumably a station of significance where, as for the papal envoys in March 519, the welcoming party is likely to have included the senior imperial officials and possibly the Caesar Justinian, as well as his prestigious rivals from the houses of Anastasius (at least Pompeius and his wife Anastasia) and of Theodosius (Olybrius and his mother Anicia Juliana). The accompanying papal adventus procession was vast. According to the papal biography ‘the whole city came out . . . with torches and crosses in honour of the blessed apostles Peter and Paul’.180 Pope John had been despatched to Justin by Theodoric the Arian Gothic king of Italy in order to seek an imperial reprieve for the Arians who were now under threat of forfeiting their churches in the East.181 Theodoric acknowledged the imperial sovereignty of Justin. He even minted gold and silver coins, at least at Rome, in the emperor’s name.182 Tensions at the court of Theodoric had been heightened by a radical change of policy in Africa, which had destabilised the 178 This answers the objection of Honoré (1978), 16 n.142 who suggests that Justinian could not have been Caesar from sometime in 525 to April 527 because the two inscriptions to laws in that period (CJ 7.39.7 [1 Dec.525] and CJ 9.19.6 [1 Dec.526]) do not include him. Legislation was reserved for Augusti. While a Caesar was never included in imperial laws, on at least one occa­ sion a Caesar does appears in an encyclical letter, namely that of Basiliscus Augustus and Marcus Caesar to the patriarch of Antioch in 475 (Evagrius, HE 3.4 with Whitby (1999), 133–5). After the mid-fifth century, a Caesar does not appear on coinage either (cf. Rösch [1978], 37). 179 This dating for the church is argued in detail in Croke (2006), and Chapter 8 (207–45), which takes account of the critique of Bardill (2017) who would date the church later in the reign of Jus­ tinian Augustus (530s) when Justinian and Theodora no longer lived in the Palace of Hormisdas. It is reinforced by the revised chronology and revaluation of Theodora’s support for Monophy­ sites in Menze (2008), 219–28. The traditional date for the church is the early to mid-530s, was proposed originally by Mango (1972) and (1975); then Bardill (2000) and (2017) 180 Liber Pontificalis 55 (trs. Davis [1989], 49). On the procession and its significance: Vitiello (2005). 181 Greatrex (2000b), 78–9. 182 Kent (1971), 67–74.

198

JUSTINIAN UNDER JUSTIN

religious equilibrium. When Hilderic succeeded Trasamund as king of the Vandals in 523 he had his predecessor’s wife, the sister of Theodoric, banished and subse­ quently put to death. He also suspended the persecution of Catholics and sought a rapprochement with the emperor at Constantinople. He was said to have been at some stage a ‘guest-friend’ of Justinian.183 Soon the image of Justin also appeared on the Vandal coinage.184 Theodoric clearly feared similar overtures to Justin from the senatorial aristocracy of Italy and when it was discovered that the consuls of 523 Boethius and Symmachus could be accused of such behaviour the Gothic king reacted strongly. They were put to death probably late in 525 while Pope John was travelling to Constantinople.185 The oft-repeated claim that Justinian was already planning the reconquest of Italy in the wake of Theodoric’s hostile actions, but took another decade to bring his dream to fruition, is wishful thinking.186 Again, it is predicated on the entrenched proposition that Justin was incapable of conduct­ ing his own relations with the kings of Italy and Africa. Pope John remained in the imperial capital for several weeks and on Easter Sunday he took precedence over the patriarch as principal celebrant of the liturgy, conducting it in Latin according to the Roman ritual.187 The homage paid by Pope John to the emperor Justin irked Theodoric.188 Likewise, it may be surmised that the honour paid by the pope to Anicia Juliana irked Justinian. Pope John will have spent time with Justinian in what was then the only church of Sts Peter and Paul in the city. It had been constructed just recently at Justinian’s request and on his estate. Relics had been supplied from Rome by John’s papal predecessor Hormis­ das and John now provided new gifts for the church. Justinian could now reap the benefit of his association with the papacy, but so could Juliana. Her church of St Polyeuktos was the most elaborate and magnificent in the capital. It too could boast a papal connection through its mosaic of Constantine and Pope Sylvester. The competition for papal attention and kudos was part of a wider, quite intense rivalry between Justin’s nephew and the great-granddaughter of Theodosius II.189 Pope John died not long after returning to Italy and Theodoric himself not long after that. The king of Italy was evidently displeased with the impact of the legation. He did not live to see one of its tangible benefits, namely the later law which reinforced persecution of heretics but which specifically excluded Gothic 183 Procopius, Wars 3.9.5. 184 Wroth (1911), 13. 185 There is a considerable literature on this topic, and the date of the execution of Boethius and Sym­ machus is disputed, but see Moorhead (1992), 219–35; Amory (1997), 216–21. 186 Stein (1949), 253–4; Vasiliev (1950), 327 (‘. . . such a plan had already been definitely formed in the head of Justinian’); Anastos (1985), 137; Rosen (1999), 773. 187 Marcellinus, Chron. s.a. 525 (ed. Mommsen, MGH.AA. XI, 102). 188 Liber Pontificalis 55 (trs. Davis [1989], 49–50) with Ensslin (1951), 128–34 (explaining the homage paid by Justin to John as the normal mode of relations conducted between emperor and patriarch at Constantinople), and Rosen (1999), 774. 189 Explained in Harrison (1989), 40, 139, cf. Fowden (1994), 275; Capizzi (1997), 78–91; James (2001), 158; McClanan (2002), 94–8; Connor (2004), 114.

199

JUSTINIAN UNDER JUSTIN

federates.190 In the extant documentation recording the papal visit to the imperial capital, along with subsequent events, there is actually no mention of Justinian at all, not even by Marcellinus, who had worked closely with Justinian until recently. This is all the more surprising, given that by late 525 Justinian had been elevated to the position of Caesar. Justin is the one who receives papal anointing. He is still very much in charge. The same impression is evident from the personal credit given for his support for those whose lives and property were devastated by the flood in Edessa in April 525,191 as well as by the urban conflagration at Antioch.192 They were both places Justin knew first-hand. He also provided similar support for cities racked by other natural disasters. There were earthquakes at Anastasius’ hometown of Dyrrachium in Epirus, at Corinth and at Anazarbos in Cilicia.193 In all these cases Malalas describes the event in terms of the direct involvement of Justin in dispensing relief funds. Further, both Edessa and Anazarbos were renamed ‘Iustinopolis’.194

Justinian Augustus, 527 By the beginning of 526 Justinian was in a more secure political position than ever before. He was now the designated successor to Justin. There were no serious rivals. The nephews of Anastasius, especially Hypatius, had accepted the regime of Justin and his adopted son – at least for the time being. Hypatius had spent most of the reign at Antioch as magister militum. It was not long after, in spring/early summer 526,195 that Justin sent Probus, another nephew of Anastasius, to Crimea to recruit reinforcements for the Lazi but he was unsuccessful. Meanwhile, in 526/7 another contingent under Peter was dispatched to support the Iberian king Gurgenes196 but again the forces were insufficient and Gurgenes fled to Constan­ tinople, whereupon Peter was recalled by Justin.197 Justin’s recent setbacks in attempting to defend the Lazi and Iberians against Persian attack heightened his resolve to strike directly at Persian territory. Leading the charge into Persarme­ nia were Sittas and Belisarius, who had been members of Justinian’s bodyguard while he was magister militum, that is, until he became Caesar in 525. They now held positions as local commanders in the east.198 At first their sortie met with some success but they were later repulsed by the Persians. Procopius of Caesarea, 190 191 192 193 194 195 196 197 198

CJ 1.5.12, with Greatrex (2000b), 79. Malalas 17.15 (Thurn 345). Malalas, 17.14 (Thurn 344). Malalas, 17.15 (Thurn 344–5); Theophanes, A.M. 6014 [de Boor 168] (Dyrrachium and Corinth) and 6017 [171.14–28] (Anazarbos). Edessa: Malalas, 17.15 (Thurn 345); Anazarbos: . Theophanes, Chron. AM 6017 (de Boor 171). Ps.-Zachariah Mytilene, HE 12.7: Greatrex (1998), 144. For the date: Greatrex (1998), 145. Vasiliev (1950), 269–71, Greatrex (1998), 139–42 (dating Gourgenes’ request to 524/5). The only scholar to have attempted to identify their positions at this time has been Greatrex (1998), 146 n.26, who plausibly proposed that Sittas was dux Tzanicae. A post for Belisarius is

200

JUSTINIAN UNDER JUSTIN

shortly to become Belisarius’ secretary, is the main source of information for all these raids. In describing them he attributes due authority to Justin.199 So, it is evi­ dent that up to the end of 526 Justin can be seen as exercising military command as Augustus without hindrance or impairment. While Probus was doing battle in the Caucasus, Justinian’s cousin Germanus, magister militum per Thraciam was being despatched by Justin to Thessalonica to confront the Slavs.200 At the same time, word was reaching Constantinople of another catastrophic event in the East, this time a powerful earthquake in Antioch in May 526. It was like nothing experienced in living memory. More detailed reports arrived later and the emperor learnt that up to 300,000 had perished in the quake including the city’s patriarch Euphrasius. Justin was shocked. Antioch was a city he knew and had once inhabited. As Malalas recorded, ‘when he heard of God’s benevolent chastisement of man, he was afflicted with great sorrow; the games were not held in Byzantion and, when the time of Holy Pentecost came, he entered the church without his crown, weeping and wearing a purple cloak along with all the senators who also were wearing purple’.201 Then he set about making provision for Antioch’s rebuilding, dispatching his representatives and writing frequently. John Malalas was in the city at the time and could see the result of Justin’s relief contributions. He may even have come to know Justin’s special envoys Phokas and Asterios. In any event, it is clear from Malalas’ account that in mid-526 Justin was in total control and directly involved in imperial business. Justinian is never mentioned. Compounding the emperor’s grief at the destruction of Antioch was the pro­ gressive worsening of an old war wound. By March 527 there was sufficient concern over the now seventy-four-year-old Justin’s health and presumably his effectiveness as emperor for the senate to once again petition him to hand over more power to Justinian. This time they wanted him to proclaim the Caesar as his full co-emperor (Augustus). Justin responded by duly crowning Justinian Augustus on 1 April 527 in the triclinium of the palace and three days later he was rati­ fied in the Delphax by the assembled soldiers and officials. Theodora was elevated to Augusta at the same time.202 Peter the Patrician has preserved a retrospective truncated account of the ceremony in the Delphax, the very room where Vitalian had been hacked to death seven years before. It was presided over by the vener­ able aristocrat Tatian, as magister officiorum.203 Peter begins by noting that Justin had fallen seriously ill and he had been requested by the senate to make Justinian

199 200 201 202 203

more difficult to posit. They were clearly no longer ‘bodyguards of Justinian’, as Procopius labels them. Procopius, Wars 1.12.20–1. Procopius, Wars 3.40.5. Malalas, Chron. 17.17 (Thurn 350). Augusta: Malalas, Chron. 17.18 (Thurn 351). PLRE 2, 1054–5 (‘Tatianus 2’). He succeeded Symmachus by 520, was later replaced himself by Licinius, then subsequently took on the office again by April 527.

201

JUSTINIAN UNDER JUSTIN

an Augustus.204 That is why the ceremony was held inside the palace rather than in the hippodrome. Meier has argued, however, that the absence of the populace or their representatives from this ceremony indicates Justinian’s resolute assertion of God’s exclusive role in conferring power on him, thereby denying any popular acclaim.205 This is a difficult argument to sustain given Peter’s clear statement about Justin’s health and the fact that Justinian was already Caesar. Further, there is no good reason to ignore the explicit statement of the generally well informed Zonaras, namely that Justinian was separately acclaimed by the people in the hippodrome.206 Peter’s account provides, in fact, the first definite indication of the state of Justin’s health. He seems to have been relatively fit and active until then. Now his days were definitely numbered. This point marks the beginning of Justinian’s reign, which lasted until November 565. He always dated the years of his reign from 1 April 527 being obliged at one stage to specify it with legal force.207 The palace of Hormisdas now became the house of the new Augustus and the new Augusta with the nearby church of St Sergius possibly providing the setting for their imperial liturgy. How long this joint rule would last nobody quite knew. Four months may have been longer, or much shorter, than expected when Justinian became Augustus. The joint emperors proved to be very industrious for the four months of their rule. Firstly, they issued at Constantinople (and Thessalonika) a considerable amount of gold coinage clearly depicting them both as Augusti. Moreover, on the obverses of the various gold coins minted at Constantinople there is signifi­ cant variation in details (globe, cross, throne, spelling of ‘Justinian’ etc.) in the depiction of the two emperors. To numismatists, such variation is explained as the result of a sudden proclamation and the ‘unusual steps taken to hasten its production’,208 a further indication of the swift onset of Justin’s deteriorating condition. As emperors Justin and Justinian issued several laws and decrees,209 204 On the Ceremonies. 95 (Reiske 432.20–433.9, trs. Moffatt and Tall [2012], 432–3): kai. auvto.j ga.r evn no,sw| mega,lh| kate,keito( kai. parekalei/to avpo. th/j sugklh,tou eivj to. poih/sai auvto.n basile,a (433.1–3) Others confirm Justin’s ill health at this point and its role in precipitating Justinian’s elevation with senatorial urging (Cyr. Scyth., vita Sabae. 68 [Schwartz 170.6]; Theophanes, A.M. 6019 [de Boor 173]). It was not a ceremony induced by a sudden medical emergency, as suggested by Dagron (2003), 69. 205 Meier (2003b), 122–3 and (2004), 10–12. 206 Zonaras, 14.40 (Büttner-Wobst 151) with Ravegnani (1989), 24. To all of this might be added the precedent of Leo II, already a Caesar, being made Augustus in 473 on the death of Leo I. 207 Novel 47 (537). 208 Metcalf (1988), 26, cf. Bellinger (1966), 92. 209 There are four definite laws of Justin and Justinian (CJ 1.31.5, 1.5.12, 1.15.2; 12.19.15) but others have been attributed to 527 (CJ 3.11.12; 7.62.36; 4.20.16; 5.3.19; 12.19.13, 12.19.14) some of which could fall in the period 1 April to 1 August. There were also decrees against factional rioters (Malalas, Chron. 17.18 [Thurn 351]). Notable too is the rescript issued to the clergy of the church of St John in Pamphylia in response to their request for imperial support against the purloining of local resources by Roman soldiers (Diehl [1893], 501–20). Justin’s own role in the legislation under his sole name from 518 is brought out thoroughly by Bassanelli (1971), 119–216.

202

JUSTINIAN UNDER JUSTIN

beginning on 22 April 527 with a special imperial rescript designed to transfer to the emperors from the magister officiorum responsibility for filling vacancies in the scholae palatinae.210 The most controversial of their laws was that against her­ etics which reverted to the harsher line characteristic of the beginning of Justin’s reign when he was still under the influence of Vitalian. Justin himself had gener­ ally taken a more tolerant approach between mid-520 and 527.211 It is thought, perhaps correctly, that the law reflects Justinian’s more forceful attitude to dealing with heretics.212 If so, we would have here the first certain sign of Justinian’s over­ riding influence on imperial business. That is not to say, however, that such laws deliberately foreshadow the burst of energy a few years later leading to the Digest, the Code and the Institutes.213 The aftermath of the Antiochene earthquake, now almost a year ago, remained serious. A delegation led by Zacharias, the new comes Orientis in Antioch, came to Constantinople to petition the emperors for further assistance. They returned to Antioch with considerable imperial largesse and more was to follow.214 From Antioch, at about the same time in the summer of 527, Justin and Justinian mobil­ ised the magister militum Libelarius against Nisibis but his assault failed. Libel­ arius was cashiered and replaced by Hypatius. By now, it was evident that Justin’s illness was life-threatening. A contemporary explains that ‘the most sacred Justin became ill from the ulcer which he had on his foot, for he had been struck there in battle by an arrow and that is what caused the danger to his life’.215 This severe physical impediment was the cause of his illness. There is no indication of any mental deficiency or impairment, least of all the senility ascribed to him by mod­ ern scholars from the very start of his reign in July 518.216 Nor is there evidence that before 527 the wound in his foot had interfered with his capacity as emperor. Justin eventually died on 1 August 527, exactly four months since he had taken on Justinian as his co-emperor. There is no doubt he died of natural causes. The contemporaries Malalas, Marcellinus and Cyril of Scythopolis record his death without comment, while Procopius notes that Justin died ‘a natural death’ at Con­ stantinople. Otherwise, Procopius’ approach to Justin is not to be trusted. Although in the Secret History Procopius combines into a single reign all the years of Justin and Justinian,217 at the same time he exposes the shallowness of his claim by clearly and consistently differentiating between their reigns,218 CJ 1.31.5. Meier (2003b), 200–1; Vasiliev (1950), 239–41. Vasiliev (1950), 241–50. So Vasiliev (1950), 391. Malalas, Chron. 17.22 (Thurn 353). Malalas, Chron 17.23 (Thurn 353). Later writers who were dependent on Malalas, for instance John of Nikiu 90. 47, transposed the war-wound from Justin’s foot to his head. 216 E.g. Stein (1949), 222; Browning (1987), 23. 217 Secret History, 1.19. 218 Procopius denotes the commencement of Justin’s reign (Secret History 6.11, 6.17, 19.), dates particular events to Justin’s time (11.5) including Justinian’s courting of Theodora (12.29). Then

210 211 212 213 214 215

203

JUSTINIAN UNDER JUSTIN

even going so far as distinguishing two successive phases of Justinian’s power: first (pro,teron) in terms of his due authority under Justin, and later (u[steron) with full imperial power.219 It is an accurate distinction also found in the Wars where Procopius notes that while Justinian did ‘not yet’ exercise imperial power under Justin he still wielded influence commensurate with his legal authority (kat v evxousi,an).220 This view is repeated in the encomiastic context of the Buildings where Procopius justifies including buildings from Justin’s period in the Justini­ anic list because ‘Justinian administered the government also during his uncle’s reign but on his own authority (kat v evxousi,an)’.221 What Procopius means is that during Justin’s reign Justinian held his own positions of high authority as magister militum and Caesar. That is very different from saying that from July 518 Justin­ ian enjoyed full imperial power and influence which has been the predominant modern interpretation of Procopius on Justin. The point of Procopius attributing all of Justin’s reign to Justinian in the Secret History is purely rhetorical. It makes the target of his abuse responsible for all disasters, adversity, crimes, injustices and loss of life during those years.222 For example, the assassinations of Amantius in 518 and Vitalian in 520 are laid at Justinian’s door. While it is possible, as we have seen, that Justinian did in fact play some role in both murders it was an easy claim to make over thirty years later. Likewise, Procopius includes in his inventory of divine retribution against Justin­ ian all those natural disasters that had occurred in Justin’s reign. This enabled him to magnify divine disapproval of Justinian.223 Following the evident thrust of Procopius’ approach to Justin, especially the combination of his reign with that of Justinian, modern scholars have generally assumed that when he reckons years of Justinian’s reign in the Secret History Procopius must be counting from

219 220 221 222

223

he records that Justin ruled for nine years (19.8), as well as noting the conclusion of his rule in August 527 (9.54). Similarly, Procopius notes the commencement of Justinian’s reign (8.4; 11.1; 18.36, 24.20, 25.5, 26.15). Moreover, three times he links the beginning of Justinian’s reign with his consort Theodora (8.53, 9.54, 30.27). Certainly, in those instances Procopius can only mean to say that Justinian’s reign began in April 527. Procopius, Secret History. 18.45: ~O VIoustinianou/ pro,teron ~Rwmai,wn dioikoume,nou th/n polutei,an kai. u]steron th.n auvtokra,tora avrch.n e;contoj) Procopius, Wars 3.9.5. Procopius, Buildings 1.3.5. Procopius, Secret History 6.20–5. It has been suggested by Meier (2003b), 88 n.154, that other authors, notably John of Ephesus and the later Chronicle of Zuqnin, also collapse the reigns of Justin and Justinian. However, the references cited only support calculating the period of perse­ cution from 518 to 566/7 (John’s date of writing), not the amalgamation of two separate reigns. Similarly, elsewhere Meier (2003b), 124 with n.28, cites Cyril of Scythopolis, vita Sabas 68 as evidence of the fact that Justinian was the power behind the throne since 518. Again, Cyril says no such thing, only that when Justinian was made emperor by Justin in 527 he had a good under­ standing of imperial business based, as would only be expected, on his experience as ‘patrician, consul and general’. Procopius, Secret History 18.37 (tou,tou ga.r VRwmai,wn dioikoume,nou ta. pra,gmata). There were floods at Edessa (18.38), along the Nile (39) and at Tarsus (40); earthquakes at Antioch (41) and Anazarbus (41) plus other cities (42) including Amasia and Corinth.

204

JUSTINIAN UNDER JUSTIN

518. In the Wars Procopius frequently records a particular event as occurring in a certain imperial year of Justinian,224 but it is clear that he is counting from 518, not from 527, which was how the emperor himself decreed that his reign was to be counted.225 In other words, Procopius’ tendentious purpose in the Secret History, attributing all evils to him whether from the reigns of Justin or his own, means that when he says he is writing in the 32nd year of Justinian he must be reckoning from 518.226 Given the Secret History’s internal inconsistencies and contradictions in describing the reign of Justin and the role of Justinian during that period, not to mention its highly polemical nature, all Procopius’ claims about Justin need to be taken with careful scepticism. The success of Procopius’ technique depended on sufficient uncertainty, distortion and hazy memory among his audience about events now some three decades ago. Naturally enough, later generations right down to the present have found it even harder to resist Justin’s misrepresentation by Procopius.

Recapitulation For almost all his reign Justin can be shown to have taken an active role in direct­ ing and supervising his generals, promulgating legislation, dealing with matters of religious policy, conducting relations with other states and potentates, and in responding to natural disasters. The rapid movement towards settlement of the Acacian schism with Rome in the weeks and months immediately after his elec­ tion in July 518 show an emperor personally engaged in events and in full control. On the contrary, there is no indication that in the early years of Justin’s reign, his nephew Justinian held power and influence beyond what might be expected from his own high official positions and privileged family connection. Moreover, Justin was not advancing his nephew’s career, instantly and hastily. Justinian remained in the same position he held under Anastasius, that is, as one of the forty imperial guards. Until his death in July 520 the general Vitalian was the more powerful figure. By now Justinian had been promoted to an equivalent position as court general but was soon under suspicion. As a result of his partisanship for the vio­ lent Blues he almost forfeited his life in 523, a low-point at which he appeared particularly powerless and vulnerable. It was around the same time that he was thwarted by the empress Euphemia from marrying his beloved Theodora. He was simply not strong enough to outflank, ignore or override Euphemia’s influence 224 Wars 1.16.10 (year 4); 1.22.17 (year 6); 2.3.56 (year 13); 2.5.1 (year 13); 2.28.11 (year 19); 2.30.48 (year 23); 3.12.1 (year 7); 4.14.6 (year 10); 4.19.1 (year 13); 4.21.1 (year 17); 4.28.41 (year 19); 5.5.1 (year 9); 5.14.14 (year 11); 8.15.12 (year 25); 8.33.26 (year 26). 225 Novel 47 (537). 226 This claim underpins the argument that the Secret History should be dated to 550/1: Greatrex (1994), 101–14 and (2003), 45–67; Evans (1996b), 301–13; Signes Codoñer (2003), 47–82. The most thorough case is made by Kaldellis (2009). The argument that the Secret History should be dated to 558/9 because the 32 years of Justinian’s reign to date begin, like the Wars, in 527 was made by Scott (1987) 215–21; Croke (2005c) 405–32.

205

JUSTINIAN UNDER JUSTIN

with Justin. Justinian’s status and power were subsequently increased in measured steps, with the senate urging and Justin resisting at each point. The bestowal of the title of nobilissimus in 524 and his elevation to Caesar in 525 marked the key transitional phases in Justinian’s rise to power. Generations of historians have been misled by the special pleading of Procopius who in his Secret History sought to combine the reigns of both Justin and Jus­ tinian into a single uninterrupted whole. The contemporary documentation from Justin’s reign, especially the laws and the correspondence in the Collectio Avellana, tell a different story. Justinian’s authority during the reign of Justin from July 518 to April 527 was not abrupt and absolute, but grudging and gradual. At the same time, Justinian’s experience under Justin had enabled him to accumulate invaluable insight into how an emperor conducted the daily business of managing the court, government, military, bureaucracy and the delicate religious positions among competing bishops. That explains why his sole reign from August 527 commenced with such a flurry of religiously inspired legislation and decisions.227

227 Leppin (2011), 90–1.

206

8 JUSTINIAN, THEODORA AND THE CHURCH OF SAINTS SERGIUS AND BACCHUS*

The emperor Justinian was an avid builder of churches. To this day his name immediately evokes at least two world-renowned edifices which symbolise the piety and power of the imperial builder – Hagia Sophia in Constantinople and San Vitale in Ravenna.1 One of the grand surviving churches of Justinian, Sts Sergius and Bacchus at Constantinople, now the mosque Küçük Ayasofya Camii, was built within the Hormisdas palace complex where he once lived.2 There has been considerable debate about when it was built and the precise circumstances surrounding its construction. The prevailing consensus, first formulated by Mango over thirty years ago, is that Justinian’s role in its construction was subsidiary. Instead, Justinian’s wife, the empress Theodora, provided the impetus for the church. She had it constructed rapidly in the early 530s, designing it to serve the liturgical needs of a large group of refugee monks persecuted for their opposition to the Christological definition of the council of Chalcedon. These adherents of the belief in Christ’s single nature, later known as Monophysites, were driven from their eastern abodes before being welcomed at Constantinople by Theo­ dora and housed in the palace of Hormisdas. Before long they needed their own church.3 This reconstruction of events has been used to give greater prominence and independence to Theodora in imperial religious policy, and to suggest that

* This chapter was originally published in Dumbarton Oaks Papers 60 (2006), 25–63 and has been reprinted by permission of Dumbarton Oaks Research Library and Collection, Trustees for Harvard University. I am most grateful to Jonathan Bardill for helpful advice on an earlier version of the original paper. This revised version takes account of Bardill’s subsequent proposals on the genesis and later dating of the church (Bardill [2017]). 1 On Justinian as church builder: Krautheimer (1986), 225–7; Alchermes (2005), 343–75, with essen­ tial background in Mango (1985), 35–8. 2 Details in Ebersolt and Thiers (1913), 21–51 (fundamental); Van Millingen (1912), 62–83; Janin (1969), 451–5; Sanpaolesi (1961), 116–80 (richly illustrated, including architectural cross-sections); Müller-Wiener (1977), 177–83; Mathews (1971), 42–51 and (1976), 242–59. Note also Mango (1985), 58–9; Grossmann (1989), 153–9; Svenshon and Stichel (2000), 389–409; Svenshon (2013), 113–28 (with digital recreations of space); Bardill (2017), 62–130; Ousterhout (2019), 111–19. 3 Mango (1972); Krautheimer (1974), 251–3; Mathews (1974), 22–9; Mango (1975), 385–92; Bardill (2000), 1–11. Earlier views: Van Millingen (1912), 65; Ebersolt and Thiers (1913), 23–4.

207

JUSTINIAN, THEODORA AND THE CHURCH

Justinian was both easily manipulated by her and lukewarm about enforcing his own doctrinal decrees in favour of orthodoxy.4 The following contribution to the scholarly discussion on Sts Sergius and Bac­ chus proceeds, firstly, by focussing on Justinian’s church building during the years in which he resided at the Hormisdas palace (Section I), then by demonstrating that the church of Sts Sergius and Bacchus cannot have been established at any stage for the refugee Monophysite monks in the palace (II–IV). Next, it moves to propose a construction date for the church commencing well before Justinian became Augustus in April 527 (V), which leads to consideration of the genesis of Justinian’s church (VI). Recent research by Connor, Bardill and Shahid has drawn attention to the stylistic and other affinities between Sts Sergius and Bacchus and the earlier church of St Polyeuktos built by Anicia Juliana, Constantinople’s most distinguished woman at the time,5 while Bardill’s redating of St Polyeuktos to the early 520s6 indirectly throws new light on the context in which Justinian con­ structed Sts Sergius and Bacchus. This research is now expanded by arguing that the impetus for the construction of Sts Sergius and Bacchus is to be found in Justinian’s political rivalry with Anicia Juliana in the early to mid-520s, rather than in any attempt to be seen as open to monophysites.7 This earlier date better explains not only the political context of the church’s founding, but also its loca­ tion and function. Insufficient attention has been paid to the church’s dedicatory inscription and other key texts, as well as to the political and religious background of the actual period of Justinian’s residency in the Hormisdas palace between 518 and 527. Significantly different implications follow from a church conceived and largely built by the orthodox emperor Justinian on his own property during the reign of his uncle Justin, compared to one ascribed to the Monophysite Theodora several years later (after 532), thereby making it contemporary with the imperial church of Hagia Sophia), but when the imperial couple had no longer lived in the palace of Hormisdas for at least five years.

Justinian and his churches, 518–527 Justinian first came to notice, and first came to be involved in the building of churches, during the reign of his uncle Justin who secured the imperial throne in July 518. At the time of Justin’s coronation his adopted son Justinian was one of the emperor’s own personal bodyguards (candidati).8 Justin was in his mid-60s 4 The impact of Mango’s thesis can be measured in later studies that take it for granted, and use it to reinforce the powerful role of Theodora: Evans (1996a), 110–11 and (2002), 73–4, 81–3, 106–7; Noethlichs (1999), 702; James (2001), 150–2; McClanan (2002), 100–2. 5 Connor (1999), 479–527; Bardill (2000), 4; Shahid (2003), 475–6 and (2004), 343–55. 6 Bardill (2004), 62–4, 111–16. 7 As argued by Bardill (2017), 83–6. 8 Victor Tonnenensis, Chron. 101: s.a. 518 (ed. Hartmann, CCh 173a, 33 = Mommsen, MGH.AA XI: 196); Petrus Patricius in Const.Porph., On the Ceremonies 1.93 (ed. Reiske, 428.4, trs. Moffatt and Tall [2012], 428).

208

JUSTINIAN, THEODORA AND THE CHURCH

at the time and Justinian in his late 30s but still unmarried. Being the emperor’s nephew and now bearing the honorary title of illustris, he required his own man­ sion. Shortly afterwards, by mid-520, Justinian was promoted to the position of magister militum praesentalis. At this stage he was clearly living in the palace of Hormisdas which was constructed not long after the foundation of the capital in the 330s and was located between the hippodrome and the Sea of Marmara, in the district long known as ta Hormisdou.9 It is likely that Justinian occupied the Hormisdas mansion from the time of Justin’s accession, or thereabouts. Like other such mansions at Constantinople (of which there were a considerable number by the 520s) that of Hormisdas must have been a compact collection of buildings and gardens. No physical traces of the palace remain. All we know about it is that it had a large heptaconch dining hall (triklinos),10 which would make it similar to the mansions of Lausus and Antiochus whose remains can be identified.11 It has been remarked of these imposing edifices that ‘we cannot help being struck by their enormous proportions and ceremonial character’.12 Doubtless such aristocratic estates, including that of Hormisdas, resembled the mansions at Rome which struck a visitor from Egyptian Thebes as medium-sized cities in themselves, each with fora, temples, fountains and baths. He was even prompted to verse: ‘one house is a town; the city hides ten thousand towns’.13 Once he occupied the Hormisdas palace in 518 or early 519, Justinian lost no time in maximising the opportunity to build a memorable and significant church on his new property.14 On 29 June 519 he inquired of Pope Hormisdas as to whether the pope could arrange for some relics of Saints Peter and Paul, as well as St Laurence, to be forwarded to him at Constantinople. He wanted them for the ‘Church of the Holy Apostles’ he was then constructing ‘hic in domo nostra’, that is to say, as part of the Palace of Hormisdas.15 Taking no chances, he arranged for the papal legates then in Constantinople to further press his claim on the pope. They suggested that the relics of each saint be sent in a separate casket.16 Pope 9 Janin (1964), 358–9. 10 Innocent of Marona, Epistula de collatione cum Severianis habita, 4–6 (ACO, 4.2 [169]): ‘in uenerabili palatio suo quod cognominatur Hormisdae . . . prima itaque die conuenimus in eodem eptaconco triclinio . . .’. 11 Greatrex and Bardill (1996), 171–97; Bardill (1997), 67–93. 12 Mango (1990), 28. 13 Olympiodorus, frag.41.1 (Blockley, 204–5 = Photius, Bibl., 80) :eivj do,moj a;stu pe,lei\po,lij a;stea muri,a keu,qei) 14 Janin (1964), 358–9; Guilland (1951), 294–303. 15 CA 187.5 (645): ‘praesumentes autem de beatitudinis uestrae beniuolentia paternam dilectionem nimium petimus, quatenus reliquiis sanctorum apostolorum tam nos quam basilicam eorum hic in domo nostra sub nomine praedictorum uenerabilium constructam illustrare et illuminare large dignemini, cognoscentes, quod nullum nobis maius nec munus nec beneficium praestare potestis, domine beatissime pater, quam si hanc nostrum petitionem adimpleueretis’. 16 CA 218.1 (679): ‘Filius uester magnificus uir Iustinianus res conuenientes fidei suae faciens basilicam sanctorum apostolorum, in qua desiderat et beati Laurentii martyris reliquias esse, con­ stituit: sperat per paruitatem nostrum ut praedictorum sanctorum reliquias celeriter concedatis . . .

209

JUSTINIAN, THEODORA AND THE CHURCH

Hormisdas replied positively on 2 September 519, so that Justinian had the rel­ ics for his church by the end of the year, at least he had relics of Peter and Paul, but perhaps not Laurence which he fails to mention any more.17 A contemporary witness, Procopius from Caesarea, confirms that near the palace of Hormisdas Justinian built his church which came not to be called the ‘Church of the Holy Apostles’, as he may have originally intended even though Constantine had built a church of that name elsewhere in the city, but the ‘Church of Peter and Paul’ whose relics he had received from Rome. It was of longitudinal basilica design at a time when ecclesiastical architecture at Constantinople was moving away from such design.18 Its dedicatory Greek epigram, presumably on the outside entrance, singles out the building’s beauty and splendour, as well as indicating that it was dedicated by ‘Justinian’ with no other epithet, thereby suggesting it was com­ pleted before his consulship in 521, and possibly even before he became magister militum in mid-520.19 Otherwise, the consulship might well have been included as a title for Justinian, as well as his generalship, which we find recorded in the dedication of the church of Theodorus around the same time.20 It has been further proposed, possibly correctly, that the church was built rather hastily.21 Still, in building a church on what was for the moment his own property Justinian was doing no more than other aristocrats before him, most recently Anicia Juliana with her magnificent church of St Polyeuktos.22 While he lived in the palace of Hormisdas, Justinian’s advancement continued. There was no more senior court position available for him than magister militum, but he was progressively installed with the highest honorific titles. Step by step, his status was increased.23 In 523 he acquired the title of patricius24 which finally put him on a par with the surviving nephews of the previous emperor Anastasius as well as with Olybrius, Anicia Juliana’s son for whom she still harboured impe­ rial ambitions. Shortly afterwards Theodora, not yet Justinian’s wife, was made patricia. They were both now living in the palace of Hormisdas. Two years later

17 18 19 20 21 22 23 24

(3) (680) unde si et beatitudini uestrae uidetur, sanctuaria beatorum Petri et Pauli secundum morem ei largiri praecipite . . . petit et de catenis sanctorum apostolorum, si possible est, et de craticula beati Laurentii martyris’. CA 190.4 (648): ‘beatissimorum uero apostolorum Petri et Pauli sanctuaria, sicut religiosissimo quaesitis affectu, per harum portitorem sub omni ueneratione transmisimus, optantes, orationibus eorum mentis uestrae oblatio et desideria gratiae sint diuinitatis accepta’. Procopius, Buildings 1.4.1, with Bardill (2004), 34. Anthologia Palatina, 1.8, l.2. Anthologia Palatina, 1.97, l.4: stratih/j h`gh,tori pa,shj; 1.98, l.2: megasqene,oj stratia,rcou. For his inclusion in the dedication of Theodorus: Alan Cameron (1976a), 278–9. Mango (1972), 189. Cf. Mango (1986a), 127–8. For details: Croke (2005c), 405–32 and (2007) and Chapter 7, 169–206. Victor Tonnenensis, Chron. 107: s.a. 523 (ed. Hartmann, CCh 173a, 35 = Mommsen, MGH.AA XI, 197). Other references to Justinian as patrician: Cyril Scyth., vita Sabas 68 (ed. Schwartz 170.20ff); Chronicon Edessenum s.a. 836 [CSCO 3.4, 10]; John of Ephesus, vitae 13 [PO 17.2, 189]); John of Nikiu, 90.16–18.

210

JUSTINIAN, THEODORA AND THE CHURCH

he received the rare title of nobilissimus25 that was an honour confined to children or, in this case, adopted children of the emperor. Then in 525, at the request of the senate, the ageing emperor Justin reluctantly elevated his nephew to the posi­ tion of junior emperor or Caesar.26 To have been inaugurated as Caesar in 525 Justinian must have been considered the designated heir who now possessed suf­ ficient authority and presence to be the next Augustus. The Caesar was in effect a ‘second emperor’ (deu,teroj basileu,j), having all the trappings and apparel of an emperor but not the full crown. The title and position did constitute, however, the unassailable guarantee of succession.27 Becoming Caesar entailed a distinc­ tive inaugural ritual involving the senior emperor and all the court officials.28 A Caesar immediately created a separate court ceremonial with associated dignitar­ ies. To express his elevated status, a Caesar needed his own palace for himself and family, his own staff and resources, his own ceremonial. So, in 525 the palace of Hormisdas, where Justinian had already lived for seven years or so, took on a new significance and possibly required modification as the palace of a reigning Caesar. In April 527, with Justin now deteriorating in health, Justinian became co­ emperor (Augustus), then sole emperor a few months later following Justin’s death in August 527. Justinian always dated his reign, however, from 1 April 527. As Augustus for the next 38 years Justinian continued to build churches in Constanti­ nople and beyond, including Holy Apostles and St Irene, Sts Cosmas and Damian, and the grandest and most ambitious of all, Hagia Sophia. Moreover, his name was associated with nearly every building project carried out during his reign. As he decreed in 538, any new church should reflect the emperor’s majesty and piety.29 During the nine years of Justin’s reign Justinian was actively involved in the building and refurbishment of other churches at Constantinople besides Sts. Peter and Paul. Sometimes he must have been engaged with several projects simultane­ ously. Some of them, mostly constructed in the fourth century, commemorating local or nearby martyrs, Justinian rebuilt completely or refurbished. They include St Acacius, St Plato, St Mocius, St Thekla (near the Harbour of Julian) and St Thyrsus, as well as others outside the city, notably the monastery of St Theodore

25 Marcellinus, Chron., s.a. 527 (ed. Mommsen, MGH AA XI. 102); Zonaras 14.5.37 [150.15]). For the ceremony surrounding the conferring of the title: Const.Porph., On the Ceremonies, 1.44 (53), ed. Reiske, 225, trs. Moffatt and Tall (2012), 225. 26 Victor Tonnenensis, Chron. 109: s.a.525 (ed. Hartmann, CCh 173a, 35 = Mommsen, MGH AA XI, 197) says that in 525 the senate compelled Justin, against his will, to proclaim Justinian as Caesar. 27 Caesar: Jones (1964), 322–3; Guilland (1967a), 25–43; Rösch (1978), 36–7. The Caesar’s role is neatly summarised in vita Marcelli. 34 (ed. Dagron [1968], 316–17): ~O de. tou/to e;cwn to. avxi,wma deu,tero,j evstin basileu,j avlourgi,da kata,cruson kai. ta. a;lla th.j basilei,aj para,shma evnduo,menoj di,ka mo,nou tou/ stefanou/ & o[per evpi. tou Faraw. kai. tou/ ~Iwse.f avkhkoa,men) ~O kai/sar toi,nun kai. a;cri peri,estin o` basileu.j pa,nta koinῇ pra,ttwn met vau/tou/( kai. teleuth,santoj monoj th.n basilei,an krateῖ mhdeno.j de. pro. au/tou/ mhde.n su.n auvtw/| tolmw/ntoj evkei,nhn labei/n) 28 Coronation: Const.Porph., On the Ceremonies, 1.43 (52), ed. Reiske 217–25, trs. Moffatt and Tall (2012), 217–25. 29 Justinian, Novel 67, cf. Procopius, Buildings 1.8.5.

211

JUSTINIAN, THEODORA AND THE CHURCH

at Resion and St Theodota’s church at the Hebdomon.30 In addition, from the same period he is also credited with other substantial constructions such as the Church of the Virgin at Blachernai, one of the most celebrated churches dedicated to the Virgin in the Byzantine world. According to Procopius, Justinian altered and improved the original basilical building by giving it a dome supported by columns forming a semicircle.31 There is also another church not mentioned by Procopius, the Melete, for which Justinian is given some credit in its dedicatory epigrams dating from Justin’s reign.32 At some point during his imperial career, which had begun in 525, the church of Sts Sergius and Bacchus was built within the Hormisdas precinct where Justinian lived between c. August 518 and August 527. Overlooking the Sea of Marmara, a little south of the hippodrome, the church still survives although considerably modified over the years.33 In terms of design the church is octagonal but built within a rectangle and enclosed by a dome, although the present dome may be much later.34 The irregularity of the building, both in orientation and symmetry, can be explained by either sloppy workmanship or the exigencies of accommodat­ ing the design to the pre-existing church that it abutted.35 Procopius clearly credits Justinian with the construction of the church and describes its interior as shining with polychrome marble and splendid mosaic decoration standing out on a gold background.36 Its marble and mosaics have disappeared but its inscription hon­ ouring St Sergius remains intact. To the doyen of modern students of Byzantine architecture, ‘Sergius and Bacchus represents one of Byzantium’s most innovative architectural creations’.37 Even in its present decoratively spare form this church is still one of the city’s finest monuments. Yet, a teasing question remains: when was it built? It is generally agreed that Sts Sergius and Bacchus cannot have been completed any earlier than August 527 when Justinian became sole Augustus. This is the clear implication of the church’s entablature inscription that mentions Justinian as ‘emperor’ (skhptou/coj). Nor can the date of construction be later than the Con­ stantinople synod of bishops in May 536, the first datable documentary record 30 Procopius, Buildings 1.4.25–9: Acacius (Janin [1969], 18); Plato (Janin [1969], 418); Mocius (Janin [1969], 367–71); Thyrsus (Janin [1969], 257); Theodore (Janin [1969], 157–8); Thecla (Janin [1969], 149); Theodota (Janin [1969], 153). 31 Janin (1969), 169–79. 32 Anthologia Palatina, 1.97–8 with Feissel (2000), 88–9. For Theodorus and the dating of his church to exactly 520: Cameron (1976c), 274–83. 33 For detailed description: van Millingen (1912), 70–83; Ebersolt and Thiers (1913), 21–51; Mathews (1971), 242–59; Bardill (2017), 105–22, Possibly Procopius was referring to the two churches within the Hormisdas mansion when he accused Justinian of wasting the treasury’s resources on ‘senseless buildings on the sea’ (Secret History 19.6, cf. 8.7–8). 34 Bardill (2017), 105–18, but disputed by Ousterhout (2018), 104 n.4 who considers the dome original. 35 Mango (1986a), 59. 36 Buildings 1.4.1–8. 37 Ousterhout (2018), 112.

