The Roman Imperial Court in the Principate and Late Antiquity 9780192865236, 0192865234

Citation preview

The Roman Imperial Court in the Principate and Late Antiquity

The Roman Imperial Court in the Principate and Late Antiquity Edited by

CAILLAN DAVENPORT AND MEAGHAN McEVOY

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Oxford University Press 2023 The moral rights of the authors have been asserted All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2023938867 ISBN 978–0–19–286523–6 DOI: 10.1093/oso/9780192865236.001.0001 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

For Brian Jones and Paul Tuffin

Acknowledgements This book has its origins in an extended panel on ‘The Roman Imperial Court from the Antonines to the Theodosians’ which we organized at the Celtic Conference in Classics, hosted by University College, Dublin, in June 2016. We would like to thank the convenors (Douglas Cairns, the late Anton Powell, Alan Ross, and Alexander Thein) for running such a superb event, which proved to be the perfect occasion for our own mini-conference. We are also grateful to those scholars who gave papers in Dublin and contributed to the lively discussion, but whose work does not appear here: Paul Jarvis, Paul Roche, Benet Salway, and Alexander Skinner. The present book, therefore, is not a publication of proceedings, but a volume which has grown and evolved in subsequent years along with our own burgeoning interest in the field of court studies. We would like to thank our contributors for their patience and good humour during the many years of revisions, as well as the inevitable delays caused by our having twins in a global pandemic. We owe a special debt of gratitude to Ben Kelly, who has been a model of scholarly generosity and collegiality. He invited Caillan to be a part of his own imperial court project with Angela Hug at York University, and accepted our own invitation to contribute to this volume. Our work has benefited enormously from engagement with the York University project (The Roman Emperor and his Court c.30 – 300, Cambridge, 2022), and Ben very kindly gave very helpful feedback on the Introduction and Chapter 14 of this volume. Our Editors at OUP, Charlotte Loveridge, Jamie Mortimer, and Kayley Gilbert have offered vital support and assistance throughout the process, and the peer reviewers provided exactly the right sort of helpful advice and suggestions. We extend our thanks as well to our project manager Thomas Deva and to our copy-editor Juliet Gardner. This volume would not have been possible without the incredibly generous support of Hartmut Leppin at the Goethe-Universität in Frankfurt am Main. He sponsored Meaghan’s Alexander von Humboldt Fellowship in 2013–16 and Caillan’s in 2019–21, along with Meaghan’s Visiting Researcher status for the second visit. He even very kindly allowed us to use his own university office when Covid restrictions prevented more than one person from occupying an office at a time, providing us with precious working space. We would also like to thank many other colleagues in the Abteilung für Alte Geschichte, including Marius Kalfelis, Simone Mehr, Muriel Moser-Gerber, Sebastian Weinert, and Alexander Weiss, for putting up with our German, sharing offices, or providing support during our

viii



stays in Frankfurt, especially during the 2019–21 visit when Covid left us and our newborn twins separated from our family. In Australia, we have benefited from the support of Macquarie University, our former institution, for allowing us to spend an extended period in Germany for research. Caillan’s chapter and early work on the project was funded by an Australian Research Council Discovery Early Career Researcher Award (DE150101110). The School of Literature, Languages, and Linguistics at the Australian National University provided a publication subsidy to cover the cost of some of the images. Ann Moffatt and Maxeme Tall very generously gifted us a complete set of the new French edition of the Book of Ceremonies, which enabled us to use the text and commentary during the final stages of our work. Our parents, Anne and Francis McEvoy and Narelle and Tony Davenport, have provided valuable childcare support over the past year. Our twin boys, Alaric and Hamish, are the joys of our life, and we are very sorry that this volume has at times distracted us from the important business of singing Incy Wincy Spider, reading Pig the Pug, building Duplo, and kicking the ball around the garden. Finally, we would like to dedicate this book to our respective Honours thesis supervisors, Brian Jones (Caillan) and Paul Tuffin (Meaghan), who set us up for our research careers all those years ago.