212

JUSTINIAN, THEODORA AND THE CHURCH

of the church’s existence. Mango contended that the church’s date was closer to 536, since it was begun only after Justinian and Theodora had vacated the Palace of Hormisdas in 527, and was therefore ‘not a precursor, but a contemporary’ of Hagia Sophia.38 He proposed further that the church of Sts Sergius and Bacchus was built by the empress Theodora expressly for the use of Monophysite monastic refugees who had fled persecution in the east.39 According to John of Ephesus a large number of these arrived at Constantinople at some stage and were settled in the imperial palace of Hormisdas where Justinian and Theodora no longer dwelt. The case against Mango’s position was immediately sketched in summary form by Krautheimer,40 then Mango produced a detailed response which has largely won the day.41 While Krautheimer had identified most of the key weaknesses in Mango’s proposal, their impact was lost because he never developed his objections systematically and in detail. Bardill later acknowledged the strength of some of Krautheimer’s arguments, while reinforcing Mango’s thesis of a church founded to meet the needs of refugee Monophysite monks.42 Where Mango had argued, however, that the church of Sts Sergius and Bacchus is to be identified with the martyrium located by John of Ephesus in a large room of the palace of Hormisdas, Bardill has proposed that the new church was not the martyrium but the separate vaulted building which John describes as replacing the martyrium after it had collapsed. The new church was needed, so it is argued, because the Monophysite monks in the palace of Hormisdas were now without a hall for prayer, while its elaborate design is to be explained by ‘Justinian’s desire to court the Monophy­ site dignitaries who were then [i.e. in 532] present in Constantinople’.43 Given this clear impetus for the construction of Sts Sergius and Bacchus, it is dated by Mango and Bardill to the early 530s.44 Subsequently, Shahid sought to downplay the role of Theodora in the church’s construction, proposing instead that its ratio­ nale and its dedication to the eastern military saint Sergius were provided by the outbreak of Justinian’s war against the Persians, particularly the commemoration of the Roman victory over the Persians at Dara in 530. However, Shahid continued to see the church as playing a key role in Theodora’s promotion of Monophysites and Justinian’s quest for doctrinal reconciliation with them.45 Most recently, Bardill has re-framed the argument that it was built in the period after 532 and for overtly Monophysite purposes, which would make the Church of Sergius and Bacchus an exact contemporary project with the much larger and 38 39 40 41 42 43 44

Mango (1972), 192, cf. (1975), 386 (c. 535), 392 (between 531 and 536). Mango (1975), 392. Krautheimer (1974), 251–3. Mango (1975), 385–92. Bardill (2000), 1–11. Bardill (2000), 9. Dates proposed: towards 536 (Mango [1972], 192); 535 or earlier (Mango [1975], 386); after 531 (Mango [1975], 394); possibly 527–533 (Bardill [2000], 4); 530–6, possibly 530–3 (Bardill [2000], 9–10). Most recently Bardill (2017), 81 (‘532–6’). 45 Shahid (2003), 479–80.

213

JUSTINIAN, THEODORA AND THE CHURCH

more elaborate Hagia Sophia, so that Justinian was watching the two churches rise to completion at precisely the same time.46 This latest case, interpreting the church as ‘symbolic of religious reconciliation’, rests essentially on linking two specific records – Procopius’ account of the church, and a later description of the 532 bishops meeting in the Palace of Hormisdas – to demonstrate that (1) the Church of Sergius and Bacchus can only have commenced after the Palace of Hormisdas was linked to the Great Imperial Palace, and (2) that occurred only after the Chalcedonian/Monophysite discussions held in the Palace of Hormisdas in 532, in fact immediately after. However, neither text can bear the explanatory weight put upon it.47 Firstly, Bardill misinterprets Procopius Buildings 1.4 by necessarily requiring the text to date the joining of the palaces (Hormidas and Imperial Palace) before the construction of the church. He claims that such an interpretation is simply the ‘natural inference’ to draw from Procopius’ statement.48 However, the Procopian statement is parenthetical, rather than strictly sequential. He then cites Buildings 1.10 to propose that the Palace of Hormisdas was disfigured by fire during the Nika riots (January 532) and Justinian was obliged to rebuild it, so that the ‘neces­ sity of extensive repair work on nearby parts of the Great Palace provided an ideal opportunity to connect the two’.49 To the best of our knowledge, the Palace of Hormisdas was well distant from the areas fired during the riots.50 Bardill next goes on to date the joining to mid-532, precisely, on the basis of an anonymous and fragmentary 8th or 9th century Syriac manuscript, which provides an account of the discussion that occurred in 532 in the Palace of Hormisdas. The document says (c.3) the parties met in ‘the hall known as Beth Hormisdas which is today joined to the Palace’. Bardill accepts the contention of the document’s editor that the joining of the palaces occurred ‘shortly afterwards’, that is shortly after the meeting, but this need not be so.51 Moreover, the attribution of this statement to one of the discussion participants (John bar Aphtonia) writing his account of pro­ ceedings shortly after is tenuous. The best we can say is that the 8th/9th century Syriac fragment represents the perspective of a later writer’s description of the 532 discussions even though it may have drawn on the contemporary record of John bar Aphtonia. The fact remains that we do not know when and how the pal­ aces were joined. Since it had evidently not happened by the time of the doctrinal discussions held there in 532 it must be dated sometime after 532, though not immediately after as suggested by Bardill. What the connection involved and how long it took to plan and construct are unclear but it is an uphill distance from the Hormisdas palace to the Great Palace. The connection may have been the same as 46 47 48 49 50 51

Bardill (2017). Bardill (2017), 67–70. Bardill (2017), 67–71, cf. Roques (2011), 122. Bardill (2017), 73. Westbrook (2011), 27. Bardill (2017) 73, following Brock (1981), 92 n.17.

214

JUSTINIAN, THEODORA AND THE CHURCH

the aerial ‘Galleries of Marcian’ that was built later in the reign of Justinian (say early 550s).52 If that is the case, then the ‘Marcian’ in question may be Justinian’s nephew who first appears in the record as a court general in 563.53 In any event, it is not necessary to have the palaces joined for the refugee monks to find a home there. So, the question remains that if the Church is to be dated on the basis of its function for monophysite monks the chronology of the monks arrival in Constan­ tinople requires closer examination. The notion that this residency of the Monophysite monks in the palace of Hor­ misdas gave rise to the original construction of the church of Sts Sergius and Bacchus depends entirely on a single chapter of the vitae of John of Ephesus. In interpreting it both Mango and Bardill have assumed that John is describing events from the late 520s into the 530s, and that by then there were already very large numbers of Monophysite monks in Constantinople requiring accommodation.54 This is a mistaken assumption. Neither Mango nor Bardill has distinguished suf­ ficiently between the two distinct phases in the persecution of Monophysites in the east, each with quite different consequences for the persecuted: one in the early to mid-520s, the other from 536/7. For both phases, John of Ephesus is the essential and contemporary witness. Clarifying the nature and timing of the imperial action against the Monophysites reopens the fundamental questions about the date, pur­ pose and original context of the church.

Persecution of Monophysites in the 520s John of Ephesus (at this stage a monk at Amida), who spent part of his life at Constantinople, provided an eye-witness account of how Justinian’s palace of Hor­ misdas was transformed into a monastery at an unspecified date. He had known the city over a long period since his first visit to Constantinople in 535. In a chap­ ter of his Lives of the Eastern Saints, written between 565 and 567, John describes ‘the community of blessed men which was gathered together in the royal city by the believing queen at the time of the persecution out of many peoples and various local tongues’.55 He goes on to say that there were about 500 monks in the city constituting a striking array of ascetics from east and west, from Syria and Arme­ nia, Cappadocia and Cilicia, Isauria and Lycaonia, Asia and Alexandria and from 52 As proposed by Bolognesi (2000), 231, cf. Kostenec (2005), 33. 53 PLRE 3, 821–3 (‘Marcianus 7’). 54 Mango (1972), 192 n.10 (refugee monks in Constantinople from 527); (1975), 386 (from 531); Bardill (2000), 6 (from 527), similarly in Voöbus (1958), 212. The notion that 500 monks were occupying the Palace of Hormisdas as early as 531 appears to have originated with Duchesne (1915), 59 and (1925), 81. Krautheimer (1974), 252 had pointed out that such a large number of monks could not have been concentrated in Constantinople before 537, but he did not elaborate. Bardill (2017), 82 finds this position ‘plausible’ and notes that, separately, Menze (2008), 23 came to the same conclusion, that is, monophysite refugee monks were not flooding into Constantinople by the early 530s. 55 John of Ephesus, vitae 47 (PO 18. 676–7).

215

JUSTINIAN, THEODORA AND THE CHURCH

Constantinople itself. To enter the palace of Hormisdas, where a large number of them resided, was to step ‘into a great and marvellous desert of solitaries, and mar­ vel at their numbers and wonder at their appearance, and the same men’s honoured old age, and be affected by the crucifixion of their bodies and their practices of standing and the same men’s spiritual songs which are heard from all sides, and at their marvellous canticles and their melancholy voices which were performed and uttered in all the chambers and courts and cells and halls of that palace’. There were displaced stylites, recluses and solitaries all trying to find peace and isola­ tion within the space of the palace. It was crowded but orderly, with the monks in one of the main halls functioning as one convent and governed accordingly. Other parts of the palace were occupied by solitaries in cells. What took place there were ‘severe labours and protracted fasts and constant vigils and perpetual prayers, as well as celebrations and descents of the Spirit in every place’.56 No wonder these ascetics attracted considerable local interest. Certain implications of John’s detailed and reliable account need immediate elucidation: (1) in accordance with the strictly segregated monastic rule of the time, the Hor­ misdas monastery was for males only, functioning as a single monastic house under a single leader;57 (2) also in accord with monastic strictures and Justinian’s own decree it was set up as a single open sleeping area with common meals,58 requirements ideally met by a large open hall area such as that at Hormisdas, as well as with special provision for solitaries through the creation of private cells by partitions and other means; (3) the community was a diverse one which can only have been brought together over time and which can only be the result of threats of persecution reach­ ing as far as the isolated solitaries of the Syrian desert. It was clearly not the result of a single massive influx from a single region, nor was it a makeshift emergency arrangement or refugee camp;59 (4) once established, its novelty and spiritual power attracted a constant stream of visitors. The manifest holiness demonstrated by the residents of the tempo­ rary Hormisdas monastery impressed Chalcedonians and Monophysites alike. Moreover, Theodora is given credit for turning over the imperial residence, although John acknowledged that she had the clear support of Justinian. The 56 John of Ephesus, vitae 47 (PO 18. 677). 57 Justinian, Novel 5.3 (535); reiterated in Novel 133.1 (539) and Novel 123.36 (544), with Alivisatos (1913), 105–12; Hatlie (2007), 143–50. 58 Justinian, Novel 5.3 (535): volumus enim nullum monasterium sub dicione nostra constitutum sive paucorum, monachos, qui ibi sunt, divisos ab alterutris esse et propriis habitationibus uti, sed com­ muniter quidem eos comedere sancimus. Dormire vero omnes in communi, unoquoque quidem in quadem propria stratura iacente, in domo vero uno collocatos . . .’ 59 As proposed by Krautheimer (1974), 252.

216

JUSTINIAN, THEODORA AND THE CHURCH

emperor himself, according to John who was in a position to know, was also a frequent visitor to the monastery; (5) at the time of writing his vitae (later 560s) John was himself the head of the Monophysite community resident throughout Constantinople,60 so his description is an informed one. The first persecution of Monophysites by the imperial government began early in the reign of Justin. From the outset, the emperor’s policy was clearly focussed on restoring and maintaining orthodoxy of belief. The immediate exile in 518 of Severus, patriarch of Antioch, symbolised the court’s intent. Further pressure was applied by Pope Hormisdas, seeking episcopal endorsement of his libellus. Yet, it is clear that the approach of Justin himself to Monophysites in the eastern provinces was not all that heavy-handed. The successive patriarchs of Antioch, however, took a much harder line.61 As John of Ephesus explains, the settlement of the long-running schism with Rome in 519 created new resolution for the orthodox emperor Justin as he sought to bring the east to heel. John describes opposition at Constantinople in general terms before explaining events in the East.62 The persecution was sudden and violent with troops deployed to arrest and imprison monks.63 At John’s abode of Amida the bishop Mare was exiled to Petra. John clearly blames Paul, the new Antiochene patriarch, for the severity of the perse­ cution throughout the east. Then he elaborates on the impact of Paul’s actions on the monasteries, again giving the impression that the result was momentous with monks scattered far and wide throughout the east.64 Paul’s aggressive approach to suppressing the Monophysites was ended when he was replaced in 521 by Euphra­ sius who continued the persecution until the devastating earthquake in May 526. Indeed, some saw his demise as divine retribution. At Edessa, meanwhile, monks continued to be expelled by the general Pharesmanes and his troops and eventu­ ally settled in Mardin,65 while, by increasing the number of ordinations, Bishop John of Tella supported those driven into the desert.66 John’s testimony is rein­ forced by another detailed first-hand account, that of the writer of the Church History known as pseudo-Zachariah.67 It was only in or just after August 527, 60 Michael the Syrian, Chronicle (Chabot [1899], 2.257, 268, 353), with Honigmann (1951), 210. 61 John of Ephesus, vitae 13 (PO 17.187) with Vasiliev (1950), 226ff; van Rompay (2005), 241–4. 62 Ps.-Dionysius of Tel-Mahre, 17–18 (trs. Witakowski [1996b]) taken more or less verbatim from the second (lost) part of the Ecclesiastical History of John of Ephesus, cf. Witakowski (1996a), 181–210. 63 Ps.-Dionysius of Tel-Mahre, 24–6 (trs. Witakowski [1996b], 25–7); John of Ephesus, vitae 5 (PO 17.95–105) – 521 persecution: Frend (1972), 247–9. 64 John of Ephesus, vitae 24 (PO 18.512–21): John of Tella; HE Second part, fragment (trs. Van Douwen and Land [1889], 217–19); Ps.-Dionysius of Tel-Mahre, 19–24 (trs. Witakowski [1996b], 21–4). 65 John of Ephesus, HE Second part, fragment B (trs. Van Douwen and Land [1889], 219–20); Ps.Dionysius of Tel-Mahre, 27–9 (trs. Witakowski [1996b]). 66 John of Ephesus, vitae (PO 18.514ff) with Harvey (1990), 100–3. 67 Ps-Zachariah Mytilene, HE 8.5.

217

JUSTINIAN, THEODORA AND THE CHURCH

when Justinian and Theodora had vacated the palace of Hormisdas to occupy the main imperial palace, that the persecution subsided. The influence of Theodora was clearly crucial in the decision that those who had been exiled for years were now allowed to return to their monasteries.68 Again, according to John of Ephesus, ‘after about six and a half years, thanks to the endeavours of the empress Theo­ dora, Belisarius received an order to allow them [monks of Edessa] to come back to their monastery’. That takes us to 528/9 when Belisarius was magister militum in the East. The only time Constantinople comes into view during these years is for the monk Stephen who had joined the exiled bishop Mare at Petra. Around 524/5 he was sent to Constantinople by Mare to negotiate a more congenial place of exile. By now Theodora ‘the patrician’ was married to the court general Justinian and they both lived in the palace of Hormisdas.69 Her position and influence with Justinian and the emperor Justin led to Mare being resettled in Egypt, not in Con­ stantinople. Stephen accompanied him. When Mare died in 529, Stephen once more set out for Constantinople, this time in order to secure Theodora’s permis­ sion to return Mare’s body to Amida.70 A couple of years later, that is in the early 530s, Theodora despatched an imperial messenger (magistrianos) with a letter inviting Stephen himself to ‘come up to the capital in order to be with her in the palace because of his eloquence and his conversation and wisdom, and moreover because he also lived a pure life and after the manner of a solitary’. John of Ephe­ sus had probably seen Theodora’s letter, because he was planning with Stephen to journey to Egypt at the time. Stephen then decided to go by way of Constanti­ nople and John accompanied him as far as Antioch where they parted.71 As events turned out, Theodora prevailed upon Stephen to remain in the capital where John met up with him once more in 535. There Stephen had set up a renowned cell for himself, dispensing blessings and charity to all, not unlike the example of Daniel the Stylite decades before. Stephen is the first monk known to have been protected by Theodora in Constantinople and this happened in the early 530s. He was never part of a large community of Monophysite monks. Instead, he remained the court’s celebrity solitary until he succumbed to the plague in 542.72 Not only were monks on the move after 519, but many bishops too. John of Ephesus includes in his account of the persecution of the 520s a list of all the bishops who were forced to leave their sees at that time. Most found refuge in neighbouring territory or joined Severus in the more sympathetic environment of

68 Ps-Zachariah Mytilene, HE 8.5; Ps.-Dionysius of Tel-Mahre, 29 (trs. Witakowski [1996b], 30); Chronicon ad annum 1234, 54 (Chabot [1937], 151.12–14) with Harvey (1990), 67–8. 69 Theodora took an early interest as patricia (John of Ephesus, vitae, 13 [PO 17.189–90]); PsZachariah Mytilene HE 8.5; Ps.-Dionysius of Tel-Mahre, 30–2 (trs. Witakowski [1996b], 30–2). 70 John of Ephesus, vitae 13 (PO 17.195); Ps.-Dionysius of Tel-Mahre, 31–2 (trs. Witakowski [1996b], 31–2). 71 John of Ephesus, vitae 13 (PO 17.207). 72 John of Ephesus, vitae 13 (PO 17.211–12).

218

JUSTINIAN, THEODORA AND THE CHURCH

Alexandria.73 Of the 54 exiled bishops he notes that only two arrived in the impe­ rial capital: Thomas of Damascus and Theosebeios of Ephesus, who died only three days after his arrival there.74 Otherwise, Constantinople is not mentioned in the course of these years of persecution. Given both its sheer distance and the religious policy of the emperor Justin, the imperial capital was not easily acces­ sible, nor was it likely to be a welcoming place of refuge in the 520s. There is no evidence of any large scale arrival of Monophysite monks there during these years. Occasionally, individual bishops such as Thomas and monks such as Ste­ phen found their way to the capital, by invitation or coercion. Yet, up to 530 or so, there was simply no need to find a suitable home in Constantinople for hundreds of displaced Monophysite monks. Even if large numbers of monks had arrived in the imperial capital during the height of the persecution in the years 519 to 526, there would not have been room for them in the Palace of Hormisdas that was then occupied by Justinian, Theodora and their substantial household.

Persecution after the synod of Constantinople in 536 From about 530, when Justinian relaxed the court’s hostile approach to Mono­ physites, the emperor was focussed on solving the problem of doctrinal dissent and disharmony by finding a path to unity between orthodox and Monophysites. He conscientiously sought a single theological formula that would be meaningful and acceptable to both sides.75 As part of this quest he invited two Constantinople bishops recently returned from exile,76 and arranged further consultations between theologians. The most significant of these discussions took place over a three-day period in 532 in the palace of Hormisdas.77 This is the first mention of Justinian and Theodora’s former residence since they moved into the imperial palace in 527. It was obviously a convenient location for such discussions in that it could accommodate the disputants and their entourages for a few days, although some of them may have stayed as long as a year. There is no hint of the place already being crowded with refugee monks, nor that they had been temporarily accommo­ dated elsewhere. The discussions involved six eminent orthodox bishops as well as the six Monophysite bishops. That does not make the palace a special preserve of Monophysites by 532, as suggested by Mango.78 If nothing else, Justinian was anxious to be seen as scrupulously even-handed. He was hoping the Monophysites 73 John of Ephesus, vitae 13 (PO 17.189–90). 74 Ps.-Dionysius of Tel-Mahre, 20 (trs. Witakowski [1996b], 20); Chron.846, 171.32–173.29. For the bishops and their sees: Honigmann (1951), 146–8. 75 For background: Maraval (1998), 399–409; Allen (2000), 811–34; Gray (2005), 215–38. 76 Ps-Zachariah Mytilene HE 9.15; Chron.846, 169.34–170.11. 77 Harvard Syr.22.3: ‘After this the order (came) for the two parties to assemble in the hall known as Beth Hormisdas, which is today joined to the Palace’ (Brock [1981], 92–3). The location is also mentioned in the record of the conversations circulated by Innocent of Marona, Epistula de collatione cum Severianis habita, 4–6 (ACO 4.2, 169). 78 Mango (1975), 392.

219

JUSTINIAN, THEODORA AND THE CHURCH

could be persuaded to change sides. Some did. The exiled patriarch of Antioch, Severus, was not among the bishops who met in the heptaconch dining hall of the palace of Hormisdas in 532. Eventually he was pressured by the emperor and his wife to join in discussions at Constantinople and arrived late in 534, his safety guaranteed by Theodora.79 Shortly after, the attempt to create some sort of unity collapsed. Two laws issued in March 533 had targeted adherents of the doctrines of Nestorius and Eutyches, but had left most Monophysites untouched.80 According to the hypotheses advanced by Mango and Bardill, it is in exactly this period of relative quiet between 531 and 536, in fact after the accession of Justinian in 527, that the Monophysite monks flooded into Constantinople from the East, thereby creating the need for a makeshift monastery for hundreds of them in the palace of Hormisdas, as well as the construction of a new church dedicated to St Sergius to meet their liturgical needs.81 In fact, Mango proposed that the church was built precisely for those Monophysites who had taken part in the dis­ cussions with the orthodox in 532, hence its construction stretched over the next few years.82 That is to say, Sergius and Bacchus was constructed to satisfy six vis­ iting bishops but with the promise of hundreds of monks to follow. More recently, Bardill has proposed that any connection between Justinian and the monophysites of the east would be sufficient impetus for the church’s building, not necessarily a large influx.83 In any event, the period from 531 to 536 is the least likely time for such an influx of monks. There was relative peace in the east during those years. John of Ephesus, for example, describes how at Amida the monks returned to their mon­ asteries in the late 520s, remaining there for several years before persecution was renewed. As Frend summed it up, ‘The years 531–536 are the years of the great truce between adherents and opponents of Chalcedon throughout the east, when persecution ceased and each side rested on its position’.84 An exception was the stylite Z’ura who came from Amida to Constantinople in 535 after being forced down from his column. He was accompanied by ten disciples. After disputatious meetings with Justinian and Theodora, as well as local nobles and bishops, he was provided with a villa at Sycae and later removed into Thrace.85 The emperor and empress were not sympathetic. 79 John of Ephesus, HE 3.3.8; vitae 48 (PO 18.687); Ps-Zachariah Mytilene, HE 9.19; Chron.846, 223; Evagrius HE 4.10 (protected by Theodora). 80 CJ 1.1.6, 7. 81 Mango (1975), 386; Bardill (2000), 6. 82 Mango (1975), 392: ‘As a further gesture of good will, the splendid martyrium of Mar Sergius was built for their benefit, hence between 531 and 536.’ Cf. Bardill (2000), 6. 83 Bardill (2017), 83: ‘If in the period in which Justinian’s church was built, we can detect any con­ nection between the Palace of Hormisdas and the non-Chalcedonians, then such a connection should not be overlooked . . . if a connection could be demonstrated between the construction of the church and the [532 Chalcedonian/Monophysite] discussions it would certainly provide a [84] plausible explanation for the church’s location, dedication and the timing of its construction’. 84 Frend (1972), 269. 85 John of Ephesus, vitae 2 (PO 17.21–35) with Harvey (1990), 84.

220

JUSTINIAN, THEODORA AND THE CHURCH

The situation was to change in 536 after the synod at Constantinople in May/ June led by the city’s new patriarch Menas. He had recently replaced Anthimus who had been forced to resign after failing to prove his unswerving support for the Chalcedonian cause. The conclusion of the synod was decisive: Severus was again exiled, others too.86 The synod also nudged Justinian into a firmer doctrinal posi­ tion, which is reflected in a detailed imperial decree giving unprecedented force and clarity to the imperial intention to outlaw the opponents of Chalcedon, espe­ cially Anthimus, Severus and Peter of Apamea.87 On 2 May, at the first session of the council called to condemn the recently deposed patriarch of Constantinople Anthimus, it was discovered that he could not be located. Certain members of the council were then designated to search for him. They reported back to the second session on 6 May that their search had proved fruitless, including their inquiries at the Church of Peter and Paul in the Hormisdas quarter.88 Thereupon a new search party was formed, with the same result reported to the third session on 10 May, although they did not specify that they had looked anywhere within the Hormisdas palace complex. Yet another search party was then commissioned which reported back to the fourth session on 21 May. Theagenes, bishop of Synada, in Phrygia, explained to the assembled bishops and others that they had started at St Sergius’ church in the Hormisdas quarter because Anthimus was known to have lived there before he became patriarch,89 that is, before 535 but possibly in the late 520s. They found the church closed. Others confirmed Theagenes’ report.90 Anthimus was condemned in absentia. He did not resurface to make his peace with Justinian until after the death of Theodora in 548. While Justinian’s law condemned him to oblivion, she had been hiding him in the palace.91 In support of a construction date for Sergius and Bacchus after 532, Bardill is obliged to construe Anthimus’ hiding in the church of Sergius and Bacchus in 536 as due to its being ‘symbolic of religious reconciliation’.92 However, the facts are that Anthimus had been one of the orthodox, not Monophysite, participants in the 532 conversation in the dining room of the palace of Hormisdas. Further, he 86 Stein (1949), 380–3; Honigmann (1951), (on Severus); Grillmeier (1995), 346–55; Maraval (1998), 406–7. 87 Justinian, Novel 42. 88 ACO 3, 80 (159.12–28) and 82 (160.3): evn toi/j ~Ormi,sdou evn tw/| a`giw| avpostoleiw| /| 89 ACO 3, 111. (174.36–8): evn tw/| septw/| euvkthri,w| tou/ a`gi,ou ma,rturoj Sergi,ou evn toi/j ~Ormi,sdou kai/ zhtou,ntwn h`mw/n :Anqimon to.n eu,labe,staton evpedei,cqh h`mi/n oi=koj e;nqa pro,teron ta.j diatriba.j evle,geto e;cein This may refer to the period immediately before he became patriarch but, more likely, it refers to an even earlier period before Anthimus became bishop of Trebizond which would clearly mean the church was completed much earlier. Since he took part in the discussions in the Hormisdas palace in 532 as the orthodox bishop of Trebizond (Letter of Innocent in ACO 3, 169.10], it could have been in the late 520s. Perhaps Anthimus had been a monk in the monastery there. There was evidently no question about his orthodoxy at that stage, when he was renowned for his ascetic qualities (Ps-Zachariah Mytilene, HE 9.19). 90 ACO 3, 118 (175.20–1); 115 (176.4–10). 91 John of Ephesus, vitae 48 (PO 18.685–7). 92 Bardill (2017), 85.

221

JUSTINIAN, THEODORA AND THE CHURCH

had taken part as the bishop of Trebizond and was summoned to the capital from there. In 535 he was summoned again by Justinian this time to become patriarch of Constantinople. The emperor considered him an eminently orthodox bishop at that time.93 If the church was only built after 532, that is to say, while Anthimus was actually far away at Trebizond, then the contemporary record of the council, which states plainly that the bishops knew where to look to find him, makes little sense. The search party went straight to the Church of Sergius and Bacchus because they suspected he would seek refuge in his former monastery. In other words, to have once been a monk at Peter and Paul/Sergius and Bacchus, before becoming bishop of Trebizond no later than 531, the simple ascetic Anthimus must have done so well before 532 when he was back in his former neighbourhood.94 What this indi­ cates is that the church, was already functional by then, whereas on the later dating (532–6) it would only be recently completed if Anthimus fled there in May 536. As noted above, after the synod at Constantinople in 536 the emperor promul­ gated a new law on 6 August which severely restricted the activities of Mono­ physites, expelling Anthimus, Severus, Peter and the troublesome Zu’ra. The situation for Monophysites now changed rapidly throughout the east, as Krau­ theimer observed.95 The law specified that, at risk of strict punishment, followers of Severus were not to engage in discussion on the faith, nor to baptise or offer communion to anyone even if they requested it. This applied both in Constantino­ ple (in hac regia nostra civitate) and elsewhere. Any church where they preached would be confiscated from them as was only just.96 It was the patriarch of Antioch, Ephraem, who acted most swiftly and harshly throughout his jurisdiction. At Amida the local bishop Abraham bar Kaili was particularly zealous in seeking out Monophysites, then torturing and expelling them. The persecution quickly spread throughout the eastern provinces. Many were exiled and pressured. As pseudoZachariah expressed it, ‘and others they hunted and drove from country to country, among them the monks’,97 while John of Ephesus reports the new outbreak of persecution after the Constantinopolitan council of May 536 with greater attention to detail than previously, distinguishing this persecution from the previous one.98 He even wrote a separate treatise about this persecution.99 An obvious implica­ tion of John’s eye-witness record is that the persecuted Monophysites, especially 93 For his arrival in Constantinople from Trebizond in 532: Brock (1981), 62 (Harvard Syr. 22, 2). 94 Precisely when Anthimus left Constantinople to take up his position in Trebizond is not known, but it must surely have been no later than 531, but likely much earlier (cf. Menze [2008], 196). 95 Krautheimer (1974), 252. 96 Justinian, Novel 42.3.1–2. 97 HE 10.1, cf. Evagrius, HE 4.11; Jacob of Edessa, Chronicon (Brooks [1905], 242) with van Rom­ pay (2005), 247. 98 John of Ephesus, HE Second part, fragment C (Van Douwen and Land [1889], 221–3); John of Ephesus., vitae, 24 (PO 18 522–4), 35 (18.620–1), 58 (19.570–1); Ps.-Zachariah Mytilene HE 10.1; Ps.-Dionysius of Tel-Mahre, 32–44 (trs. Witakowski [1996b], 32–41). 99 John of Ephesus, vitae 35 (PO 18.607); HE Second part, fragment C (Van Douwen and Land [1889], 221).

222

JUSTINIAN, THEODORA AND THE CHURCH

the monks and nuns, were quickly exiled. In contrast to the situation in the 520s, they scattered far and wide. John of Tella, for example, was arrested and brought to Antioch, where he soon died a martyr’s death.100 Other monks driven out of Amida were kept moving in a sort of tent city.101 On this occasion, again unlike the previous persecution in the 520s, many sought refuge in the imperial capital despite the lengthy and arduous journey. Constantinople was now an attractive option for refugees because they knew that Theodora could offer sympathetic succour and Justinian would at least tolerate her actions, despite the intention and force of his own law. First, there arrived Mari, Sergius and Daniel who all died in the imperial capital. Others such as Isaac and Paul continued at Constantinople to carry out their humble work of serving the poor and sick, while James impressed Theodora by the sanctity exhibited in his cell.102 Later, Hala from Edessa sought the succour of Theodora ‘who had been gathering together persecuted men from all quarters and looking after them, in that they had been placed by her in the palace called Hormisda’.103 Besides the prov­ inces under the jurisdiction of the bishop of Antioch, persecution was also felt in Egypt. In December 536 a large number of monks arrived in Constantinople from Alexandria. They accompanied Theodosius the Monophysite bishop of Alexan­ dria, who came to Theodora before being exiled to Thrace.104 Theodosius soon returned to Constantinople as head of the Monophysite community there, while some of the Alexandrian monks had remained in Constantinople where they estab­ lished monasteries and churches.105 Some even resided in the monastery set up in the palace of Hormisdas. Also from Egypt came Mare the solitary who settled in his cell at Sycae106 and John of Hephaestopolis who sought expert help for an ill­ ness. Theodora housed him in the mansion of Anthemius.107 As others continued to arrive from various parts, in 539 Justinian charged the newly appointed quaesitor with examining the credentials of monks and nuns who had recently arrived in Constantinople.108 Such scrutiny was then reinforced by another law forbidding the entry of any monk into the city without a letter of authorisation from his local patriarch.109 Following the initial influx in 537/8, the arrival of refugee monks at Constantinople was now to be carefully monitored and regulated.

100 John of Ephesus, vitae 24 (PO 18.522–3). 101 Ps.-Dionysius of Tel-Mahre, 30 (trs. Witakowski [1996b], 30). 102 Mari, Sergius and Daniel: John of Ephesus, vitae 42, (PO 18.657); Isaac: vitae 45 (PO.18. 669); Paul: vitae (PO 18.675); James: vitae, 49 (PO 18.691); Thomas: vitae 21 (PO 17.208). 103 John of Ephesus, vitae 33 (PO 18.600). Hala’s expulsion arose from the persecution commencing in 536 (PO 18.598 with n.1). 104 John of Ephesus, vitae 25 (PO 18.528–9); Ps.-Zachariah Mytilene, HE 10.1. 105 Victor Tonnenensis, Chron. 126: s.a. 540 (ed. Hartmann, CCh 173a, 41 = Mommsen, MGH.AA XI, 199). 106 John of Ephesus, vitae 36 (PO 18.624–41). 107 John of Ephesus, vitae 25 (PO 18.531). 108 Justinian, Novel 80. 109 Justinian, Novel 86.8.

223

JUSTINIAN, THEODORA AND THE CHURCH

John of Ephesus and the Hormisdas palace monastery John of Ephesus himself first set foot, as a visitor, in Constantinople in 535 and returned there in 540. He was housed in the mansion of Probus, the only surviv­ ing nephew of the emperor Anastasius and a well-known Monophysite.110 This mansion was probably also set up as a monastery just like those of Hormisdas and Anthemius. While at the house of Probus he was joined for two years by the bishop Simeon who died there. Many others he saw pass through in search of ordination.111 John records that in 565/6 persecution was finally visited on the monasteries of both men and women in and around Constantinople. Some of them were well populated, ‘particularly so were those for females who had been expelled in the beginning from Antioch, Isauria, Cilicia, Cappadocia and other regions, and who had been brought together by the empress Theodora whose soul is at rest. Some of these communities numbered in excess of three hundred’.112 Otherwise, John does not mention any migration of holy women to Constanti­ nople from Amida and elsewhere, although that clearly occurred. As we have seen, however, he does devote a separate chapter to ‘the great and marvellous desert of solitaries’ which was housed in the palace of Hormisdas, throughout every room and hall. There were monks from Alexandria as well as from his own region and further afield.113 According to John, writing in the period 565 to 567, thirty years have elapsed since this monastic community first started to form.114 In other words, it began in the aftermath of the persecution in 536, but not earlier. For now, the emperor and empress had access to an unprecedented concentration of sanctity and spiritual power. In the years between 536 and her death in 548 Theodora had presided over the consolidation at Constantinople of hundreds of monks set up in various places, including at the palace of Hormisdas. Since ecclesiastical law required that mon­ asteries be communal and open, the spacious aristocratic mansions of the imperial capital were well suited to accommodating large numbers of monks. We therefore find monastic groups in the houses of Probus, Placidia and Anthemius, as well as in the imperial palace and the mansion of Hormisdas. Theodora’s support for them was widely noted, but it was measured and it never contradicted Justinian’s policies.115 Looking back, John describes the ‘concourse of believers’ gathered in the imperial capital by the 540s116 and speculates that Theodora was ‘perhaps appointed queen by God to be a support for the persecuted’ who were able to find 110 John of Ephesus, vitae 10 (PO 17.157). 111 John of Ephesus, vitae 24 (PO 18.522); other visitors included Leontius (vitae 39 [18.646] and Abraham (vitae 40 [648]). 112 HE Part 3, 1.10 (trs. Brooks [1936], 5). 113 John of Ephesus, vitae 47 (PO 18.677–81). 114 John of Ephesus, vitae 47 (PO 18.680–1). 115 For example, by Evagrius, HE 4.10. For the Monophysite tradition on Theodora: Harvey (2001), 1–31, with the observations in Pazdernik (1994), 271–81; Foss (2002), 141–9. 116 John of Ephesus, vitae 58 (PO 19.225).

224

JUSTINIAN, THEODORA AND THE CHURCH

refuge at Constantinople.117 Even so, Theodora knew there were limits. She could shelter and sustain, but she had no freedom or authority to advocate and promote. The emperor’s overriding strategy was to use this limited support of refugee ascet­ ics as a public policy of containment and ecclesial suffocation. Confining them to designated mansions was part of that containment process. In addition, ordina­ tions were forbidden, so that at one stage Theodora had to admonish John of Hep­ haestopolis: ‘Remain still and keep quiet like your companions and do not make priests in this city’.118 The very decree which set the persecution in motion in 536 severely restricted the activities of Monophysites at Constantinople.119 This Monophysite monastery which was established in the Hormisdas palace after 536 has been identified by Mango and Bardill with another monastery which was already in the precinct of Hormisdas before 536, but not in the palace itself. In the documentary record of the synod at Constantinople in May/June 536 a certain Paul is listed as the leader of the monastery of ‘the holy apostles Peter and Paul and the martyrs Sergius and Bacchus’ (presbu,teroj kai. h`gou,menoj tw/n a`gi,wn Pe,trou kai. Pau,lou tw/n avposto,lwn( Sergi,ou kai. Ba,kcou tw/n martu,rwn)) that is to say, a monastery associated with the two contiguous churches built on Justin­ ian’s Hormisdas estate.120 Paul formed part of a contingent of the heads of local monasteries that sought and were granted permission to present special petitions in person to the first two sessions of the synod on 2 May and 6 May, 536. Their petitions were strongly supportive of the orthodox position. As a public advocate of orthodoxy before the synod Paul is highly unlikely to have been, or to be sus­ pected of having been, an anti-Chalcedonian. Nor is there any need to assume that he must have been one, however secretly. At Constantinople the abbots of monasteries were approved and appointed by the patriarch.121 It is improbable that in the early years of Justinian’s reign a patriarch would have appointed anyone of dubious orthodoxy to head a monastery in a property intimately associated with the orthodox emperor. So, Mango’s resort to the contention that Paul was actually a Monophysite but had cunningly changed sides, or was merely an ‘orthodox fig­ urehead’ for the purposes of the synod, is unnecessary,122 as is Bardill’s claim that Paul was ‘forced to sign’ the acts of the synod of 536.123 No duress was involved. Moreover, Paul’s signature was not to be found on the acta of the synod at all. 117 John of Ephesus, vitae 25 (PO 18.529) with Duchesne (1915), 57–79. 118 John of Ephesus, vitae 25 (PO 18.534). Theodora provides bishops for the Arabs James and Theodore: John of Ephesus, vitae 50 (PO 19.154–5). 119 Justinian, Novel 42.3.2. 120 See ACO 3, 46.3–4. Otherwise the location of Paul’s monastery is designated more simply as the monastery of ‘Peter near the palace’ (tou/ avpostovlou Pe,trou plhsi,on tou/ palati,ou) merely to distinguish it from other local monasteries named after Peter: 129.4; 158.3; 164.44; 173.22, cf. 144.39 (no location). 121 Justinian, Novel 5.9 (535). 122 Mango (1975), 389, cf. 392: ‘nominally orthodox abbot’; Bardill (2000), 7, contra Mathews (1974), 24. 123 Bardill (2000), 10.

225

JUSTINIAN, THEODORA AND THE CHURCH

Only the bishops present were eligible to sign. What Paul did sign was the peti­ tions presented to the synod on behalf of the city’s orthodox monasteries. The reason that Paul has been viewed as a Monophysite manqué by Mango and Bardill is their contention that by the time of the synod in May 536 the palace of Hormisdas had already been turned into a Monophysite monastery for hundreds of refugee monks and that this can have been the only monastery on the property. However, the conversion of the palace into a monastery for Monophysites cannot have commenced by May 536, as we have seen, so there is no difficulty in Paul being what he manifestly is in the record of the 536 synod – an orthodox head of an orthodox monastery. In addition, the monastery led by Paul was one associ­ ated with both Sts Sergius and Bacchus and Sts Peter and Paul, thereby implying a single monastery serving both the churches already established on the imperial estate and plainly associated with the orthodox by 536.124 Accordingly, on the Monophysite hypothesis, as Bardill recognised, the church of Peter and Paul must also have been set aside for the use of the Monophysites,125 which is decidedly unlikely given its strictly orthodox and Roman associations as the repository of relics of Sts Peter and Paul supplied by Pope Hormisdas himself. The mansion of Hormisdas remained an imperial residence, or at least was kept in imperial use, long after Justinian and Theodora moved to the nearby imperial palace.126 At some point after 532 the palace of Hormisdas was structurally con­ nected to the imperial palace complex,127 and continued to require the overall supervision of an official designated the curator palatii Hormisdas. Nevertheless, for a considerable part of Justinian’s reign the palace functioned as a monastery for Monophysites, although over time it declined in numbers. During that time, so John explains, the community experienced two undated disasters that illustrated God’s providence for the occupants. The first was the collapse, under the weight of a pressing crowd, of the floor of the large heptaconch dining hall where the doctrinal discussions had been held in 532. It was now being used as a chapel. The din and damage were enormous. The worst was feared, but nobody was killed. Such obvious divine favour increased the honour of the monastic community at the Hormisdas palace, at least in John’s view. Justinian had the space rebuilt as

124 Krautheimer (1974), 252; Mathews (1971), 24 see it as a separate monastery serving both churches and dating from their construction and located nearby, but Mango thinks otherwise. Bardill (2017), 100 follows Krautheimer and Mathews. Monasteries could serve more than one church. 125 Bardill (2000), 10. 126 The courtier Andrew was imprisoned there early in Justin II’s reign (John of Ephesus, HE 3.2.9 [Brooks (1936), 49–50]), while the wife and daughters of the new Caesar Tiberius were lodged there in the later 570s. Tiberius would spend his evenings there with his family before returning to the imperial palace in the morning (John of Ephesus, HE 3.3.7 [Brooks (1936), 87]; 3.6.28 [Brooks (1936), 258]). 127 Procopius, Buildings 1.4.1–2. The connection must be dated sometime after 532 since it was not completed by the time of the doctrinal discussions held there in 532 (Brock [1981], 92 n.17), though not immediately after.as suggested by Bardill (2017), 73.

226

JUSTINIAN, THEODORA AND THE CHURCH

a portico.128 According to Bardill, the word for ‘portico’ in the account of John of Ephesus (azga) is better translated as a ‘vaulted roof’ and can thereby refer to the church of Sergius and Bacchus.129 If so then before May 536, indeed before 533 according to Bardill, (1) the Hormisdas palace monastery must have been well established and full to overcrowded, (2) the hall subsequently collapsed, and (3) the new church of Sergius and Bacchus built to replace it was already com­ pleted. In reality this incident, the collapse of the dining room and the erection of its replacement structure, cannot have occurred until the monks were wellestablished there, that is, sometime after 536 at the earliest which is too late for the replacement structure to be identified with the church of St Sergius. Furthermore, apart from the compressed chronology required for Bardill’s hypothesis, John says explicitly that Justinian’s new construction replaced the fallen hall (triklinos) in the very same space within the palace of Hormisdas, not elsewhere outside it. The church, however, was physically separate from the palace as is made clear in the record of the 536 synod.130 So the replacement structure built in the palace cannot be the church of Sts Sergius and Bacchus, even though the triklinos may well have been turned into a vaulted chapel as Bardill has proposed. The other miracle story adduced by John concerns a fire which ‘God wrought in the place where this holy community was’. The ensuing narrative is some­ what disjointed and, as so often in John’s vitae, it is difficult to be certain of the exact locational sequence of his story. He begins by noting that ‘after the death of the believing Theodora the queen [June 548], when this community had also lost some of its earlier numbers, the adversaries were stirred up with envy against it and induced the king [Justinian] to eject them from the former place and remove them to another place belonging to the crown called the House of Urbicius’. John begins this whole chapter by referring to ‘the community of the blessed men which was gathered together in the royal city by the believing queen at the time of the persecution, out of many peoples and various local tongues . . .’. Theodora had indeed been responsible for protecting a large number of monks in Constantinople, over 500 according to John, and they were spread across different locations. John then goes on to describe one of them, possibly the largest, which was in the palace of Hormisdas. Most of his chapter is subsequently devoted to that particular community. In the course of his account John explains that following Theodora’s death in 548 the emperor Justinian honoured his promise to continue to succour the Mono­ physite monks. Indeed, ‘he supported the remnant of them even to the present

128 John of Ephesus, vitae 47 (PO 18. 681–3). 129 Bardill (2000), 8–9 but acknowledging the tenuous nature of his hypothesis (‘We cannot of course be certain . . . but if we grant that it was. . . . But if . . . then . . .’). 130 Bardill (2017), 99–100 seeks to argue that the church was somehow built within the palace itself. Apart from the difficulty of visualising such a situation, the reference he cites evn toi/j ~Ormi,sdou (ACO 3, 159, 160 and 174) is not a reference to the actual palace building but simply to the district called ‘Hormisdas’ (ta Hormisdou), cf. Janin (1964), 358–9.

227

JUSTINIAN, THEODORA AND THE CHURCH

time which is the year 877 [= 565/6], that is the remnant that was left of it to the present time’. In other words, not long before his death in November 565 Justinian was still protecting what John himself called a ‘remnant’ of the monastic commu­ nity which had first settled into the palace of Hormisdas nearly thirty years ago. Doubtless after Theodora’s death Justinian found it difficult to keep the promise he had made to his wife, not only because of his own laws but also in the face of orthodox hostility towards the Monophysites in Constantinople. Hence John’s introduction to the story of the fire, in which he appears to say that, in response to unnamed ‘adversaries’, Justinian was induced ‘to eject them from the former place and remove them to another place’ which, from the sequence of John’s narrative, means from the palace of Hormisdas to the palace of Urbicius near the Strategion.131 So the remnant of the original Monophysite monastic community was now being supported by Justinian at the palace of Urbicius, with the palace of Hormisdas being reclaimed by the emperor. Even so, the ‘adversaries’ were not satisfied and at this new location they tried to corrupt the mainly old monks by introducing into the community married couples and ‘others who were not chaste’. Next, John explains the consequence of this development – ‘whereas these men thought to defile the saints’ dwelling God purified it by a sudden fire, in that fire fell and burnt the whole of the place, only a small portion of it escaping’.132 What John is describing appears to be nothing less than a fire that sometime after 548 burnt down most of the mansion of Urbicius, incinerating at least some of the provocative females who had come to lodge there. Finally, he comments on the aftermath as follows: ‘And so at last it was given to the martyrs’ chapel of the holy Mar Sergius and a monastery was built on that spot; and it remains to the present time to the glory of God’.133 Again, the sense is that on the site of the burntout palace of Urbicius was built a small monastery for the remaining Monophysite monks, which in 566 was still functional. Less clear is what John means by ‘it was given to the martyrs’ chapel of the holy Mar Sergius’ at a date which must have been sometime in the 550s. Either it was attached to the church of St Sergius at the palace of Hormisdas sometime after the Monophysite monks had been expelled from there by the orthodox Justinian or, more likely, to some otherwise unknown church of Sergius closer to the Monophysite monastery at ta Ourbikiou.134 If the former, that would still not make St Sergius a church exclusively for Monophy­ sites, either then or earlier. In this reading of John’s account, the fire does not occur in the same place as the previous disaster, the collapse of the hall in the palace of Hormisdas. Mango and Bardill, however, have interpreted John as setting both stories in the same venue, 131 132 133 134

For its location: Janin (1964), 400; Berger (1987), 404–6. John of Ephesus, vitae 47 (PO 18.683–4). John of Ephesus, vitae 47 (PO 18.684). Perhaps that near the Cistern of Aetius which commemorated its inauguration on 29 November each year, that is, assuming it was already constructed by the mid-sixth century (Janin [1969], 470).

228

JUSTINIAN, THEODORA AND THE CHURCH

so that the fire affected the community of Hormisdas and burnt down most of that palace presumably after the hall (identified by Bardill as the Church of Sergius and Bacchus) had been rebuilt following its previous collapse. Both Mango and Bardill state very clearly that the removal of the monks to the palace of Urbicius followed, not preceded, the fire.135 This cannot be so. John is quite explicit on the point and he was there at the time.136 Mango considers that St Sergius’ church was built for the monks, while the new monastery was redesigned to replace the burnt out sections of the Hormisdas palace. Bardill proposes that the new monastery now constructed was for both churches and was in fact the first permanent monas­ tic structure on the site, that is, on the questionable assumption that the monastery led by Paul in 536 was actually the Monophysite one within the physical confines of the palace. The alternative, a possibility also recognised by Bardill, is far more likely: ‘that the Monophysites who were lodged in the palace of Hormisdas had no link at all with the monastery of Sts Peter and Paul and Sts Sergius and Bacchus. But in that case an explanation would have to be found for the existence of an Orthodox mon­ astery of Sts Peter and Paul and Sts Sergius and Bacchus not far from a community of refugee Monophysite monks . . .’.137 Quite so. The explanation for separate monasteries is self-evident: the orthodox monastery over which Paul presided in 536 served both of the churches built side by side by Justinian and must have been located nearby; the Monophysite monastery was established after 536 within the halls of the palace of Hormisdas. It did not require, nor did it enjoy, access to the church of Sts Sergius and Bacchus. Its chapel was the main hall of the palace rebuilt in a more appropriate form, as Bardill has argued, following its collapse. The long-standing monastery of Sts Sergius and Bacchus and Sts Peter and Paul remained what it had always been, a monastery for orthodox monks. It was to the Church of Sts Sergius and Bacchus that Pope Vigilius fled in 546/7 after Justinian sent soldiers to arrest the pope for imposing a period of penance on the patriarch Menas. The emperor’s men tried to drag him out but Vigilius, who was a strong man, hung onto the altar columns and eventually pulled them down, whereupon Justinian relented.138 Since Theodora was still alive at that time, the Monophysite monks were occupying the palace of Hormisdas, but clearly not the Church of Sergius. The Pope would hardly have sought security and refuge in a Monophysite stronghold. Thus, a connection with Monophysite refugees does not form part of the foundation story of St Sergius and Bacchus’ church after all. We have to look elsewhere to locate the inspiration, as well as the background and likely date of construction, for Justinian’s church.

135 136 137 138

Mango (1972), 192 and (1975), 386; Bardill (2000), 7. As noted, for example, by Janin (1964), 400. Bardill (2000), 10–11. Malalas, Chron. 18.111 (Thurn 411); Theophanes, AM 6039 (de Boor 225); Nicephoros Callistus, HE 17.26 (PG, 45.281–284A).

229

JUSTINIAN, THEODORA AND THE CHURCH

The St Sergius entablature inscription Churches such as that of Sts Sergius and Bacchus were erected over a period of several years. Where there are precise indications such as inscriptions, monograms or brickstamps at one or more levels, then different parts of a structure can be dated to different years. Otherwise, there is little certain knowledge about the organisation and speed of building such a complex, richly decorated structure. Hagia Sophia, perhaps best known because of its current state of preservation, took over five years to build even with the imperial inclination for speed, a relatively uncluttered site, and the capacity to concentrate all the best available craftsmen on the task.139 The construction of other churches would have taken just as long or longer. For instance, Anicia Juliana’s church of St Polyeuktos, despite being smaller than Hagia Sophia, needed fifteen years, about ten of which were solidly devoted to building.140 Sts Sergius and Bacchus may well have taken seven to eight years, certainly two to three years at the very least. The columns at ground level, for instance, would appear to have been carved with monograms sometime before those at the gallery level.141 Generally speaking, dedicatory inscriptions and carved epigrams represent the latest phase of a building’s decoration. The elaborate epigrams of St Polyeuktos were carved only towards the end of construction when the interior was being fin­ ished off.142 Likewise, it would have been towards the end of the construction of St Sergius and Bacchus that the dedicatory epigram was carved into the octagonal nave entablature. The inscription, with each line separated by a palmette, pro­ claims the patronage of Justinian and his wife Theodora, as follows:143 :Alloi me.n basilh/ej evtimh,santo qano,,ntaj avne,raj( w-n avno,nhtoj e;hn po,noj) h`me,teroj de. euvsebi,hn skhptou/coj VIoustiniano.j ave,xwn Se,rgion aivglh,enti do,mwi qera,ponta gerai,ei Cristou/ paggene,tao) to.n ouv puro.j avtmo.j avna,ptwn( ou, xi,foj( ouvc e`te,rh basa,nwn evta,raxen avna,gkh( avlla. Qeou/ te,tlhken u`pe.r Cristoi/o damh/nai ai;mati kerdai,nwn do,mon ouvrano,n) avll v evni. pa/sin koirani,hn basilh/oj avkoimh,toio fula,xoi kai. kra,toj auvxhseie qeostefe,oj Qeodw,rhj( h=j no,oj euvsebi,hi faidru,netai( h-j po,noj aivei. avktea,ntwn qrepth/rej avfeide,ej eivsi.n avgw/nej) 139 140 141 142 143

Mainstone (1988), 150–7; Mango (1985), 18; Van Millingen (1912), 28–9; Taylor (2003), 212–55. Harrison (1986), 223 and (1989), 71. Garipzanov (2018), 171–2. Harrison (1989), 80; Connor (1999), 505. Text: Mercati (1925), 205 which improves on Kaibel (1888), 1064 (p. 478) and van Millingen (1912), fig. 20 (by A. E. Henderson). Translation: Mango (1972), 190, with an earlier translation in van Millingen (1912), 73. Bardill (2017), 88 provides a valuable diagrammatic layout of the inscription.

230

JUSTINIAN, THEODORA AND THE CHURCH

Other sovereigns have honoured dead men whose labour was unprofit­ able, but our sceptered Justinian, fostering piety, honours with a splendid abode the Servant of Christ, Begetter of all things, Sergius; whom not the burning breath of fire, nor the sword, nor any other constraint of tor­ ments disturbed; but who endured to be slain for the sake of Christ, the God, gaining by his blood heaven as his home. May he in all things guard the rule of the sleepless sovereign and increase the power of the Godcrowned Theodora whose mind is adorned with piety, whose constant toil lies in unsparing efforts to nourish the destitute. This intricately carved epigram in St Sergius’ church was obviously sanctioned by Justinian and Theodora. Presumably, they were also responsible for its precise content, although its recondite language required the services of a professional Greek versifier.144 Yet, despite their close involvement with the text, Justinian and Theodora are described in only the most general terms. There is no indication of their precise imperial rank and titulature. Justinian is just ‘our sceptered Justinian’ (h`me,teroj . . . skhptou/coj) and ‘the sleepless sovereign’ (basilh/oj avkoimh,toio),145 while Theodora is ‘god-crowned’ (qeostefe,oj). Each of these is an uncommon epithet, especially in the verse epigrams on churches and other imperial buildings. Skhptou/coj is an Homeric appellation (e.g. Iliad 2.86; Odyssey 2.231) familiar to the educated onlookers of Constantinople who could have seen it on a statue in the baths of Zeuxippos, for example (AP, 2 by the poet Christodorus). As for qeostefe,oj it represents an early use of what was later to become a standard part of Byzantine imperial titulature. The only other extant instance from this period is applied to Justinian (qeostefe,j) on his statue at Antioch in Pisidia.146 Mango was inclined to see in the epigram a greater emphasis on Theodora’s role than on Justinian’s, and therefore a more active involvement by her in the planning and construction of the church,147 but this is an exaggeration. It is no more Theodora’s church than Justinian’s. The two patrons are evenly weighed or, if anything, Jus­ tinian receives more emphasis. He alone is signalled out as ‘royal’. Theodora is singled out for her piety and support for the destitute, Justinian for his piety and for constructing a church for Sergius. In any case, her pious activities in support of the destitute are otherwise well attested.148

144 Reinforced by the brief analysis in Mary Whitby (2006), 183–4. 145 Mango (1975), 388–9, relying on Procopius, Secret History 12.20, 27 and John Lydus, On the Magistrates 3.5.5, considered that might be a literal reference to Justinian’s restless energy but it is more likely just a literary topos (Krautheimer [1974], 253, cf. Bardill [2000], 4). 146 Zuckerman (2002), 244. 147 Mango (1972), 190 and (1975), 389. Ebersolt and Thiers (1913), 26 had earlier seen the Theodora reference as implying a monastery. 148 Theodora’s well known reputation: Procopius, Buildings.1.95–10; Secret History 17.5 (convent of repentance for reformed prostitutes); Malalas, Chron. 18. 24 (Thurn 368); John of Nikiu 93.3; Justinian, Novel 14 (1 December 535). The reference in the epigram to Theodora is taken by

231

JUSTINIAN, THEODORA AND THE CHURCH

In the case of Sts Sergius and Bacchus, the emperor Justin’s nephew and his wife could claim credit for the church not least because it was built on their prop­ erty. The way they are described in the epigram, however, would appear to suggest that at the time Justinian was indeed emperor or Augustus (‘sceptred’), while The­ odora was empress or Augusta (‘God-crowned’). Even so, this need not necessar­ ily mean, as it has always been construed, that the epigram can only be dated after August 527 when Justinian became sole emperor following the death of Justin. It could equally apply to the period between April and August 527 when he ruled as co-emperor with Justin and when the imperial couple still lived in the Palace of Hormisdas as Augustus and Augusta. If the epigram reflects the time when Justinian was only joint emperor and Theodora his consort, then there may be a particular significance in the way Theodora is described. In so publicly and per­ manently calling on Sergius to increase the power of Theodora (kra,toj auvxhseie qeostefe,oj Qeodw,rhj) there may lie the expectation that one day she would be the wife of a sole emperor, thereby enjoying unrivalled power. On the other hand, this may be nothing more than a rhetorical topos. While it was argued previously that a further indication of the church’s date of construction could be found in the several monograms of Justinian and Theodora on the capitals this is not so. Given Bardill’s analysis of the monograms after the restoration process in 2003–6, they are not sufficiently distinctive in shape or style to denote a date of composition, except they must be at or later than April 527 because they do refer to the Augustus and Augusta149 In short, the entablature (reinforced by the monograms) permits a date for their carving as early as April 527, but how much earlier could the church have been planned and construction commenced? It has always been considered highly unusual that Justinian would have begun to build a church such as Sts Sergius and Bacchus within a mansion he no lon­ ger occupied, that is, on the assumption that it was more or less entirely built after 527.150 This problem evaporates if the church was planned and at least com­ menced, if not largely completed, before the death of the emperor Justin in August 527. In that case, it was designed and built in the very estate occupied by Justinian and Theodora then and for the foreseeable future. As Caesar in 525, then Augustus in April 527, Justinian could be assured of one day becoming sole emperor in succession to his uncle. When the church of Sts Sergius and Bacchus was being planned, however, and throughout the years of its construction, he could not know how long Justin would remain as emperor and occupant of the imperial palace. Mango (1972), 191; Bardill (2000), 4; Shahid (2003), 479–80 as an exclusive reference to support for Monophysites which is an unnecessary and unlikely supposition. 149 Bardill (2017), 76–81, with summary table (79), which supersedes the account of the monograms in Swainson (1895), 106–8. 150 Mango (1975), 388; Bardill (2000), 4 and (2017), 81, 83 (‘. . . one is left wondering why the imperial couple chose to build the church in the Palace of Hormisdas in particular, rather than elsewhere, and why they built it at a specific time, having (apparently) neglected the Palace of Hormisadas and its church of Sts Peter and Paul from 527 until sometime after 532’) and 84 (‘. . . even though they had been satisfied to have no direct access to it for the preceding five years’).

232

JUSTINIAN, THEODORA AND THE CHURCH

As we have seen, it was probably only after Justin came to the throne in 518 that Justinian moved into the mansion of Hormisdas. Almost immediately he started building the church of Saints Peter and Paul. It is not impossible, given their jux­ taposition and common atrium, that both churches were part of a single original plan, that is, in 518. In any event they were built close together in time. If, as sug­ gested here, the inscription could have been carved as early as April 527 and the church took seven years or so to build, then it would have been designed in the early 520s, about the same time that the church of Sts Peter and Paul was being finished (c.520) and slightly before San Vitale.151 If, however, it was built in only two to three years, then a planning and commencement date around 524/5 has to be considered. So, if 527 is the latest possible date for the church’s completion, then it is very likely to have been planned and commenced while Justinian and Theodora were living in the palace of Hormisdas as eminent dignitaries (patri­ cians) and, from 525, as an imperial couple. The palace was not just the ‘private house of an ordinary citizen’, as Mango contended.152 The church of Sts Sergius and Bacchus was planned befitting the patrician or new imperial status of Justin­ ian and Theodora. If the church was conceived no later than the mid-520s, then the impetus for its construction needs to be sought in those years. Entering the church for the first time and observing the epigram’s crisp uncial lettering must have been an arresting experience, for the verses begin by challeng­ ing the honour of other unspecified imperial churches at Constantinople. As the viewer read the words, silently or out loud, he must have asked himself which are the ‘other sovereigns’ being denigrated here, and who were the ‘dead men whose labour was unprofitable’. While these phrases might be considered as pure literary affectation without intending allusion to any specific imperial or saintly figures, it is hard to believe that, in the politically and ideologically loaded context of church patronage in sixth-century Constantinople, Justinian had no particular targets in mind. In seeking to distinguish his church of Sts Sergius and Bacchus from those of previous emperors, Justinian was making a pointed comparison through his contrastive me.n . . . de. verses. The :alloi basilh/ej were surely meant to be self-evident, but who could they be? As has been observed recently by Connor, Bardill and Shahid, certain senti­ ments and phrases in the Sergius and Bacchus epigram reflect those in the inscrip­ tion set out in a comparable location around Anicia Juliana’s celebrated church of St Polyeuktos which had been constructed not too long before, as is discussed below.153 Justinian’s epigram could not boast the long imperial genealogy that suffuses the verses of Juliana. Instead, he sought to assert superiority in terms of his dedicatee’s spiritual efficacy. Sergius is a worthy saintly patron for an imperial church, Polyeuktos less so. It is a deliberate contrast, a conscious riposte. Justinian may well have been playing on common sentiment. Polyeuktos was a relatively 151 Mango (1985), 76–9; Alchermes (2005), 365–6. 152 Mango (1975), 389. 153 Connor (1999), 511–12; Bardill (2000), 4; Shahid (2003), 476–80, cf. Cesaretti (2004), 279–81.

233

JUSTINIAN, THEODORA AND THE CHURCH

obscure saint, a Roman soldier at Melitene in Armenia who refused to sacrifice to the emperor in the 250s and suffered martyrdom. His cult grew up among the Palestinian monks in which milieu the empress Eudocia first encountered it. This was probably the inspiration for her establishing a church in his honour in Con­ stantinople.154 Yet, despite Polyeuktos’ martyrdom there was doubt about whether he ever became a baptised Christian.155 Against this uncertainty Justinian was able to proclaim the credentials of Sergius on the entablature inscription: ‘the Servant of Christ, Begetter of all things, Sergius; whom not the burning breath of fire, nor the sword, nor any other constraint of torments disturbed; but who endured to be slain for the sake of Christ, the God, gaining by his blood heaven as his home’. Like Polyeuktos, Sergius was also a military saint and martyr, but one who was now established as a powerful protector possessed of a rapidly expanding cult throughout the East.156 At a time when Justinian was still struggling to assert his political legitimacy and authority, he was also claiming supremacy for his new church and for its dedicatee St Sergius.157 There could be no doubt which military saint now afforded the better spiritual potential and protection. Traditionally, the dedication has been assumed to be linked to the establishment of the church for the use of Monophysite monks, that is, since Sergius’ cult was most popular in the area where Monophysites predominated, then the choice must have been influenced accordingly.158 The real justification for the choice of patron saint would appear to lie, however, in the connection between Justinian and the east. ‘Justinian’s interest in the Syro-Mesopotamian martyr has been given short shrift’, according to Elizabeth Key Fowden,159 by which she means that it is mis­ leading to concentrate on Theodora’s connection to Sergius without recognising

154 Pizzone (2003), 121; Shahid (2004), 347–8. 155 Acta Sanctorum, February, vol. II, 651–52. 156 Fowden (1999), 102–5; Shahid (2003), 472–4. The name ‘Sergius’, a traditional Roman name (cf. Shahid (2003), 473–4), grew in popularity through the fifth and sixth century. There are 2 ‘Sergii’ in PLRE 1, 9 in PLRE 2 and 55 in PLRE 3. In contrast, the name ‘Polyeuktos’ was taken by few. There are none in PLRE 1 and 2 and only one in PLRE 3, cf. Poidebard and Mouterde (1949), 113–14. 157 When Justinian’s church was dedicated it may have been known only as St Sergius’, since its entablature inscription mentions only this single saint. Subsequently, the church is referred to on several occasions over an extensive period as simply ‘St Sergius’ (ACO 3, 174.37; Malalas, Chron.18.111 [Thurn 412]; Const Porph., On the Ceremonies 1.11 (Reiske 87, trs. Moffatt and Tall [2012], 87); Anthologia Palatina 1.8 [lemma]; John of Ephesus, vitae 47 [PO 18.684]). In contrast, other documents, the earliest dating from 536, refer to it as the church of ‘Sts Sergius and Bacchus’ (Procopius, Buildings 1.4.3; Patria Constantinoupoleos 3.39 (Berger [2013], 162–3); Cedrenus 401.1 (Tartaglia 623). It has been suggested that, since Bacchus connoted debauchery and decadence, the name was deliberately suppressed (Shahid [2003], 478–9). A more likely explanation for this double nomenclature is that originally the church was dedicated to St Sergius alone, but since Sergius and Bacchus had come to be so closely linked in the popular imagination and pious literature it came naturally and quickly to be referred to as ‘Sts Sergius and Bacchus’. 158 Bardill (2000), 5 and (2017), 83. 159 Fowden (1999), 132.

234

JUSTINIAN, THEODORA AND THE CHURCH

that the same was true, if not more so, for Justinian.160 Since the establishment of a major shrine at Resafa in the mid-fifth century Sergius’ cult had spread rapidly and decisively throughout the Eastern province and into Persian domains.161 As Justin came to the throne in 518, and Justinian moved into the palace of Hor­ misdas, a new stone shrine at Resafa was dedicated replacing the original mudbrick creation.162 In other words, by Justinian’s time the cult of Sergius had been encountered for several generations by Roman soldiers and officials moving back and forth between the imperial capital and the eastern frontier. Whether or not, as a result, a relic such as the thumb of Sergius was translated to Constantinople, and could be honoured in a church named after Sergius, is uncertain.163 Sergius’ popularity at Constantinople had been growing for over twenty years by the time Justinian built the church in the grounds of his palace complex. He was obviously intent on supporting this movement by solidifying the cult of Ser­ gius within the imperial court. As a soldier himself, Justin had been involved in the zone of Sergius’ influence for many years and must have been very familiar with Sergius and his cult. Nor is it impossible that Justinian had also spent part of his earlier military career in the East in his 20s and 30s, so he too may well have been in Resafa. In fact a posting in the East may lie behind the discredited story preserved in the stubborn Syrian tradition that it was there that he met his wife Theodora.164 Since it is not evident that Sergius was exclusively or predominantly a patron of Monophysites, there is no need to presume that Justinian’s church dedication was based on a desire to curry favour with Monophysite monks who had sought refuge in the imperial capital from persecution in the east.165 Nor was it necessarily planned and built to provide a talisman for Justinian’s war against the Persians, as proposed by Shahid.166 According to the chronology advanced here it was already completed when war broke out in 530.167 If the church were designed as a ‘talisman’, some explicit, or evidently implicit, statement might have been expected in Justinian’s entablature inscription. The war continued on until 532. It is possible, however, that the war accentuated the symbolic significance of an 160 Besides the church at Constantinople the emperor built a chapel of Sergius in Phoenician Ptol­ emais and a well at a monastery of St Sergius in Palestine (Procopius, Buildings, 5.9–20, 25). He and Theodora also sent to Sergius’ shrine at Resafa a jewelled cross which was to become one of the shrine’s most precious and potent treasures (Evagrius, HE 4.28). 161 Traced in detail in Fowden (1999), 101ff. 162 For the dedicatory inscription: Gatier and Ulbert (1991), 169–82. 163 Honigmann (1951), 102–3 thought so, but he may have read too much into the relevant text (cf. Gatier and Ulbert [1991], 182; Fowden [1999], 232). 164 Chronicon Anonymum ad 819 (Chabot [1920], 192.4–17). 165 As presumed by Bardill (2000), 10 and (2017), 83. Sergius had powerful supporters among both Chalcedonians and anti-Chalcedonians (Fowden [1999], 112; Shahid [2003], 468). 166 Shaid (2003), 469–72. 167 Shahid (2003), 469, 474 links the construction of the church to the outbreak of war by dating it to 527. However, in the period from 527 to 529 Justinian’s policy was to consolidate and fortify Roman territory while negotiating an enduring peace. War with Persia did not break out until 530 (as explained in Greatrex [1998], 151–65).

235

JUSTINIAN, THEODORA AND THE CHURCH

imperial church recently completed and dedicated to an eastern military saint, Sergius. The impulse and initiative for Justinian’s church must lie elsewhere and his rivalry with Anicia Juliana provides the likely source.168 The idea that, in the period before August 527, Justinian consciously set out to build a church on his own property which would emulate and rival that of Juliana deserves fuller con­ sideration than it has hitherto received.

Justinian and Juliana: St Sergius versus St Polyeuktos Justinian may have become sole emperor by August 527, but his authority was still not unchallenged. His uncle Justin, and now Justinian himself, had struggled to acquire imperial prestige for a lowly born Balkan family surrounded by two generations of aristocratic imperial relatives. In his mural depicting the story of Justin’s rise to the throne, Marinus explained that it illustrated how ‘. . . the great men, the rich and the children of important families not trust in their power, their wealth, and the importance of their noble families, but in God who raises the unfortunate man form the dung pile and places him at the head of the people’.169 Marinus’ mural was an ever-present lesson and throughout the 520s Justinian remained conscious of his imperial rivals, particularly the representatives of pre­ vious imperial dynasties. In 525 the former emperor Anastasius’ nephew Hypatius conspired to support Justinian’s adoption by the Persian king as a means of curb­ ing his potential power. Hypatius was acting out of malice towards the emperor Justin, according to Procopius.170 Another nephew, Probus, was accused in 528 of plotting against Justinian but was pardoned, while in the following year the comes excubitorum Priscus was disgraced and expelled for challenging Theodo­ ra.171 Opposition to Justinian reached its climax in January 532 during the Nika revolt which was construed at the time as a senatorial plot to topple the emperor. Justinian was rescued from the brink of fleeing the city by his resolute wife. In the highly charged aftermath of the revolt Anastasius’s nephews Hypatius and Pom­ peius were killed on imperial orders, along with numerous other senators accused of using the riot to install a new emperor. This time Probus was exiled. Also exiled was Olybrius, son of Anicia Juliana and Areobindus.172 Juliana had probably once hoped or even expected to see her son as emperor, while an oracle of the early sixth century even forecast the rule of an ‘Olybrius’.173 168 Bardill (2017), 92–4 thinks such an expectation unlikely, rather the language of the epigram sug­ gests only general literary, specifically biblical, motifs. 169 Ps.-Zachariah Mytilene, HE 8.1. 170 Greatrex (1998), 135–8. 171 Hypatius: Procopius, Wars 1.11.31; Probus: Malalas, Chron. 18.22 (Thurn 367); Priscus: Malalas, Chron. 18.43 (Thurn 377) = de ins., 45 (de Boor 171.35–172.6). 172 Probus: Malalas, Chron. 18.80 (Thurn 403); Olybrius: Malalas, Chron. 18.80 (Thurn 403). Detail in Greatrex (1996), 136–40 and (1997), 60–86; Meier (2003a), 273–300; Westbrook (2011), 33–5. Still valuable is Bury (1897). 173 Alexander (1967), 126 n.15.

236

JUSTINIAN, THEODORA AND THE CHURCH

During the Nika riot the city’s main church, the basilical Hagia Sophia, was virtually burnt down and had to be replaced. Justinian now seized the opportunity to construct a magnificent new church of Hagia Sophia with an entirely different design. One of the key factors which drove Justinian to such an ambitious design for his new Hagia Sophia was the unexpected opportunity presented to finally eclipse the church which was then the city’s largest, most sumptuous and most elaborate, namely the St Polyeuktos church of one of the emperor’s key political rivals in the 520s, Anicia Juliana. The church of Sergius and Bacchus, on the other hand, has often been cast as a smaller version of the new Hagia Sophia, hence the soubriquet ‘Little Hagia Sophia’. There is a significant similarity of design between them, which has led some to suggest that the former may well have been a prototype for the latter.174 More importantly, for our purposes here, there are also clear links between the churches of St Polyeuktos and Sts Sergius and Bacchus. It is even possible that their respective epigrams were composed by the same poet and carved by the same craftsman or craftsmen.175 Furthermore, the opening words of the Sergius and Bacchus epigram (‘other emperors . . .’) look like an intentional retort to the imperial city’s most splendid church at the time, that of St Polyeuktos.176 He is not contrasting himself with other gods or other benefac­ tors but with ‘other rulers’, others who had some claim to sovereign power. The new emperor Justinian who was born into a lowly Balkan family was proclaiming his church as superior in sacral power to that of Juliana and her distinguished line of imperial forebears.177 Moreover, as we have seen, he was also asserting the spiritual superiority of Sergius over Polyeuktos. The church of St Polyeuktos was originally constructed by the empress Eudocia, wife of Theodosius II, but most recently was the work of her great-granddaughter Anicia Juliana, daughter of the western Roman sovereign Olybrius.178 The church was believed by its original excavators to have been built by Juliana from begin­ ning to end between 524 and 527, while the construction of Sergius and Bacchus has normally been placed a few years later, making it more or less contemporary with Hagia Sophia.179 More recently, however, Bardill’s research on the brickstamps of St Polyeuktos has brought the commencement of the church construc­ tion back by nearly two decades and its completion to the early to mid-520s. On 174 It has been suggested, although there is no clear point in favour, that one of the architects of Hagia Sophia, Anthemius or Isidorus, had previously been responsible for Sergius and Bacchus: Mainstone (1988), 157; but note the caution of Mango (1985); Bardill (2017), 103. 175 It is possible that both epigrams were also circulated in manuscript form in their own right, cf. Alan Cameron (1973), 113. Even so, it seems unlikely that Juliana herself was the author of the Polyeuktos epigram, as proposed by both Fowden (1994), 275; Connor (1999), 516. Bardill (2017), 96–8 illustrates from other inscriptions how the text circulated geographically and over time. 176 Connor (1999), 511–12 n.69 (511); Shahid (2003), 478, cf. Bardill (2017), 91–2. 177 Shahid (2003), 476–8. 178 Details: Harrison (1986), 4–5 and (1989), 33–41. Bardill (2004), 125–6; Toivanen (2003–4), 127–50. 179 Harrison (1986), 4, 223 and (1989), 35.

237

JUSTINIAN, THEODORA AND THE CHURCH

the basis of his analysis of the brickstamps he has dated the bricks in the sub­ structure between 508/9 and 511/2, with those in the superstructure belonging to the period between 517/8 and 520/1.180 This means that Juliana’s magnificent feat began under Anastasius, but was only revealed in all its glory at the time Justin was emperor and his nephew Justinian was emerging at court. In other words, it was being built at exactly the same time as Justinian’s Sts Peter and Paul church. Given the connections, noted above, between the churches of St Polyeuktos and Sts Sergius and Bacchus, an earlier completion date for the former (early 520s) makes possible an earlier completion date for the latter (before August 527, or shortly thereafter). Moreover, this revised chronology, which means that Sergius and Bacchus was constructed in the Hormisdas palace grounds during Justin­ ian’s residence there, and while Juliana was still alive, makes more plausible and more explicable the notion that Justinian’s church was conceived as a response to Juliana’s. Juliana represented a potent challenge to Justinian. As the daughter of an emperor she was nobilissima (evpifanesta,th),181 in fact the only one at Constanti­ nople at the time so entitled. She boasted not only venerable aristocratic pedigree, but could claim imperial lineage for several generations, both western and eastern. There was a public reminder of her credentials in the epigrams of her churches. The inscription in St Polyeuktos is a complex one covering three imperial gen­ erations. The only member of Juliana’s family actually mentioned by name is her great-grandmother Eudocia: . . . a;nassa (AP 1.10, l.1) . . . avristw,dinoj avna,sshj (l.9). Juliana can therefore be described as having imperial blood (l.8), and accom­ plishing her feat on behalf of the souls of her imperial ancestors, as well as for those now alive and those to come (ll.74–6). There is a clear emphasis on dynastic continuity. In particular, the imperial claim of Juliana’s son Olybrius was high­ lighted by the reference to the family’s present and future members (evssome,nwn kai. evo,ntwn, AP 1.97, l.76). A similar sentiment is found in the dedicatory epi­ gram for another of Juliana’s churches, St Euphemia.182 Furthermore, according to Bardill’s revised chronology, the construction of St Polyeuktos had already commenced in 512 when she would have become empress had the attempt to pro­ mote her husband Areobindus as a replacement for the unpopular Anastasius suc­ ceeded. Her authority and prestige remained unassailable, while her well-known Chalcedonian orthodoxy was also an attraction.183 When Pope Hormisdas wrote to her from Rome on 22 July 519, he could refer to her as being ennobled by imperial blood.184 Juliana’s superior status was evident, too, in the frontispiece of 180 Bardill (2004), 62–4; 111–16. The brickstamps were originally catalogued by S. J. Hill in Har­ rison (1986), 207–25. Also Pizzone (2003), 127 n.93. 181 Malalas, Chron. 16.19 (Thurn 334), cf. Koch (1903), 99–100. 182 Anthologia Palatina, 1.12, ll. 8–10; 1.14, ll. 1–2. 183 Capizzi (1968), 214–26 and (1997), 65–91. For Juliana’s constructions: Capizzi (1977), 120–46 and (1997), 93–130; Connor (2004), 105–16. 184 CA 179.2 (635): ‘ut sicut personam uestram imperialis sanguinis uena nobilitat, ita conscientia bonorum meritorum luce praefulgeat’.

238

JUSTINIAN, THEODORA AND THE CHURCH

the Vienna manuscript of Dioscorides, where she is represented as embodying the distinctly imperial virtues of Sophia, Phronesis and Megalopsychia.185 Yet she was never an empress herself, and so could not be a;nassa,186 although such a title has mistakenly been claimed for her.187 Juliana could use both her literary patronage and her ecclesiastical building program to promote the imperial credentials of herself and her family. In display­ ing her lavish new church of St Polyeuktos at Constantinople in the 520s, Juliana claimed to have rivalled another royal builder – Solomon and his Jerusalem tem­ ple.188 With Hagia Sophia in the 530s Justinian boasted that not merely did he, too, rival Solomon, but that he surpassed him altogether.189 Justinian, not Juliana, was the builder of the new Temple of Solomon. No wonder it was rumoured at the time that Justinian himself had deliberately provoked the revolt which led to the incineration of the basilical church of Hagia Sophia in 532 and necessitated its replacement.190 Just months after the inauguration of the new Hagia Sophia, Justinian sought to head off any potential future rivals such as Juliana by making it more difficult for anyone else to build a church, or monastery or house of prayer, in Constantinople or elsewhere.191 While the function of Hagia Sophia as a delib­ erate attempt by Justinian to outdo Juliana has come to be well recognised,192 less attention has been paid, until quite recently, to the possible connections between Juliana’s church of St Polyeuktos and Justinian’s earlier church of Sts Sergius and Bacchus. As is now clear, stylistic similarities in their decoration appear to

185 Kiilerich (2001), 172–3, 185. For its artistic context: Brubaker (2002), 189–214; Connor (2004), 110–12. 186 References in PLRE 2, 636 (‘Juliana 3’). 187 Proposed by Kiilerich (2001), 169–90 but assumed by Connor (1999), 509. Kiilerich notes that in the frontispiece of the Dioscorides MS, in an epigram traceable through the internal octagonal borders of the depiction of Juliana, she is described as a;nassa On closer inspection, however, the a;nassa turns out to be a modern restoration of a now invisible original (proposed by von Premer­ stein [1903], 111). It could just as easily be patriki,a rather than a;nassa, which would conform to the way Juliana is depicted in patrician dress. Patricia was Juliana’s usual title (PLRE 2, 636 [‘Juliana 3’]), although she was also nobilissima. 188 Juliana: Anthologia Palatina 1.10, l.48: kai. sofih,n pare,lassen aveidome,vou Solomw/nojelucidated by Harrison (1983), 276–9. Juliana may also have been seeking to emulate Ezechiel’s temple, as proposed by Milner (1994), 73–81. 189 Diegesis, ed. Preger (1901), vol. 1, 105: evni,khsa, se Solomw,n. Justinian placed a statue of Solomon opposite his new church as a permanent reminder of the scale of his feat (Patria Constantinopoleos 2.40, Berger [2013], 74–5). Romanos infers that Hagia Sophia is the new temple of Solomon risen from the ashes (cant.54, strophes ka v& kb v[Trypanis 470]) which was echoed by Corippus: Salomoniaci sileat descriptio templi (Iust 2.283). 190 Procopius Buildings. 1.1.22 (anyone aware of the outcome would have justified the fire). Melodi­ cally, Romanos was able to reduce what was obviously considered an unseemly short interval to a mere day: avll venvtau/qa meta/ mi,an th/j ptw,sewj h;rxanto h`me,ran to. th/j evkklhsia/j evgeirei,sqai e;rgon (cant.54, strophe kb , 5–6 [Trypanis 470]). 191 Justinian, Novel 67 (538); Procopius, Buildings. 1.8.5. 192 Harrison (1986), 420 and (1989), 40; Mainstone (1988), 148; Connor (1999), 482, 510; James (2001), 158; McClanan (2002), 94–8.

239

JUSTINIAN, THEODORA AND THE CHURCH

link the two churches.193 Some masons and sculptors probably worked on both Polyeuktos and Sergius and Bacchus, as well as later on Hagia Sophia.194 At the same time, St Polyeuktos, and Hagia Sophia in Constantinople, along with San Vitale in Ravenna evidently share glass tesserae from the same centralised work­ shop in Constantinople, so it would not surprise to find the same tesserae pattern in Sergius and Bacchus.195 As already noted, there are also identifiable affinities between the dedicatory epigrams of Sergius and Bacchus and St Polyeuktos which suggest that Sts Sergius and Bacchus consciously sought to answer the political and dynastic challenge represented by St Polyeuktos. This manifest correlation in dedicatory epigram and architectural style between the churches of Justinian and Juliana, combined with the new earlier chronology for the construction of St Polyeuktos, requires renewed inspection of the surviving record of a direct personal encounter between Justinian and Juliana in her Poly­ euktos church. The episode was recounted by Gregory of Tours several decades later.196 As Gregory tells the story, news of the wealth of a ‘certain Juliana’ reached the ears of the emperor Justinian, whereupon he paid her a visit in the hope of per­ suading her to divert some of her wealth from the erection of churches. He wanted her resources to support, instead, the more secular purposes of war and beautifi­ cation of the city of Constantinople.197 In return, she would earn the respect and acclamation of posterity. Further, her generosity would be honorifically recorded and she would be remembered as someone who assisted in the embellishment of Constantinople (‘laudis tuae titulo praecurrente, canatur, urbem Constantinopoli­ tanam a Juliana matrona fuisse pecuniis sublevatam’). This strategy looks like an attempt to link the emperor’s own fortunes and plans to Juliana’s considerable resources and prestige. It was possibly part of a wider plan to acquire aristocratic estates, by persuasion or coercion, and devote their resources to building churches. Procopius repeatedly accuses Justinian of exactly that.198 Juliana protested that her wealth was spread across many properties but that she would gather it together for him. Meanwhile, she collected all her available gold, melted it down and had it made into gold plates to adorn the ceiling of her new church of St Polyeuktos. Sometime later, Justinian, who had already presumed imperial use for the newly

193 Strube (1984), 82, 91 n.410 suggesting that Sergius and Bacchus can be dated after 527 purely on the questionable grounds of style of capital relief, and Pasquini Vecchi (1999), 109–44 (highlight­ ing the oriental influences on the sculpture); Krautheimer (1986), 225–6. 194 Harrison (1986), 414–15, 418 and (1989), 141. 195 Schibelle and McKenzie (2014). 196 Text: Krusch (1885), 105–7, trs. van Dam (1988), 124–6. Another translation, but based on an earlier and less reliable text, is in Harrison (1986), 9. The most recent, and most analytical, discus­ sion is that of Bardill (2006). 197 Gregory of Tours, Gloria martyrum 101 (Krusch [1885], 106.4–8): Ergo quia tibi potentia maies­ tatis divinae multum contulit auri, quaeso, ut nobis manum porrigas atque aliquid pecuniae commodes, ut scilicet, cum tributorum publicorum fuerit summa delata, ilico tibi quae commo­ daveris reformentur, ac in postero . . . pecuniis sublevatam. 198 Procopius, Secret History 6.18.20, 8.9, 8.31–3, 13.6 with Bardill (2004), 34.

240

JUSTINIAN, THEODORA AND THE CHURCH

assembled gold, was summoned to Juliana’s mansion to view what she had col­ lected. ‘What pitiful small wealth I have put together is ready here’, she said, ‘come and gaze on it and do what you please’. She then took the emperor into her nearby church of St Polyeuktos and asked him to fix his eyes on its newly goldplated ceiling: ‘Look up, I beseech you, at the roof of this church, most glorious Emperor, and know that my poor resources are contained in this work. You must now do what you want. I offer no opposition’ (‘Suspice, quaeso, cameram huius aedis, gloriosissime Auguste, et scito quia paupertas mea in hoc opere continetur. Tu vero quod voleris exinde facito, non adversor’). Justinian was duly shamed and praised the magnificence of the ceiling. He was about to take his leave when Juli­ ana interrupted him to explain that she had something more precious to give him than all her gold, namely a ring containing an extraordinary emerald.199 Gregory concludes his story by explaining how Polyeuktos had intervened to assist Juliana in outwitting the greedy Justinian, just as the saint always did in such situations.200 The point of Gregory’s story is to illustrate the effectiveness of St Polyeuk­ tos in countering rapacious behaviour. As with the other stories he includes in his book on the Glory of the Martyrs, Gregory was not so much concerned with the local chronological and topographical details, but with the moral lesson to be derived from his anecdote. Gregory’s information probably originated with eastern ambassadors, traders or others passing through Gaul and is essentially authentic.201 Usually the description has been taken at face value.202 However, there are several incongruous elements of this vignette which have not been ade­ quately explained. There is no indication of date, for example, except that it must 199 Juliana’s remark to Justinian: ‘Accipe, imperator sacratissime, hoc munusculum de manu mea, quod super pretium huius auri valere censetur’, Gregory of Tours, Gloria martyrum, 101 (Krusch [1885], 106.32–3). The ring symbolises, as so often in literature, the locus of power. Juliana is acknowledging that her power and the status deriving from her imperial line are now formally passed to Justinian, thereby relinquishing any future support for the installation of her son Oly­ brius. It may be a claim once made by Justinian as part of the drawn-out process of securing and advertising his own legitimacy. The ring suddenly linked Justinian’s own house to that of Theodosius, while the whole encounter vindicated the spiritual virtue of St Polyeuktos (Harrison [1989], 40). This ring is possibly that worn by Nero when presiding as emperor at gladiatorial combats which would only have strengthened its symbolic potency (Cf. Pliny, Historia Naturalis 37.64). For an explanation of the stone and its connection with Nero: Epiphanius of Constantia, de XII gemmis rationalis 12 (CA 244, 747–9). If this was originally Nero’s ring it was probably still being used by emperors up to the time of Juliana’s father, Olybrius. 200 Gregory of Tours, Gloria martyrum 101 (Krusch [1885], 105.20–4, 107.2–5): ‘Apud Con­ stantinopolim vero magno cultu Polioctus martyr colitur, pro eo praecipuae quod, cum magnis virtutibus polleat, in periuribus tamen praesens ultor exsistit. Nam quicumque, ut adsolet, occul­ tum scelus admiserit, et data suspitione ad hoc perductus fuerit templum, aut statim quod admisit virtute martyris perterritus confitetur aut, si periuraverit, protinus ultione divina perculitur. . . . Unde non est dubium, etiam in hac re martyris huius intercessisse virtutem, ne opes locis sanctis et pauperibus delegatae in illius transferrentur dominatione, cuius non fuerant studio congregatae’. 201 Gregory’s information on the emperors Justin II and Tiberius is generally sound: Averil Cameron (1975b), 421–6. 202 For example by Mango and Ševčenko (1961) as well as Harrison (1986), 420 and (1989), 40.