Table of Contents List of Figures Abbreviations Contributors

Introduction: Connecting Courts Caillan Davenport and Meaghan McEvoy

xi xiii xv

1

PART I: RITUAL AND SPATIAL DYNAMICS 1. Great King or Civilis Princeps? Monarchical Ideals and Daily Interaction in the Reign of Antoninus Pius Christoph Michels

41

2. Changing the Guard: Guard Units and Roman State Ceremonial from the First to the Fourth Century Christian Rollinger

56

3. Cities, Palaces, and the Tetrarchic Imperial Courts Verena Jaeschke and Caillan Davenport 4. The Court in Constantinople Facing the Death of the Emperor Audrey Becker

75 105

PART II: INDIVIDUAL AND COMMUNITY AT COURT 5. Was the Roman Imperial Court an ‘Emotional Community’? Benjamin Kelly

121

6. Jurists as Courtiers from Augustus to Justinian Jill Harries

142

7. Court Construction and Regime Change in the Mid-Fourth Century 156 Kevin Feeney 8. Sharing the Imperial Limelight: The Age of the Magister Militum Meaghan McEvoy 9. Representatives and Co-Rulers: Imperial Women and the Court in Late Antiquity Anja Busch

172

203

x

  

PART III: THE POLITICS OF ACCESS 10. Beyond the Veil: Athanasius at the Court of Constans Fabian Schulz

221

11. Dynamics of Power: The Nestorian Controversy, the Council of Ephesus of 431, and the Eastern Imperial Court Daniëlle Slootjes

240

12. Splendid Isolation: Secluded Emperors and the Spectre of Oriental Despotism Martijn Icks

262

13. Envisioning Audiences at the Roman Imperial Court Caillan Davenport

278

PART IV: COMPARATIVE PERSPECTIVES 14. The Evolution of the Roman Imperial Court in Historical Context Caillan Davenport and Meaghan McEvoy

309

Bibliography Index

359 401

List of Figures 2.1 Sestertius of the Emperor Nero (RIC I² 105). Münzkabinett, Staatliche Museen zu Berlin, 18,220,895. © Münzkabinett, Staatliche Museen zu Berlin. Photograph by Dirk Sonnenwald.

61

3.1 Fresco on the south wall of the Imperial Cult Chamber at Luxor. © Manfred Bail/imageBROKER/agefotostock.

87

3.2 The Arch of Galerius in Thessaloniki. Photo by Caillan Davenport.

92

3.3 Interior of the Basilica of Constantine in Trier. Photo by Caillan Davenport.

94

9.1 Bronze coin of Eudoxia (RIC X Arcadius 102, Nicomedia), Dumbarton Oaks BZC.1948.17.1122. © Dumbarton Oaks, Byzantine Coins and Seals Collection, Washington DC.

207

9.2 Solidus of Pulcheria (RIC X Theodosius II 205, Constantinople), Dumbarton Oaks BZC.1948.17.1182. © Dumbarton Oaks, Byzantine Coins and Seals Collection, Washington DC.

214

10.1 Constantius II as depicted in the Codex-Calendar of 354. MS Barb.lat.2154. pt.B (f. 13r). © Bibliotheca Apostolica Vaticana.

233

13.1 Martini, Simone (1284–1344): Saint Martin and the Miracle of the Fire. Assisi, Church of San Francesco. © 2022. Photo Scala, Florence.

295

14.1 Ivory depicting a relic translation in Constantinople, Trier Cathedral Treasury. © akg-images.

315

14.2 The south-eastern face of the base of the Theodosian Obelisk, Constantinople. Photo by Meaghan McEvoy.

328

14.3 Mosaic depicting the Emperor Justinian and his court, Basilica of San Vitale, Ravenna. © Christine Webb, Alamy Stock Photo.