241

JUSTINIAN, THEODORA AND THE CHURCH

be towards the end of Juliana’s life.203 She is a frail elderly lady (‘senex’) who needs Justinian’s physical support on the short walk from her nearby mansion to her church.204 Since Juliana died in 528 and Justinian is described as emperor (‘imperator’, ‘Augustus’) in this episode, it would appear to be dated between April 527 and the end of 528 (at the latest).205 Given that, on Bardill’s new chronology, the St Polyeuktos church was ded­ icated well before Justinian became Augustus in April 527, Gregory’s account presents a dilemma: either the implied date is wrong, or his description of how the ceiling came to be gold-plated is wrong. To explain, if Gregory’s implied date is to be followed, then the church of St Polyeuktos (c.521) was already built when Justinian, as emperor, first heard about Juliana’s wealth (527/8), so its glittering ceiling must have been an unscheduled afterthought provoked by the greed of Justinian.206 Certainly, two phases of decoration several years apart (520/1 and 527/8) are not impossible. Yet, the internal chronology of Gregory’s narrative necessitates a considerable period of time for the putative second phase alone. Weeks and months surely passed between Juliana promising to show Justinian her remaining wealth spread across all her properties and subsequently having the ceiling of St Polyeuktos covered with gold plaques (tabulae). Quite apart from the craftsmanship involved, the ceiling was both the most central and the least accessible part of the building. Erecting and dismantling the necessary scaffold­ ing would have interrupted the essential use of the church for a lengthy period, not to mention the special provisions required to avoid damaging the building’s exquisite internal fabric. Since, on the implicit chronology of Gregory’s account, there can only have been a few months between Justinian becoming Augustus (April 527) and the death of Juliana (sometime in 527/528) it makes for a very tight fit. Further, imputing to Gregory a two-phase decoration is fundamentally problematic. It is more likely that the scene should be set earlier c.521 before Justinian became emperor.207 The epigram set out at St Polyeuktos clearly describes its golden ceiling, thereby implying that it was part of the church’s original construction.208 If so, 203 There is no evidence for claiming that Justinian here sought out Juliana ‘on his accession’ (Har­ rison [1989], 40). 204 The mansion must have been close to the church, possibly on the Mese near the house of Eudo­ cia. It may even be identified with the house of Eudocia, as proposed in the ingenious study of Magdalino (2001), 58–9. 205 Juliana’s death is placed by Cyril of Scythopolis (normally punctilious about dates) between the death of Justin I in August 527 (vita Sabae 68 [Schwartz 170.15]) and Saint Sabas’ 91st year in 529 (vita Sabae 69 [Schwartz 171.7]). 206 Bardill (2004), 62, 111–16. 207 Bardill (2006), 360. 208 Anthologia Palatina, 1.10, ll.57. (crusofo,rou avkti,naj averta,zousi kalu,ptrhj) with Mango and Ševčenko (1961), 245 and Connor (1999), 491, 502. The dome of Hagia Sophia in Edessa was also covered in gold mosaic, cf. Palmer and Rodley (1988), 131, 158–9. The ceiling of St Polye­ uktos was not domed, as Harrison thought, but a trussed flat roof, as explained by Bardill (2006), 361–6.

242

JUSTINIAN, THEODORA AND THE CHURCH

then it must refer to the final phase of S. Polyeuktos, c.521. In that event Gregory simply applied the imperial title retrospectively, or as the obvious way of des­ ignating Justinian, and when Justinian petitioned Juliana to offer her wealth for ‘our’ use (‘. . . dum vos quietos esse volumus, dum patrias defensare studemus, dum gentes nobis placamus, dum solatia diversorum dando conquirimus’), he was not merely using an imperial plural. Rather he was referring to a period when he was closely associated with the emperor Justin, his uncle and adoptive father. This is strikingly similar to the language of imperial edicts and of Justinian’s letters in the early 520s.209 As magister militum and Caesar Justinian was busily involved in the construction of churches during those very years (520–7). Per­ haps his plea to Juliana was to dedicate some of her enormous funds to restoring the church of the Virgin at Blachernai, for instance, or possibly even the church of Saints Peter and Paul at the Hormisdas palace which was definitely built in the early 520s. The story picked up in Constantinople by Gregory’s informants, even though it became distorted in the telling and re-telling, reflects a well-known encounter which lay at the heart of the rivalry between Justinian and Juliana. Furthermore, Gregory’s tale of the efficacy of St Polyeuktos would appear to explain Jus­ tinian’s riposte in the entablature epigram in Sts Sergius and Bacchus, namely that Polyeuktos was a useless saint unworthy of an imperial church dedication (qano,ntaj/avne,raj( w-n avno,nhtoj e;hn po,noj) ll.1–2). The fact that apparently no more Byzantine churches were dedicated to St Polyeuktos after Juliana’s may also be partly explained by the influence of Justinian’s pointed criticism of his spiritual status.210 Gregory’s moral preoccupation continually drives his narrative. As a result, he seems not to realise that Juliana was not just another Byzantine grande dame (‘Juliana quaedam urbis illius matrona’). She was in fact a nobilissima, the city’s most eminent woman, apart from empress Theodora, that is, and the mastermind of what was for a brief period its most magnificent church. Justinian is said to have only discovered about Juliana’s wealth by hearsay (‘Cum ad imperatorem Justinianum fama facultatis eius, multis narrantibus, pervenisset’). However, since it is simply fanciful to imagine that Justinian learned about Juliana’s wealth only after he became Augustus in 527, this constitutes another reason for dating Gregory’s story to c.521. Justinian and Juliana had much in common in terms of religious enthusiasm and involvement. Indeed, they worked together in 518/519 corresponding with Pope Hormisdas about the emerging reunification of the eastern and western churches. Juliana would have been among the distinguished crowd that welcomed the papal envoys to Constantinople and formed a candle­ light adventus procession into the city. She wrote to Hormisdas on 22 April 519 209 For example: CA 162.2 (614): ‘pro nobis quoque mandatorum uestrorum custodibus . . . nosque supplices’, cf. 187.5 (645); 191.1 (645); 235.5 (716). 210 There was one other fifth-century church of Polyeuktos at Constantinople, although there is some uncertainty about its location (Janin [1969], 420).

243

JUSTINIAN, THEODORA AND THE CHURCH

in terms not dissimilar to Justinian’s letter on the same day.211 Their letters were delivered to the pope by the same envoy. Hormisdas replied to them both at the same time, reminding Juliana that she was ennobled by imperial blood.212 She later wrote again to Hormisdas as did Justinian.213 In the wake of the settlement with Rome in 519 Justinian finished his church of Sts Peter and Paul, incorporating relics he had sought directly from the Pope. His church was clearly designed to reflect the newly re-established ecclesial unity of East and West. Juliana’s church may well have been designed to do the same.214 In 525/6 Juliana greeted Pope John on his arrival in Constantinople and doubt­ less introduced him to the mosaic in her church which depicted the baptism of Constantine. This particular scene symbolised the unity of East and West that was exemplified by Juliana and her family.215 It also highlighted her family’s promo­ tion of the notion of inheriting the mantle of Constantine as Christian emper­ or.216 By now, however, she had a rival. At this very time, the emperor Justin was being acclaimed by the people of Constantinople as the ‘new Constantine’ for his role in restoring ecclesial unity with the bishop of Rome.217 His nephew Justinian was closely coupled with him in the promotion of a religious policy which allied emperor and pope, emperor and patriarch. Justinian’s vigorous activity in building and restoring churches during the reign of his uncle, including the construction of Sts Sergius and Bacchus, was designed to entrench his political and religious standing. That the entablature inscription can be construed as referring to any specific individuals, particularly those of imperial blood, has now been questioned by Bardill. While conceding that there was doubtless rivalry between Justinian and Juliana, he insists that the inscription is best read scripturally with reference to the Book of Wisdom in particular.218 Further, any reference to Juliana’s family is irrelevant if the church was constructed in the period 532 to 536 rather than earli­ er.219 On the other hand, it is argued above that the church of Sergius and Bacchus cannot have been built in the period from 532 to 536, let alone to meet the specific liturgical needs of a large community of refugee Monophysite monks living in the neighbouring Palace of Hormisdas and sponsored by the empress Theodora. Instead, it was built in the mid-520s, completed not long after April 527, while the emperor and his wife were living in the Palace of Hormisdas and when his political rivals were still real. This means that, apart from invoking the Book of Wisdom and other literary precedents as Bardill has demonstrated, the inscription 211 212 213 214 215 216 217 218 219

CA 164 (615]), with Pizzone (2003), 122–32. CA 179 (635). CA 198 (657–8). Bardill (2004), 115–16. Fowden (1999), 278ff. Pizzone (2003), 130–1. ACO 3, 27 (74.23). Bardill (2017), 96. Bardill (2017), 87.

244

JUSTINIAN, THEODORA AND THE CHURCH

could be seen by contemporaries as a deliberate foil to the church building of a political rival.

Conclusion The church of Sts Sergius and Bacchus on the estate of the palace of Hormis­ das was not planned and built in the early 530s, at the initiative of the empress Theodora, to serve the liturgical needs of refugee Monophysite monks. Nor was it constructed to promote and commemorate Justinian’s war with the Persians which commenced in 530. Instead, it was planned and built by Justinian while he was actually resident at the palace of Hormsidas in the mid-520s and while he was consolidating his career as patrician, Caesar and, from April 527, Augustus. It was about the same time that the church with which it has been most frequently linked, San Vitale at Ravenna, was designed. It was also just a few years after St Polyeuktos and a few years before Hagia Sophia. The backdrop to the construction of Sts Sergius and Bacchus was a period of insecurity for Justinian who was still establishing the imperial credibility of his Balkan family in the wake of venerable imperial rivals, particularly the family of Anicia Juliana. The church is best explained as a programmatic response to Juliana and the overt imperial ideology of her St Polyeuktos church. Bardill’s new chronology would also suggest that Sts Sergius and Bacchus was the first church for which Justinian had responsibility after the completion of St Polyeuktos. His more decisive response was to come in the early 530s with Hagia Eirene and especially Hagia Sophia where the rivalry is made so explicit. Yet that opportunity only arose from the unforeseen catastrophe of the Nika riot which resulted in the devastation of churches which were standing firm and indestructible at the time when Sts Sergius and Bacchus was built. In the inscribed entablature epigram of St Sergius the emperor Justinian boldly claims to have surpassed other emperors, while his choice of the military saint Sergius is seen to surpass their chosen saints. Juliana’s Polyeuktos was Justinian’s obvious target. In the period c.525–527 his new church of Sts Sergius and Bacchus was his best and only chance to respond. The church was designed and built as one befitting the status and ceremonial of the Caesar (from 525) and his consort, then (from April 527) the Augustus and his Augusta. As events transpired, it was to fulfil this imperial role for only a brief period. Disconnecting the church from Theodora’s assistance for refugee monks years later in the under-utilised palace puts into sharper perspective the narrowness of her support, and the extent to which her religious activities were cir­ cumscribed by Justinian. Assigning the church of Sts Sergius and Bacchus to the period when Justinian occupied the Hormisdas palace highlights its ceremonial function and clearly restores the initiative for its construction to Justinian himself.

245

9 JUSTINIAN THE ‘SLEEPLESS EMPEROR’*

The emperor Justinian (525–65) is usually portrayed as an energetic and active man, forever reforming and reconquering. His reign was the second longest of any Roman emperor after those of Augustus (44 BC–14 AD) and Theodosius II (402– 450). Only Basil II (976–1025) would subsequently rival his tenure. In effect, Justinian held imperial power from 525 when he became Caesar until his death in November 565, that is, just over forty years. What is less often remarked is that in all those forty years he hardly ever left the imperial palace, let alone Constantinople itself. He reinforced the sedentary model of the early Byzantine emperor, originally established by Arcadius and Theodosius II a century before. The city was his cosmos, the place where the whole world was concentrated and encapsulated. The emperor never needed to reach out to his empire; it continually came to him. Even though Justinian, like his predecessors, almost never left the imperial city, his contemporaries considered him to be an endlessly busy emperor who was preoccupied day and night with the affairs of empire. He is therefore characterised as the ‘sleepless emperor’. Indeed, this notion of Justinian as ‘sleepless’ was a virtue consciously created and promoted by emperor and court. In the contemporary propaganda and satire of Justinian’s reign imperial insomnia oscillated between virtue and vice, beginning with Procopius whose picture of Justinian has shaped and coloured all subsequent interpretations from the sixth century to the present, but particularly since the discovery of his Secret History in the 17th century.1

Virtue and vice That sleeplessness was deliberately cast by Justinian as an imperial virtue, and could be described accordingly, is suggested by the claims of the bureaucrat and courtier John the Lydian. John characterises Justinian in 551 as ‘the most * This chapter originally appeared in G. Nathan and L. Garland (eds), Basileia. Essays on Imperium and Culture (Melbourne 2011), 103–8, and is reprinted with minor changes by permission of Brill (Leiden). 1 Croke (2019), 74–89.

246

JUSTINIAN THE ‘SLEEPLESS EMPEROR’

indefatigable of all emperors’,2 and ‘for the most part keeping tireless vigil against the foe and taking pains to brave the first danger on our behalf’,3 and who makes ‘ceaseless efforts’ for the Praetorian Prefecture to which John himself belonged.4 The historian Procopius of Caesarea, by contrast, who may well have been a friend of John, viewed Justinian’s sleeplessness very differently. In the encomiastic context of Procopius’ Buildings, his elaborate inventory of the various constructions and reconstructions attributable to Justinian, written in the mid-550s, attention was drawn to the emperor’s austere personal habits. He criticised them as unbecoming any imperial official, let alone an emperor, because they were not healthy practices.5 Clearly, for Procopius good governance required a balanced diet and a good night’s sleep. Justinian, however, would rise from his bed ‘at early dawn and keeping watch over the State, and constantly managing its affairs by word and deed from early dawn to midday and equally into the night. And although he went to his couch late in the night, he immediately rose again, as if he could not endure his bed’.6 John the Lydian was possibly one of the first readers of Procopius’ Secret History7 which is a searing account of the emperor, his wife Theodora and senior officials. In particular, in the Secret History Procopius is more sharply critical of the motive and outcomes of Justinian’s sleeplessness. ‘He was not given to sleep (ahypnos) as a general thing’.8 so Procopius wrote, ‘and during his two-day fast for Easter the emperor would sleep about one hour for he made it his task to be constantly awake and to undergo hardships and to labour for no other purpose than to contrive constantly and every day more grievous calamities for his subjects’.9 Justinian’s nocturnal habits are a further subject of Procopius’ vitriol in the Secret History. He claims to report the gossip of palace insiders that ‘some of those who were present with the emperor at very late hours of the night presumably and held conferences with him, obviously in the palace, men whose souls were pure, seemed to see a sort of phantom spirit unfamiliar to them in place of him’.10 Procopius is probably referring here to the testimony of local bishops and theologians who kept the emperor engaged in late night theological discussions. ‘For one of these asserted’, so Procopius continues, ‘that he would rise suddenly from the imperial throne and walk up and down there . . . and the head of Justinian would disappear suddenly, but the rest of his body seemed to keep making these same long circuits. . . . Later, however, when his head had returned to the John Lydus, On the Magistrates, 3.55.1. John Lydus, On the Magistrates, 2.15.2. John Lydus, On the Magistrates, 3.39.1. Procopius, Buildings 1.7.7. Procopius, Buildings 1.7.8–9. At least as argued by Kaldellis (2004b), 11. There is no indication that the Secret History was available to anyone during Procopius’ lifetime. 8 Procopius, Secret History 13.28. 9 Procopius, Secret History 13.32, cf. Buildings 1.7.5. 10 Procopius, Secret History 12:25. 2 3 4 5 6 7

247

JUSTINIAN THE ‘SLEEPLESS EMPEROR’

body he thought to his surprise that he could fill out that which a moment before had been lacking’. Procopius admits that this gruesome reportage is only hearsay and then proceeds to add that someone else who stood beside the emperor saw his face become just a mass of flesh before the facial features reappeared.11 Next he adds the story of a monk who came to see Justinian but refused to proceed to his audience because he saw not an emperor seated on the throne but ‘the Lord of the Demons’.12 ‘And how could this man fail to be some wicked demon’, Procopius concludes, ‘he who never had a sufficiency of food or drink or sleep but walked about the palace at unreasonable hours of the night’.13 There is yet more. Procopius goes so far as to report the gossip that Justinian’s mother once confided that her son was not the offspring of her husband Sabbatius, but of a demon. When she was conceiving Justinian she believed she was having intercourse with a demon who quickly disappeared, as in a dream.14 The demonic emperor became a recurrent explanatory motif for Procopius.15 It is even cited to explain natural disasters.16 Subsequently, Procopius’ relentless and demonic characterisation of the ascetic and sleepless Justinian has seeped, however unconsciously, into most of the modern literature on the emperor. Discounting the rhetoric of demonology, Justinian becomes the ageing and incurable insomniac, bad-tempered, despotic and out of touch because of lack of sleep. For Justinian’s critical contemporaries and occasional courtiers such as Procopius sleeplessness would appear to be an unnatural imperial vice.

Sleeplessness as vigilance Sleeplessness, more positively ‘Vigilantia’, had never been one of the canonical imperial virtues officially advertised on imperial coinage and other public media, although individual emperors such as Trajan could be so described by their panegyrists.17 Justinian, however, appropriated the notion of sleeplessness and vigilance as an imperial virtue and it recurs throughout his reign. The earliest instance is to be found in an inscription in the Church of Saints Sergius and Bacchus at Constantinople which was built by Justinian in the mid-520s just as he was coming to the throne. Carved into the decoration of the church, for all to see forever after, was a key phrase the new emperor chose to describe himself – ‘sleepless 11 12 13 14 15 16 17

Procopius, Secret History 12.23. Procopius, Secret History 12.26. Procopius, Secret History 12.27. Procopius, Secret History 12.18–19. Relevant commentary and elucidation in Averil Cameron (1985), 56–7; Brodka (2004), 35–9. Procopius, Secret History 18.1–4, 36, 37. Wallace-Hadrill (1981), 298–323. In taking exception to this claim, Bardill (2017), 92 misunderstands the ‘canonical imperial virtues’ which were limited and founded in Roman political ideology. That is not to say that previous emperors were not praised for their vigilance or for keeping awake to control businesss.

248

JUSTINIAN THE ‘SLEEPLESS EMPEROR’

emperor’ (βασιλέως ἀκοιμήτοιο).18 Justinian is here advertising his vigilance. This essential self-conscious attribute later appears in many of Justinian’s laws. For instance, in launching the publication of his new legal textbook, the Institutes, in November 534 it was noted that Justinian’s industrious oversight (‘summis vigiliis’) enabled the project to be brought to completion. Moreover, it is clear that most of the work was done by the leading lawyers Tribonian, Theophilus and Dorotheus. Once they had drafted the work, however, Justinian read it carefully and issued it with the full force of an imperial law.19 The prefaces to Justinian’s laws, especially the more fully preserved Novels, show the emperor regularly reminding the Romans that his toil is unceasing on their behalf: It falls to our lot to spend every day and night considering with all vigilant care, how some benefit pleasing to God might be bestowed by us upon our subjects. We take this vigilance seriously, so much so that we exercise it all day, using nights just as much as days, on such plans as will ensure our subjects welfare, and their freedom from every care; we take upon ourselves their concerns on all matters.20 Here again the sense of endless energy and vigilance is emphasised, together with the implication of sleeplessness (‘days and nights’). In another law directed to reforming the habits of the gardeners of Constantinople Justinian makes clear that he cannot afford to be distracted by frivolous or insubstantial matters, given the sheer volume of imperial business: Our purpose is that by means of this divine pragmatic law of ours, and of the direction that will take place on it by your excellency, we should in future remain free from vexation on such matters, and that concerns like this should not make their away into all our other cares on the realm’s behalf, as well. There is no aspect of our realms, great or small, that is outside our concern; we traverse everything in our mind’s eye, and we wish nothing to remain disordered, undisciplined or controversial.21 Once more, Justinian assures his subjects that no matter is too small for his attention and that he is fully immersed in the business of the court.

18 Text of the inscription in Mercati (1925), 205 which improves on the more accessible Kaibel (1888), 1064 (478). For the date and context see Croke (2006), 25–63; Bardill (2017), 62–130 and Chapter 8, 207–45. 19 Justinian, Institutes, Praef. 20 Justinian, Novel 8, praef. (535) (trs. Miller and Sarris [2018], 128), cf. similar sentiments in Novels 46 (537), 80 (539), 81 (539), 86 (539), 93 (539) and 114 (541). 21 Justinian, Novel 64 (trs. Miller and Sarris [2018], 482).

249

JUSTINIAN THE ‘SLEEPLESS EMPEROR’

Similarly, Justinian in September 535 established the new position of ‘Prefect of the People’ (praetor plebis/praetor tōn demōn). In the preamble to his statute the emperor explained that there used to be a position called ‘Prefect of the Night Watch’ (praefectus vigilum/nykteparchos) who was responsible for maintaining law and order in the darkness hours. However, this position has become redundant since its functions are just as necessary by day as by night and there was conflict between the different magistrates responsible separately for the day and for the night. So, he blurs the distinction, noting that he has come to the conclusion that ‘in future, no-one at all is to be called ‘night-prefect’, because they are going to be dealing with all unacceptable actions, by day as well as by night’.22 In addition to the emperor’s own laws, Justinian’s perpetual watchfulness was also reflected in the advice given to him early in his reign by the deacon Agapetus. Writing sometime in the 530s, he exhorted Justinian on how to be a good ruler. One way, among many other things, was to be ‘like a helmsman. Agapetus goes on to explain what he meant by this phrase, namely that ‘the many-eyed intellect of the emperor remains ever vigilant (agrypnei), holding secure the rudder of good government and firmly pushing back the torrents of lawlessness . . .’.23 Later on, reinforcing the emperor’s claim in addressing the peccadilloes of the Constantinople’s gardeners quoted above, he advises Justinian that no issue is too small to warrant his attention: ‘you will best administer your good kingship if you strive to oversee everything and allow nothing to escape notice. For there is nothing small for you, however small it appears in comparison with your affairs’.24

Holy vigilance By Justinian’s time being sleepless evoked significant spiritual power and virtue. Moreover, Agapetus might have been hinting at that when addressing Justinian. It seems likely that Agapetus was somehow connected to the monastery of St John Studius founded at Constantinople in the 460s, where there were the famous ‘sleepless monks’ (akoimetoi) who prayed and sang psalms in at least Greek, Latin and Syriac in rostered eight-hour shifts, thereby ensuring that some monks were at prayer in each language every minute of every day.25 The ‘sleepless emperor of the Sergius and Bacchus inscription inevitably invites comparison with the ‘sleepless monks’. By Justinian’s day the ‘Sleepless Monks’ came to be publicly renowned as powerful supporters of orthodoxy so that an emperor who originally portrayed himself as both orthodox and sleepless (akoimetos) would benefit from being so overtly associated with them.26 Over time, however, Justinian came to strongly oppose these monks.27 22 23 24 25 26 27

Justinian, Novel 13.1 (trs. Miller and Sarris [2018], 174). Agapetos, Advice to the Emperor Justinian, 2 (trs. Bell [2009], 100). Agapetos, Advice to the Emperor Justinian, 26 (trs. Bell [2009], 108). Bell (2009), 9. Taft (1993), 171–4. Hatlie (2007), 141–2.

250

JUSTINIAN THE ‘SLEEPLESS EMPEROR’

Meanwhile, in a law designed to upgrade and enforce monastic discipline in Constantinople and elsewhere in 539, the emperor noted that monks could be supervised all day and all night because in monasteries some monks would always be awake while others were sleeping. Their main function, however, was to pray but for their prayers to be effective they needed to be virtuous, pious and averse to the slightest temptation. If so, Justinian continues, ‘surely all will be well with the armies, there will be stability in the cities, the earth will bear us harvests and the sea will yield its own, because their prayer brings God’s favour on the whole realm’.28 Another aspect of watchfulness and sleeplessness evident in the time of Justinian was the way the night-time hours were increasingly claimed not just for monastic prayer in the case of the ‘Sleepless Monks’ but also for public prayer and liturgy, including outdoor processions. The role of the all-night vigil (agrypnia) had expanded well beyond its introduction at Constantinople in the fourth century. In Justinian’s day there were night vigils for the eve of Easter and many other key feasts including those for local saints. The contemporary hymns of Romanos illustrate the role and nature of the night vigils, as well as how they involve the congregation inside and outside churches.29 Being sleepless meant not only being prayerful but also being vigilant, the capacity to keep watch especially in the darkness hours. The laws of Justinian demonstrate this trait overtly. His vigilance was a positive attribute. Indeed, vigilantia was a useful asset for an emperor and in Justinian’s case his reputation for watchfulness was reinforced by a family name. He was vigilant by blood as well as by habit. Vigilantia was Justinian’s mother’s name, as it was that of his sister in Constantinople, the mother of his successor Justin II. Another relative of Justinian may have been the comes domesticorum Vigilantius.30 As head of the palace guard and a close associate of the emperor it is very likely that Vigilantius was a trusted ally of the emperor. That this family association with vigilance and watchfulness was highlighted by Justinian and his panegyrists is evident from the way it is exploited by the African poet Corippus in his treatment of Vigilantia as the mother of the emperor Justin II. In fact, Corippus invests Justin with the same Justinianic virtue of vigilance and links it to a traditional motif of the sleeplessness of the poet himself.31 ‘Vigilantia’ and ‘Sapientia’, otherwise Sophia the wife of Justin, are cited by Corippus as the inspiration for his poem. In fact, they may have been his principal informants on the events he describes inside the palace.32 By casting himself as sleepless Justinian turned vigilance into a virtue, a necessary attribute of a caring, prayerful and watchful emperor.

28 29 30 31 32

Justinian, Novel 133.5 (trs. Miller and Sarris [2018], 886). Frank (2006), 59–78. Proposed in PLRE 3, 1376 (‘Vigilantius’). Corippus, In laudem Justini, pr. 21ff and 1.8ff, with Dewar (1993), 211–23. Garland (1999), 41.

251

JUSTINIAN THE ‘SLEEPLESS EMPEROR’

Conclusion The ‘watchful ruler’ was a literary and political type that could be traced back to Homer33 and, in the Roman tradition, to Vergil’s Aeneas,34 who was perhaps also reflecting the emperor Augustus – a ruler available to sign a letter at any time of day or night.35 While panegyrists such as Pliny (second century) and Mamertinus (fourth century) could laud the vigilance of the emperors Trajan and Julian respectively, Justinian was evidently the first emperor to describe himself as ‘sleepless’ and to publicise the virtue. His model may have been that of the sleepless monks in the monastery of Studius so that, just as the sleepless monks were perpetually at prayer on behalf of God’s people, Justinian promoted the notion of an emperor perpetually at work for the benefit of his subjects. What Justinian’s widespread promotion of his imperial vigilance enabled Procopius to do was to deliberately invert its value and meaning. As with so much of the Secret History, Procopius turned an imperial Christian virtue on its head, transforming it into a demonic vice.36 So, in the Secret History the vigilant and sleepless Justinian became a figure of derision and scorn, whereas for John the Lydian the emperor Justinian hardly ever slept because he was being forever vigilant and hard working for his courtiers and his people, the very model of a ‘sleepless emperor’. Whether true or not, this contemporary idea of the ever-watchful emperor implied that Justinian worked long hours and slept little. He probably did.

33 34 35 36

Homer, Iliad. 9.9 ff. Vergil, Aeneid, 1.305. Suetonius, Augustus 50. Kaldellis (2004a), 155–6.

252

10 J U S T I N I A N ’ S C O N S TA N T I N O P L E *

Constantinople, Monday 11 August 559. An unusual event is unfolding in the coolness of the early morning. The emperor Justinian and his vast entourage of officials, servants, pack horses and carts are assembling outside the walls of the imperial capital. The overseer of the occasion has left his description.1 They are returning from Selymbria (modern Silivri) about 65km to the west, along the coast by the Sea of Marmara. Justinian is now a portly figure in his late seventies and has been Roman emperor for over thirty years. In all that time he has hardly set foot outside the imperial city. For the past few months, however, the aged sovereign and his court have been residing in a local imperial palace at Selymbria for extraordinary and urgent reasons. They had been engaged in the restoration of the Thracian Long Wall, which ran for sixty kilometres across the full length of the peninsula and constituted the outer defences of the capital.2 The wall had been seriously damaged in an earthquake in December 557, and later overrun by a menacing band of Cotrigur Huns. Justinian’s veteran general Belisarius had been despatched to repulse the Huns, and the ensuing peace facilitated the imperial expedition to Selymbria.3 Relocating the court over such a distance for five months had involved a massive exercise in preparation and logistical support. The dozens of pack animals and carts carrying the imperial perquisites set off in procession back to Constantinople and advanced ponderously along the Via Egnatia from Selymbria to the Hebdomon, the imperial suburb just outside the city. The road used to be boggy and treacherous but Justinian had previously upgraded it into a firm dual carriageway. Along the road they passed through two main towns which, either earlier or just recently, had been reinforced on imperial orders: Athyras (modern Büyük

* This chapter first appeared under the same title in M. Maas (ed.), Cambridge Companion to Justinian (2005), 60–86 and is reprinted with permission of Cambridge University Press. It has been corrected and updated, where necessary. 1 Peter the Patrician, On the Ceremonies, in Haldon (1990), 139. 2 Mango (1995), 9–18; Crow and Ricci (1997), 235–62; Crow (2007), 397–410. 3 Agathias, Histories 5.14.6–20.8; Malalas, Chronicle 18.129 (Thurn 421); Theophanes, Chronicle A.M. 6051 (de Boor 231–2).

253

J U S T I N I A N ’ S C O N S TA N T I N O P L E

Çekmeçe) and Rhegion (modern Küçük Çekmeçe), each laying astride a deep lagoon. Over the latter stretch of water Justinian had replaced the wooden bridge with a stone one,4 while he had reinforced the walls of Athyras and built a cistern there.5 Within a couple of days the imperial expedition will have reached the pal­ ace at the Hebdomon (modern Bakiröy), at the seventh milestone from the centre of Constantinople. The Hebdomon was originally a military parade-ground and until the time of Leo I (457–74) emperors were proclaimed there but it too had suf­ fered greatly from the 557 earthquake. Now it was an important imperial residence and formed an integral part of a city. In fact, Justinian had built the Iucundinae palace and often stayed there, as he will have on this occasion.6 He also built or restored a number of saintly shrines and around 550 he completed the church of St John the Baptist at the Hebdomon.7 By the early morning of 11 August Justinian’s party stood outside the walls of Constantinople, by the Gate of Charisius. The emperor’s arrival was announced and orchestrated. This was the traditional rite of welcome to a Roman city, the adventus. On entering the city he was formally greeted by the City Prefect and other senatorial dignitaries, possibly too he was ritually acclaimed by the Blue and Green circus factions in their colourful billowing costumes.8

Location and layout Inside the gate of Charisius a sparkling panorama opened up before Justinian.9 It was virtually the same vista he had encountered when he first set foot inside the city. Across the undulating hills and valleys before him the horizon was punctured by statues of previous emperors (Constantine, Theodosius I, Marcian) resting atop tall columns, along with the glittering domes of many churches, especially the emperor’s own renowned constructions of St Eirene and Sts Sergius and Bac­ chus in the far distance. The greatest of all, however, St Sophia, was temporarily diminished. Its dome had weakened and later collapsed in the same earthquake in December 557 that had affected the Long Wall and it would take several more years to reconstruct. The morning summer sun would have highlighted the city’s watery surrounds on three sides, forming a ‘garland around the city’, as Procopius described it:10 the Sea of Marmara to the right, the Bosporus straight ahead, and the narrower stretch of the Golden Horn to the left. Perched on a peninsula at the 4 For the renovation work: Procopius, Buildings 4.8.17 with Avramea (2002), 70–1. 5 Procopius, Buildings 4.8.18. 6 Procopius, Buildings 1.11.16 (imperial palace at Iucundianae). The monogrammed capitals of Jus­ tinian and Theodora in the Istanbul Archaeological Museum may come from the palace. Justinian was certainly resident at Hebdomon when he issued laws from there in May 539 (Novel 87) and August 543 (Novel 118). 7 Hebdomon: Janin (1964), 446–9; Berger (1988), 681–4. 8 Plausibly suggested by Alan Cameron (1976b), 250–1. 9 Reconstructed in a digital model in Yoncaci-Arslan (2018), 23. 10 Procopius, Buildings 1.5.10.

254

J U S T I N I A N ’ S C O N S TA N T I N O P L E

extremity of Thrace, Constantinople was strategically positioned at the narrow crossing from Europe to Asia and watched over the profitable sea-lane between the Mediterranean and the Black Sea. The sea was visible from almost anywhere in the city. It was the highway conducting food supplies, building materials and luxury goods to the capital. It also linked the city with suburbs on the other side of both the Bosporus and the Golden Horn, all now part of the modern city of Istanbul. In Justinian’s time the suburbs of Hebdomon, Chalcedon and Sykai were considered part of the imperial city.11 The praetorian prefect maintained a fleet of three fast boats for crossing the straits on imperial business,12 while in 528 Jus­ tinian joined Sykai to Constantinople by a new bridge across the Golden Horn.13 Being surrounded by water also made the capital vulnerable to maritime attack, however, thereby necessitating enveloping sea walls from the mid-5th century.14 The formidable walls protecting Constantinople on its landward side were com­ pleted in 413, and consisted of three layers of defence: two lines of wall, a castel­ lated inner wall with a smaller curtain wall, and a substantial moat. At intervals along the wall were towers and gates, patrolled by soldiers. Much of the wall system is still standing and part of it has been the subject of a controversial res­ toration.15 On the steep ground at the northern end of the wall, but just beyond it, there stood the palace complex of Blachernae, a subject of recent archaeological investigation.16 The nearby Church of the Virgin was the original construction in the area and was completed by the empress Pulcheria around the mid-fifth century. Subsequently, in the reign of Leo I (457–74) a new chapel was built to take solemn possession of the Virgin’s robe which was to become such a powerful Byzantine talisman.17 Around the turn of the sixth century a palace was added to enable the emperor to spend time there. Justinian had himself rebuilt the church during the reign of his uncle Justin in the 520s and was a frequent visitor to Blachernai.18 Normally the emperor returning from Selymbria and points nearer the city would enter through the ‘Golden Gate’ close by the sea.19 On this occasion Justin­ ian had his own reason to break with protocol and enter the city through the more northerly gate. A straight, porticoed path lay ahead. As the procession advanced, the senators, soldiers and Justinian on horseback passed on their left the large open Cistern of Aetius, which was one of several such water storage facilities 11 Mango (1996), 325. 12 John the Lydian, On the Magistrates 2.14. 13 Chronicon Paschale, 618.14–17 (Dindorf) – this was probably the bridge of St Callinicus at the head of the Golden Horn (Janin [1964], 240–3). 14 Exposure to sea: Magdalino (2000), 209–26. Sea walls: Foss and Winfield (1986), 70–2. 15 Janin (1964), 265–83; Müller-Wiener (1977), 286–307; Foss and Winfield (1986), 41–70. 16 For discovery of remnants of the building and relation to palace and church at Blachernae: Dark and Özgumüs (2013), 66–81. 17 The very complicated traditions of the Marian garments, and how they came to Blachernae, are examined closely in Shoemaker (2008), 53–74. 18 Procopius, Buildings, 1.3.3 with Müller-Wiener (1977), 301–7. 19 Bardill (1999a), 671–96, cf. Mango (2000), 181–6.

255

J U S T I N I A N ’ S C O N S TA N T I N O P L E

built on the city’s high ground in the early 420s.20 Today it is a football stadium. Stretching across the landscape on both sides of the main thoroughfare were the city’s extensive vegetable gardens. Soon the imperial party was passing through the wall of Constantine that marked the original perimeter of the city when it was first constructed in the 320s. The wall was now in a state of disrepair. From that point the city suddenly became more crowded, busy and dense. Not far from the wall was the Church of the Holy Apostles. This grandiose cruciform edifice with multiple domes had been completed by Justinian in 550 as a replacement for the original church built by Constantius II (337–61). Next to it, was the mausoleum built for himself by Constantine (305–337) where most of Justinian’s imperial predecessors now rested in marble sarcophagi, probably including some of those which currently grace the forecourt of the Istanbul Archaeological Museum. Jus­ tinian had built a new mausoleum nearby and his wife Theodora who died in 548 was its first occupant.21 The reason the emperor had taken this route through the city back to his palace was precisely so that he could spend time at the tomb of Theodora and light candles to honour her memory.22 At the Church of the Holy Apostles the imperial party was swelled by the addi­ tion of another band of senators and officials including the eunuchs of the impe­ rial bedchamber (cubicularii). All of them would have been in the appropriate ceremonial dress reflecting their carefully differentiated rank and status. They joined Justinian as he emerged from the Church of the Holy Apostles where the imposing Aqueduct of Valens came into view. It had been conducting the city’s water across a valley since its construction in the mid-fourth century.23 Nearby was the magnificent church of St Polyeuktos, built in the early 520s by one of Justinian’s aristocratic rivals the imperial-blooded Anicia Juliana. When Justin­ ian first arrived in Constantinople he will have passed the large basilica church of St Polyeuktos built by the empress Eudocia. Her granddaughter Juliana had demolished the original church and constructed what at the time (early 520s) was a building of unprecedented magnificence. The excavation of its remains last cen­ tury has given some sense of its grandeur and opulence. As Justinian passed, he may well have recalled a personal encounter with Juliana when the magnificent building was just being finished.24 As they advanced, just to the right, the imperial party could see the column of the emperor Marcian (451–7). Now they had virtu­ ally reached it, possibly at a small forum of Marcian.25 Next en route was the key junction where the course Justinian had been following from the Gate of Charisius converged with the other route he might normally have been expected to follow, namely that through the Golden Gate. This route would have led him through 20 Details in Crow et al. (2008), 129. 21 Procopius, Buildings 1.4.9–24, with Dark and Özgumüs (2013), 85–96; Karydis (2020), 99–130 and the 1940s lectures of architectural historian, Paul Underwood (2020), 369–94. 22 Haldon (1990), 139. 23 Details in Crow et al. (2008), 118–24 (portion inside city walls). 24 Described by Gregory of Tours, On the Glory of the Martyrs 101, with Bardill (2006), 346–60. 25 Suggested by Mango (1986b), 117–36, but more or less retracted in Mango (2000), 177.

256

J U S T I N I A N ’ S C O N S TA N T I N O P L E

the Forum of Arcadius with its column and statue,26 and the Forum Bovis27 then direct to the ‘Kapitolion’ or Capitol (at modern Laleli).28 Again this was originally a Constantinian construction, modelled on the Capitol at Rome. Nearby stood many statues and the monument of ‘brotherly love’ (Philadelphion) part of which survives in the embracing porphyry tetrarchs now in Venice.29 Peter the Patrician’s description of the procession continues: As [Justinian] entered the Mese, there met him the domestikoi protektores, the seven Scholai with their tribounoi and komites, all wearing white mantles and with candles, standing to right and left; and along with them magistrianoi, phabrikesioi, the body of eparchs and the Eparch [of the City], silversmiths and all the merchants and every guild, so that from the Kapitolion to the Chalke of the palace the Mese was quite simply entirely filled, and the emperor’s horse could only just pass through. . . . All the officials and senators processed on foot from the Holy Apostles, and the koubikoularioi followed also on foot.30 Justinian and his senatorial party then set off from the Capitol down the city’s main artery, the so-called ‘Middle Street’ (Mese), encountering along the way the serried ranks of the imperial officials itemised by Peter the Patrician: the emper­ or’s personal staff (protektores domestikoi), the full complement of over 5,000 guardsmen (scholai) in their dazzling white uniforms, the staff of the magister Peter (magistrianoi), and the Prefects including the Prefect of the City who had previously welcomed him at the Gate of Charisius. Also present were the city’s trade guilds: the skilled craftsmen responsible for the parade – armour (fabriskesioi), the silversmiths and money-lenders, and all the other merchants. It was a colourful and crammed procession, framed by the long porticos of the street.31 Indeed, there was barely room for the emperor’s horse to plod a path on a street designed to carry chariots in both directions. Above the packed crowd, visible to all, moved the emperor.32 Justinian’s ceremonial horseback entry to the imperial capital in 559 reflects his triumphal posture on the famous gold medallion in the British Museum and similarly on the Barberini Ivory (Paris). Moving along the double porticoed street the emperor’s party soon passed through the massive triumphal arch which led into the Forum of Theodosius (or Forum Tauri), built to commemorate the triumph of Theodosius I (379–95) over the Goths in the 380s. The arch supported statues of both Honorius (393–423) – facing the west, and Arcadius (383–408) – facing the east, and by Justinian’s day 26 27 28 29 30 31 32

Janin (1964), 71–2; Müller-Wiener (1977), 250–3; Berger (1988), 7; Taddei (2009). Janin (1964), 69–71; Müller-Wiener (1977), 253–4; Berger (1988), 348–50. Mango (2000), 177. Müller-Wiener (1977), 266–7. Haldon (1990), 139, 141. M. Mango (2001), 44–5. Haldon (1990), 139, 141.

257

J U S T I N I A N ’ S C O N S TA N T I N O P L E

the forum contained a large number of other statues. Deliberately imitating the Forum of Trajan at Rome, this forum’s centrepiece was an imperial column with sculpted spiral reliefs depicting Theodosius’ military triumphs.33 It was beneath this column that foreign ambassadors were ceremonially welcomed to Constanti­ nople and it was here that the emperor was welcomed home whenever he returned from the west, as Justinian did on this occasion. After being saluted, Justinian progressed further through the crowds until he reached the circular-shaped Forum of Constantine. By Justinian’s day the Forum of Constantine was simply ‘The Forum’ and formed a central location for most imperial processions. Within this area with its massive gates at each end stood a tribunal to review troops, a senate house and many statues notably one of Apollo from Troy.34 The central monument of the Forum was a porphyry column of the emperor, known in its present form as the ‘Burnt Column’. On top of it stood a statue of Constantine, later replaced by a cross.35 Beneath it was buried the statue of Pallas, the city’s traditional talisman.36 The double-storeyed porticoed Mese led out from the Forum of Constantine to the imperial forecourt, Augusteon. Just before reaching the Augusteon Justinian passed by the main public entrance to the hippodrome, a large edifice pivotal to the life of the city. The hippodrome was the focus of regular programs of chariot racing and other entertainments ‘all of which’, so Agathias observed, ‘tend to have a profoundly disturbing effect on the minds of the young’.37 A considerable proportion of the city’s population could be accommodated in the hippodrome to cheer on their favourite charioteer as he circled the arena’s axis and their pre­ ferred faction, predominantly Blue or Green. Justinian was the patron of the circus and sponsored the games.38 He presided over the races from the imperial box (kathisma) which was also the place where he was presented to the city as the new emperor in April 527.39 Very early in his reign (528) Justinian had remodelled the imperial box to make it more spacious and more conspicuous. At the same time, he improved the view for the senatorial spectators.40 The kathisma of the hippodrome was connected to the imperial palace through a spiral staircase. As they reached the end of the Mese Justinian’s entourage skirted the Augus­ teon square and approached the entrance to the imperial palace. The Augusteon of 559 was considerably different to the original ceremonial space laid out by 33 In general: Janin (1964), 64–8; Müller-Wiener (1977), 258–65; Berger (1988), 323–7. For relevant excavations: Naumann (1976), 117–41. 34 On the forum and its many statues: Janin (1964), 62–4; Müller-Wiener (1977), 255–6; Berger (1988), 288–309. 35 Janin (1964), 79; C. Mango (1993a), 1–6. 36 Dagron (1974), 39. 37 Agathias, Histories 5.21.4. 38 Cameron (1976b), 22, 217. 39 For the hippodrome: Dagron (1974), 320–47; Guilland (1969), vol. 1, 369ff; Müller-Wiener (1977), 64–71; Dagron (1984), 161–90; Herrin (1991), 214. 40 Marcellinus, Chronicle s.a. 528 (ed. Mommsen, MGH AA, XI, 102), with Croke (1995), 124.

258

J U S T I N I A N ’ S C O N S TA N T I N O P L E

Constantine and different again to the one modified in 459 by the addition of porticoes. It had been destroyed in 532 during the Nika riots and had been com­ pletely rebuilt by Justinian. It was now smaller than before and functioned more as a courtyard to the church of Hagia Sophia on its north-eastern side. On the Augusteon’s other sides stood the senate house, although by now senators mainly met with the emperor in the palace, and the Baths of Zeuxippos both of which had been rebuilt by Justinian after 532. The Augusteon’s main feature was a statue of Justinian on top of a high column erected in 543. The bronze statue was of the emperor on horseback holding an orb in his hand and facing east towards Persia. He was dressed in the martial uniform of Achilles, replete with ceremonial head­ dress (tufa).41 At last, Justinian reaches the Chalke or ‘brazen house’, the grand entrance ves­ tibule of the imperial palace and his ceremonial return from Selymbria comes to an end. The accompanying senators and chamberlains have now dispersed, the emperor and his household staff are home. Peter the Patrician concludes: As the emperor enters the Chalke, the admissionalis stands there with a protektor and the triumphator, and calls out the triumphal salute; and the magistros ordered this to take place [at this point], since the emperor did not enter via the Golden Gate.42 The triumphal return of Justinian to his palace sees him greeted by the admissionalis, the official who controlled entrance to the imperial presence, along with one of the protectores and the triumphator. The whole ceremonial was planned and supervised by the magister officiorum, Peter the Patrician, who tells us that he himself now ordered the ‘triumphal salute’ to be given at this point.43 Had Jus­ tinian returned to the city through the Golden Gate, the ritual of triumph would have been performed there instead. The salute was accentuated and amplified by the emperor standing beneath the magnificent mosaics that covered the ceiling of the Chalke. They represented his earlier triumphs over the Vandals and the Goths.44 Along with two equestrian statues imported from Ephesus, they formed part of the lavish decoration of the Chalke which had been completely rebuilt by Justinian after 532.45 Only recently have the foundations of the Chalke, if that is what they are, been unearthed.46 Beyond the Chalke, the imperial palace complex spread out across the ter­ raced hillside and down to the water. Justinian’s palace which had evolved over the decades was a sprawling collection of buildings and gardens connected by 41 Malalas, Chronicle 18.94 (Thurn 408); Theophanes, Chronicle, A.M. 6036 (de Boor 224) with Mango (1993b), 1–8. 42 Haldon (1990), 139–41. 43 Haldon (1990), 139–41. 44 Procopius, Buildings 1.10.15–18. 45 Malalas, Chronicle, 18.85 (Thurn 404). 46 Girgin (2008).

259

J U S T I N I A N ’ S C O N S TA N T I N O P L E

porticoed walkways. The palace’s major spaces were the consistorium (a central room where the emperor sat in majesty to receive his visitors), the Tribunal of the 19 Couches or Delphax (extensive terrace for special assemblies and investitures), the Magnaura (audience hall) and the Triklinos (large banquet room). The palace housed quite a horde, being home not only to the Emperor and his family but also to his senior officers and their families.47 Justinian’s main contribution was to rebuild and refurbish the Chalke and neighbouring structures.48 Nothing much of the palace has ever been unearthed except for some striking mosaics, but recent excavations have located some of its substructures and there are expectations of more to come.49 Justinian was now back in the imperial palace where he spent most of his time with court officials as they dealt with a constant stream of appellants and special visitors, plus meetings with senior bureaucrats and generals. The systematic and notable centralisation of power that characterised Justinian’s reign created a busy life for the emperor.50 In fact, Justinian was singled out by contemporaries for his diligence. The ‘sleepless emperor’ could even be caricatured for spending long hours in the palace, day and night, on imperial business.51 There were numerous documents for examination and briefing, new appointments, discussion and deci­ sion making, the reception of ambassadors, and imperial banquets. In between those obligations there was constant and extensive correspondence. Following the long interlude at Selymbria Justinian evidently made himself even scarcer than usual. A few weeks later (September 559) the rumour took flight that the emperor had died but the city was soon illuminated to demonstrate he was still alive.52 Justinian had simply returned to the familiar hectic round of imperial business, which had dominated his life for more than thirty years now and would continue to preoccupy him until his death in November 565, except for his long journey in 563 to the shrine of St Michael at Germia (mod. Gümüskonak).53

Governance and sustenance Inside the walls of Justinian’s Constantinople lived around 500,000 people. As the imperial capital, the city was dominated by the needs and demands of the emperor, his court and his palace. It was styled by Justinian in his laws the ‘royal city’, or ‘our fortunate city’. Justinian, like his predecessors, took a close interest in the 47 McCormick (2000), 141. 48 For the palace: Janin (1964), 106–21; Müller-Wiener (1977), 229–30. The most significant work on the palace is now Westbrook (2019), although not taken account of here. 49 Dark and Özgumüs (2001), for a revaluation of previous excavations: Bardill (1999b), 216–30 and, overall, Westbrook (2019). 50 Maas (1992), 14–15. 51 Procopius, Wars 7.32.9; Secret History, 13.28–33; John the Lydian, On the Magistrates 2.15 with Chapter 9, 246–52. 52 Theophanes, Chronicle A.M. 6053. 53 Details in Croke (forthcoming b).

260

J U S T I N I A N ’ S C O N S TA N T I N O P L E

city and occasionally made laws and other pronouncements designed to improve its financing and functioning. By Justinian’s day the emperor never led his troops into battle but was still commander-in-chief of the army. His two senior generals (magistri militum praesentales) attended the court at Constantinople while most of the soldiers they commanded were stationed outside the city in nearby towns. So too, the majority of the emperor’s palace guard (scholae palatinae), who lined the Mese to welcome him home in 559, was also stationed outside the city. The palace and the emperor himself were mainly protected by the excubitores who could also be deployed outside the palace to provide an armed escort for those summoned to the imperial presence, to help restore order or prevent a serious civil disturbance from threatening the palace and the emperor. They were prominent, for example, during the Nika riots in January 532,54 as well as in dealing with dangerous situations in 547, 561 and 563.55 Sometimes they provided an armed escort for those summoned to the imperial presence such as the feisty monk Z’ura who had been lodged in a monastery for Monophysite monks at Sykai.56 Constantinople had its own civic administration, modelled on that of Rome. The city was originally divided into 14 regions. Responsibility for local affairs in each region rested with a curator (or ‘regionarch’) and five vicomagistri who watched over the region at night and who reported to the City Prefect. In general, the prefect was a senior member of the imperial court and had the authority to convene the senate. He had a large staff, a praetorium close to the palace and his own carriage.57 When in April 563 the City Prefect Procopius was replaced by Andrew we catch a glimpse of the new prefect emerging from the grand entrance to the palace, the Chalke, in the prefectural carriage.58 Justinian’s Constantinople was a city of consumers. Providing sufficient food and water for all the inhabitants of the enclosed capital was a permanent challenge. A disruption to routine could quickly bring disaster. The city’s water supply origi­ nated over 100km away in the mountains of Thrace and in streams closer in. The water was channelled through a long and complex system of aqueducts and tun­ nels all the way to Constantinople where it was stored in large cisterns and distrib­ uted around the city’s fountains and baths. Water was stored in the large open-air cisterns of Aetius (c.420s), Aspar (c.460) and Mocius (c.500) which progressively supplied the growing city.59 They were complemented by numerous covered cis­ terns strategically located.60 Justinian’s City Prefect Longinus was responsible 54 Chronicon Paschale, 626 (Dindorf) and later riots in 547 (Malalas, Chronicle, 18.99 [Thurn 409]), 561 (Malalas, Chronicle 18.135) and 563 (Malalas, Chronicle 18.151 [Thurn 431]). 55 547: Malalas, Chronicle 18.99 (Thurn 409); 561: Malalas, Chronicle 18.135 (Thurn 423–4); 563: Malalas, Chronicle 18.151 (Thurn 431). 56 John of Ephesus, vitae, 2 (PO 17.1, 28). 57 Dagron (1974), 277–82; Theophanes, Chronicle A.M. 6055 (de Boor 237–9) – prefect’s carriage in 563. 58 Theophanes, Chronicle A.M. 6055. (de Boor 237–9). 59 Ward et al. (2017). 60 Mango (1995), 9–18; Bono et al. (2001), 1325–33, and (for detail) Crow et al. (2008).

261

J U S T I N I A N ’ S C O N S TA N T I N O P L E

for the construction of the Basilica Cistern (Yerebatanserai) which still impresses visitors to Constantinople.61 The city also boasted numerous private and public baths including the Baths of Dagisthaeus completed by Justinian in 528,62 as well as the largest and most elaborate of all – the Baths of Zeuxippos restored by Justin­ ian after the Nika riots.63 The city’s water supply was put under pressure every summer, sometimes seri­ ously.64 In November 562 a drought in the city’s Thracian catchment was creating a water shortage and thirsty people were fighting each other at the city’s public fountains. By the following August (563) the prolonged drought had created an even more serious situation. Now the city’s baths were closed and people were actually ‘murdered at the fountains’, so desperate had some become.65 The impe­ rial court had avoided exposing itself to such pressure. One of the consequences of the chaos unleashed by the Nika riots in 532 was the difficulty of providing for those barricaded inside the palace complex. As a result Justinian ensured that in future the palace always had its own self-sufficient water supply from its own internal cistern.66 At the same time the emperor had a private granary and bakery built inside the palace, thereby reducing dependence on the city’s grain supplies. All Constanti­ nople’s grain was imported with most coming from Egypt’s reliable annual yield. This supply line was also a long one involving an annual armada of ships loaded with grain setting out for Constantinople and relying on favourable winds to reach their destination. Their safe arrival always brought assurance to the city. A quick turnaround was required for the ships, forever rushing to make two to three return trips per year. Docking the grain ships required four to five kilometres of wharves across the harbours on both the Sea of Marmara and the Golden Horn.67 In recent years the harbours have become more precisely located, while the har­ bour of Theodosius, a main one for the era of Justinian, has been accidentally discovered and excavated at Yenikapi with a total of 37 sunken ships discovered. A few of these nautical carcasses were probably once the very ships bringing grain annually to Justinian’s Constantinople.68 No less enormous was the infrastructure required to unload the ships, store and transport the grain, then to bake the city’s daily bread. The grain was unloaded quickly into nearby warehouses then distrib­ uted through a system of public and private bakeries throughout the city. Near the harbour of Theodosius were the Theodosian and Alexandrian storehouses (Horreum Theodosianum and Horrea Alexandrina). In effect, Constantinople operated 61 Malalas, Chronicle 18.17 (Thurn 364); Chronicon Paschale 619 (Dindorf). 62 Chronicon Paschale, 618 (Dindorf). 63 Procopius, Buildings 1.10 with Janin (1964), 222–4; Müller-Wiener (1977), 51; Berger (1987), 378–9. 64 Procopius, Buildings 1.11.10. 65 Malalas, Chronicle 18.139 (Thurn 425); Theophanes, Chronicle A.M. 6055 (de Boor 237–9). 66 Malalas, Chronicle 18.71 (Thurn 394–400); Chronicon Paschale 629 (Dindorf). 67 Mango (1990), 120; Magdalino (2000), 211. 68 Harbours: Dark (2005); Yenikapi: Pulak et al. (2015); Kocabaş (2015).

262

J U S T I N I A N ’ S C O N S TA N T I N O P L E

a duplex structure of food supply: one was a government-supplied daily ration to the city’s inhabitants distributed at a designated point within each city region; the other was the private baking of bread for those not entitled to the government ration.69 Monitoring the levels of stored grain was crucial to the city’s survival and civic stability. While this would have been done daily by the City Prefect’s staff, it was important for the emperor to demonstrate public assurance too. Each year, towards the end of summer when current supplies were dwindling, Justinian would set out on an elaborate expedition to inspect the granaries, accompanied by the City Prefect, and then pronounce reassurance to the city.70 There were times, however, when the grain supply was seriously deficient. In May 555, for instance, a severe bread shortage lasted for three months until the first ships arrived from Alexan­ dria. Justinian was at the hippodrome entertaining the Persian ambassador when the chanting circus factions confronted the emperor in the imperial box with the hungry people’s plight, whereupon he had the leaders of the Blue faction punished by the City Prefect for their insolence.71 To give the Alexandrian grain ships an even surer trip, and minimise the chance of a bread shortage at Constantinople, Justinian had offload and storage facilities built on the island of Tenedos with the grain gradually being moved on smaller boats to Constantinople.72 Besides grain, the imperial capital also had to import all its wine and oil. There were separate storehouses for these commodities, which arrived in large earthen­ ware jars, and they formed part of the same distribution system, that is, by a peri­ odic ration for those entitled or by private purchase.73 Although not as serious as a deficiency of grain, shortages of wine and oil were also momentous events for the city’s inhabitants. Justinian had to deal with a wine shortage in 546 and had expe­ rienced a scarcity of oil during the reign of his uncle Justin in 524.74 Complement­ ing the imported grain, wine and oil were other foods supplied locally. There was an abundance of fish in the city’s waterways.75 Livestock, including cattle, were evidently traded in the open markets of the Forum Tauri and the Strategion while a variety of meat was consistently available.76 Both fish and meat were sold by butchers in the macellum, a walled area bordered by shops.77 In February 545, resulting from a miscalculation in the date for the start of Lent, Justinian ordered that meat be sold for an extra week but people were suspicious of the emperor’s motives and boycotted the butchers. Some even tried to spoil the meat set out for Durliat (1995), 21–33. On horrea: M. Mango (2000), 192–3. Const. Porph., On the Ceremonies 2.51 (Reiske 699–701 trs. Moffatt and Tall [2012], 699–701]). Malalas, Chronicle 18.121 (Thurn 418); Theophanes, Chronicle A.M. 6048 (de Boor 230). Procopius, Buildings 5.1.10, with Barnes (2006), 167–77. Morrisson (2002), 206. Wine: Malalas, Chronicle 18.95 (Thurn 408); Theophanes Chronicle A.M. 6038 (de Boor 225); oil: Marcellinus, Chronicle s.a. 524 (MGH AA, XI, 102). 75 John the Lydian, On the Magistrates, 3.62. 76 Magdalino (2000), 214–15. For the location of the Strategion: Westbrook (2013). 77 M. Mango (2000), 193–4. 69 70 71 72 73 74

263

J U S T I N I A N ’ S C O N S TA N T I N O P L E

sale by throwing dust and chalk over it.78 As for vegetables, they were plentiful all year round being mainly grown in the gardens between the walls of Constantine and Theodosius, as well as beyond the walls.79 Justinian’s Constantinople was highly combustible. Compact wooden struc­ tures, combined with permanent hearths for various essential purposes such as baking, smelting of metals and glass-blowing, made fire a perpetual risk and it could spread rapidly. Under the Prefect’s control were over 500 collegiati or fire wardens, appointed by the various guilds. When fire broke out the cry to summon them went up – ‘on the double everyone’.80 Justinian decreed in 535 that the new Praetor of the People should be assisted by 30 firefighters to help rescue property from the flames and with authority to arrest anyone found looting a burning build­ ing.81 There were major fires in April 550, December 560 (Harbour of Julian), October 561 (Caesarius quarter) and December 563 (destroying the hospital of Sampson, the atrium of Hagia Sophia, two monasteries near St Irene and part of the church’s narthex).82 Sometimes, fire resulted from rioting crowds such as the most spectacular conflagration during the Nika riots in 532 when so much of the monumental centre of the city was burned down. One eye-witness looked back on a city ‘uninhabitable because of dust, smoke and stench of materials being reduced to ashes, striking pathetic dread in those who beheld it’.83 Even more destructive than these frequent blazes were the numerous earth­ quakes the city experienced during the reign of Justinian. Constantinople is part of an active seismic zone. Tremors in November 533 brought a large crowd into the Forum of Constantine where they prayed all night for the city to be spared. No one was hurt.84 There were other minor earthquakes and tremors in 540/1, 545, 547, 551, 554, and 555,85 but 557 was the year of seismic terror. A quake on Monday 16 April frightened the city but caused no damage,86 while another on Saturday 19 October had the same effect.87 Disaster finally struck, however, at 78 79 80 81 82 83 84 85

86 87

Malalas, Chronicle 18.96 (Thurn 408); Theophanes, Chronicle A.M. 6038. (de Boor 225). Justinian, Novel 64 (538) with Koder (1985), 49–56. John the Lydian, On the Magistrates 1.50. Justinian, Novels 13. 5 and 4 (1). 550: Malalas, Chronicle 18.108 (Thurn 411); 560: Theophanes, Chronicle A.M. 6053 (de Boor 234–5); 561: Theophanes, Chronicle A.M. 6054 (de Boor 235–7); 563: Theophanes, Chronicle A.M. 6056 (de Boor 240). John the Lydian, On the Magistrates 3.70. Romanos was prompted by the fires to write one of his famous kontakia (Topping [1978]).Other fires resulting from riots occurred in 548 (Malalas, Chronicle 18.105 [Thurn 413]) and 561 (Malalas, Chronicle 18.135 [Thurn 423–4]). Malalas, Chronicle 18.77 (Thurn 402), Chronicon Paschale 629.10–20 (Dindorf). 540/1: Preserved in pseudo-Dionysius of Tel-Mahre, Chronicle, trs. Witakowski (1996b), 82; 545: Theophanes, Chronicle A.M. 6038 (de Boor 235); 547: Malalas, Chronicle 18.102 (Thurn 410); Theophanes, Chronicle A.M. 6040 (de Boor 226); 551: Agathias, Histories 2.15.1; 554: Malalas, Chronicle 18.118 (Thurn 416); Theophanes, Chronicle A.M. 6034 (de Boor 222), 6046 (de Boor 229); 555: Theophanes, Chronicle A.M. 6047 (de Boor 229–30). Malalas, Chronicle 18.123 (Thurn 419); Theophanes, Chronicle A.M. 6049 (de Boor 230–1). Theophanes, Chronicle A.M. 6051.

264

J U S T I N I A N ’ S C O N S TA N T I N O P L E

midnight on 14 December when a very large earthquake rocked Constantinople and its environs causing both the weakening of the dome of Hagia Sophia which collapsed the following May and the damage to the Thracian Long Wall which allowed the Cotrigur Huns to pass through early in 559. The city walls of Theo­ dosius and Constantine were severely damaged, along with churches and other buildings. Many lives were lost. In a sympathetic gesture, Justinian went without his imperial crown for forty days.88 Every year thereafter the earthquake of 557 was commemorated at Constan­ tinople with a liturgy of supplication.89 In an undated law issued to the people of the city Justinian reminded them that loose public morals were the cause of God’s anger manifest in any earthquake,90 hence the need for propitiation. It was a message he reinforced in March 559 before setting out for Selymbria to deal with the consequences of divine wrath at the Long Wall.91 Agathias who lived through the 557 quake at Constantinople observed that it had an immediate impact on people’s behaviour: the rich gave to the poor, doubters turned to prayer, the vicious became suddenly virtuous, but soon they all lapsed back to their former ways.92 The need for propitiation subsided until the next annual reminder. Besides earthquakes, thunder and lightning in June 548 killed people in their sleep and part of the column of Arcadius in his forum (Xerolophos) was shorn off.93 There were violent winds, plus thunder and lightning again, on 19 July 555 which toppled the cross on the Rhesion gate (Yeni Mevlevihane Kapisi) of the land walls;94 and more thunder and lightning on 13 July 556 produced ‘many casualties’.95 All these natural disasters ensured regular rebuilding and refurbishment in the city.96 They also disrupted the regular routine of public life and created a demand for medical support and care. Justinian’s Constantinople had the best possible physicians and some dedicated places for nursing the sick and injured. When the plague struck, however, in 542 the city could barely cope with the horrendous loss of life and the pall of death which engulfed it. The plague returned in 558.97 Funeral processions had to be abandoned since there were so many corpses requir­ ing rapid disposal. The city’s dead were usually buried in cemeteries inside the Theodosian walls, or in underground hypogaea.98

88 Malalas, Chronicle 18.124 (Thurn 419); Theophanes, Chronicle A.M. 6050 (de Boor 231–2); Agathias 5.3.1–6.9. 89 Details in Croke (1981), 124–5. 90 Justinian, Novel 77. 91 Justinian, Novel 141. 92 Agathias, Histories 5.5.4–5. 93 Malalas, Chronicle 18.103 (Thurn 410); Theophanes, Chronicle A.M. 6041 (de Boor 226). 94 Theophanes, Chronicle A.M. 6047 (de Boor 229–30). 95 Theophanes, Chronicle A.M. 6048 (de Boor 230). 96 Mango (1986b), 124–5. 97 Malalas, Chronicle 18.127 (Thurn 420); Theophanes, Chronicle A.M. 6050 (de Boor 231–2); Agathias, Histories 5.10.1–7. 98 Dagron (1991), 157–61.

265

J U S T I N I A N ’ S C O N S TA N T I N O P L E

Another of the responsibilities of the City Prefect was the maintenance of civic order and control of crime. An habitual challenge for the prefect was the rivalry maintained by the Blue and Green circus factions which regularly erupted into violence. The most famous occasion was the Nika riots in 532.99 In 547 during the city’s anniversary celebrations on 11 May the factions got out of control and the imperial guard was called in, resulting in heavy loss of life.100 Much the same occurred in May 562.101 Other factional riots broke out in October 562102 and in 563.103 There were yet other riots including one over the debasement of coinage in 553, which forced Justinian to back down.104 Contemporaries were unnerved by the frequency and magnitude of such urban violence and looked for a decisive response from the emperor and the City Prefect. In his first-hand account of the Nika riots, Justinian’s official Marcellinus singles out the rioters as ‘criminal citizens’ who may be contrasted to the ‘good citizens’ who avoid throwing stones and who salute the public punishment of the perpetra­ tors of violence.105 Procopius strongly criticised Justinian for letting the factions get out of hand,106 John the Lydian explained the need for the emperor to take action against violence and looting in the city,107 while John Malalas observed the general unpopularity of factional strife.108 Still, clear action in response to these concerns was rare.109 On his accession, Justinian issued a comprehensive decree outlawing violence110 and in 535 sought to curb civil unrest by a broadly based attempt to stem arson, looting and theft in particular. This involved upgrad­ ing the prefect of the night watch to the role of praetor plebis.111 Evidently, this law proved less than effective so in 539 he decided to strike more directly at the root of the trouble by ruling that only officially registered arms manufacturers can make arms and they may not sell them to any private person for any reason whatsoever.112 By the time of Justinian, population density was aggravating civic tensions, potentially threatening security or self-sufficiency. Constantinople was firmly fixed in its physical limits. It could not afford to take on more inhabitants than it could support through its present mechanisms and resources, nor could it encroach on the open space required for cultivation. Still the city was like a magnet, constantly 99 100 101 102 103 104 105 106 107 108 109 110 111 112

Malalas, Chronicle 18.71 (Thurn 394–400), with Greatrex (1997), 60–86; Bury (1897), 92–119. Malalas, Chronicle 18.99 (Thurn 409); Theophanes, Chronicle A.M. 6039 (de Boor 225–6). Malalas, Chronicle 18.135 (Thurn 423–4), with Cameron (1976b), 94. Malalas, Chronicle 18.138 (Thurn 425); Theophanes, Chronicle A.M. 6055 (de Boor 237–9). Theophanes, Chronicle A.M. 6055 (de Boor 237–9). Malalas, Chronicle 18.117 (Thurn 415). Marcellinus, Chron. 532 (ed. Mommsen, MGH AA, XI, 103, trs. Croke [1995], 44). Procopius, Secret History 7. John the Lydian, On the Magistrates 2.30.5–6, cf. 1.50.9. Malalas, Chronicle 18.151 (431–2). Cameron (1976b), 272–3. Malalas, Chronicle 17.18 (Thurn 351); Chronicon Paschale., 617.1–6 (Dindorf). Justinian, Novel 13. Justinian, Novel 85.

266

J U S T I N I A N ’ S C O N S TA N T I N O P L E

attracting supplicants and ambitious young men from everywhere. Each new gen­ eration of lawyers and soldiers set their sights on Constantinople. Many hoped and contrived to stay if they could. The city’s delicate equilibrium of supply and demand for essential goods and services needed to be kept under strict observa­ tion. It is clear from the legislation of Justinian that the city and the emperor were constantly wrestling with a potential imbalance caused by crowds of complainants and refugees coming to Constantinople.113 Some were even obtaining free rations illegally.114 The number of farmers was a special concern. Yet, the incursions of the Huns, Goths and Bulgars in the Danubian and Balkan provinces inevitably propelled small farmers towards the capital in search of a different fortune.115 Justinian issued many laws designed to address this concern and in 539 established a new position (quaesitor) with special responsibility for it. The quaesitor and his assistants shall ascertain from all foreigners entering the city their identity, origin and purpose in coming to Constantinople. Their motives must be carefully scruti­ nised and, if litigious, then they must return home once their case has been heard.116 As for those visitors who satisfactorily completed their business and stayed, Jus­ tinian and Theodora built a hospice to shelter many of them until they had a place to lodge.117 Justinian also tried to deal with those who claimed that they had been prevented from returning home by severe weather or illness.118 Bishops were sin­ gled out for special treatment. They can no longer come to Constantinople without good reason and written explanation from their archbishop,119 an instruction later re-iterated and expanded by forcing a visiting bishop seeking an audience with the emperor to apply through the patriarch of Constantinople.120 Of particular concern during Justinian’s reign was the large number of monks arriving in Constantinople fleeing persecution in the east, particularly after 536 when they were outlawed by imperial decree for dissenting from the orthodox Christology. Monks from Alexandria, headed by their bishop Theodosius, soon established monasteries and churches in the city.121 Others arrived later from vari­ ous parts and in 539 Justinian charged the newly appointed quaesitor with exam­ ining the credentials of monks and nuns who had arrived in Constantinople.122 Such scrutiny was then reinforced by another law forbidding the entry of any monk into the city without a letter of authorisation from his local patriarch.123 113 Justinian, Novel 8, praef. (535) – ‘priests and town councillors, officials, landowners, farmers’; 102 (536); 69.1 (538) cf. Procopius, Buildings 1.11.24–5. 114 Justinian, Novel 88.2 (539). 115 Procopius, Secret History 4.3.2, 6.1–3; Wars 7.16.12–13, cf. Beck (1965), 11–45. 116 Justinian, Novel 80. See further Beck (1972), 17–18. 117 Procopius, Buildings 1.11.26–7. 118 Justinian, Novel 49 pr. (537). 119 Justinian, Novel 6.3 (535). 120 Justinian, Novel 123.9 (544). 121 John of Ephesus, vitae (PO 18.326–7); ps.-Zachariah, HE 10.1; Victor of Tunnuna, Chronicle s.a. 540 (ed. Hartmann, CCh 173a, 126 = Mommsen, MGH.AA, XI 199). 122 Justinian, Novel 80. 123 Justinian, Novel 86.8.

267

J U S T I N I A N ’ S C O N S TA N T I N O P L E

In the decade or so between 536 and her death in 548 the empress Theodora had taken responsibility for giving succour at Constantinople to hundreds of mono­ physite monks whom she resettled in various under-utilised mansions, including at the imperial palace of Hormisdas. Although her support for them was widely noted, it was always measured and it never contradicted Justinian. Following the initial influx in 537/8, the arrival of refugee monks at Constantinople was care­ fully monitored and regulated by Justinian.

Society and community As a centre of power and influence Justinian’s Constantinople attracted people from a diversity of linguistic, cultural and regional backgrounds.124 The city streets resounded with a cacophony of Latin, Greek and Syriac, Aramaic and Armenian, Coptic and Ethiopic, Gothic and Hunnic, Persian and Arabic. Justinian and his family were Latin-speaking Illyrians, while the diplomat Peter the Patri­ cian was a Greek-speaking Illyrian although born in Dara on the Persian frontier. Among the Africans were the rhetorician and grammarian Priscian and the bishop Victor. There were Johns from Lydia and Cappadocia, as well as John Malalas from Antioch.125 From Armenia came the generals Narses and Artabanes. The pre-eminent general Belisarius was a Latin speaker from the Balkans but his sec­ retary Procopius was a highly educated Greek speaker from Palestinian Caesarea. Procopius, Peter the Patrician and many others, including Justinian himself, were at home in both Greek and Latin. Increasingly, however, Greek was becoming the city’s common language so those who had mastered Latin such as John the Lydian could feel aggrieved that its use was declining even in the prefectural office.126 In 535 Justinian decided that Greek should thereafter be the ‘language of state’. It was the inevitable outcome of a long process.127 How this rich variety of individuals, families and clans was spread throughout the city is difficult to determine. However, the Notitia of Constantinople – an offi­ cial or quasi-official document from the reign of Theodosius II – provides a sound guide to the basic structure of the city and the spread of its population even in Jus­ tinian’s day.128 It enables us to conclude that the areas around the imperial palace (Regions I and II) and in the western part of the city (Regions X and XI) were rela­ tively open and populated more by the city’s aristocracy; while labourers, crafts­ men and unemployed lived around the ports and fora and along the Mese. Regions IV to VIII comprised most of the areas of the craftsmen surrounding the porticoes of the Mese and the maritime districts, both around the harbour of Julian and along the Golden Horn. The Notitia also records how many free-standing houses or domus there were in each region with the greatest number being in Region XI, an 124 125 126 127 128

Rapp (2002), 153–72. Croke (1990), 10. John the Lydian, On the Magistrates 3.42, 68. Dagron (1989), 29–76. Anderson (2016).

268

J U S T I N I A N ’ S C O N S TA N T I N O P L E

elegant area surrounding the church of the Holy Apostles.129 The least populated was Region V, which took in the Forum of Theodosius, the baths of Eudocia and Honorius, the cistern of Theodosius, the Troad, Valentinianic and Constantinian granaries as well as oil stores. This region also had the highest proportion of free bread outlets and the highest number of public bakeries. Sixth-century Constantinople was a mélange of diverse communities, sepa­ rated or bound by religion, language and regional identity. The Chalkoprateia, for example, was a Jewish neighbourhood with its own synagogue130 and there were also definite communities of Goths who naturally tended to congregate together and probably had their own shops and churches. During Justinian’s time the Alex­ andrians in the imperial capital, including the transplanted monks, had divided in response to the two current rivals for the patriarchal throne – Theodosius and Gainas. The ‘Gaianitae’ and ‘Theodosianitae’ built their own separate monasteries and churches in the city.131 Then there were the Cappadocians who, like the Isau­ rians of a previous generation, were always unpopular at Constantinople. A law of Justinian singled them out as being a nuisance132 and John the Lydian repeated some unkind anti-Cappadocian doggerel in common circulation.133 Latin was the language of a considerable proportion of the city’s population. Yet there were several quite different Latin speaking communities. Firstly, there was the traditional senatorial aristocracy with strong western connections. The Theo­ dosian imperial dynasty with branches in both east and west survived at Constan­ tinople in the family of Anicia Juliana, the daughter of the former western emperor Olybrius. She had long been resident there as patron of litterateurs and holy men, as well as being responsible for the construction of the grand church of St Polyeu­ ktos. She had also been involved in supporting papal embassies in Constantinople at the outset of Justin’s reign. Latin studies flourished in her milieu.134 Many other such families also lived on in Constantinople. Symmachus, consul in 485, had a mansion in the centre of Constantinople burnt out during the Nika riots.135 There was certainly a steady flow of noble refugees from Italy during the war against the Goths in the 540s and 550s. Cassiodorus, Cethegus and others were among them. Another large group of Latin speakers in Constantinople was from Africa. Indeed it was due to the pressure in Constantinople of African nobles, disenchanted or proscribed during the Vandal overlordship, that Justinian launched his expedition in 533.136 The historian Procopius, the chronicler and courtier Marcellinus, as well as the exiled African bishop Victor of Tunnuna, all resident in Constantinople, 129 Magdalino (2001), 53–69. 130 Janin (1964), 44. 131 Victor of Tunnuna, Chronicle s.a. 540 (ed. Hartmann, CCh 173a, 126 = Mommsen, MGH.AA, XI 199). 132 Justinian, Novels 30.5.1(535). 133 John Lydus, On the Magistrates 3.57.2. 134 Momigliano (1956), 249–90. 135 Chronicon Paschale 623.7 (Dindorf). 136 ps.-Zachariah of Mytilene, HE 9.17.

269

J U S T I N I A N ’ S C O N S TA N T I N O P L E

refer to the presence in the city of many victims of Vandal persecution who had had their tongues cut out.137 Like others, the Africans tended to live together and be laid to eternal rest together. When bishop Theodorus, for example, died in 565 he was buried alongside other victims of Huneric’s persecution who had passed away in the city long ago.138 The third significant segment of Latin speakers were refugees and veterans from the Balkan provinces of Illyricum, where so many still had families and friends only a few days’ travel away.139 The family of Justinian was numbered among them. Justin I, uncle of Justinian, left his hometown of Bederiana and came to Constantinople to find a worthwhile career in the army. He was followed later by his sister, Justinian’s mother, and others family members later still.140 Most of the known Illyrians at Constantinople were military men but they were strongly engaged with theological and religious issues, especially in support of the ortho­ dox Chalcedonian position and the authority of the bishop of Rome.141 The Constantinople of Justinian’s time was crowded and boisterous, as it had long been and would remain. It was a city with buildings stacked so close together that there was very little room for those both inside and outside, while compact crowds of people and animals jostled each other.142 An earthquake engendered panic because there was no nearby space for people to congregate in safety.143 Such density exacerbated social and communal friction arising from ethnic, sport­ ing and religious antagonisms and rivalries.144 It also impacted on sanitation, light and neighbourly relations, as is evident from the earliest when laws needed to be introduced regulating the distance between buildings.145 A law of 538 refers to the proverbial ‘evil neighbour’ who devises ways of getting around the prohibition on building closer than 100 feet from the sea in certain areas.146 Building regulations, however, did little to contain the overcrowding problem.

Ritual and ceremonial The pomp and majesty of the imperial court dominated Constantinople in the time of Justinian. The city functioned as the theatre for an elaborate and colourful ritual with every movement of the imperial personage around the city carefully planned 137 Marcellinus, Chronicle s.a. 484.2 (MGH AA, XI, 93); Procopius, Wars 2.8.4; Victor Tonnenensis, Chronicle s.a. 479 (ed. Hartmann, CCh 173a, 51 = Mommsen, MGH AA, XI, 189) also CJ 1.27.1, 4 (April 534) – noted by the lawyer Evagrius (Historia Ecclesiastica 4.14). 138 Victor of Tunnuna, Chronicle s.a. 566/7 (ed. Hartmann, CCh 173a, 173 = MGH AA, XI, 206). 139 Buildings 4.4.2. 140 Procopius, Wars 4.16.12–13. 141 Croke (2001), 88–101. 142 Zosimus, New History 2.35. 143 Agathias, Histories 5.3.6. 144 In Beck (1973), 17 and (1965), 25–6. 145 Dagron (1974), 529–30 (including references). 146 Justinian, Novels 63, cf. 165 (no date).

270

J U S T I N I A N ’ S C O N S TA N T I N O P L E

and stylised. The city had been adapted progressively to this ceremonial role ever since emperors became permanently established there from the time of Theodo­ sius I (379–95) and Arcadius (383–408) such that an emperor’s main movements were wholly within the city itself.147 There now developed a new form of impe­ rial self-representation focussed on the traditional virtues of the Roman imperial image. Moreover, during the fifth century Constantinople’s imperial ritual became increasingly liturgified and sacralised.148 Even the normal imperial movements in and around the city had become ritualised by elaborating the Roman adventus (arrival) and profectio (leaving). Justinian’s return from Selymbria in 559 exem­ plifies the new convention, as does Theodora’s profectio in August 528 to the hot springs at Pythia near Pylai (modern Yalova) where she was accompanied by an entourage of 4000 including leading aristocrats and the officials of the imperial bedchamber.149 Since Constantinople was now the city where emperors were born, baptised, married, crowned, and were buried, each of these occasions gave rise to signifi­ cant public ceremonial. Records survive of both Justinian’s coronation and his funeral. Since he had already been crowned Caesar (junior emperor) by his uncle in 525 the ceremony was much simpler when he became Augustus on 1 April 527, with Theodora made Augusta at the same time. It took place inside the palace, at the Tribunal of the 19 Couches, with the patriarch offering prayers and Justin crowning him as the assembled palace guards, court officials and soldiers looked on.150 His funeral, by contrast, in November 565, was a more public and more elaborate affair. Corippus describes the chanting, the music and incense, the dense crowds lining the Mese mourning as the bier of Justinian passed in solemn pro­ cession to his resting place beside his wife in the mausoleum at the Church of the Holy Apostles.151 Theodora’s funeral in June 548 would have involved a similar splendid procession. The rhythm of the city’s ceremonial was dictated by the rhythm of the annual calendar. In fact, the inhabitants of Justinian’s Constantinople knew four different years simultaneously and marked the start of each of them with due ceremony and celebration: 1 January (consulship); 1 April (Justinian’s regnal year); 11 May (birthday of the city); 1 September (tax year or indiction, liturgical year).152 To minimise the confusion of overlapping years Justinian decreed in 537 that the official method of reckoning and recording time on legal and other documents was to be the year of his reign.153 Traditionally, the consulship had been a great Roman 147 See Croke (2010) and Chapter 1, 6–28. 148 Dagron (1974), 91–2, 405–9, 454–8, 495; Patlagean (1977), 215; MacCormack (1981), 62ff; Diefenbach (2002), 21–47; Van Nuffelen (2012), 183–20. 149 Malalas, Chronicle 18.25 (Thurn 368); Theophanes, Chronicle A.M. 6025 (de Boor 186). 150 Peter the Patrician, in Const. Porph., On the Ceremonies, 1.95 (Reiske 422–3, trs. Moffatt and Tall [2012], 422–3). 151 Corippus, Panegyric on Justin 3.1–61. 152 Janin (1966), 73. 153 Justinian, Novel 47 (537).

271

J U S T I N I A N ’ S C O N S TA N T I N O P L E

festival at which the new consuls for the year could demonstrate their generosity with lavish displays and handouts to the populace. Justinian’s own consulships in 521 and 528 were said to be the most lavish ever,154 although the diptych he issued in 521 was scarcely ornate.155 In attempting to control the consulship more tightly Justinian declared that he would be a perpetual consul and he had the year-long duties for others streamlined, compressing them into a single week.156 By 542 the consulship had lapsed altogether.157 Throughout the liturgical year Constantinople witnessed regular public ceremo­ nial and liturgy to mark feast days commemorating saints and martyrs, previous emperors, the opening of churches, earthquakes and other natural events, all of which engendered an official public proclamation, normally by the City Prefect.158 Many of these occasions involved religious processions.159 Indeed, most of the liturgical processions recorded in later Byzantine calendars were already insti­ tutionalised at Constantinople by Justinian’s day.160 The common features of all these processions and liturgical re-enactments was the role of the imperial house­ hold and patriarch, integrated with the populace in a common cause of propitiation and celebration. The ritual and ceremonial of Justinian’s Constantinople centred on the fora which were settings for imperial events with each new emperor seek­ ing to eclipse his predecessors by giving his forum prominence on processional routes. This ritualised use of the city’s main public spots played a major role in focussing the energies and interests of the people in the liturgical celebrations of the day.161 In the early years of Justinian’s reign the most conspicuous ceremonial was devoted to the traditional celebration of triumph. Justinian could boast triumphs over the Persians and Bulgars in 530,162 but most spectacular of all were the tri­ umphs in 534 and 541 when the Kings of the Vandals and Goths respectively, along with their precious possessions, were paraded before the emperor and people of Constantinople.163 These triumphs became the highlights of Justinian’s reign. There are many other instances of the emperor celebrating at Constantinople news of an imperial victory, mainly local and short-term, such as receiving the keys of Verona and Brescia in 562.164 Still, the emperor did not need to win military 154 521: Marcellinus, Chronicle s.a. 521 (ed. Mommsen, MGH AA, XI, 101–2); 528: Chronicon Paschale 627 (Dindorf). 155 Cutler (1984), 75–115. 156 Justinian, Novels 105. 157 CLRE 7–12. 158 McCormick (1986), 190–5. 159 Janin (1966), 68–9 for the processions and Averil Cameron (1987), 106–36 for the context of the Book of Ceremonies. 160 Croke (2001), 122–4. 161 Dagron (1977) 1–25; Baldovin (1987), 182ff, 211–14; Croke (2001), 116–24; Berger (2001), 27–61. 162 Croke (1980), 188–95. 163 McCormick (1986), 65–7, 124–9. 164 Malalas, Chronicle 18.140 (Thurn 425); Theophanes, Chronicle A.M. 6055 (de Boor 237–9).

272

J U S T I N I A N ’ S C O N S TA N T I N O P L E

victories to be portrayed as victorious. He was ever-victorious, a triumphal ruler. It was simply part of the imperial ideology so that when Justinian returned to the capital in 559 he entered and progressed triumphally through the city as if he had personally defeated the Cotrigur Huns and repulsed them by force of arms. He was even acclaimed as ‘victor’. This was the first time a church had been included in the ritual of triumph.165 Considerable ceremonial and time was also taken up by Justinian receiving and entertaining foreign potentates. The Persian king never travelled to Constantino­ ple but most lesser rulers did. At the beginning of Justinian’s reign in 527 Grepes, king of the Heruli was honoured in the imperial capital and sponsored in baptism by the emperor.166 A deputation ‘of the flower of the [Lazi] nation’s nobility’ was sent to Justinian and received the royal insignia from him.167 The same pattern was followed in 528 by Grod, king of the Huns168 then the Hun Askoum.169 In 534 the Iberian king Zamanarzos ‘accompanied by his wife and senators’ came to the city. Justinian lavished him with gifts while Theodora ‘gratified his wife with jewel­ lery of all kinds decorated with pearls’.170 The Persian king’s ambassadors were always particularly honoured and entertained at Constantinople as Isdigousnas (or Zich) and his family found in 547/8171 and again in 555.172 In 549 Justinian introduced the Ethiopian ambassador to the hippodrome crowd,173 and at other times received legations from the Heruli,174 the Huns,175 the Lombards176 and the Avars whose appearance created quite a stir in the city.177 Popes from ‘old Rome’ were entertained by Justinian too. Pope John came to Constantinople with great ceremony in 526 and celebrated Easter there, Pope Agapetus did likewise in 536 and Pope Vigilius in 547. None of them lived to see Rome again. Justinian’s singular contribution to Constantinople was the building of churches and monasteries. He built or rebuilt 33 of them. In the 520s, just before he came to the throne and just after, he built both the churches of Sts Peter and Paul and Sts Sergius and Bacchus on his estate at the palace of Hormisdas,178 and was responsible for the restoration of the church at Blachernae.179 In the 530s there fol­ lowed the replacement churches of Saint Irene and, most magnificent of all, Hagia 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179

McCormick (1986), 67. Malalas, Chronicle 18.6 (Thurn 356). Procopius, Wars 3.13.1–3, 15.2. Malalas, Chronicle 18.14 (Thurn 360–1). Malalas, Chronicle 18.21 (Thurn 366). Theophanes, Chronicle A.M. 6027 (de Boor 216). Procopius, Wars 2.28.39–41. Agathias, Histories 4.30.8. Malalas, Chronicle 18.106 (Thurn 411); Theophanes, Chronicle A.M. 6042 (de Boor 226–7). Procopius, Wars 6.15.30–3. Procopius, Wars 8.4.12–13. Procopius, Wars 8.27.2–4. Malalas, Chronicle 18.125 (Thurn 420); Theophanes, Chronicle A.M. 6050 (de Boor 231–2). Croke (2006); Bardill (2017); Ousterhout (2018). See further, above pp. 207–45. Procopius, Buildings 1.3.3–5.