341

14.4 Ivory diptych of the Augusta Ariadne, Kunsthistorisches Museum, Vienna. © akg-images.

351

Abbreviations Abbreviations used in this volume usually follow the conventions of the fourth edition of the Oxford Classical Dictionary (OCD⁴) and the three volumes of the Prosopography of the Later Roman Empire (PLRE I, II, III). Where there are differences between these two reference works, we have used the PLRE version. We have also employed the following abbreviations in addition to, or in variation from, the above works: CEL CPG Dio Eus. SC Fronto Ad amicos Ad Ant. Imp. Ad Ant. Pium Ad M. Caes. Ad Verum Imp. De fer. Als. De nep. am. HA IAph2007

Jul. Ep. Lact. DMP Lenel, Pal. LSA Musurillo, Acta Nest. Herac. OED OLD

P. Cugusi (ed.) 1992–2022. Corpus Epistularum Latinarum: papyris, ostracis, tabulis servatarum. Florence. M. Geerard, F. Glorie, J. Noret (eds) 1974–2023. Clavis Patrum Graecorum. Turnhout. Cassius Dio Eusebius, De Sepulchro Christi, in I. A. Heikel (ed.) 1902. Eusebius Werke I. Leipzig. M. P. J. van den Hout (ed.) 1988. M. Cornelii Frontonis Epistulae. Second edition. Leipzig. Epistulae ad amicos Epistulae ad Antoninum Imperatorem Epistulae ad Antoninum Pium Epistulae ad Marcum Caesarem Epistulae ad Verum Imperatorem Epistulae de feriis Alsiensibus Epistulae de nepote amisso Historia Augusta J. Reynolds, C. Roueché, G. Bodard (eds) 2007. Inscriptions of Aphrodisias (2007), available at https://insaph.kcl.ac.uk/insaph/ iaph2007/ Julian, Epistulae, in W. C. Wright (ed.) 1923. Julian. Volume III. Loeb Classical Library 157. Cambridge, MA. Lactantius, De Mortibus Persecutorum, in J. L. Creed (ed.) 1984. Lactantius. De Mortibus Persecutorum. Oxford. Lenel, O. (ed.) 1889. Palingenesia Iuris Civilis. Leipzig. R. R. R. Smith, B. Ward-Perkins (ed.) 2012. Last Statues of Antiquity Database, available at http://laststatues.classics.ox.ac.uk H. Musurillo (ed.) 1954. The Acts of the Pagan Martyrs. Acta Alexandrinorum. Oxford. Nestorius, Liber Heraclidis, in G. R. Driver and L. Hodgson (eds) 1925. Nestorius. The Bazaar of Heraclides. Oxford. Oxford English Dictionary P. G. W. Glare (ed.) 1982. The Oxford Latin Dictionary. Oxford.

xiv



P. Aphrod. Lit. P. Panop. Beatty Theoph. Cont. TMI vdH²

J.-L. Fournet (ed.) 1990. Hellénisme dans l’Égypte du VIe siècle. La bibliothèque et l’oeuvre de Dioscore d’Aphrodité. Cairo. T. C. Skeat (ed.) 1964. Papyri from Panopolis in the Chester Beatty Library Dublin. Dublin. I. Bekker (ed.) 1938. Theophanes Continuatus. Bonn. Thompson Motif Index, in S. Thompson (ed.) 1955–8. Motif-Index of Folk-Literature. Copenhagen. M. P. J. van den Hout (ed.) 1988. M. Cornelii Frontonis Epistulae. Second edition. Leipzig.