273

J U S T I N I A N ’ S C O N S TA N T I N O P L E

Sophia which was dedicated on 27 December 537 commencing with a procession from the church of Anastasia in which Justinian advanced with the people on foot while the patriarch Menas was conveyed in the imperial carriage.180 A similar procession took place on 28 June 550 for the dedication of the new church of the Holy Apostles. Again Menas occupied the imperial carriage juggling on his knees three caskets containing the relics of Andrew, Luke and Timothy, a scene reminis­ cent of that on the famous Trier Ivory. Another empty imperial carriage preceded him.181 Presumably Justinian was processing along the Mese on foot and would be conveyed back to the palace in the carriage. The process was repeated once more the following year, this time for the dedication of the church of St Irene at Sykai, on the site now occupied by Arap Camii. The patriarch Menas began from Hagia Sophia sitting in the imperial carriage with relics for the new church on his knee, with Justinian evidently using another carriage.182 Then there was the dedication of the church of Theodora in August 562183 and a few months later on 24 Decem­ ber was the grand occasion of the rededication of Hagia Sophia with its secure new dome.184 This was the event that gave rise to the celebrated encomium of Paul the Silentiary.185 Churches were also the engines of social philanthropy and, besides these, Justinian was the founder of the hospitals of Sampson and Eubulus, while the hospital of Narses was also built during this period.186 Justinian’s Constantinople was a vibrant and volatile city, noisy, smelly and colourful. As the imperial capital, it provided the stage for his long and sedentary reign. The emperor possessed the city, he dominated it and secured it. During Jus­ tinian’s reign the empire changed dramatically in political, economic and social terms but the emperor did not fundamentally change the life or organisation of Constantinople. When Justinian first entered the city it was much the same size as he left it sixty to seventy years later. He had tried to ensure that the population was carefully monitored and its expansion strictly controlled. His particular contribu­ tion to the city itself was to enlarge its ritual and to integrate the growing num­ ber of churches into its ceremonial life, ‘a change which announces the Middle Ages’.187 The widespread destruction of the Nika riots in 532 provided an unfore­ seen opportunity to refurbish and reshape much of the ceremonial core of the city to this end. Most of what we know of Justinian’s Constantinople comes from 180 Malalas, Chronicle 18.86 (Thurn 405); Theophanes, Chronicle A.M. 6030 (de Boor 217). For orientation on Hagia Sophia: Mathews (1971), 77–102; Mainstone (1988); Pentcheva (2017); Ousterhout (2019), 199–217; Croke (2021), 139–67. 181 Malalas, Chronicle 18.109 (Thurn 412); Theophanes, Chronicle A.M. 6042 (de Boor 226–7); Procopius, Buildings 1.4.17ff. 182 Malalas, Chronicle 18.113 (Thurn 414–15); Procopius, Buildings, 1.7; Theophanes, Chronicle A.M. 6044 (de Boor 228). 183 Malalas, Chronicle 18.137 (Thurn 424). 184 Malalas, Chronicle 18.143 (Thurn 429); Theophanes, Chronicle A.M. 6055 (de Boor 237–9). 185 Macrides and Magdalino (1988), 47–82. Part translation of Paul in Bell (2009), 189–212. 186 References and analysis in Magdalino (2007), 31–2. 187 Mango (2004), 52, cf. Magdalino (2002), 530.

274

J U S T I N I A N ’ S C O N S TA N T I N O P L E

the contemporary literary records, including the first attempt to codify imperial ceremonial such as the adventus of Justinian returning from Selymbria in August 559. In recent years, there has been an efflorescence of interest, new excavations, texts, and translations, as well as much new research on Constantinople including the sixth century city. Justinian’s Constantinople is slowly coming to life.

275

BIBLIOGRAPHY

(a) Frequently cited sources Candidus edition and translation, R. C. Blockley (1983) The Fragmentary Classicising Historians of the Later Roman Empire: Eunapius, Olympiodorus, Priscus and Malchus, vol. 2. Liverpool, 463–73. Cedrenus edition, L. Taraglia (2016) Georgii Cedreni Historiarum Compendium. Rome. Chronicon Paschale edition, L. Dindorf (1832) translation, Michael Whitby and Mary Whitby (1989) Chronicon Paschale 284–628 AD. Liverpool. Cyril of Scythopolis, Vita Sabae, edition, E. Schwartz (1939) Kyrillos von Skythopolis, Texte und Untersuchungen, vol. 49. Leipzig, translation, R. M. Price (1991) Cyril of Scythopolis: Lives of the Monks of Palestine. Kalamazoo. Evagrius HE edition, J. Bidez and L. Parmentier (1898) The Ecclesiastical History of Evagrius with the Scholia. London, translation, Michael Whitby (2000) The Ecclesiasti­ cal History of Evagrius Scholasticus. Liverpool. John Lydus De magistratibus populi Romani, edition and translation, M. Dubuisson and J. Schamp (2006) Jean le Lydien. Des Magistratures de l’état romain. Paris. John of Antioch edition and translation, S. Mariev (2008) Iohannis Antiochenii fragmenta que supersunt omnia, CFHB 47. Berlin. Jordanes edition, Th. Mommsen (1882) Iordanis Romana et Getica, MGH.AA, vol. 5.1. Berlin. Malalas edition J. Thurn (2000) Ioannis Malalae Chronographia, CFHB 35. Berlin, trans. E. Jeffreys, M. Jeffreys and R. Scott (1986) The Chronicle of John Malalas. Melbourne. Malchus edition and translation, R. C. Blockley (1993) The Fragmentary Classicising His­ torians of the Later Roman Empire: Eunapius, Olympiodorus, Priscus and Malchus, vol. 2. Liverpool, 402–55. Marcellinus edition, Th. Mommsen (1894) Chronica minora saec. IV. V. VI. VII, vol. 2, MGH.AA, 11, 37–108. Berlin, translation, B. Croke (1995) The Chronicle of Marcel­ linus. Sydney. Priscus edition and translation, R. C. Blockley (1983) The Fragmentary Classicising His­ torians of the Later Roman Empire: Eunapius, Olympiodorus, Priscus and Malchus, vol. 2. Liverpool, 222–377. Ps. Zachariah of Mytilene edition, E. W. Brooks (1919–1921) Historia ecclesiastica Zacha­ riae Rhetori uulgo adscripta, 2 vols, CSCO Scriptores Syri, 38–39. Louvain, translations,

276

BIBLIOGRAPHY

G. Greatrex et al. (2010) Historia ecclesiastica the Chronicle of Pseudo-Zachariah Rhetor: Church and War in Late Antiquity. Liverpool. Socrates HE edition, C. G. Hansen (1995) Berlin, translations, P. Périchon and P. Maraval (2004–7) Socrate de Constantinople. Histoire Ecclesiastique, 4 vols. Paris. Sozomen HE edition, J. Bidez and G. C. Hansen (1960). Berlin. translations, A. Festugière, with B. Grillet and G. Sabbah (1983–2008) Sozomène, Histoire ecclésiastique, 4 vols. Paris. Theophanes Theophanis Chronographia, edition C. de Boor (1885). Leipzig, translations, C. Mango and R. Scott (1997) The Chronicle of Theophanes Confessor. Oxford. Victor Tonnenensis edition, C. Hartmann (2002) Victoris Tunnunensis Chronicon cum rel­ iquiis ex Consularibus Caesaraugustanis, CCh 173a. Turnhout = Th. Mommsen (1894) Chronica minora saec. IV. V. VI. VII, vol. 2, MGH.AA, 11, 178–220. Berlin. Zonaras edition, Th. Büttner-Wobst (1897) Ioannis Zonarae. Annales, vol. 3. Berlin.

(b) Modern scholarship Alchermes, J. (2005) ‘Art and Architecture’, in ed. M. Maas, Cambridge Companion to the Age of Justinian. Cambridge: Cambridge University Press, 343–75. Alemany, A. (2000) Sources on the Alans: A Critical Compilation. Leiden: Brill. Alexander, P. (1967) The Oracle of Baalbek: The Tiburtine Sybil in Greek Dress. Washing­ ton, DC: Dumbarton Oaks. Alföldi-Rosenbaum, E. (1968) ‘Portrait Bust of a Young Lady of the Time of Justinian’, Metropolitan Museum Journal 1, 19–40. Alivisatos, H. S. (1913) Die kirchliche Gesetzgebung des Kaisers Justinian I. Berlin, rp. in (1973). Aalen: Scientia. Allen, P. (1981) Evagrius Scholasticus the Church Historian. Louvain: Spicilegium Sacrum Lovaniense. Allen, P. (1992) ‘Contemporary Portrayals of the Empress Theodora (527–548)’, in eds. B. Garlick and P. Allen, Stereotypes of Women in Power: Historical Perspectives and Revisionist Views. New York: Greenwood Press, 93–103. Allen, P. (2000) ‘The Definition and Enforcement of Orthodoxy’, in eds. A. Cameron, B. Ward-Perkins and M. Whitby, Cambridge Ancient History, vol. 14. Cambridge: Cam­ bridge University Press, 811–34. Amory, P. (1997) People and Identity in Ostrogothic Italy, 489–554. Cambridge: Cam­ bridge University Press. Anastos, M. (1985) ‘The Emperor Justin I’s Role in the Restoration of Chalcedonian Doc­ trine, 518–519’, Byzantina 13, 127–39. Anderson, B. (2016) ‘Social Clustering in 5th c. Constantinople: The Evidence of the Noti­ tia’, Journal of Roman Archaeology 29, 494–508. Angelova, D. (2004) ‘The Ivories of Ariadne and Ideas about Female Imperial Authority in Rome and Early Byzantium’, Gesta 43, 1–15. Angelova, D. (2015) Sacred Founders: Women, Men, and Gods in the Discourse of Impe­ rial Founding, Rome through Early Byzantium. Berkeley, Los Angeles and London: Uni­ versity of California Press. Antonopoulos, P. (1990) Pe,troj Patri,kioj o` Buzanti,noj diplwma,thj avxiwmatou/coj kai, suggrafe,aj. Athens: Historikes Ekdoseis St. D. Vasilopoulos.

277

BIBLIOGRAPHY

Athanassiadi, P. (2006) ‘Antiquité tardive: Construction et déconstruction d’un modèle historiographique’, Antiquité tardive 14, 311–24. Avramea, A. (2002) ‘Land and Sea Comunications, Fourth-Fifteenth Centuries’, in ed. A. Laiou, The Economic History of Byzantium: From the Seventh through the Fifteenth Century. Washington, DC: Dumbarton Oaks, 57–90. Baldovin, J. F. (1987) The Urban Character of Christian Worship: The Origins, Develop­ ment, and Meaning of Stational Liturgy. Rome: Pont. Institutum Studiorum Orientalium. Baldwin, B. (1977) ‘Malchus of Philadelphia’, Dumbarton Oaks Papers 31, 91–107. Baldwin, B. (1989) ‘Illiterate Emperors’, Historia 38, 124–6. Bandy, A. (1983) trans. On Powers, or, the Magistracies of the Roman State/Ioannes Lydus. Philadelphia: Edward Mellen Press. Bardill, J. (1997) ‘The Palace of Lausus and Nearby Monuments in Constantinople: A Topographical Study’, American Journal of Archaeology 101, 67–93. Bardill, J. (1999a) ‘The Golden Gate in Constantinople: A Triumphal Arch of Theodosius I’, American Journal of Archaeology 103, 671–96. Bardill, J. (1999b) ‘The Great Palace of the Byzantine Emperors and the Walker Trust Excavations’, Journal of Roman Archaeology 12, 216–30. Bardill, J. (2000) ‘The Church of Sts. Sergius and Bacchus in Constantinople and the Monophysite Refugees’, Dumbarton Oaks Papers 54, 1–11. Bardill, J. (2004) Byzantine Brickstamps. Oxford: Oxford University Press. Bardill, J. (2006) ‘A New Temple for Byzantium: Anicia Juliana, Solomon, and the Gilded Ceiling of the Church of St Polyeuktos, Constantinople’, Late Antique Archaeology 3, 339–70. Bardill, J. (2017) ‘The Date, Dedication, and Design of Sts. Sergius and Bacchus in Con­ stantinople’, Journal of Late Antiquity 10, 62–130. Barker, J. W. (1966) Justinian and the Later Roman Empire. Madison, WI: The University of Wisconsin Press. Barnes, C. L. H. (2006) ‘The Ferries of Tenedos’, Historia 55, 167–77. Barsanti, C. (1995) ‘Il foro di Teodosio I a Constantinopoli’, in eds. A. Iacobini and E. Zanini, Arte profana e arte sacra a Bisanzio. Rome: Àrgos, 9–50. Bassanelli, G. (1971) ‘La legislazione processuale di Giustino 1’, Studia et Documenta Historiae Iuris 37, 119–216. Bassett, S. (2004) The Urban Image of Late Antique Constantinople. Cambridge: Cam­ bridge University Press. Bauer, F. A. (1996) Stadt, Platz und Denkmal in der Spätantike. Untersuchungen zur Aus­ stattung des öffentlichen Raums in den spätantiken Städten Rom, Konstantinopel und Ephesos. Mainz: Philipp von Zabern. Bauer, F. A. (2001) ‘Urban Space and Ritual: Constantinople in Late Antiquity’, Acta ad Archaeologiam et Artium Historiam pertinentia 15, 27–61. Baynes, N. H. (1925) ‘The Vita S. Danielis Stylitae’, English Historical Review 40, 397–402. Baynes, N. H. (1955a) ‘The Finding of the Virgin’s Robe’, in ed. N. H. Baynes, Byzantine Studies and Other Essays. London: The Athlone Press, 240–7. Baynes, N. H. (1955b) ‘The Spiritual Defenders of Constantinople’, in ed. N. H. Baynes, Byzantine Studies and Other Essays. London: The Athlone Press, 248–60. Beaucamp, J., Briquel-Chatonnet, F. and Julien, R. C. (1999/2000) ‘La persécution des chrétiens de Nagran et la chronologie himyarite’, Aram 11/12, 15–43.

278

BIBLIOGRAPHY

Beck, H.-G. (1965) ‘Konstantinopel. Zur Sozialgeschichte einer frühmittelalterlichen Hauptstadt’, Byzantinische Zeitschrift 58, 11–45, rp. in H.-G. Beck (1972) Ideen und Realitäten in Byzanz. London. Beck, H.-G. (1973) ‘Grossstadt-Probleme: Konstantinopel vom 4–6 Jahrhundert’, in ed. H.-G. Beck, Studien zur Frühgeschichte Konstantinopels. Munich: Institut für Byzan­ tinistik und Neugriechische Philologie der Universität, 3–19. Begass, C. (2016) ‘Property and Power of the Senatorial Aristocracy of the Eastern Roman Empire in the Fifth and Sixth Centuries’, Journal of Late Antiquity 9, 462–82. Begass, C. (2018) Die Senatsaristokratie des oströmischen Reiches, ca. 457–518. Proso­ pographische und sozialgeschichtliche Untersuchungen. Munich: C. H. Beck. Bell, P. N. (2009) Three Political Voices from the Age of Justinian. Liverpool: Liverpool University Press. Bellen, H. (1981) Die germanische Leibwache der römischen Kaiser des julisch-claudischen Hauses. Wiesbaden: Steiner. Bellinger, A. (1966) ‘Byzantine Notes 3: The Gold of Justin and Justinian I’, American Numismatic Society, Notes and Monographs 12, 90–2. Bellinger, A. (1992) Catalogue of the Byzantine Coins in the Dumbarton Oaks Collection and in the Whittemore Collection: Anastasius I to Maurice, 491–602. Washington, DC: Dumbarton Oaks. Berger, A. (1988) Untersuchungen zu den Patria Konstantinupoleos. Bonn: Habelt. Berger, A. (1997) ‘Regionen und Straßen im frühen Konstantinopel’, Istanbuler Mitteilun­ gen 47, 349–414. Berger, A. (2000) ‘Streets and Public Spaces in Constantinople’, Dumbarton Oaks Papers 54, 161–72. Berger, A. (2001) ‘Imperial and Ecclesiastical Processions in Constantinople’, in ed. N. Necipoglu, Byzantine Constantinople: Monuments, Topography and Everyday Life. Leiden-Boston-Cologne: Brill, 73–87. Berger, A. (2013) Accounts of Medieval Constantinople: The Patria. Washington, DC: Dumbarton Oaks. Bersanetti, G. M. (1943/4) ‘Basilisco e l’imperatore Leone I’, Rendiconti della Pontificia Accademia di Archeologia 20. Blockley, R. (1981) The Fragmentary Classicising Historians of the Later Roman Empire, vol. 1. Liverpool: Francis Cairns. Blockley, R. (1982) ‘Roman-Barbarian Marriages in the Late Empire’, Florilegium 4, 63–79. Blockley, R. (1992) East Roman Foreign Policy. Leeds: Francis Cairns. Boak, A. (1919) ‘Imperial Coronation Ceremonies of the Fifth and Sixth Centuries’, Har­ vard Studies in Classical Philology 30, 39–40. Boak, A. (1924) Two Studies in Later Roman and Byzantine Administration. New York: Macmillan. Bolgiani, F. (1980) ‘Decadenza di Roma o tardo Antico?: Alcune riflessioni sull’ultimo libro di Henri-Irénée Marrou’, in La Storiografia ecclesiastica nella tarda antichità: Atti del convegno tenuto in Erice (1978). Messina: Centro di studi umanistici, 535–88. Bolognesi, E. (2000) ‘Il Gran palazzo’, Bizantinistica 2, 197–242. Bolognesi Recchi-Franceschini, E. (2008) ‘The Scholae of the Master of Offices as the Palace Praetorium’, Anatolia Antiqua 16, 231–57. Bolognesi Recchi-Franceschini, E. and Featherstone, M. (2002) ‘The Boundaries of the Palace: De Ceremoniis II.13’, Travaux et Mémoires 14, 36–47.

279

BIBLIOGRAPHY

Bono, P., Crow, J. and Bayliss, R. (2001) ‘The Water Supply of Constantinople: Archaeol­ ogy and Hydrogeology of an Early Medieval City’, Environmental Geology 40, 1325–33. Börm, H. (2010) ‘Heerscher und Eliten’, in eds. H. Börm and J. Wiesehöfer, Commutatio et contentio: Studies in the Late Roman, Sasanian, and Early Islamic Near East. Düs­ seldorf: Wellem, 158–98. Brandes, W. (1993) ‘Familienbande? Odoaker, Basiliskos und Harmatios’, Klio 75, 407–37. Bremmer, J. (1998) ‘An Imperial Palace Guard in Heaven: The Date of the Vision of Doro­ theus’, Zeitschrift für Papyrologie und Epigraphik 75, 82–8. Brennan, P. (1996) ‘The Notitia Dignitatum’, in Les littératures techniques dans l’antiquité romaine. Geneva: Fondation Hardt, 147–78. Brennan, P. and Barlow, J. (2001) ‘Tribuni scholarum palatinarum c.A.D. 353–364: Ammi­ anus Marcellinus and the Notitia Dignitatum’, Classical Quarterly 51, 237–54. Brière, M. and Graffi, F. (1975) eds. and trans. Les Homiliae cathedrales de Sévère d’Antioche. Homélies XVIII à XXV, PO, 37/1. Turnhout: Brepols. Brock, S. (1981) ‘The Conversations with the Syrian Orthodox under Justinian (532)’, Orientalia Christiana Periodica 47, 87–121. Brodka, D. (2004) Die Geschichtsphilosophie in der spätantiken Historiographie. Frank­ furt a.M.: P. Lang. Brooks, E. W. (1893) ‘The Emperor Zenon and the Isaurians’, English Historical Review 8, 209–38. Brooks, E. W. (1905) trans. Jacob of Edessa, Chronicon, CSCO Scriptores Syri, 6. Louvain: Peeters. Brooks, E. W. (1915) ed. and trans. A Collection of Letters of Severus of Antioch from Numerous Syriac Manuscripts, PO, 12. Paris: Firmin-Didot. Brooks, E. W. (1936) trans. Iohannis Ephesini Historiae Ecclesiasticae pars Tertia, CSCO Scriptores Syri, 55. Louvain: Peeters. Brown, P. (1982) The Cult of the Saints. Chicago, IL: University of Chicago Press. Brown, P. (1995) Authority and the Sacred. Cambridge: Cambridge University Press. Brown, P. (1997) ‘The World of Late Antiquity Revisited (with others)’, Symbolae Osloenses 72, 5–90. Brown, P. (2011) ‘The Field of Late Antiquity’, in ed. D. Hernandez de la Fuente, New Perspectives on Late Antiquity. Newcastle upon Tyne: Cambridge Scholars Publishing, 6–22. Browning, R. (1987) Justinian and Theodora, 2nd edn. London: Thames and Hudson. Brubaker, L. (1997) ‘Memories of Helena: Patterns of Matronage in the Fourth and Fifth Centuries’, in ed. L. James, Women, Men and Eunuchs: Gender in Byzantium. London: Routledge, 52–75. Brubaker, L. (2002) ‘The Vienna Dioskorides and Anicia Juliana’, in eds. A. Littlewood, H. Maguire and J. Wolschke-Bulmahn, Byzantine Garden Culture. Washington, DC: Dumbarton Oaks, 189–214. Brubaker, L. and Tobler, H. (2000) ‘The Gender of Money: Byzantine Empresses on Coins (324–802)’, Gender and History 12, 572–94. Budge, E. A. W. (1909) Texts Relating to Saint Mêna of Egypt and Canons of Nicaea in a Nubian Dialect. London: Longman. Burgess, R. W. (1993) The Chronicle of Hydatius and the Consularia Constantinopolitana. Oxford: Oxford University Press. Burgess, R. W. (1993/4) ‘The Accession of Marcian in the Light of Chalcedonian Apolo­ getic and Monophysite Polemic’, Byzantinische Zeitschrift 86/87, 47–68.

280

BIBLIOGRAPHY

Burgess, R. W. (2003) ‘The Passio S. Artemii, Philostorgius and the Dates of the Invention and Translation of the Relics of Andrew and Luke’, Analecta Bollandiana 121, 5–36. Burgess, W. D. (1990) ‘Isaurian Names and the Ethnic Identity of the Isaurians in Late Antiquity’, Ancient World 21, 109–21. Burgess, W. D. (1992) ‘Isaurian Factions in the Reign of Zeno the Isaurian’, Latomus 51, 874–80. Bury, J. B. (1886) ‘A Note on the Emperor Olybrius’, English Historical Review 1, 507–9. Bury, J. B. (1897) ‘The Nika Riot’, Journal of Hellenic Studies 17, 92–119. Bury, J. B. (1907) ‘The Ceremonial Book of Constantine Porphyrogennetos’, English His­ torical Review 22, 209–27 and 426–48. Bury, J. B. (1911) The Imperial Administrative System in the Ninth Century. London: H. Frowde. Bury, J. B. (1912) ‘The Great Palace’, Byzantinische Zeitschrift 21, 210–25. Bury, J. B. (1919) ‘Justa Grata Honoria’, Journal of Roman Studies 9, 1–13. Bury, J. B. (1923a) A History of the Later Roman Empire, vol. 1. London: Macmillan and Co. Bury, J. B. (1923b) A History of the Later Roman Empire, vol. 2. London: Macmillan and Co. Cameron, A. (1972) Review of Frank (1969) Classical Review, n.s. 22, 136. Cameron, A. (1973) Porphyrius the Charioteer. Oxford: Oxford University Press. Cameron, A. (1975a) ‘The Empress Sophia’, Byzantion 65, 5–21, rp. in Cameron, A. (1981) Continuity and Change in Sixth-Century Byzantium. London: Ashgate. Cameron, A. (1975b) ‘The Byzantine Sources of Gregory of Tours’, Journal of Theological Studies 26, 421–6, rp. in Cameron, A. (1981) Continuity and Change in Sixth-Century Byzantium. London: Ashgate. Cameron, A. (1976a) ‘Theodorus triseparchos’, Greek Roman and Byzantine Studies 17, 269–86. Cameron, A. (1976b) Circus Factions. Oxford: Oxford University Press. Cameron, A. (1976c) ‘The Authenticity of the Letters of St Nilus of Ancyra’, Greek Roman and Byzantine Studies 17, 181–96. Cameron, A. (1976d) ed. and trans. Corippus, In laudem Iustini Augusti minoris. London: Athlone. Cameron, A. (1978) ‘The House of Anastasius’, Greek, Roman and Byzantine Studies 19, 259–76. Cameron, A. (1981) Continuity and Change in Sixth-Century Byzantium. London. Cameron, A. (1982a) ‘The Empress and the Poet: Paganism and Politics at the Court of Theodosius II’, Yale Classical Studies 27, 217–90. Cameron, A. (1982b) ‘The Death of Vitalian’, Zeitschrift für Papyrologie und Epigraphik 48, 93–4. Cameron, A. (1985) Procopius. London: Duckworth. Cameron, A. (1987) ‘The Construction of Court Ritual: The Byzantine Book of Ceremo­ nies’, in eds. D. Cannadine and S. Price, Rituals of Royalty. Power and Ceremonial in Traditional Societies. Cambridge: Cambridge University Press, 106–36. Cameron, A. (1993a) with J. Long and L. Sherry, Barbarians and Politics at the Court of Arcadius. Berkeley, Los Angeles and London: University of California Press. Cameron, A. (1993b) The Mediterranean World in Late Antiquity AD 395–600. London: Duckworth. Cameron, A. (2000) ‘Justin I and Justinian’, in eds. A. Cameron, B. Ward-Perkins and M. Whitby, Cambridge Ancient History, vol. 14. Cambridge: Cambridge University Press, 63–85.

281

BIBLIOGRAPHY

Cameron, A. (2002) ‘The “Long” Late Antiquity: A Late Twentieth Century Model’, in ed. T. P. Wiseman, Classics in Progress. Oxford: Oxford University Press, 165–91. Cameron, A. (2003) ‘Ideologies and Agendas in Late Antique Studies’, in eds. L. Lavan and W. C. Bowden, Late Antique Archaeology 1: Theory and Practice in Late Antique Archaeology. Leiden: Brill, 3–21. Cameron, A. (2005) ‘History and the Individuality of the Historian: The Interpretation of Late Antiquity’, in eds. C. Straw and R. Lim, The Past before Us: The Challenge of Historiographies of Late Antiquity. Turnhout: Brepols, 69–78. Cameron, A. (2008) ‘A. H. M. Jones and the End of the Ancient World’, in ed. D. Gwynn, A. H. M. Jones and the Later Roman Empire. Leiden: Brill, 232–49. Cameron, A. (2016) ‘Late Antiquity and Byzantium: An Identity Problem’, Byzantine and Modern Greek Studies 40, 27–37. Cameron, A. and Herrin, J. (1984) eds. Constantinople in the Early Eighth Century: The Parastaseis syntomoi chronikai. Leiden: Brill. Capizzi, C. (1968) ‘Anicia Giuliana (462ca–530ca), Ricerche sulla sua famiglia e la sua vita’, Rivista di studi byzantini e neoellenici 15, 214–26. Capizzi, C. (1969) L’Imperatore Anastasio. Rome: Pont. Institutum Orientalium Studiorum. Capizzi, C. (1973) ‘L’intervento di Anicia Giuliana nell’unione ecclesiastica del 519’, in Atti del Congresso internazionale sulle relazione fra le due Sponde adriatiche. Lecce: Le edizioni del lavoro, 51–86. Capizzi, C. (1977) ‘L’attività edilizia di Anicia Giuliana’, in Collectanea Byzantina, Orien­ talia Christiana Analecta, 204. Rome: Pont. Institutum Orientalium Studiorum, 120–46. Capizzi, C. (1994) Giustiniano I tra politica e religione. Messina: Rubbettino. Capizzi, C. (1997) Giuliana la committente. Milan: Jaca. Cesaretti, P. (2004) Theodora: Empress of Byzantium. New York: Vendome Press. Chabot, J.-B. (1899) ed. and trans. Chronique de Michel le Syrien, patriarche jacobite d’Antioche, 1166–1199. 4 vols. Paris: Ernest Leroux. Chabot, J.-B. (1920) ed. and trans. Chronicon Anonymum ad 819, CSCO Scriptores Syri, 81. Paris: J. Gabalda. Chabot, J.-B. (1937) ed. and trans. Chronicon ad annum 1234, CSCO Scriptores Syri, 56. Louvain: Peeters. Chew, K. (2006) ‘Virgins and Eunuchs: Pulcheria, Politics and the Death of Emperor Theo­ dosius II’, Historia 55, 207–27. Clark, E. A. (2004) History, Theory, Text. Historians and the Linguistic Turn. Cambridge, MA: Harvard University Press. Clauss, M. (1980) Der magister officiorum in der Spätantike. Munich: C. H. Beck. Clover, F. M. (1978) ‘The Family and Early Career of Anicius Olybrius’, Historia 27, 169–96. Coleman-Norton, P. (1966) Roman State and Christian Church, 3 vols. London: SPCK. Connor, C. (1999) ‘The Epigram in the Church of Hagios Polyeuktos in Constantinople and Its Byzantine Response’, Byzantion 69, 479–527. Connor, C. (2004) Women of Byzantium. New Haven: Yale University Press. Conrad, L. (2000) ‘Zeno, the Epileptic Emperor: Historiography and Polemics as Sources of Realia’, Byzantine and Modern Greek Studies 24, 61–81. Courtois, C. (1955) Les Vandales et l’Afrique. Paris: Arts et métiers graphiques. Cracco Ruggini, L. (1989) ‘Arnaldo Momigliano e il tardoantico’, in ed. L. Cracco Rug­ gini, Omaggio ad Arnaldo Momigliano. Como: Edizioni New press, 159–84.

282

BIBLIOGRAPHY

Cracco Ruggini, L. (1993) ‘Il tardoantico: per una tipologia dei punti critici’, in Storia di Roma, 3.1. Turin: Einaudi, xxxiii–xlv. Crawford, P. (2019) Roman Emperor Zeno, the Perils of Power Politics in Fifth-Century Constantinople. Barnsley: Pen and Sword. Croke, B. (1980) ‘Justinian’s Bulgar Victory Celebration’, Byzantinoslavica 41, 188–95, rp. in Croke, B. (1992) Christian Chronicles and Byzantine History, 5th–6th Centuries. London: Ashgate. Croke, B. (1981) ‘Two Early Byzantine Earthquakes and Their Liturgical Commemora­ tion’, Byzantion 51, 112–47, rp. in Croke, B. (1992) Christian Chronicles and Byzantine History, 5th-6th Centuries. London: Ashgate. Croke, B. (1982) ‘The Date of the Anastasian Long Wall in Thrace’, Greek, Roman and Byzantine Studies 23, 59–78, rp. in Croke, B. (1992) Christian Chronicles and Byzantine History, 5th–6th Centuries. London: Ashgate. Croke, B. (1983) ‘Basiliscus the Boy-Emperor’, Greek, Roman and Byzantine Studies 24, 81–91, rp. in Croke, B. (1992) Christian Chronicles and Byzantine History, 5th–6th Centuries. London: Ashgate. Croke, B. (1990) ‘Malalas, The Man and His Work’, in ed. E. M. Jeffreys, with B. Croke and R. Scott, Studies in John Malalas. Sydney: Australian Association for Byzantine Studies, 1–25. Croke, B. (1995) The Chronicle of Marcellinus: Translation and Commentary. Sydney: Australian Association for Byzantine Studies. Croke, B. (2001) Count Marcellinus and His Chronicle. Oxford: Oxford University Press. Croke, B. (2003) ‘The Imperial Reigns of Leo II’, Byzantinische Zeitschrift 96, 559–75. Croke, B. (2005a) ‘Dynasty and Ethnicity: Emperor Leo I and the Eclipse of Aspar’, Chi­ ron 35, 147–203. Croke, B. (2005b) ‘Leo I and the Palace Guard’, Byzantion 75, 1–35. Croke, B. (2005c) ‘Procopius’ Secret History: Rethinking the Date’, Greek, Roman and Byzantine Studies 46, 405–32. Croke, B. (2006) ‘Justinian, Theodora, and the Church of Saints Sergius and Bacchus’, Dumbarton Oaks Papers 60, 25–63. Croke, B. (2007) ‘Justinian under Justin: Reconfiguring a Reign’, Byzantinische Zeitschrift 100, 13–56. Croke, B. (2010) ‘Reinventing Constantinople: Theodosius I’s Imprint on the Imperial City’, in eds. S. McGill, C. Sogno and E. Watts, From the Tetrarchs to the Theodosians. Cambridge, MA: Cambridge University Press, 241–64. Croke, B. (2011) ‘Justinian the ‘sleepless emperor’, in eds. G. Nathan and L. Garland, Basilieia: Essays in Imperium and Culture in Honour of E. M. and M. J. Jeffreys. Bris­ bane: Australian Association for Byzantine Studies, 103–8. Croke, B. (2014) ‘Dynasty and Aristocracy in the Fifth Century’, in ed. M. Maas, Cambridge Companion to the Age of Attila. Cambridge, MA: Cambridge University Press, 102–26. Croke, B. (2015) ‘Ariadne Augusta. Shaping the Identity of the Early Byzantine Empress’, in eds. G. D. Dunn and W. Mayer, Christians Shaping Identity from the Roman Empire to Byzantium. Leiden and Boston: Brill, 293–320. Croke, B. (2019) ‘Procopius, from Manuscripts to Books 1400–1850’, Histos Supplement 9, 1–173. Croke, B. (2021) ‘Looking, Listening and Learning: Justinian’s Hagia Sophia’, in eds. S. Gador-Whyte and A. Mellas, Hymns, Homilies and Hermeneutics in Byzantium. Leiden: Brill, 139–67.

283

BIBLIOGRAPHY

Croke, B. (forthcoming a) ‘The Political Appropriation of Archangel Michael’. Croke, B. (forthcoming b) ‘Emperor and Archangel: Justinian at Germia’. Crow, J. (1995) ‘The Long Walls of Thrace’, in eds. C. Mango and G. Dagron, Constanti­ nople and Its Hinterland. London: Ashgate, 109–24. Crow, J. (2007) ‘The Anastasian Wall and the Lower Danube Frontier before Justinian’, in ed. L. Vagalinski, The Lower Danube in Antiquity. Sofia: Bulgarian Academy of Sci­ ences, 397–410. Crow, J. (2019) ‘Constantinople in the Long Sixth Century’, in eds. I. Jacobs and H. Elton, Asia Minor in the Long Sixth Century: Current Research and Future Directions. Oxford: Oxbow, 167–80. Crow, J., Bardill, J. and Bayless, R. (2008) The Water Supply of Byzantine Constantinople. London: Society for the Promotion of Roman Studies. Crow, J. and Ricci, A. (1997) ‘Investigating the Hinterland of Constantinople: Interim Report on the Anastasian Long Wall’, Journal of Roman Archaeology 10, 235–62. Cutler, A. (1984) ‘The Making of the Justinian Diptych’, Byzantion 54, 75–115. Dagron, G. (1968) ‘La Vie ancienne de saint Marcel l’Acémète’, Analecta Bollandiana 86, 271–321. Dagron, G. (1970) ‘Les moines et la ville: Le monachisme à Constantinople jusqu’au Con­ cile de Chalcédoine’, Travaux et Mémoires 4, 229–76. Dagron, G. (1974) Naissance d’une capitale. Paris: Presses Universitaires de France. Dagron, G. (1977) ‘Le Christianisme dans la ville byzantine’, Dumbarton Oaks Papers 31, 1–25. Dagron, G. (1982) ‘Le fils de Léon 1er (463). Témoignages concordants de l’hagiographie et de l’astrologie’, Analecta Bollandiana 100, 271–5. Dagron, G. (1984) Constantinople imaginaire. Paris: Presses Universitaires de France. Dagron, G. (1989) ‘Aux origines de la civilisation byzantine: langue de culture et langue d’état’, Revue Historique 489, 29–76. Dagron, G. (1991) ‘“Ainsi rien n’échappera à la réglementation”, État, Église, corpora­ tions, confréries: à propos des inhumations à Constantinople (IVe–Xe siècles)’, in eds. V. Kravari, J. Lefort and C. Morrisson, Hommes et richesses dans l’Empire byzantin, vol. 2. Paris: P. Lethielleux, 157–61. Dagron, G. (2003) Emperor and Priest: The Official Office in Byzantium. Cambridge: Cam­ bridge University Press. Dark, K. (2005) ‘The Eastern Harbours of Early Byzantine Constantinople’, Byzantion 75, 152–163. Dark, K. and Özgumüs, F. (2001) ‘The Last Roman Imperial Palace? Rescue Archaeology in Istanbul’, Minerva 12, 52–5. Dark, K. and Özgumüs, F. (2013) Constantinople: Archaeology of a Byzantine Megalopo­ lis. Oxford: Oxbow. Daube, D. (1967) ‘The Marriage of Justinian and Theodora: Logical and Theological Reflections’, Catholic University of America Law Review 16, 380–99. Davis, R. (1989) The Book of Pontiffs (Liber Pontificalis). Liverpool: Liverpool University Press. Dawes, E. and Baynes, N. H. (1977) Three Byzantine Saints. London: Mowbrays. Delbrück, R. (1933) Spätantike Kaiserporträts. Berlin and Leipzig: de Gruyter. Delbrueck, R. (2009) Dittici consolari tardoantichi, ed. M. Abbatepaolo. Bari: Edipuglia, updated translation of Die Consulardiptychen und verwandte Denkmäler, 2 vols. Berlin, 1929.

284

BIBLIOGRAPHY

Delehaye, H. (1913) ‘De fontibus vitae S. Danielis Stylitae’, Analecta Bollandiana 32, 225–7. Delehaye, H. (1923) Les Saints stylites. Brussels and Paris: Picard. Delmaire, R. (1984) ‘Les dignitaires laïcs au Concile de Chalcedoine: notes sur la hiérar­ chie et les préséances au milieu du Ve siècIe’, Byzantion 54, 141–75. Delmaire, R. (1995) Les institutions du Bas-Empire romain. Paris: Les Editions du CERF. Del Medico, H. (1955) ‘Le couronnement d’un empereur byzantin vu par un juif de Con­ stantinople’, Byzantinoslavica 16, 43–75. Demandt, A. (1970) ‘Magister Militum’, RE Suppl. 12, 553–790. Demandt, A. (1980) ‘Der spätrömische Militäradel’, Chiron 10, 609–36. Demandt, A. (1986) ‘Die Kelch von Ardabur und Anthusa’, Dumbarton Oaks Papers 40, 113–7. Demandt, A. (1989) Die Spätantike. Munich: C. H. Beck. Demandt, A. (2007) Die Spätantike, 2nd edn. Munich: C. H. Beck. den Boeft, J., den Hengst, D. and Teitler, H. C. (1987) Philological and Historical Com­ mentary on Ammianus XX. Groningen: E. Forsten. De Simone, G. F. and Russell, B. (2019) ‘The Late-Antique Eruption of Vesuvius in AD 472 and Its Impact from the Bay of Naples to Aeclanum’, Journal of Roman Archaeol­ ogy 32, 359–89. Dewar, M. (1993) ‘Corippus on the Wakefulness of Poets and Emperors’, Mnemosyne 46, 211–23. Diefenbach, S. (2002) ‘Zwischen Liturgie und civilitas. Konstantinopel im 5. Jahrhundert und die Etablierung eines städtischen Kaisertums’, in ed. R. Warland, Bildlichkeit und Bildorte von Liturgie. Schauplätze in Spätantike, Byzanz und Mittelalter. Wiesbaden: Reichert, 21–47. Diehl, C. (1893) ‘Rescrit des empereurs Justin et Justinien en date du 1er Juin 527’, Bulletin de Correspondance hellénique 17, 501–20. Diehl, C. (1901) Justinien et la civilisation Byzantine, vol. 1. Paris: Leroux. Dijkstra, J. and Greatrex, G. (2009) ‘Patriarchs and Politics in Constantinople in the Reign of Anastasius (with a Re-Edition of O.Mon.Epiph.59)’, Millennium 6, 227–30. Doran, R. (1992) The Lives of Simeon Stylites. Kalamazoo: Cistercian Publications. Drescher, J. (1947) Three Coptic Legends: Hilaria, Archellites, the Seven Sleepers. Cairo: Imprimerie de l’Institut Français d’Archéologie Orientale. Duchesne, L. (1915) ‘Les protégés de Théodora’, Mélanges d’archéologie et d’histoire de l’école française de Rome 35, 57–80. Duchesne, L. (1925) L’église au VIe siècle. Paris: E. de Boccard. Durliat, J. (1995) ‘L’approvisionnement de Constantinople’, in eds. C. Mango and G. Dag­ ron, Constantinople and Its Hinterland. London: Ashgate, 19–34. Durry, M. (1938) Les cohortes prétoriennes. Paris: E. de Boccard. Dziurdzik, T. (2017) ‘Ethnic Units in the Late Roman Army? The Case of the Equites Dal­ matae’, Studia Europaea Gnesnensia 16, 447–62. Ebersolt, J. and Thiers, A. (1913) Les églises de Constantinople. Paris: Ernest Leroux. Effenberger, A. (1996) ‘Überlegungen zur Aufstellung des Theodosios-Obelisken im Hip­ podrom von Konstantinopel’, in ed. B. Brenk, Innovation in der Spätantike. Wiesbaden: Reichert, 207–83. Elsner, J. (2002) ‘The Birth of Late Antiquity: Riegl and Strzygowski in 1901’, Art History 25, 338–79. Elton, H. (1996) Warfare in Roman Europe AD 350–425. Oxford: Oxford University Press. Elton, H. (2000a) ‘Illus and the Imperial Aristocracy under Zeno’, Byzantion 70, 393–407.