Contributors Audrey Becker is Professor of Roman History at the University of Franche-Comté, France. Her research focuses on imperial power from Constantine to Justinian and on social sciences as applied to Late Antiquity. Her publications include Les relations diplomatiques romano-barbares en Occident. Acteurs, fonctions, modalités (De Boccard, 2013) and Dieu, le souverain et la cour. Stratégies et rituels de légitimation du pouvoir impérial et royal dans l’Antiquité tardive et au haut Moyen Âge (Ausonius, 2022). Anja Busch completed her doctorate within the framework of the International Research Training Group ‘Political Communication from Antiquity to the Present’ at the universities of Frankfurt am Main and Bologna. She then worked as a lecturer in Mainz and as a research assistant in Bonn. Since 2018, she has been working on a project funded by the German Research Foundation on political murders in Hellenistic monarchies at the University of Hamburg. Caillan Davenport is Associate Professor of Classics and Head of the Centre for Classical Studies at the Australian National University. He is the author of A History of the Roman Equestrian Order (Cambridge, 2019), which won the Royal Historical Society’s Gladstone Prize, and the co-editor of Fronto: Selected Letters (Bloomsbury, 2014) and Emperors and Political Culture in Cassius Dio’s Roman History (Cambridge, 2021). Kevin Feeney is a scholar of the political culture of the late Roman Empire. His work concentrates on the role of the emperor and the imperial government in late Roman society, particularly at moments of crisis or transition. He is especially interested in the ways that political legitimacy was constructed and shifts in the rhetoric of power across the imperial period. He is currently writing a monograph on the accession of the emperor in Late Antiquity. Jill Harries is Emerita Professor of Ancient History at the University of St Andrews. Her many publications span the fields of Late Antiquity and Roman law, including Sidonius Apollinaris and the Fall of Rome (Oxford, 1994), Law and Empire in Late Antiquity (Cambridge, 1998), Law and Crime in the Ancient World (Cambridge, 2007), and Imperial Rome  284–363: The New Empire (Edinburgh, 2012). Martijn Icks is a Lecturer in Ancient History at the University of Amsterdam specializing in the history of the Roman Empire. His research interests include Roman imperial representation, especially with regard to the (in)visibility and (in)accessibility of imperial power, as well as character assassination as a historical and cross-cultural phenomenon. Among other works, he has published The Crimes of Elagabalus (I.B. Tauris, 2011), Character Assassination throughout the Ages (co-editor, Palgrave, 2014), and Character Assassination and Reputation Management: Theory and Applications (co-author, Routledge, 2021).

xvi



Verena Jaeschke obtained her doctorate in Cultural Studies from the Europa-Universität Viadrina. It has now been published as Sedes Imperii. Architektur und Herrschaftsrepräsentation in den tetrarchischen Residenzstädten (wbg Academic, 2020). Benjamin Kelly is Associate Professor of History at York University, Toronto. He is the author of Petitions, Litigation, and Social Control in Roman Egypt (Oxford, 2011) and the co-editor, with A. G. Hug, of The Roman Emperor and his Court: ca. 30 BC—ca. AD 300. Volume 1: Historical Essays. Volume 2: A Sourcebook (Cambridge, 2022). Meaghan McEvoy is Senior Lecturer in History at the Australian National University. She is the author of Child Emperor Rule in the Late Roman West, AD 367–455 (Oxford, 2013) and co-editor of three thematic journal issues, Imperial Presence in Late Antique Rome (2nd–7th Centuries) for Antiquité Tardive (with S. Corcoran and M. Moser, 2017), Envisioning the Roman Emperor in Speech and Word in Late Antiquity for the Journal of Late Antiquity (with J.W. Drijvers, 2021), and Shaping Christian Politics in Late Antiquity for the Journal of Late Antiquity (with R. Flower and R. Whelan, 2022). Christoph Michels is a Heisenberg Fellow at Westfälische Wilhelms-Universität Münster. After studying history, classical archaeology, and art history at the Ruhr-Universität Bochum, he was awarded his doctorate within the framework of the International Research Training Group ‘Political Communication from Antiquity to the 20th Century’ in a cotutelle procedure at the universities of Innsbruck and Frankfurt am Main (2008). In his dissertation thesis, he dealt with the Hellenistic kingdoms of Bithynia, Pontos, and Cappadocia, and the attempts of their rulers to integrate themselves into the Hellenistic world of states. In 2016, he completed his habilitation at RWTH Aachen University with a thesis on the relationship between governmental practice and monarchical representation in the Roman Principate, using the example of Antoninus Pius. Christian Rollinger is Reader for Ancient History at the University of Trier, where he completed his PhD on amicitia in Roman elite society, and his Habilitation on late antique court ceremonial. His research deals mostly with the social, economic, and cultural history of the late Roman republic and early imperial period, Hellenistic and Roman monarchic power and its performance, as well as classical receptions in modern videogames. He is co-editor (with F. Carlà-Uhink) of The Tetrarchy as Ideology: Reconfigurations and Representations of an Imperial Power (Franz Steiner, 2023) and is currently preparing the manuscript of his next book, Zeremoniell und Herrschaft in der Spätantike. Die Rituale des Kaiserhofs in Konstantinopel (Franz Steiner, 2024), for publication. Fabian Schulz is a Classicist and Ancient Historian educated in Berlin, Oxford and Paris. After working at Free University Berlin and the Heidelberg Academy of Sciences and Humanities he joined the University of Tübingen, where he worked as a senior researcher for over a decade. His main areas of interest are late Roman and early Greek history. He recently completed his Habilitation with a thesis on Greek and Roman concepts of East and West. Daniëlle Slootjes holds the Chair of Ancient History at the Department of History, European Studies and Religious Studies, at the Faculty of Humanities of the University of Amsterdam. She is the author of The Governor and his Subjects in the Later Roman Empire (Brill, 2006), and the co-editor of Aspects of Ancient Institutions and Geography: Studies in Honor of Richard J. A. Talbert (Brill, 2014) and Leadership, Ideology and Crowds in the Roman Empire of the Fourth Century (Franz Steiner, 2020).

Introduction Connecting Courts Caillan Davenport and Meaghan McEvoy

1. Introduction In the middle of August 355, news reached the Emperor Constantius II at Milan that one of his generals, Silvanus, had declared himself emperor at Colonia Agrippina (modern-day Cologne) on the Rhine frontier.¹ According to the historian Ammianus Marcellinus, Constantius was so alarmed by this challenge to his authority that he called the members of his imperial council to his side in the small hours of the night.² The council agreed that the emperor should recall Ursicinus, who had recently fallen from favour and been accused of treason, because he was the most qualified general to defeat Silvanus’ usurpation.³ Ammianus described Ursicinus’ arrival at court and into the imperial presence as follows: After Ursicinus had been summoned there by the master of admissions (magister admissionum)—the honourable procedure—and he had entered the consistorium, he was offered the purple much more courteously than on previous occasions. who was the first of all emperors to establish that the emperor should be adored (adorari) with this foreign ceremony in the manner of kings, whereas we have read that before this time principes were greeted (salutatos) in the same manner as other officials.⁴

In this passage, Ammianus revealed that he regarded the ceremonial procedures of the imperial court in his own day as fundamentally different from those of the early empire. This distinction is encapsulated by the change from the ceremony of salutatio (greeting the emperor, usually with a kiss on the lips or cheek to indicate ¹ All dates in this book are  unless otherwise noted. ² Amm. 15.5.17–18. ³ For his disgrace, see Amm. 15.2.1–6. ⁴ Amm. 15.5.18: et per admissionum magistrum—qui mos est honoratior—accito eodem, ingresso consistorium offertur purpura multo quam antea placidius. omnium primus extero ritu et regio more instituit adorari, cum semper antea ad similitudinem iudicum salutatos principes legerimus (trans. Davenport). Caillan Davenport and Meaghan McEvoy, Introduction: Connecting Courts In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0001