285

BIBLIOGRAPHY

Elton, H. (2000b) ‘The Nature of the Sixth-Century Isaurians’, in eds. S. Mitchell and G. Greatrex, Ethnicity and Culture in Late Antiquity. London: Duckworth, 293–307. Elton, H. (2009) ‘Imperial Politics at the Court of Theodosius II’, in eds. A. Cain and N. Lenski, The Power of Religion in Late Antiquity. Burlington, VT: Ashgate, 133–42. Elton, H. (2018) The Roman Empire in Late Antiquity. Cambridge: Cambridge University Press. Engelhardt, I. (1974) Mission und Politik in Byzanz. Ein Beitrag zur Strukturanalyse Byz­ antinischer Mission zur Zeit Justins und Justinians. Munich: Institut fur Byzantinistik und Neugriechische Philologie der Universität. Ensslin, W. (1925) ‘Leo I’, RE 12, 1947–62. Ensslin, W. (1932) ‘Tatianus 4’, RE 4.2, 467–8. Ensslin, W. (1942/6) ‘Zur Frage nach der ersten Kaiserkrönung durch den Patriarchen und zur Bedeutung dieses Aktes im Wahlzeremoniell’, Byzantinische Zeitschtift 42, 101–15 and 369–72. Ensslin, W. (1951) ‘Papst Johannes I als Gesandter Theoderichs des Grossen bei Kaiser Justinos I’, Byzantinische Zeitschrift 44, 26–134. Ernesti, J. (1998) Princeps christianus und Kaiser aller Römer. Theodosius der Grosse im Lichte zeitgenössischer Quellen. Paderborn: Ferdinand Schöningh. Errington, R. M. (1996a) ‘The Accession of Theodosius I’, Klio 78, 438–53. Errington, R. M. (1996b) ‘Theodosius and the Goths’, Chiron 26, 1–27. Errington, R. M. (1997a) ‘Church and State in the First Years of Theodosius I’, Chiron 27, 21–72. Errington, R. M. (1997b) ‘Christian Accounts of the Religious Legislation of Theodosius I’, Klio 79, 398–443. Errington, R. M. (2000) ‘Themistius and His Emperors’, Chiron 30, 861–904. Errington, R. M. (2006) Roman Imperial Policy from Julian to Theodosius. Chapel Hill: University of North Carolina Press. Evans, J. A. S. (1996a) The Age of Justinian: The Circumstances of Imperial Power. London: Routledge. Evans, J. A. S. (1996b) ‘The Dates of Procopius’ Works: A Recapitulation of the Evidence’, Greek, Roman and Byzantine Studies 37, 301–13. Evans, J. A. S. (2002) The Empress Theodora: Partner of Justinian. Austin, TX: University of Texas Press. Evans Grubbs, J. (2002) Women and the Law in the Roman Empire. London: Routledge. Faedo, L. (1982) ‘Il complesso monumentale del Foro di Teodosio a Constantinopoli’, Corsi di cultura sull’arte ravennate e bizantina 29, 159–68. Faedo, L. (1998) ‘Teodosio, Temistio e l’ideologia ercula nella Nea Rome: A propósito dell’arco del Forum Tauri’, Mitteilungen des Deutschen Archäologischen Instituts (Röm. Abt.) 105, 315–28. Featherstone, J. (1996) ‘Life of St. Matrona of Perge’, in ed. and trans. A.-M. Talbot, Holy Women of Byzantium: Ten Saints’ Lives in English Translation. Washington, DC: Dumbarton Oaks. Feissel, D. (1999) ‘Deux grandes familles isauriennes du Ve siècle d’après des inscriptions de Cilicie Trachée’, Mitteilungen zur Christlichen Archäologie 5, 9–17. Feissel, D. (2000) ‘Les Édifices de Justinien au témoignage de Procope et de l’épigraphie’, Antiquité Tardive 8, 88–9. Feissel, D. (2006) ‘Dates et durées de règne selon Malalas, de Théodose II à Justinien’, in eds. A. Augusta-Boularot, J. Beaucamp, A.-M. Bernardi and E. Caire, Recherches sur la

286

BIBLIOGRAPHY

Chronique de Jean Malalas, vol. 2. Paris: Association des Amis du Centre d’Histoire et Civilisation Byzantine, 187–96. Festugière, A. (1961) Les moines d’Orient, vol. 1. Paris: Les Éditions du Cerf. Festy, M. (2020) with M. Vitiello, Anonyme de Valois II. L‘Italie sous Odoacre et Théodoric. Paris: Les Belles Lettres. Fiebiger, O. (1909) ‘Excubitores’, RE 6.2, 1577. Fisher, E. A. (1984) ‘Theodora and Antonina in the Historia Arcana: History and/or Fic­ tion?’, in eds. J. Peradotto and J. P. Sullivan, Women in the Ancient World: The Arethusa Papers. Albany, NY: State University of New York Press, 287–313. Foss, C. (1987) ‘St Autonomus and His Church in Bithynia’, Dumbarton Oaks Papers 41, 187–98. Foss, C. (2002) ‘The Empress Theodora’, Byzantion 72, 141–76. Foss, C. and Winfield, D. (1986) Byzantine Fortifications. Pretoria: University of South Africa. Fowden, E. K. (1999) Barbarian Plain: Saint Sergius between Rome and Iran. Berkeley, Los Angeles and London: University of California Press. Fowden, G. (1994) ‘Constantine, Silvester and the Church of S. Polyeuctus in Constanti­ nople’, Journal of Roman Archaeology 7, 274–84. Fowden, G. (2014) Before and after Muhammad: The First Millennium refocussed. Princeton: Princeton University Press. Frank, G. (2006) ‘Romanos and the Night Vigil in the Sixth Century’, in ed. D. Krueger, A People’s History of Christianity, vol. 3: Byzantine Christianity. Minneapolis: Fortress, 59–78. Frank, R. I. (1969) Scholae Palatinae: The Palace Guards of the Later Roman Empire. Rome: American Academy. Frend, W. H. C. (1972) The Rise of the Monophysite Movement. Cambridge: Cambridge University Press. Garipzanov, I. (2018) Graphic Signs of Authority in Late Antiquity and the Early Middle Ages, 300–900. Oxford: Oxford University Press. Garland, L. (1999) Byzantine Empresses Women and Power in Byzantium, AD 527–1204. London: Routledge. Gatier, P.-L. and Ulbert, T. (1991) ‘Eine Tursturzinschrift aus Resafa-Sergiupolis’, Dam­ aszener Mitteilungen 5, 169–82. Gautier, E. F. (1935) Genséric. Roi des Vandales. Paris: Payot. Gauthier, G. (1998) Justinien. Le rêve impérial. Paris: France Empire. Geary, P. J. (2015) ‘The Discourse of Herrschaft as the Practice of Herrschaft in the Fifth Century’, Medieval Worlds 1, 5–15. Gelzer, M. (1927) ‘Altertumswissenschaft und Spätantike’, Historische Zeitschrift 135, 173–87. Geyssen, J. (1998) ‘Presentations of Victory on the Theodosian Obelisk Base’, Byzantion 68, 47–55. Giardina, A. (1999) ‘Esplosione di tardoantico’, Studi storici 40, 157–80. Gillett, A. (2002) ed. On Barbarian Identity: Critical Approaches to Ethnicity in the Early Middle Ages. Turnhout: Brepols. Gillett, A. (2003) Envoys and Politicial Communication in the Late Antique West, 411–533. Cambridge: Cambridge University Press. Girgin, C. (2008) ‘La porte monumentale trouvée dans les fouilles près de l’ancienne prison de Sultanahmet’, Anatolia antiqua. Eski Anadolu 16, 259–90.

287

BIBLIOGRAPHY

Goffart, W. (1988) The Narrators of Barbarian History. Princeton: Princeton University Press. Gračanin, H. (2006) ‘The Huns and South Pannonia’, Byzantinoslavica 64, 29–76. Gray, P. (2005) ‘The Legacy of Chalcedon’, in ed. M. Maas, Cambridge Companion to the Age of Justinian. Cambridge: Cambridge University Press, 215–38. Greatrex, G. (1994) ‘The Dates of Procopius’ Works’, Byzantine and Modern Greek Stud­ ies 18, 101–4. Greatrex, G. (1996) ‘Flavius Hypatius, quem vidit validum Parthus sensitque timendum: An Investigation of His Career’, Byzantion 66, 120–42. Greatrex, G. (1997) ‘The Nika Riot: A Reassessment’, Journal of Hellenic Studies 117, 60–86. Greatrex, G. (1998) Rome and Persia at War. Leeds: Francis Cairns. Greatrex, G. (2000a) ‘The Background and Aftermath of the Partition of Armenia in A.D. 387’, Ancient History Bulletin 14, 35–48. Greatrex, G. (2000b) ‘Justin I and the Arians’, Studia Patristica 34, 72–81. Greatrex, G. (2003) ‘Recent Work on Procopius and the Composition of Wars VIII’, Byz­ antine and Modern Greek Studies 27, 45–67. Greatrex, G. (2010) ‘The Fall of Macedonius Reconsidered’, Studia Patristica 44, 125–32. Greatrex, G. and Bardill, J. (1996) ‘Antiochus the Praepositus, a Persian Eunuch at the Court of Theodosius II’, Dumbarton Oaks Papers, 171–97. Gregory, T. (1982) Vox Populi: Popular Opinion and Violence in the Religious Controver­ sies of the Fifth Century A.D. Columbus, OH: Ohio State University Press. Grierson, P. (1962) ‘The Tombs and Obits of the Byzantine Emperors (337–1042)’, Dumbarton Oaks Papers 16, 3–63. Grierson, P. and Mays, M. (1992) Catalogue of Late Roman Coins in the Dumbarton Oaks Collection and in the Whittemore Collection: From Arcadius and Honorius to the Acces­ sion of Anastasius. Washington, DC: Dumbarton Oaks. Grig, L. and Kelly, G. (2012) Two Romes: Rome and Constantinople in Late Antiquity. Oxford: Oxford University Press. Grillmeier, A. (1995) with T. Hainthaler, Christ in Christian Tradition, vol. 2, part 2. London: Mowbray. Grosse, R. (1920) Römische Militärgeschichte von Gallienus bis zum Beginn der byzan­ tinischen Themenverfassung. Berlin: Weidmann. Grossmann, P. (1989) ‘Beobachtungen zum ursprünglichen Grundriß der Sergios- und Bakchoskirche in Konstantinopel’, Istanbuler Mitteilungen 39, 153–59. Gruen, E. (2013) ‘Did Ancient Identity Depend on Ethnicity? A Preliminary Probe’, Phoe­ nix 67, 1–22. Guilland, R. (1951) ‘Études sur le palais du Boukoléon. Le palais d’Hormisdas’, Byzanti­ noslavica 12, 210–24, rp. in Guilland (1969) Etudes de topographie de Constantinople byzantine. Berlin: Adolf Hakkert, 294–303. Guilland, R. (1967a) ‘Le César’, in Recherches sur les institutions byzantines, vol. 2. Ber­ lin: Akademie-Verlag, 25–43. Guilland, R. (1967b) ‘Les patrices byzantins du VIe siècle’, in Recherches sur les institu­ tions byzantines, vol. 2. Berlin: Akademie-Verlag, 132–61. Guilland, R. (1969) ‘Le Stama ou Pi’, in ed. R. Guilland, Études de topographie de Con­ stantinople byzantine. Berlin: Adolf Hakkert, 451–7. Haacke, W. (1939) Die Glaubensformel des Papstes Hormisdas im Acacianischen Schisma. Rome: Aedes Universitatis Gregorianae.

288

BIBLIOGRAPHY

Haarer, F. (2006) Anastasius I: Politics and Empire in the Late Roman World. Cambridge: Francis Cairns. Hahn, G. (1984) ‘Die Münzprägung für Aelia Ariadne’, in eds. W. Hörandner et al., Byzan­ tios: Festschrift für Herbert Hunger. Vienna: E. Becvar, 101–6. Haldon, J. (1984) Byzantine Praetorians: An Administrative, Institutional and Social Sur­ vey of the Opsikion and Tagmata, c.580–900. Bonn: Habelt. Haldon, J. (1990) Constantine Porphyrogenitus Three Treatises on Military Expeditions, CFHB Series Vindobonensis, 28. Vienna: Österreichischen Akademie der Wissenschaften. Harries, J. (2013) ‘Men without Women: Theodosius’ Consistory and the Business of Gov­ ernment’, in ed. C. Kelly, Theodosius II: Rethinking the Roman Empire in Late Antiquity. Cambridge: Cambridge University Press, 67–89. Harrison, R. (1983) ‘The Church of St. Polyeuktos in Istanbul and the Temple of Solomon’, Harvard Ukrainian Studies 7, 276–9. Harrison, R. (1986) Excavations at Saraçhane in Istanbul, vol. 1. Princeton: Princeton University Press. Harrison, R. (1989) A Temple for Byzantium: The Discovery and Excavations of Anicia Juliana’s Palace-Church in Istanbul. London: H. Miller. Harvey, S. A. (1990) Asceticism and Society in Crisis. Berkeley, Los Angeles and London: University of California Press. Harvey, S. A. (2001) ‘Theodora the “Believing Queen”: A Study in Syriac Historiographi­ cal Tradition’, Hugoye: Journal of Syriac Studies 4, 1–31. Hatlie, P. (2007) The Monks and Monasteries of Constantinople ca.350–850. Cambridge: Cambridge University Press. Heather, P. (1991) Goths and Romans 332–489. Oxford: Oxford University Press. Heather, P. (1994) ‘New Men for New Constantines? Creating an Imperial Elite in the Eastern Mediterranean’, in ed. P. Magdalino, New Constantines: The Rhythm of Imperial Renewal in Byzantium, 4th–13th Centuries. Aldershot: Ashgate, 11–33. Heather, P. (1997) ‘Late Antquity and Early Medieval West’, in ed. M. Bentley, Companion to Historiography. London: Routledge, 69–87. Heather, P. (2005) The Fall of the Roman Empire: A New History. London: Pan Books. Heather, P. (2010) ‘Liar in Winter: Themistius and Theodosius’, in eds. S. McGill, C. Sogno and E. Watts, From the Tetrarchs to the Theodosians. Cambridge: Cambridge University Press, 185–214. Heather, P. and Moncur, D. (2001) Politics, Philosophy, and Empire in the Fourth Century: Select Orations of Themistius. Liverpool: Liverpool University Press. Hebblewhite, M. (2020) Theodosius and the Limits of Empire. London: Routledge. Herrin, J. (1991) ‘Byzance: le palais et la ville’, Byzantion 61, 213–30. Herrin, J. (2004) Women in Purple: Rulers of Medieval Byzantium. Princeton: Princeton University Press. Herrin, J. (2013) Unrivalled Influence: Women and Empire in Byzantium. Princeton: Princeton University Press. Heucke, C. (1997) ‘Die Herrschaft des oströmischen Kaisers Zenon: Ein Beispiel für Inte­ gration?’, in eds. M. Beer, M. Kintzinger and M. Krauss, Migration und Integration: Aufnahme und Eingliederung im historischen Wandel. Stuttgart: Steiner, 45–54. Hoffmann, D. (1969) Das Spätrömische Bewegungsheer und die Notitia Dignitatum. Düs­ seldorf: Rheinland-Verlag. Hohlfelder, R. L. (1984) ‘Marcian’s Gamble: A Reassessment of Eastern Imperial Policy toward Attila, AD 450–453’, American Journal of Ancient History 9, 54–69.

289

BIBLIOGRAPHY

Holmes, W. G. (1912) The Age of Justinian and Theodora, 2nd edn. London: G. Bell and sons. Holt, P. et al. (1977) The Cambridge History of Islam, vol. 1A. Cambridge: Cambridge University Press. Holum, K. G. (1982) Theodosian Empresses: Women and Imperial Dominion in Late Antiquity. Berkeley and Los Angeles: University of California Press. Holum, K. G. and Vikan, G. (1979) ‘The Trier Ivory, “Adventus” Ceremonial, and the Rel­ ics of St. Stephen’, Dumbarton Oaks Papers 33, 113–33. Honigmann, E. (1949/1950) ‘Le cubicularius Urbicius’, Revue des études byzantines 7, 47–50 and 212–3 (‘Notes additionelles’). Honigmann, E. (1951) Évêques et évêchés monophysites d’Asie antérieure au VIe siècle. Louvain: Durbecq. Honoré, T. (1975) ‘Some Constitutions Composed by Justinian’, Journal of Roman Studies 65, 107–23. Honoré, T. (1978) Tribonian. London: Duckworth. Humphries, M. (2012) ‘Valentinian III and the City of Rome (AD 425–55): Patronage, Politics, and Power’, in eds. L. Grig and G. Kelly, Two Romes: Rome and Constantinople in Late Antiquity. Oxford: Oxford University Press, 174–9. Hunger, H. (1978) Die hochsprachliche profane Literatur der Byzantiner, vol. 1. Munich: C. H. Beck. Hunt, T. (2018) ‘The Influence of French Colonial Humanism on the Study of Late Antiq­ uity: Braudel, Marrou, Brown’, International Journal of Francophone Studies 21, 255–78. Hurbanič, M. (2012) ‘A Topographical Note Concerning the Avar Siege of Constantinople the Question of the Localization of St. Callinicus Bridge’, Byzantinoslavica 70, 15–24. Icks, M. (2014) ‘The Inadequate Heirs of Theodosius. Ancestry, Merit and Divine Blessing in the Representation of Arcadius and Honorius’, Millennium 11, 69–100. Jacobs, I. (2012) ‘The Creation of the Late Antique City: Constantinople and Asia Minor during the “Theodosian Renaissance”’, Byzantion 82, 113–64. James, E. (2008) ‘The Rise and Function of the Concept “Late Antiquity”’, Journal of Late Antiquity 1, 20–30. James, L. (2001) Empresses and Power in Early Byzantium. Leicester: Leicester University Press. James, L. (2002) ‘Goddess, Whore, Wife or Slave? Will the Real Byzantine Empress Please Stand Up?’, in ed. A. Duggan, Queens and Queenship in Medieval Europe. London: Boydell and Brewer, 123–40. James, L. (2011/12) ‘Making a Name: Reputation and Imperial Founding and Refounding in Constantinople’, Wiener Jahrbuch für Kunstgeschichte 60/61, 63–72. Janin, R. (1964) Constantinople byzantine, 2nd edn. Paris: Institut Français d’Études Byz­ antines. Janin, R. (1966) ‘Les processions religieuses à Byzance’, Revue des études byzantines 24, 69–88. Janin, R. (1969) Les églises et les monastères, vol. 3 of La géographie ecclésiastique de l’Empire byzantin, part 1. Le siège de Constantinople et le patriarcat oecuménique, 2nd edn. Paris: Institut Français d’Études Byzantines. Jeffreys, E. (1990a) ‘Chronological Structures in Malalas’ Chronicle’, in ed. E. Jeffreys, with B. Croke and R. Scott, Studies in John Malalas. Sydney: Australian Association for Byzantine Studies, 138–43.

290

BIBLIOGRAPHY

Jeffreys, E. (1990b) ‘Malalas in Greek’, in eds. E. Jeffreys, with B. Croke and R. Scott, Studies in John Malalas. Sydney: Australian Association for Byzantine Studies, 245–68. Jobst, W. and Gurtner, C. (1997) ‘Mosaikenforschung 1995. Bericht uber die Arbeiten am Palastmosaik in Istanbul’, Anzeiger der phil.-hist. Klasse der Osterreichische Akademie der Wissenschaften 133, 11–27. Jobst, W. and Vetters, H. (1992) eds. Mosaikenforschungen in Kaiserpalast von Konstanti­ nopel. Vorbericht über das Forschungs- und Restaurierungsproject am Palastmosaik in den Jahren 1983–1988. Vienna: Österreichischen Akademie der Wissenschaften. Johnson, M. (1991a) ‘On the Burial Places of the Theodosian Dynasty’, Byzantion 61, 330–9. Johnson, M. (1991b) ‘On the Burial Places of the Valentinian Dynasty’, Historia 40, 501–6. Johnson, S. F. (2012) ‘Preface: On the Uniqueness of Late Antiquity’, in ed. S. F. Johnson, The Oxford Handbook of Late Antiquity. Oxford: Oxford University Press, i–xxix. Jones, A. H. M. (1964) The Later Roman Empire 284–602: A Social, Economic and Admin­ istrative Survey. Oxford: Basil Blackwell. Kaegi, W. (1981) Byzantine Military Unrest 471–843. Amsterdam: Hakkert. Kaibel, G. (1888) Epigrammata Graeca. Berlin: G. Reimer, rp. in (1965), Hildesheim: Olms. Kaldellis, A. (2004a) Procopius of Caesarea: Tyranny, History and Philosophy at the End of Antiquity. Philadephia, PA: University of Pennsylvania Press. Kaldellis, A. (2004b) ‘Identifying Dissident Circles in Sixth-Century Byzantium: The Friendship of Prokopios and Ioannes Lydos’, Florilegium 21, 1–17. Kaldellis, A. (2005) ‘The Works and Days of Hesychios the Illoustrios of Miletos’, Greek, Roman and Byzantine Studies 4, 381–403. Kaldellis, A. (2009) ‘The Date and Structure of Prokopios’ Secret History and His Pro­ jected Work on Church History’, Greek, Roman and Byzantine Studies 49, 585–616. Kaldellis, A. (2010) Prokopios the Secret History with Related Texts. Indianopolis: Hackett. Kaldellis, A. (2018) ‘Leo I, Ethnic Politics and the Beginning of Justin I’s Career’, Zbornik radova Vizantološkog instituta 55, 9–17. Kaldellis, A. (2019a) Romanland. Cambridge, MA: Belknap Press. Kaldellis, A. (2019b) Byzantium Unbound. Amsterdam: Arc Humanities Press. Karydis, N. (2020) ‘Justinian’s Church of the Holy Apostles’, in eds. M. Mullett and R. G. Ousterhout, The Holy Apostles: A Lost Monument, a Forgotten Project and the Present­ ness of the Past. Washington, DC: Dumbarton Oaks, 99–130. Kaster, R. (1988) Guardians of Language: The Grammarian and Society in Late Antiquity. Berkeley, Los Angeles and London: University of California Press. Kelly, C. (2008) Attila the Hun: Barbarian Terror and the Fall of the Roman Empire. London: The Bodley Head. Kelly, G. (2003) ‘The New Rome and the Old: Ammianus Marcellinus’ Silences on Con­ stantinople’, Classical Quarterly 53, 588–607. Kelly, J. N. D. (1975) Jerome. London: Duckworth. Kennedy, D. (1978) ‘Some Observations on the Praetorian Guard’, Ancient Society 9, 275–301. Kent, J. P. C. (1971) ‘The Coinage of Theodoric in the Names of Anastasius and Justin I’, in ed. R. A. G. Carson, Mints Dies and Currency: Essays Dedicated to the Memory of Albert Baldwin. London: Methuen, 67–74. Kent, J. P. C. (1991) ‘The Empress Ariadne: What’s in a Name?’, Nordisk umismatisk Arsskrift/Nordic Numismatic Journal 51, 35–44.

291

BIBLIOGRAPHY

Kent, J. P. C. (1993) ‘“Concordia” Solidi of Theodosius I: A Reappraisal’, Numismatic Chronicle 153, 77–90. Kent, J. P. C. (1994) Roman Imperial Coinage, vol. 10. London: Spink and Son. Keppie, L. J. F. (1996) ‘The Praetorian Guard before Sejanus’, Athenaeum 84, 101–24. Kiel-Freytag, A. (2010) ‘Betrachtungen zur Usurpation des Illus und des Leontius (484– 488 n.Chr.)’, Zeitschrift fur Epigraphik 74, 291–301. Kiilerich, B. (1998) The Obelisk Base in Constantinople: Court Art and Imperial Ideology. Rome: Bretschneider. Kiilerich, B. (2001) ‘The Image of Anicia Juliana in the Vienna Dioscurides: Flattery or Appropriation of Imperial Imagery’, Symbolae Osloenses 76, 169–90. Klein, H. A. (2006) ‘Sacred Relics and Imperial Ceremonies at the Great Palace of Con­ stantinople’, in ed. F. A. Bauer, Visualisierungen von Herrschaft. Istanbul: Deutsches Archäologisches Institut, 79–99. Kocabaş, U. (2015) ‘The Yenikapı Byzantine-Era Shipwrecks, Istanbul, Turkey: A Prelimi­ nary Report and Inventory of the 27 Wrecks Studied by Istanbul University’, Interna­ tional Journal of Nautical Archaeology 44, 5–38. Koch, P. (1903) Die Byzantinischen Beamtentitel von 400 bis 700. Jena: UniversitätsBuchdruckerei G. Neuenhahn. Koder, J. (1985) ‘Fresh Vegetables for the Capital’, in eds. C. Mango and G. Dagron, Con­ stantinople and Its Hinterland. London: Ashgate, 49–56. Kornemann, E. (1949) Weltgeschichte des Mittelmeer-Raumes, vol. 2. Von Augustus bis zum Sieg der Araber, chapters reprinted in Kornemann (1977) Geschichte der Spätan­ tike. Munich: C. H. Beck. Kos, S. (2008) ‘The Family of Romulus Augustulus’, in ed. P. Mauritsch et al., Antike Leb­ enswelten: Konstanz – Wandel – Wirkungsmacht: Festschrift für Ingomar Weiler zum 70. Geburtstag. Wiesbaden: Harrassowitz, 439–49. Kosinski, R. (2008) ‘Leo II: Some Chronological Questions’, Palamedes 3, 210–11. Kosinski, R. (2010) The Emperor Zeno: Religion and Politics. Cracow: Jagiellonian Uni­ versity. Kosinski, R. (2012) ‘Euphemios, Patriarch of Constantinople in the Years 490–496’, Jahr­ buch der österreichischen Byzantinistik 62, 66–9. Kostenec, J. (1998) ‘Studies on the Great Palace in Constantinople, 1: The Palace of Con­ stantine the Great’, Byzantinoslavica 59, 279–96. Kostenec, J. (2005) ‘Observations on the Great Palace at Constantinople: The Sanctuaries of the Archangel Michael, the Daphne Palace and the Magnaura’, Reading Medieval Studies 31, 27–55. Kraus, T. J. (2000) ‘(Il)literacy in Non-Literary Papyri from Graeco-Roman Egypt: Fur­ ther Aspects of the Educational Ideal in Ancient Literary Sources and Modern Times’, Mnemosyne 53, 322–42. Krautheimer, R. (1974) ‘Again Saints Sergius and Bacchus at Constantinople’, Jahrbuch der österreichischen Byzantinistik 23, 251–3. Krautheimer, R. (1986) Early Christian and Byzantine Architecture, 4th edn, rev. R. Kraut­ heimer and S. Ćurčić. Harmondsworth: Penguin. Krautschick, S. (1986) ‘Zwei Aspekte des Jahres 476’, Historia 35, 344–71. Krautschick, S. (1995) ‘Die unmögliche Tatsache. Argumente gegen Johannes Antioche­ nus’, Klio 77, 332–8. Krusch, B. (1885) ed. Gregorii Episcopi Turonensis Liber in Gloria Martyrum. Hannover: Hahn.

292

BIBLIOGRAPHY

Kulikowski, M. (2000) ‘The Notitia Dignitatum as a Historical Source’, Historia 49, 358–77. Laiou, A. E. (1999) ‘Women in Byzantine Society’, in ed. L. Mitchell, Women in Medieval Western European Culture. New York: Garland Publishing, 161–7. Lammert, F. (1956) ‘Mattiarii’, RE 14, 2322–3. Lane Fox, R. (1997) ‘The Life of Daniel’, in eds. M. Edwards and S. Swain, Portraits: Biographical Representation in the Greek and Latin Literature of the Roman Empire. Oxford: Oxford University Press, 175–225. Laniado, A. (1991) ‘Some Problems in Sources for the Reign of the Emperor Zeno’, Byz­ antine and Modern Greek Studies 15, 153–4. Laniado, A. (2015a) Ethnos et droit dans le monde protobyzantin, V-Ve siècle. Geneva: Droz. Laniado, A. (2015b) ‘Aspar and His Phoideratoi: John Malalas on a Special Relationship’, in eds. U. Roberto and L. Mecella, Governare e riformare l’impero al momento della sua divisione. Oriente, Occidente, Illirico. Rome: l’Ecole française de Rome. Leader-Newby, R. (2004) Silver and Society in Late Antiquity: Functions and Meanings of Silver Plate in the Fourth to Seventh Centuries. Burlington, VT: Ashgate. Lee, D. (2000) ‘The Eastern Empire: Theodosius to Anastasius’, in eds. A. Cameron, B. Ward-Perkyns and M. Whitby, Cambridge Ancient History, vol. 14. Cambridge: Cam­ bridge University Press, 33–62. Lee, D. (2012) From Rome to Byzantium, AD 363 to 565. Edinburgh: Edinburgh University Press. Lee, D. (2013) ‘Theodosius and His Generals’, in ed. C. Kelly, Theodosius II: Rethink­ ing the Roman Empire in Late Antiquity. Cambridge: Cambridge University Press, 90–108. Lenski, N. (1999) ‘Assimilation and Revolt in the Territory of Isauria, from the 1st Century BC to the 6th Century AD’, Journal of the Economic and Social History of the Orient 42, 413–65. Lenski, N. (2000) ‘The Election of Jovian and the Role of the Late Imperial Guard’, Klio 82, 492–515. Lenski, N. (2002) Failure of Empire: Valens and the Roman State in the 4th Century AD. Berkeley, Los Angeles and London: University of California Press. Leppin, H. (1996) Von Constantin dem Grossen zu Theodosius II: Das christliche Kaiser­ tum bei den Kirchenhistoriken Socrates, Sozomenus und Theodoret. Göttingen: Vanden­ hoeck and Ruprecht. Leppin, H. (2002) ‘Theodora und Justinian’, in eds. H. Temporini and G. Vitzthum, Die Kaiserinnen Roms. Von Livia bis Theodora. Munich: C. H. Beck, 437–81. Leppin, H. (2003) Theodosius der Grosse. Gestalten der Antike. Darmstadt: Primus. Leppin, H. (2011) Justinian. Eine christliche Experiment. Stuttgart: Klett-Cotta. Leszka, M. J. (1998) ‘Empress-Widow Verina’s Political Activity during the Reign of Emperor Zeno’, in ed. W. Ceran, Mélanges d’histoire Byzantine offerts à Octawiusz Jurewicz à l’occasion de son soixante-dixième anniversaire. Lodz: Wydawn. Uniwer­ sytetu Łódzkiego, 128–36. Liebeschuetz, W. (2001) ‘Late Antiquity and the Concept of Decline’, Nottingham Medi­ eval Studies 14, 1–11. Liebeschuetz, W. (2004) ‘The Birth of Late Antiquity’, Antiquité Tardive 12, 253–61. Liebeschuetz, W. (2008) ‘Afterword: A. H. M. Jones and the Later Roman Empire’ in ed. D. Gwynn, A. H. M. Jones and the Later Roman Empire. Leiden: Brill, 251–66.

293

BIBLIOGRAPHY

Lilie, R.-J. (1995) ‘Die Krönung des Kaisers Anastasios I (491)’, Byzantinoslavica 56, 3–12. Lilie, R.-J. (1998) ‘Die krönungsprotokolle des Zeremonienbuchs und die Krönung Kaiser Leons I’, in ed. C.-F. Collatz, Dissertatiunculae criticae. Festschrift für Günther Chris­ tian Hansen. Würzburg: Königshausen u. Neumann, 395–408. Lippold, A. (1972) ‘Zeno 17’, RE Suppl. 10A, 149–203. Lippold, A. (1980) Theodosius der Grosse und seine Zeit. Munich: C. H. Beck. Lizzi Testa, R. and Marconi, G. (2019) The Collectio Avellana and Its Rervivals. Edin­ burgh: Cambridge Scholars Publishing. Lucchelli, T. M. and Rohr Vio, F. (2012) ‘Augustae, le donne dei principi. Riflessioni su Augustae’, Athenaeum 100, 496–511. Maas, M. (1992) John Lydus and the Roman Past. London: Routledge. MacCormack, S. (1981) Art and Ceremony in Late Antiquity. Berkeley, Los Angeles and London: University of California Press. MacGeorge, P. (2002) Late Roman Warlords. Oxford: Oxford University Press. MacMaster, T. (2017) ‘Afterword: Why the Seventh Century? The Problems of Periodiza­ tion acros Cultures?’, in eds. A. Gnasso, E. Intaglia, T. MacMaster and B. Morris, The Long Seventh Century. New York: Peter Lang, 291–8. Macrides, R. and Magdalino, P. (1988) ‘The Architecture of Ekphrasis: Construction and Context of Paul the Silentiary’s Poem on Hagia Sophia’, Byzantine and Modern Greek Studies 12, 47–82. Maenchen-Helfen, O. J. (1973) The World of the Huns. Berkeley, Los Angeles and London: University of California Press. Magdalino, P. (2000) ‘The Maritime Neighbourhoods of Constantinople: Commercial and Residential Functions, Sixth to Twelfth Centuries’, Dumbarton Oaks Papers 54, 209–26. Magdalino, P. (2001) ‘Aristocratic Oikoi in the Tenth and Eleventh Districts of Constan­ tinople’, in ed. N. Necipoglu, Byzantine Constantinople: Monuments, Topography and Everyday Life. Leiden-Boston-Köln: Brill, 53–69. Magdalino, P. (2002) ‘Medieval Constantinople: Built Environment and Urban Develop­ ment’, in ed. A. Laiou, The Economic History of Byzantium: From the Seventh through the Fifteenth Century. Washington, DC: Dumbarton Oaks, 529–37. Magdalino, P. (2007) ‘Medieval Constantinople’, in Studies on the History and Topography of Byzantine Constantinople. Aldershot and Burlington, VT: Ashgate, 1–111. Magliaro, L. A. (2013) Arriana. La garante della porpora. Milan: Jaca. Mainstone, R. (1988) Hagia Sophia: Architecture, Structure and Liturgy of Justinian’s Great Church. London: Thames and Hudson. Maisano, R. (1995) Temistio. Discorsi. Turin: UTET. Manders, E. and Slootjes, D. (2020) Leadership, Ideology and Crowds in the Roman Empire of the Fourth Century AD. Stuttgart: Steiner. Mango, C. (1959) The Brazen House. Copenhagen: Munksgaard. Mango, C. (1972) ‘The Church of Saints Sergius and Bacchus at Constantinople and the Alleged Tradition of Octagonal Palatine Churches’, Jahrbuch der österreichischen Byz­ antinistik 21, 189–93, rp. in C. Mango, Studies on Constantinople. Aldershot: Ashgate. Mango, C. (1975) ‘The Church of Sts. Sergius and Bacchus Once again’, Byzantinische Zeitschrift 68, 385–92, rp.in C. Mango, Studies on Constantinople. Aldershot: Ashgate. Mango, C. (1985) Byzantine Architecture. New York: Rizzoli. Mango, C. (1986a) The Art of the Byzantine Empire 312–1453. Toronto: University of Toronto Press.

294

BIBLIOGRAPHY

Mango, C. (1986b) ‘The Development of Constantinople as an Urban Centre’, in The 17th International Byzantine Congress: Main Papers. New Rochelle, NY: A.D. Caratzas, 117–36, rp. in C. Mango, Studies on Constantinople. Aldershot: Ashgate. Mango, C. (1991) ‘The Palace of Marina, the Poet Palladas and the Bath of Leo the Wise’, in Euphrosynon: Aphieroma ston Manole Chatzidake. Athens: Tameío Archaiologikṓn Pórōn kai Apallotriṓseōn, 321–30. Mango, C. (1993a) ‘Constantine’s Column’, in ed. C. Mango, Studies on Constantinople. Aldershot: Ashgate. Mango, C. (1993b) ‘The Columns of Justinian and His Successors’, in ed. C. Mango, Stud­ ies on Constantinople. Aldershot: Ashgate. Mango, C. (1995) ‘The Water Supply of Constantinople’, in eds. C. Mango and G. Dagron, Constantinople and Its Hinterland. London: Ashgate, 9–18. Mango, C. (1996) ‘La banlieue de Constantinople à l’époque byzantine’, Mélanges de l’École française de Rome 108, 363–5. Mango, C. (1997) ‘The Palace of the Boukoleon’, Cahiers Archéologiques 45, 41–50. Mango, C. (2000) ‘The Triumphal Way of Constantinople and the Golden Gate’, Dumbar­ ton Oaks Papers 54, 173–88. Mango, C. (2004) Le développement urbain de Constantinople, rev. edn. Paris: De Boc­ card. Mango, C. and Ševčenko, I. (1961) ‘Remains of the Church of St. Polyeuktos at Constan­ tinople’, Dumbarton Oaks Papers 15, 243–7. Mango, M. (2000) ‘The Commercial Map of Constantinople’, Dumbarton Oaks Papers 54, 173–88. Mango, M. (2001) ‘The Porticoed Street at Constantinople’, in ed. N. Necipoglu, Byzan­ tine Constantinople: Monuments, Topography and Everyday Life. Leiden-Boston-Köln: Brill, 29–51. Maraval, P. (1985) Lieux, saints et pèlerinages d’Orient. Paris. Maraval, P. (1998) ‘La politique religieuse de Justinien’, in ed. L. Pietri, Les Églises d’Orient et d’Occident, Histoire du Christianisme, vol. 3. Paris, 399–409. Maraval, P. (1999) L’empereur Justinien. Paris: Les Éditions du cerf. Marcone, A. (2000) ‘La tarda antichità e le sue periodizzazioni’, Rivista storica italiana 112, 318–34. Marcone, A. (2003) ‘Un treno per Ravenna: Il contributo di Arnaldo Momigilano agli studi tardoantichi’, Antiquité Tardive 10, 291–8. Marcone, A. (2008) ‘A Long Late Antiquity? Considerations on a Controversial Periodiza­ tion’, Journal of Late Antiquity 1, 4–19. Marcone, A. (2020) ‘Late Antiquity: Then and Now’, in ed. H. T. Lopes, Antiquity and Its Reception: Modern Expressions of the Past. London: InTechOpen, 1–13. Markus, R. A. (2009) ‘Between Marrou and Brown: Transformations of Late Antiquity’, in eds. P. Rousseau and M. Papoutsakis, Transformations of Late Antiquity: Essays for Peter Brown. Farnham and Burlington, VT: Ashgate, 1–13. Marrou, H. (1949) St Augustin et la fin de la culture antique. “Retractatio”. Paris: E. de Boccard. Martin, D. B. and Miller, P. C. (2005) The Cultural Turn in Late Ancient Studies. Durham, NC: Duke University Press. Mastrolorenzo, G., Palladino, D. M., Vecchio, G. and Taddeucci, J. (2002) ‘The 472 AD Pollena Eruption of Somma-Vesuvius (Italy) and Its Environmental Impact at the End of the Roman Empire’, Journal of Volcanology and Geothermal Research 113, 19–36.

295

BIBLIOGRAPHY

Mateos, J. (1962–3). Le Typicon de la Grande Église, 2 vols. Rome: Pont. Institutum Ori­ entalium Studiorum. Mathews, T. F. (1971) The Early Churches of Constantinople. University Park, PA: The Pennsylvania State University Press. Mathews, T. F. (1974) ‘Architecture et liturgie dans les premières églises palatiales de Constantinople’, Revue de l’art 24, 22–9. Mathews, T. F. (1976) The Byzantine Churches of Istanbul: A Photographic Survey. Uni­ versity Park, PA and London: The Pennsylvania State University Press. Mathisen, R. (1986) ‘Patricians as Diplomats in Late Antiquity: Extraordinary Senatorial Status in the Late Fifth Century’, Byzantinische Zeitschrift 79, 35–49, rp. in R. Mathisen (1991) Studies in the History, Literature and Society of Late Antiquity. Amsterdam. Matthews, J. F. (1975) Western Aristocracies and Imperial Court A.D. 364–425. Oxford: Oxford University Press. Matthews, J. F. (1989) The Roman Empire of Ammianus. London: Duckworth. Matthews, J. F. (2012) ‘The Notitia Urbis Constantinopolitanae’, in eds. L. Grig and G. Kelly, Two Romes: Rome and Constantinople in Late Antiquity. Oxford: Oxford Uni­ versity Press, 81–115. McClanan, A. (1996) ‘The Empress Theodora and the Tradition of Women’s Patronage in the Early Byzantine Empire’, in ed. J. H. McCash, The Cultural Patronage of Medieval Women. Athens, GA: University of Georgia Press, 50–72. McClanan, A. (2002) Representations of Early Byzantine Empresses: Image and Empire. London: Palgrave. McCormick, M. (1986) Eternal Victory. Cambridge: Cambridge University Press. McCormick, M. (1997) ‘Emperors’, in ed. G. Cavallo, The Byzantines. Chicago, IL: Uni­ versity of Chicago Press, 230–54. McCormick, M. (2001) ‘The Emperor and Court’, in eds. A. Cameron, B. Ward-Perkins and M. Whitby, Cambridge Ancient History, vol. 14. Cambridge: Cambridge University Press, 135–63. McEvoy, M. (2010) ‘Rome and the Transformation of the Imperial Office in the Late Fourth–Mid-Fifth Centuries AD’, Papers of the British School at Rome 78, 151–92. McEvoy, M. (2013) Child Emperor Rule in the Late Roman West AD 367–455. Oxford: Oxford University Press. McEvoy, M. (2014) ‘Between the Old Rome and the New’, in eds. D. Dzinjo and K. Parry, Byzantium, Its Neighbours and Its Cultures. Brisbane: Australian Association of Byzan­ tine Studies, 245–67. McEvoy, M. (2016) ‘Becoming Roman? The Not So Curious Case of the Ardaburii and Aspar’, Journal of Late Antiquity 9, 483–511. McEvoy, M. (2019) ‘Celibacy and Survival in Court Politics in the Fifth Century AD’, in ed. S. Tougher, The Emperor in the Byzantine World. London: Routledge, 115–34. McGill, S., Sogno, C. and Watts, E. (2010) eds. From the Tetrarchs to the Theodosians. Cambridge: Cambridge University Press. McLynn, N. (2010) ‘Moments of Truth: Gregory of Nazianzus and Theodosius I’, in eds. S. McGill, C. Sogno, and E. Watts, From the Tetrarchs to the Theodosians. Cambridge: Cambridge University Press, 215–40. Meier, M. (2003a) ‘Die Inszenierung einer Katastrophe: Justinian und der Nika-Aufstand’, Zeitschrift für Papyrolgie und Epigraphik 142, 273–300. Meier, M. (2003b) Das andere Zeitalter Justinians. Kontingenzerfahrung und Kontingenz­ bewältigung im 6. Jahrhundert n.Chr. Göttingen: Vandenhoeck & Ruprecht.

296

BIBLIOGRAPHY

Meier, M. (2004) Justinian: Herrschaft, Reich und Religion. Munich: C. H. Beck. Meier, M. (2009) Anastasios I: Die Entstehung des Byzantinischen Reiches. Stuttgart: Klett-Cotta. Meier, M. (2010) ‘Ariadne: der “Rote Faden” des Kaisertums’, in ed. A. Kolb, Augustae: Machtbewusste Frauen am römischen Kaiserhof? Herrschaftsstrukturen und Herrschafts­ praxis II. Berlin: de Gruyter, 277–82. Meier, M. (2017) ‘A Contest of Interpretation: Roman Policy toward the Huns as Reflected in the “Honoria Affair” (448/50)’, Journal of Late Antiquity 10, 42–61. Meischner, J. (1996) ‘Das Missorium des Theodosius in Madrid’, Jahrbuch des Deutschen Archäologischen Instituts 111, 389–432. Menze, V.-L. (2008) The Making of a Church: The Syrian Orthodox in the Shadow of Byz­ antium and the Papacy. Oxford: Oxford University Press. Mercati, S. G. (1925) ‘Epigraphica, II: Sulla tradizione manoscritta dell’inscrizione del fre­ gio dei santi Sergio e Bacco di Costantinopoli’, Atti della Pontificia Accademia Romana di Archeologia, ser. 3, 197–205. Mergiali-Sahas, S. (2001) ‘Byzantine Emperors and Holy Relics’, Jahrbuch der öster­ reichischen Byzantinistik 51, 41–60. Metcalf, W. E. (1988) ‘Joint Reign Gold of Justin I and Justinian I’, in eds. W. Hahn and W. E. Metcalf, Studies in Early Byzantine Gold Coinage. New York: American Numismatic Society, 19–27. Millar, F. (1977) The Emperor in the Roman World. London: Duckworth. Millar, F. (2006) A Greek Roman Empire: Power and Belief under Theodosius II, 408–450. Berkeley, Los Angeles and London: University of California Press. Miller, D. (1970) ‘The Emperor and the Stylite: A Note on the Imperial Office’, Greek Orthodox Theological Review 15, 207–12. Miller, D. and Sarris, P. (2018) The Novels of Justinian, 2 vols. Cambridge: Cambridge University Press. Milner, C. (1994) ‘The Image of the Rightful Ruler: Anicia Juliana’s Constantine Mosaic in the Church of Hagios Polyeuktos’, in ed. P. Magdalino, New Constantines: The Rhythm of Imperial Renewal in Byzantium, 4th–13th Centuries. Aldershot: Ashgate, 73–81. Miranda, S. (1983) ‘Étude sur le Palais sacré de Constantinople. Le Walker Trust et le palais de Daphne’, Byzantinoslavica 44, 41–9 and 196–204. Mitchell, S. (2014) A History of the Later Roman Empire, AD 284–641, 2nd edn. London: Blackwell Publishing. Moffatt, A. (1996) ‘Celebrating a Byzantine Imperial Birthday’, in eds. C. N. Constan­ tinides, N. M. Panagiotakes, E. M. Jeffreys and A. D. Angelou, Philhellen: Studies in Honour of Robert Browning. Venice: Istituto Ellenico di Studi Bizantini e Postbizantini di Venezia, 255–66. Moffatt, A. and Tall, M. (2012) trans. Constantine Porphyrogennetos: The Book of Ceremo­ nies, 2 vols. Canberra: Australian Association for Byzantine Studies. Momigliano, A. (1956) ‘Gli Anicii e la storiografia latina del VI sec. d.C’, Accademia Nazionale dei lincei. Rendiconti di Scienze morali, storiche e filologiche 11, 279–97, rp. in Secondo Contributo alla storia degli studi classici (Rome 1960), 231–54. Mommaerts, T. S. and Kelley, D. H. (1992) ‘The Anicii of Gaul and Rome’, in eds. J. Drinkwater and H. Elton, Fifth-Century Gaul: A Crisis of Identity? Cambridge: Cambridge University Press, 111–21. Mommsen, T. (1889) ‘Das römische Militärwesen seit Diocletian’, Hermes 24, 195–279. Moorhead, J. (1992) Theodoric in Italy. Oxford: Oxford University Press.

297

BIBLIOGRAPHY

Moorhead, J. (1994) Justinian. London: Longman. Morrisson, C. (2002) ‘The Sixth Century Economy’, in ed. A. Laiou, The Economic His­ tory of Byzantium: From the Seventh through the Fifteenth Century. Washington, DC: Dumbarton Oaks, 171–220. Müller-Wiener, W. (1977) Bildlexicon zur Topographie Istanbuls. Tübingen: Wasmuth. Nau, F. (1897) ‘Analyse de la seconde partie inédite de l’Histoire Ecclésiastique de Jean d’Asie, patriarche jacobite de Constantinople’, Revue de l’Orient chrétien 2, 422–93. Naumann, R. (1976) ‘Neue Beobachtungen am Theodosiusbogen und Forum Tauri in Istanbul’, Istanbuler Mitteilungen 26, 117–41. Noethlichs, K. (1999) ‘Iustinianus’, Reallexikon für Antike und Christentum 19, 668–763. Norman, A. F. (1953) ‘An Identification in Suidas’, Classical Quarterly n.s. 3, 171–2. O’Donnell, J. J. (2004) ‘Late Antiquity: Before and After’, Transactions of the American Philological Association 134, 203–213. O’Flynn, J. M. (1991) ‘A Greek on the Roman Throne: The Fate of Anthemius’, Historia 41, 122–8. Olovsdotter, C. (2011) ‘Representing Consulship’, Opuscula 4, 99–123. Oost, S. I. (1964) ‘Aetius and Majorian’, Classical Philology 59, 27–8. Oost, S. I. (1968) Galla Placidia Augusta: A Biographical Essay. Chicago, IL: University of Chicago Press. Ostrogorsky, G. (1969) History of the Byzantine State, rev. edn. New Brunswick: Rutgers University Press. Ousterhout, R. (2018) ‘Aesthetics and Politics in the Architecture of Justinian’, Travaux et Mémoires 22, 185–207. Ousterhout, R. (2019) Eastern Medieval Architecture: The Building Traditions of Byzan­ tium and Neighbouring Lands. Oxford: Oxford University Press. Painter, K. (1991) ‘The Silver Dish of Ardabur Aspar’, in eds. E. Herring, R. Whitehouse and J. Wilkins, Papers of the Fourth Conference of Italian Archaeology, vol. 2. London: Accordia Research Institute, 73–9. Palme, B. (1998) ‘Flavius Epiphanius, comes domesticorum’, Eirene 34, 98–116. Palmer, A. and Rodley, L. (1988) ‘The Inauguration Anthem of Hagia Sophia in Edessa: A New Critical Edition and Translation with Historical and Architectural Notes and Com­ parison with a Contemporary Constantinopolitan kontakion’, Byzantine and Modern Greek Studies 12, 117–67. Pasquini Vecchi, L. (1999) ‘La scultura di S. Polieucto: episodio saliente nel quadro della cultura artistica di Costantinopoli’, Bizantinistica 1, 109–44. Patlagean, E. (1977) Pauvreté économique et pauvreté sociale à Byzance, 4e–7e siècles. Paris: Mouton. Pazdernik, C. (1994) ‘Our Most Pious Consort Given Us by God: Dissident Reactions to the Partnership of Justinian and Theodora, A.D. 525–548’, Classical Antiquity 13, 271–81. Pazdernik, C. (2015) ‘The Quaestor Proclus’, Greek Roman and Byzantine Studies 55, 241–59. Pentcheva, B. (2017) Hagia Sophia. University Park, PA: The Pennsylvania State Univer­ sity Press. Peschlow, U. (1992) ‘Betrachtungen zur Gotensäule in Istanbul’, in Tesserae, Festschrift für Josef Engemann. Münster: Aschendorff, 215–28. Pfeilschifter, R. (2013) Der Kaiser und Konstantinopel. Berlin: de Gruyter. Pieler, P. (1972) ‘L’aspect politique et juridique de l’action de Chosroès propose par les Perses à Justin’, Revue internationale des droits de l’antiquité 19, 399–432.

298

BIBLIOGRAPHY

Pigonski, L. (2018) ‘Berichus and the Evidence for Aspar’s Political Power and Aims in the Last Years of Theodosius II’s Reign’, Studia Ceranea 8, 237–51. Pigulewskaja, N. (1969) Byzanz auf den Wegen nach Indien. Berlin: Akademie-Verlag. Pingree, D. (1976) ‘Political Horoscopes from the Reign of Zeno’, Dumbarton Oaks Papers 30, 133–50. Pizzone, A. M. V. (2003) ‘Da Melitene a Constantinopoli: S. Polieucto nella politica dinastica di Giuliana Anicia. Alcune osservazioni in margine ad AP 1.10’, Maia 55, 107–32. Placanica, A. (1997) Vittore da Tunnuna Chronica: Chiesa e impero nell’età di Giustiniano. Florence: Sismel. Pohl, W. (2014) ‘Goths and Huns’, in ed. J. McInerney, A Companion to Ethnicity in the Ancient Mediterranean. London: Wiley, 555–68. Poidebard, A. and Mouterde, R. (1949) ‘A propos de Saint Serge. Aviation et Epigraphie’, Analecta Bollandiana 67, 113–4. Preger, T. (1901) Scriptores originum Constantinopolitanarum, 2 vols. Leipzig: Teubner. Pulak, C., Ingram, R. and Jones, M. (2015) ‘Eight Byzantine Shipwrecks from the Theo­ dosian Harbour Excavations at Yenikapı in Istanbul, Turkey: An Introduction’, Interna­ tional Journal of Nautical Archaeology 44, 39–73. Rapp, C. (2002) ‘A Medieval Cosmopolis: Constantinople and Its Foreign Inhabitants’, in eds. J. Ma and N. Van Deusen, Alexander’s Revenge: Hellenistic Culture through the Centuries. Reykjavik: University of Iceland Press, 153–72. Rapp, C. (2005) Holy Bishops in Late Antiquity: The Nature of Christian Leadership in an Age of Transition. Berkeley, Los Angeles and London: University of California Press. Ravegnani, E. (2006) Consoli e dittici consolari nella tarda antichità. Rome: Aracne. Ravegnani, G. (1989) La corte di Giustiniano. Rome: Jouvence. Rebenich, S. (1985) ‘Gratian, a Son of Theodosius, and the Birth of Galla Placidia’, His­ toria 34, 372–85. Rebenich, S. (1991) ‘Zum Theodosiusobolisken in Konstantinopel’, Istanbuler Mitteilun­ gen 41, 455–9. Rebenich, S. (2009) ‘Late Antiquity in Modern Eyes’, in ed. P. Rousseau, A Companion to Late Antiquity. Malden, MA: Wiley, 77–92. Redies, M. (1997) ‘Die Usurpation des Basiliskos (475–6) im Kontext der aufstiegenden monophysitischen Kirche’, Antiquité Tardive 5, 211–21. Ritzerfeld, U. (2001) ‘“Omnia Theodosio cedunt subolique perenni”: Überlegungen zu Bildprogramm und Bedeutung des Theodosiusobelisken in Konstantinopel’, Jahrbuch fur Antike und Christentum 44, 168–84. Roberto, U. (2000) ‘Sullla tradizione storiografica di Candido Isaurico’, Mediterraneo Antico 3, 685–727. Robinson, C. (2010) ed. The New Cambridge History of Islam, vol 1. Cambridge: Cam­ bridge University Press. Roques, D. (2011) Procope de Cesarée Constructions de Justinen 1er. Alessandria: Edizioni dell’Orso. Rösch, G. (1978) Onoma Basileias: Studien zum offiziellen Gebrauch der Kaisertitel in spätantiker und frühbyzantinischer Zeit. Vienna: Österreichischen Akademie der Wissenschaften. Rose, M. and Aydingun, S. (2007) ‘Under Istanbul (Ancient Byzantine Port at Constanti­ nople’s Ancient Harbor)’, Archaeology 60, 34–40. Rosen, K. (1999) ‘Iustinus I’, Reallexicon für Antike und Christentum 19, 763–78.

299

BIBLIOGRAPHY

Rosi, M. and Santacroce, R. (1983) ‘The A.D. 472 “Pollena” Eruption: Volcanological and Petrological Data for This Poorly-Known, Plinian-Type Event at Vesuvius’, Journal of Volcanology and Geothermal Research, 17, 249–71. Ross, M. C., Zwirn, R., and Boyd, S. A. (2005) Catalogue of the Byzantine and Early Medieval Antiquities in the Dumbarton Oaks Collection: Jewelry, Enamels and Art of the Migration Period, vol. 2 Dumbarton Oaks: Washington, DC. Rubery, E. (2013) ‘The Vienna “Empress” Ivory and its Companion in Florence: Crowned in Different Glories?’, in eds. A. Eastmond and L. James, Wonderful Things. Byzantium through its Art. London: Ashgate, 99–114. Rubin, B. (1960) Das Zeitalter Justinians, vol. 1. Berlin: de Gruyter. Sablayrolles, R. (1996) Libertinus Miles.les cohortes de Vigiles. Paris: École française de Rome. Safran, L. (1993) ‘Points of View: The Theodosian Obelisk Base in Context’, Greek, Roman and Byzantine Studies 34, 409–35. Şahin, S. (1991) ‘Inschriften aus Seleukeia am Kalyadnos’, Epigraphica Anatolica 17, 155–63. Salamon, M. (1994) ‘Basiliscus cum Romanis Suis’, in eds. L. Mrozewicz and K. Ilski, Studia Moesiaca. Poznán: Zakład Historii Starożytnej UAM, 179–96. Salzman, M. (2002) The Making of a Christian Aristocracy: Social and Religious Change in the Western Roman Empire. Cambridge, MA: Harvard University Press. Salzman, M. (2017) ‘Emperors and Elites in Rome after the Vandal Sack of 455’, Antiquité. Tardive 26, 243–62. Sande, S. (1981) ‘Some New Fragments from the Column of Theodosius’, Acta ad artium historiam pertinantia 1, 1–78. Sanpaolesi, P. (1961) ‘La chiesa dei SS Sergio e Bacco a Constantinopoli’, Rivista dell’istituto nazionale d’archeologia e storia dell’arte 10, 116–80. Saradi, H. (1995) ‘Constantinople and Its Saints’, Studi Medievali 36, 87–110. Sarantis, A. (2019) ‘Justinian’s Novella 11: Memory and Political Propaganda in the Build Up to the Gothic War’, Early Medieval Europe 27, 494–520. Scarcella, A. (1997) La Legislazione di Leone I. Milan: A. Giuffrè. Scharf, R. (1993) ‘Der Kelch des Ardabur und der Anthusa’, Byzantion 63, 213–23. Scharf, R. (2001) Foederati. Von der völkerrechtlichen Kategorie zur byzantinischen Trup­ pengattung. Vienna: Holzhausen. Schibelle, N. and McKenzie, J. (2014) ‘Glass Tesserae from Hagios Polyeuktos, Constan­ tinople: Their Early Byzantine Affiliations’, in eds. D. Keller et al., Neighbours and Successors of Rome: Traditions of Glass Production and Use in Europe and the Middle East in the Later 1st Milennium AD. Oxford: Oxbow Books, 114–27. Schwartz, E. (1934) Publizistische Sammlungen zum Acacianischen Schisma. Munich: C. H. Beck. Scott, L. (1976) ‘Aspar and the Burden of Barbarian Heritage’, Byzantine Studies 3, 59–69. Scott, R. (1987) ‘Justinian’s Coinage and Easter Reforms and the Date of the Secret His­ tory’, Byzantine and Modern Greek Studies 11, 215–22. Seeck, O. (1876) Notitia Dignitatum accedunt Notitia urbis Constantinopolitanae et later­ culi prouinciarum. Berlin: Weidmann. Seeck, O. (1894) ‘Anagastes’, RE 1.2, 2022. Seeck, O. (1896) ‘Ardabur 2’, RE 2.1, 607–10. Seeck, O. (1896a) ‘Ardabur 1’, RE 6.1, 606–7. Seeck, O. (1907) ‘Eusebius 23’, RE 6.1, 1370.

300

BIBLIOGRAPHY

Seeck, O. (1913) ‘Hilarianus 5’, RE 7.2, 1599. Seeck, O. (1920) Geschichte des Untergangs der antiken Welt, vol. 6. Stuttgart: J. B. Metzler. Sguaitamatti, L. (2012) Der spätantike Konsulat. Fribourg: Academic Press. Shahid, I. (1989) Byzantium and the Arabs in the Fifth Century. Washington, DC: Dumbar­ ton Oaks. Shahid, I. (1995) Byzantium and the Arabs in the Sixth Century, vol. 1. Washington, DC: Dumbarton Oaks. Shahid, I. (2003) ‘The Church of Sts Sergios and Bakhos at Constantinople, Some New Perspectives’, in ed. A. Avramea et al., Byzantium State and Society: In Memory of Nikos Oikonomides. Athens: Institute of Byzantine Research, 475–6. Shahid, I. (2004) ‘The Church of Hagios Polyeuktos in Constantinople: Some Observa­ tions’, in ed. G. K. Livadas, Graeco-Arabica: Festschrift in Honour of V. Christides. Athens, 343–55. Shaw, B. D. (1990) ‘Bandit Highlands and Lowland Peace: The Mountains of IsauriaCilicia’, Journal of the Economic and Social History of the Orient 33, 237–70. Shoemaker, S. J. (2008) ‘The Cult of Fashion: The Earliest Life of the Virgin and Constan­ tinople’s Marian Relics’, Dumbarton Oaks Papers 62, 53–74. Siebigs, G. (2010) Kaiser Leo I. Das oströmische Reich in den ersten drei Jahren seiner Regierung (457–460 n. Chr.), vol. 1. Berlin and New York: De Gruyter. Signes Codoñer, J. (2003) ‘Prokops Anekdota und Justinians Nachfolge’, Jahrbuch der österreichischen Byzantinistik 53, 47–82. Sivan, H. (2011) Galla Placidia: The Last Roman Empress. New York and Oxford: Oxford University Press. Smith, J. (2000) ‘Did Women Have a Transformation of the Roman World?’, Gender and History 12, 552–71. Snee, R. (1998) ‘Gregory of Nazianzen’s Anastasia Church: Arianism, the Goths and Hagi­ ography’, Dumbarton Oaks Papers 52, 157–86. Solari, A. (1948) ‘La successione di Giustino in Bisanzio’, Accademia Nazionale dei lincei. Rendiconti di Scienze morali, storiche e filologiche 8, 340–9. Sotiroudis, P. (1989) Untersuchungen zum Geschichtswerk des Johannes von Antiocheia. Thessalonki: Aristotle University. Speck, P. (1997) ‘Beobachtungen zur Unterbasis des Theodosios-Obelisken im Hippodrom von Konstantinopel’, Boreas 20, 17–22. Speidel, M. P. (1978) Guards of the Roman Armies: An Essay on the Singulares of the Provinces. Bonn: Habelt. Speidel, M. P. (1994) Die Denkmäler der Kaiserreiter. Equites Singulares Augusti. Bonn: Habelt. Speidel, M. P. (1995a) ‘A Horse Guardsman in the War between Licinius and Constantine’, Chiron 25, 83–7. Speidel, M. P. (1995b) ‘Die Garde des Maximus auf der Theodosiusäule’, Istanbuler Mit­ teilungen 45, 131–6. Speidel, M. P. (1996) ‘Sebastian’s Strike Force at Adrianople’, Klio 78, 434–7. Speidel, M. P. (1997) Riding for Caesar. Cambridge, MA: Harvard University Press. Speigl, J. (1996) ‘Synoden im Gefolge der Wende der Religionspolitik unter Kaiser Justi­ nos (518)’, Ostkirchliche Studien 45, 8–20. Spira, A. (1967a) Gregory of Nyssa, Oratio funebris in Meletium episcopum in Gregory Nyssensis, opera omnia, vol. 9. Leiden: Brill, 441–57.

301

BIBLIOGRAPHY

Spira, A. (1967b) Gregory of Nyssa, Oratio consolatoria in Pulcheriam in Gregory Nys­ sensis, opera omnia, vol. 9. Leiden: Brill, 461–72. Spira, A. (1967c) Gregory of Nyssa, Oratio funebris in Flaccillam in Gregory Nyssensis, opera omnia, vol. 9. Leiden: Brill, 475–90. Stein, E. (1919) ‘Iustinus 1’, RE 10, 1315–28. Stein, E. (1949) Histoire du Bas-Empire, vol. 2. Paris: Desclée de Brouwer. Stein, E. (1959) Histoire du Bas-Empire, vol. 1. Paris: Desclée de Brouwer. Strube, C. (1984) Polyeuktoskirche und Hagia Sophia: Umbildung und Auflösung antiker Formen, Entstehen des Kämpferkapitells. Munich: C. H. Beck. Svenshon, H. (2013) ‘New Considerations for the Layout Design of the Sergios-Bacchus Church in Istanbul’, Architectura. Zeitschrift fur Geschichte der Baukunst 43, 113–28. Svenshon, H. and Stichel, R. H. W. (2000) ‘Neue Beobachtungen an der ehemalige Kirche der Heiligen Sergios und Bakchos (Küçük Ayasofya Camisi) in Istanbul’, Istanbuler Mitteilungen 50, 389–409. Swainson, H. (1895) ‘Monograms of the Capitals of S. Sergius at Constantinople’, Byzan­ tinische Zeitschrift 4, 106–8. Szidat, J. (1975) ‘Zu Ammian 20, 4, 21: “Excubitor nomine”’, Chiron 5, 493–4. Szidat, J. (1977) Historischer Kommentar zu Ammianus Marcellinus Buch XX-XXI. Wies­ baden: Steiner. Taddei, A. (2009) ‘La colonna di Arcadio a Constantinopoli profile storico di un monument attraverso le fonti documentarie dalle origini al’ età moderna’, Nea Rhome 6, 37–103. Taddei, A. (2019) ‘Il “revival” della colonna coclide quale strumento di propaganda all’epoca di Theodosio I (379–395). Appunti sullo stato della ricerca’, Bisanzio e l’Occidente 2, 43–58. Taft, R. (1993) The Liturgy of the Hours in East and West: The Origins of the Divine Office, rev. edn. Collegeville, MN: Liturgical Press. Tate, G. (2004) Justinien. L’épopée de l’empire d’Orient. Paris: Fayard. Taylor, R. (2003) Roman Builders: A Study in Architectural Process. Cambridge: Cambridge University Press. Thompson, E. A. (1946) ‘The Isaurians under Theodosius II’, Hermathena 68, 18–31. Thompson, E. A. (1982) Romans and Barbarians. Madison, WI: University of Wisconsin Press. Tinnefeld, F. (1971) Kategorien der Kaiserkritik in der byzantinischen Historiographie. Munich: W. Fink. Toivanen, H.-R. (2003–4) ‘The Church of St Polyeuktos: Archaeology and Texts’, Acta Byzantina Fennica 2, 127–50. Topping, E. C. (1978) ‘On Earthquakes and Fires: Romanos’ Encomium to Justinian’, Byz­ antinische Zeitschrift 71, 22–35. Tougher, S. (2017) ‘Periodization’, in ed. T. Loughran, A Practical Guide to Studying His­ tory: Skills. London: Bloomsbury Academic. Tougher, S. (2019) ed. The Emperor in the Byzantine World. London: Routledge. Trampedach, K. (2013) ‘Daniel Stylites and Leo I: An Uneasy Relationship between Saint and Emperor’, in eds. B. Dignas, R. Parker and G. Stroumsa, Priests and Prophets among Pagans, Jews and Christians. Leuven: Peeters, 185–207. Treadgold, W. (1995) Byzantium and Its Army 284–1081. Stanford. Treadgold, W. (1997) A History of the Byzantine State and Society. Stanford: Stanford University Press.

302

BIBLIOGRAPHY

Trilling, J. (1989) ‘The Soul of the Empire: Style and Meaning in the Mosaic Pavement of the Byzantine Imperial Palace in Constantinople’, Dumbarton Oaks Papers 43, 27–72. Trombley, F. R. and Watt, J. W. (2000) trans. The Chronicle of Pseudo-Joshua the Stylite. Liverpool: Liverpool University Press. Turcan, R. (1987) Vivre à la cour de Césars d’Auguste à Dioclétien. Paris: Les Belles Lettres. Underwood, P. (2020) ‘The Architecture of Justinian’s Church of the Holy Apostles, 1 and 2, in eds. M. Mullett and R. G. Ousterhout, The Holy Apostles: A Lost Monument, a Forgotten Project and the Presentness of the Past. Washington, DC: Dumbarton Oaks, 369–94. Vallejo Girvés, M. (2017) ‘Empress Verina among the Pagans’, in eds. M. Sághy and E. M. Schoolman, Pagans and Christians in the Late Roman Empire: New Evidence, New Approaches (4th–8th Centuries). Budapest: Central European University Press, 43–58. van Dam, R. (1988) Gregory of Tours: The Glory of the Martyrs. Liverpool: Liverpool University Press. Van Dam, R. (2002) Kingdom of Snow: Roman Rule and Greek Culture in Cappadocia. Philadelphia, PA: Universty of Pennsylvania Press. Vanderspoel, J. (1995) Themistius and the Imperial Court: Oratory, Civic Duty, and Paid­ eia from Constantius to Theodosius. Ann Arbor: University of Michigan Press. Van Douwen, W. J. and Land, J. P. N. (1889) Joannis Episcopi Ephesi Syri Monophysitae Commentarii de beatis orientalibus et Historiae ecclesiasticae fragmenta. Amsterdam: Müller. Van Millingen, A. (1912) Byzantine Churches in Constantinople. London: Macmillan. Van Nuffelen, P. (2012) ‘Playing the Ritual Game in Constantinople (379–457)’, in eds. L. Grig and G. Kelly, Two Romes: Rome and Constantinople in Late Antiquity. Oxford: Oxford University Press, 183–200. Van Rompay, L. (2005) ‘Society and Community in the Christian East’, in ed. M. Maas, Cambridge Companion to the Age of Justinian. Cambridge: Cambridge University Press 239–66. Vasiliev, A. A. (1950) Justin the First: An Introduction to the Epoch of Justinian the Great. Cambridge, MA: Harvard University Press. Vasiliev, A. A. (1958) History of the Byzantine Empire, vol. 1. Madison, WI: University of Wisconsin Press. Vattioni, F. (1980) ‘Aspar è nome libico o fenicio-punico?’, Die Sprache 26, 191–4. Vernadsky, G. (1941) ‘Flavius Ardaburius Aspar’, Südostforschungen 6, 38–73. Vessey, M. (1998) ‘The Demise of the Christian Writer and the Remaking of “Late Antiq­ uity”: From H.-I. Marrou’s Saint Augustine (1938) to Peter Brown’s Holy Man (1983)’, Journal of Early Christian Studies 6, 377–411. Vitiello, M. (2005) ‘“Cui Iustinus Imperator venienti ita occurit ac si beato Petro”. Das Ritual beim ersten Papst-Kaiser Treffen in Konstantinopel: Eine römische Auslegung?’, Byzantinische Zeitschrift 98, 81–96. von Haehling, R. (1988) ‘“Timeo, ne per me consuetudo in regno nascatur”: Die Germanen und der römische Kaiserthron’, in ed. M. Wissemann, Roma Renascens: Beiträge zur Spätantike und Rezeptsiongschichte. Frankfurt a.M: Peter Lang, 97–103. von Premerstein, A. (1903) ‘Anicia Juliana im Wiener Dioskorides-Codex’, Jahrbuch der kunsthisthistorischen Sammlungen 24, 105–24. Voöbus, A. (1958) A History of Asceticism in the Syrian Orient, vol. 3. Louvain: Peeters.

303

BIBLIOGRAPHY

Walker, A. (2010) ‘Numismatic and Metrological Parallels for the Iconography of Early Byzantine Marriage Jewelry. The Question of the Crowned Bride’, Travaux et Mémoires 16, 849–63. Wallace-Hadrill, A. (1981) ‘The Emperor and His Virtues’, Historia 30, 298–323. Wallis Budge, E. A. (1932) The Chronography of Gregory Abu’l-Faraj 1225–1286. London: Oxford University Press. Ward, K. A., Crapper, M., Altuğ, K. and Crow, J. (2017) ‘The Byzantine Cisterns of Con­ stantinople’, Water Science and Technology: Water Supply 17, 1499–1506. Ward-Perkins, B. (2005) The Fall of Rome and the End of Civilization. Oxford: Oxford University Press. Wenger, A. (1952) ‘Notes inédites sur les empereurs Théodose I, Arcadius, Théodose II, Léon I’, Revue des études byzantines 10, 47–59. Wessel, S. (2008) Leo the Great and the Spiritual Rebuilding of a Universal Rome. Leiden and Boston: Brill. Westbrook, N. (2007) ‘The Question of Origin in Late Antique and Early Byzantine Art-Historical Scholarship’, Journal of the Australian Early Medieval Association 3, 327–46. Westbrook, N. (2011) ‘The Account of the Nika Riots as Evidence for Sixth-Century Con­ stantinopolitan Topography’, Journal of the Australian Early Medieval Association 7, 33–54. Westbrook, N. (2013) ‘Towards the Reconstruction of the Forum of the Strategion and Its Related Roads in Early Byzantine Constantinople’, Journal of the Australian Early Medieval Association 9, 9–46. Westbrook, N. (2019) The Great Palace in Constantinople: An Architectural Interpreta­ tion. Turnhout: Brepols. Whitby, M. (1985) ‘The Long Walls of Constantinople’, Byzantion 55, 560–83. Whitby, M. (1987) ‘On the Omission of a Ceremony in Mid-Sixth Century Constanti­ nople’, Historia 36, 483–8. Whitby, M. (1999) ‘The Violence of the Circus Factions’, in ed. K. Hopwood, Organised Crime in Antiquity. London: Duckworth/Classical Press of Wales, 229–54. Whitby, M. (2000a) ‘The Army, c.420–602’, in eds. A. Cameron, B. Ward-Perkins and M. Whitby, Cambridge Ancient History, vol. 14, Cambridge: Cambridge University Press, 288–314. Whitby, M. (2000b) ‘Armies and Society in the Later Roman World’, in eds. A. Cameron, B. Ward-Perkins and M. Whitby, Cambridge Ancient History, vol. 14, Cambridge: Cam­ bridge University Press, 469–95. Whitby, M. (2006) ‘The St Polyeuktos Epigram (AP 1.10): A Literary Perspective’, in ed. S. F. Johnson, Greek Literature in Late Antiquity. Aldershot: Ashgate, 159–88. White, C. (1986) Gregory of Nazianzus, Autobiographical Poems. Cambridge: Cambridge University Press. Williams, S. and Friell, G. (1995) Theodosius: The Empire at Bay. New Haven: Yale Uni­ versity Press. Williams, S. and Friell, G. (1999) The Rome That Did Not Fall: The Survival of the East in the Fifth Century. London: Routledge. Winterling, A. (1999) Aula Caesaris. Studien zur Institutionalisierung des römischen Kai­ serhofes in der Zeit von Augustus bis Commodua (31 v. Chr–192 n. Chr). Munich: Old­ enbourg.

304

BIBLIOGRAPHY

Wirth, G. (1964) ‘Zur Datierung einiger Ereignisse in der Regierungszeit Justins I’, His­ toria 13, 376–83. Witakowski, W. (1996a) ‘The Sources of Pseudo-Dionysius of Tel-Mahre for the Second Part of His Chronicle’, in ed. J. O. Rosenqvist, Leimon. Studies Presented to Lennart Ryden on his Sixty-Fifth Birthday. Uppsala: Uppsala University, 181–210. Witakowski, W. (1996b) trans. Pseudo-Dionysius of tel Mahre Chronicle Part III. Liver­ pool: Liverpool University Press. Wolfram, H. (1988) History of the Goths. Berkeley, Los Angeles and London: University of California Press. Wood, P. (2009) ‘The Invention of History in the Later Roman World: The Conversion of Isauria in the Life of Conon’, Anatolian Studies 59, 129–39. Wood, P. (2011) ‘Multiple Voices in Chronicle Sources: The Reign of Leo I (457–474) in Book Fourteen of Malalas’, Journal of Late Antiquity 4, 298–314. Woods, D. (1996) ‘The Scholae Palatinae and the Notitia Dignitatum’, Journal of Roman Military Equipment Studies 7, 37–50. Woods, D. (1997a) ‘Ammianus and some tribuni scholarum palatinarum c. A.D. 353–64’, Classical Quarterly 47, 269–91. Woods, D. (1997b) ‘The Emperor Julian and the Passion of Sergius and Bacchus’, Journal of Early Christian Studies 5, 335–67. Woods, D. (1997c) ‘Valerius Victorinus again’, Chiron 27, 85–93. Woods, D. (2006) ‘On the Alleged Reburial of Julian the Apostate at Constantinople’, Byzantion 76, 364–71. Wortley, J. (2005) ‘The Marian Relics at Constantinople’, Greek, Roman and Byzantine Studies 45, 177–83. Wroth, W. (1908) Imperial Byzantine Coins in the British Museum, vol. 1. London: British Museum. Wroth, W. (1911) Catalogue of the Coins of the Vandals, Ostrogoths and Lombards. London: British Museum. Yoncaci-Arslan, P. (2018) ‘Registrars of Urban Movement in Constantinople: Monumental Columns and the Mese’, Annual of Istanbul Studies 7, 7–29. Zuckerman, C. (1994) ‘L’empire d’orient et les Huns, notes sur Priscus’, Travaux et Mémoires 12, 159–63. Zuckerman, C. (2002) ‘The Dedication of a Statue of Justinian at Antioch’, in eds. T. DrewBear, M. Tashalan and C. M. Thomas, Actes du Ier congrès international sur Antioche de Pisidie. Lyon: Université Lumière-Lyon 2, 243–55.

305

INDEX

Acacius, martyr 13, 212n30 Aetius, comes Isauriae 67n79 Aetius, general 34–7, 41, 44, 48–9, 126n100, 197, 203 Alypia, daughter of Anthemius 35 Ambrose, bishop Milan 7, 17, 21n116 Anagastes, general 80, 82, 83n157, 86–92, 135–6, 138 Anastasian ‘Long Wall’ 90 Anastasius, emperor (491–518) 30, 37–8, 50, 126–7, 129, 137, 143, 153–4, 162–8, 171–9, 181–3, 185, 187–8, 193–4, 196, 198, 200, 205, 210, 224, 236, 238 Anicia Juliana 38, 61–2, 171, 184–5, 194, 198–9, 208, 210, 230, 233, 236–7, 245, 256, 269 anniversaries: dies imperii 14, 16; foundation of Constantinople 11, 14, 266; saints and emperors 13–16, 19, 28n176 Anthemius, emperor (467–72) 35–7, 54, 60, 62, 79–80, 82, 84, 98, 142n46, 152, 164–5, 223–4 Aqueduct of Valens 10, 21, 256 Arcadia, daughter of Arcadius 32 Arcadia, first wife of Zeno 64, 72, 76, 137n11, 157n33 Arcadius, emperor (395–408) 7, 9, 14–16, 18–20, 22–3, 26–8, 32, 45, 50, 246, 257, 265, 271 Ardaburius, father of Aspar 41, 47 Ardaburius, son of Aspar 47, 51, 53–4, 57–8, 61–5, 69–71, 73, 75–82, 84–7, 89–91, 93–4, 96–100, 105–7, 135–9 Areobindus, general 38, 41, 47, 57, 59, 61 Areobindus, husband of Anicia Juliana 38, 61–2, 171, 236, 238 Ariadne, wife of Zeno and Anastasius 30, 37–8, 60–2, 72n102, 75–7, 81, 89,

93, 95, 98, 102, 106, 135–8, 150n89, 153–68 Arians 11–12, 57, 78, 94–5, 99, 198; see also Homoeans aristocracy 4, 19, 21, 26, 29, 44–50, 56, 103, 136, 199, 268–9 Armatus, nephew of Verina 41, 60, 100, 159–60 Arnegisclus, general 80, 87 Artabanes, general 268 Aspar, general 37–8, 41, 47–8, 51–107, 108, 110, 125, 135–9, 147, 154–5, 261 Athaulf, Gothic king 32, 41, 43–4 Attila, general and Hun king 29, 37, 41–4, 47, 57–9, 78–80, 104 Augusta 15–16, 27, 30, 41, 60, 62, 115, 147, 153, 157–68, 171, 201–2, 232, 245, 271; see also empress Aurelian, City Prefect 21–2, 25, 50 Avitus, emperor (455–6) 41, 54n13 Basiliscus, emperor (475–6) 30, 38, 41, 51, 59, 63–4, 80, 82, 84–7, 89–91, 96–8, 100, 102, 106, 135, 145, 147, 149, 153, 155, 158–9, 161, 197 Bauto, general 30, 32, 37, 46 Belisarius, general 83, 169, 200–1, 218, 253, 268 bishops: Alexandria 67, 267; Antioch 161, 167, 220, 223; Constantinople 181, 185, 219, 222; Rome 180, 198, 244, 267, 270 Burgundian/Burgundians 29, 37, 41, 43–4 Byzantine/Byzantines 77, 99, 119, 133, 140, 143, 151, 153–4, 168, 212, 231, 243, 246, 255, 272 Carthage 35, 82–3, 85, 97, 129 Celer, general 172, 174, 178–9, 182, 185

306

INDEX

ceremonial 7, 9, 11, 14–20, 22–4, 26–7, 46–7, 49–50, 62, 108–9, 119, 126, 133, 143–4, 146, 148, 152, 172, 176, 194, 196–8, 209, 211, 245, 256–9, 270–5; see also procession; ritual Chalke, imperial place 24, 109, 130, 132, 257, 259–61 Constantine I, emperor (306–37) 6, 11, 13, 17, 23, 27, 108, 117–18, 121–2, 128, 158, 173–4, 180, 199, 210, 244, 254, 259 Constantinople 1, 3–4, 6–28, 30, 32, 35, 37–8, 43, 45–6, 49–50, 51–107, 108–10, 113, 119, 123–5, 128–9, 133, 135–6, 138–43, 146, 148, 150, 153–68, 172–82, 185–203, 207–45, 246–52, 253–75; baths: Constantine 165; Arcadia 22, 72; churches: Holy Apostles 12–13, 17–18, 22, 28, 150, 159, 162, 166, 211, 256, 269, 271, 274; church of Euphemia 19, 165; Hagia Eirene 12, 20, 211, 254, 264, 273; Hagia Sophia 20, 24, 131, 180, 259, 264; St Polyeuktos 38, 199, 210, 230, 256, 269; St Paul 19; St Menas 13; St Mark 24; St Mocius 211; St Michael at Nea 165; Forty Martyrs 165; St Elias 165; Sts Peter and Paul 183, 193, 199, 209–10, 221, 226, 233, 273; Sts Sergius and Bacchus 198, 202, 207–45, 248, 273; Sts Cosmas and Damian 211; St Anastasia 12, 274; St Irene (Sykai) 274; Virgin (Eugenius) 24; Virgin (Chalcoprateia) 156; Virgin (Blachernae) 156, 166, 212, 256, 273; cisterns: Basilica 24, 262; Aetius 228n134, 225, 261; Justinian 254; Aspar 261; Mocius 261;Theodosius 269; columns: Constantine 10; Goths 25; forums: Constantine 6, 13, 22, 56, 166, 258, 264; Theodosius I 22–4, 27, 257–8, 269; Arcadius 257, 265; Strategius 26, 263; Marcian 256; the Ox 257; Justinian 272; harbours: Julian 211, 261, 264, 268; Theodosius I 262; hippodrome: 1, 11, 14–15, 22, 24, 26, 56, 94, 98, 108–9, 128, 140, 146–8, 159, 162, 166, 172–3, 189, 196–7, 202, 209, 212, 258, 263, 273; palaces: Constantine 119, 132; Arcadia 153; statues: Constantine 11, 14, 258; storehouses (horrea): 22, 262, 263n69; walls: Constantine 256, 264–5; Theodosius II 265 Constantius II, emperor (337–61) 11, 17–18, 94, 108, 256

empress see Ariadne; Eudocia; Eudoxia; Euphemia; Flaccilla; Galla; Galla Placidia; Pulcheria; Sophia; Theodora; Verina ethnicity 57, 63, 80–1, 102–5 Eucherius, son of Stilicho 32, 34 Eudocia, daughter of Valentinian III 34–5, 43, 83 Eudocia, wife of Theodosius II 32, 38, 46, 60, 62, 234, 237–8, 242n204, 256 Eudoxia, daughter of Theodosius II 34–5, 38, 43, 54n23, 60, 62, 78, 83 Eudoxia, wife of Arcadius 32 Euphemia, wife of Justin I 180–2, 185, 193, 205 Flaccilla, daughter of Theodosius II 32 Flaccilla, wife of Theodosius I 15–17, 21–2, 27 funerals, imperial 10, 16–17, 62, 265, 271 Gaiseric, Vandal king 34–5, 41, 43, 83, 85, 97 Galla, wife of Theodosius I 16–17, 22 Galla Placidia, daughter of Theodosius I 14, 30, 32, 41, 43–4 generals 29–30, 34, 37, 41, 43, 45–6, 52–3, 63, 71, 86, 105, 155, 163, 173, 176–7, 191, 205, 260–1; see also Aetius; Anagastes; Anthemius; Ardaburius; Areobindus; Arnegisclus; Artabanes; Aspar; Attila; Basiliscus; Bauto; Belisarius; Celer; Gundobad; Hypatius; John the Vandal; Jordanes; Justin I; Justinian; Majorian; Marcian; Narses; Orestes; Patricius; Plinta; Ricimer; Stilicho; Theodoric; Theodoric Strabo; Theodosius I; Vitalian; Zeno Glycerius, emperor (473–4) 37, 143n46 Goths 1, 6, 8, 10, 12, 21, 26–7, 30, 43–4, 49, 58, 69–70, 72–3, 78–84, 87, 91, 99–102, 104–6, 111, 135, 145, 257, 259, 267, 269, 272 Gratian, emperor (375–83) 14–15, 17 Gregory Nazianzen 10, 12 Gregory of Nyssa 10–11, 16 Gundobad, general and Burgundian king 37, 41 Hebdomon 15, 20, 22, 24, 56, 158, 165, 212, 253–5; church of John the Baptist 24, 165, 254

307

INDEX

Helena, mother of Constantine 13, 17, 153, 158, 180 Helenianae, district of Constantinople 25, 56 Hilderic, Vandal king 35, 199 Homoeans 13, 14; see also Arians Honoria, daughter of Galla Placidia 32, 41, 43–4 Honorius, emperor (395–423) 14–15, 18, 23, 30, 32, 45, 257, 269 Huneric, Vandal king 34–5, 41 Huns 1, 43, 47–9, 58–9, 78–82, 86–7, 100, 104, 190, 253, 265, 267, 273 Hypatius, general 50, 171, 177 Jerusalem 13, 27, 34–5, 60, 62, 181, 192, 239 John the Vandal, general 64, 177 Jordanes, general 64, 69, 87 Jovian, emperor (363–4) 9, 15, 17, 27, 120 Julian, emperor (361–3) 15, 17, 22n132, 25, 27, 38, 108, 119–20 Justin I, emperor (518–27) 4–5, 126–9, 143, 169–206, 208–9, 211–12, 217–19, 232, 235–6, 238, 243–4, 255, 269–71 Justin II, emperor (565–78) 129, 132, 154, 226n126, 251 Justinian, emperor (527–65) 3–4, 18, 35, 43, 108, 110, 122, 127, 129, 131, 143, 154, 159, 165, 167, 169–206, 207–45, 245–52, 253–75 Leo I, emperor (457–74) 4, 37–8, 51–107, 108–33, 134–5, 137–9, 142–3, 145–8, 151, 153, 156–9, 168, 196, 254–5 Leo II, emperor (474) 77, 98, 102, 132, 134–52, 153, 158–9, 196–7 Leontia, daughter of Leo I 37, 51, 61–2, 94–5, 98, 106, 155, 157–60, 165 Leontius, usurper 30, 38, 161–2 Libius Severus, emperor (461–5) 35 Longinus, brother of Zeno 38, 72, 103, 160–1, 163 Majorian, emperor (457–61) 34–5, 41, 83 Marcian, emperor (450–7) 30, 34–5, 37, 38, 41, 43–4, 48–9, 52–5, 58, 60–1, 65, 79, 86, 141, 147, 154, 156, 161, 163–4, 254, 256 Marcian, husband of Leontia, usurper 157–60, 164n70, 165 Marina, daughter of Arcadius 32, 153 Maximus, emperor (455) 34–5, 41 Maximus, usurper 7, 10n28, 12, 26–7

Narses, general 268, 274 Nepos, emperor (474–5) 37, 143n46 New Rome (Constantinople) 1, 6, 12–13, 27, 180, 198 Olybrius, emperor (472) 35, 37–8, 60, 62, 83, 237, 269 Olybrius, son of Anicia Juliana 171, 193, 198, 210, 236, 238, 241n199 Orestes, general 37, 48–9, 143n46 Patricius, general 50, 172, 178, 182 Persian/Persians 15, 18, 24–5, 27, 53, 57, 64, 69, 71, 87, 128–9, 132, 164, 174, 190, 195, 197, 200, 213, 235–6, 245, 263, 268, 272, 273 Plinta, general 46–8, 57n31 procession 10, 13, 16, 19–21, 23, 56, 109, 131, 155, 162, 173n27, 183, 188, 198, 243, 251, 253, 255, 257–8, 265, 271–2, 274; see also ceremonial; ritual Procopius, usurper 6–9, 15 Procopius of Caesarea, writer 172, 268–9; Buildings 4, 204, 210, 212, 214, 247, 254; Secret History 73–4, 123–5, 128, 169–71, 173, 180, 187, 193–4, 203, 205–6, 240, 246–8, 252, 266, 268–9; Wars 82, 85, 97, 177, 195–6, 200, 204–5, 236 Pulcheria, daughter of Theodosius I 14, 16–17 Pulcheria, wife of Marcian 14, 30, 32, 34–5, 43, 54, 62, 153, 156, 158, 161, 163–5, 255 Ricimer, general 37, 41, 47, 100n234 ritual 14–16, 19–20, 23, 25, 27, 45, 62, 108, 155, 164, 168, 194, 197, 199, 211, 259, 270, 271–4; see also ceremonial; procession Rome 1, 3, 12–13, 17, 19, 23, 30, 35, 37–8, 44, 46, 49–50, 54, 60, 79, 82, 114–18, 133, 165, 180–6, 198–9, 205, 209–10, 217, 238, 244, 257–8, 261, 270, 273 senate, Constantinople 6, 8, 10, 19–20, 26–7, 53, 56, 63, 66, 69, 94, 99, 146, 147–8, 150, 162, 172–3, 181, 183, 194–6, 201, 206, 211, 258–9, 261; Rome 44 Serena, wife of Stilicho 16, 30, 32, 43 Severus, bishop Antioch 16n57, 167, 180, 217–18, 220–2

308

INDEX

Sophia, wife of Justin II 154, 163, 168, 251 Stilicho, general 16, 30, 32, 37, 43, 46, 50, 262 Theodora, wife of Justinian 129, 154, 159, 163–5, 167–8, 169, 176, 193–4, 198, 201, 205, 207–8, 210, 213, 216, 218–36, 243–7, 254, 256, 267–8, 271, 273–4 Theodoric, general and Ostrogthic king 38, 41, 47–8, 50, 54, 79, 155, 198–9 Theodoric, Visigothic king 34, 41 Theodoric Strabo, general 58, 91, 100–1, 105 Theodosius I, emperor (379–95) 3–5, 6–28, 30, 32, 45–6, 50, 103n250, 108, 254, 257, 271

Theodosius II, emperor (408–450) 7, 16n83, 20, 24n154, 28, 30, 32, 43–7, 49–50, 52, 54–60, 62, 73, 82, 108, 121, 154, 156, 173, 199, 237, 246, 268 Valentinian III, emperor (425–455) 30, 32, 34–5, 41, 43–5, 49, 60–1 Vitalian, general 126n102, 129, 174, 177–8, 180–1, 183–9, 192, 201, 203–5 Zeno, emperor (476–491) 37–8, 41, 47, 51–3, 58, 63–5, 67–77, 81, 84–102, 103–6, 110, 123–6, 128, 135–40, 144–52, 153–5, 157–67, 174, 194 Zeno, general 59, 103

309