2

   

equality) to that of adoratio (the act of kissing the emperor’s purple cloak, an act that articulated the emperor’s superiority). This was an innovation which Ammianus and other fourth-century writers attributed to the Emperor Diocletian.⁵ Careful study of the history of Roman imperial ceremonial enables us to nuance Ammianus’ picture,⁶ but it does not invalidate his fundamental point, that reception rituals under Augustus and Constantius II were governed by different rules and expectations, which were indicative of the transformation of the court between the Principate and Late Antiquity. We can set beside this picture of transformation a passage reflecting on the fundamental continuities of the court experience. This comes from the pen of a man with intimate knowledge of the institution, the Emperor Marcus Aurelius. Born into a Roman senatorial family, Marcus had become a favourite of Hadrian when he was but a boy. Subsequently, at age sixteen, he was adopted on Hadrian’s orders by Antoninus Pius, before ascending to the purple himself in 161, aged thirty-nine. When, in his later years, he came to ponder the circle of life, Marcus wrote: Constantly reflect that all the things which happen now have happened before: reflect too that they will happen again in the future. Have in your mind’s eye whole dramas with similar settings, all that you know of from your own experience or earlier history—for example, the whole court of Hadrian, the whole court of Antoninus, the whole court of Philip, Alexander, Croesus. All the same as now: just a different cast.⁷

This rather sombre passage reminds us that there were many essential similarities shared not only by the courts of Roman emperors, but by monarchical courts in all pre-modern societies.⁸ These include the competition for access to the monarch and to win his favour;⁹ rivalries between different ranks and groups of relatives, advisors, and officials;¹⁰ the use of ceremonial to delineate hierarchies of status and power and to bind the court together as a community;¹¹ and the constant analysis and critique of the monarch’s every move.¹² This means that even though the procedure of greeting the emperor underwent changes across the imperial period, some sort of ritual was always present. A fourth-century courtier would

⁵ Matthews 1989: 244–7. ⁶ Alföldi 1970; Davenport 2022. ⁷ M. Aur. Med. 10.27: καὶ ὅλα δράματα καὶ σκηνὰς ὁμοειδεῖς, ὅσα ἐκ πείρας τῆς σῆς ἢ τῆς πρεσβυτέρας ἱστορίας ἔγνως, πρὸ ὀμμάτων τίθεσθαι, οἷον αὐλὴν ὅλην Ἁδριανοῦ καὶ αὐλὴν ὅλην Ἀντωνίνου καὶ αὐλὴν ὅλην Φιλίππου, Ἀλεξάνδρου, Κροίσου· πάντα γὰρ ἐκεῖνα τοιαῦτα ἦν, μόνον δι᾿ ἑτέρων (trans. Hammond 2006). ⁸ On these similarities, see Duindam 1995: 192; Wallace-Hadrill 1996: 283; 2011: 91–2; Winterling 1999: 7. ⁹ Winterling 1997a: 15–16; Elias 2006: 113, 127. ¹⁰ Walthall 2008: 8–9; Geary et al. 2015: 189–90; Duindam 2016: 195–200, 214–16. ¹¹ Duindam 1995: 103–4; 2016: 200–4, 255–73; Elias 2006: 90–3, 151. ¹² Duindam 2003: 226.



3

have longed to kiss the purple of Constantius II as much as their counterparts would have wanted to kiss the lips of Marcus Aurelius several centuries before.¹³ This volume examines aspects of the Roman imperial court in both the Principate and Late Antiquity. These are modern terms of convenience which are conventionally given to the period from Augustus to the death of Carinus in the late third century (the Principate), and from Diocletian to the age of Justinian, in the mid-sixth century (Late Antiquity). The aim of examining the court over this time span of some six hundred years is motivated by our belief that by analysing both the early and late Roman Empire—periods which are usually separated in studies of the imperial court¹⁴—we can more successfully tease out continuities, changes, and connections. The problem of how to precisely delineate and analyse changes at monarchical courts is one that has preoccupied the leading minds in the field of aulic studies, such as Norbert Elias and Jeroen Duindam, for the very reason outlined by Marcus Aurelius above—that monarchical courts often share very similar features.¹⁵ What is necessary, as both Elias and Duindam have observed in their own distinctive ways, is the precise examination of the historical conditions which dictated the features and environment of courts at specific points in time. This approach can help us discriminate between genuine long-term developments and short-term changes which were introduced and then subsequently fell into abeyance. The present book does not attempt to cover all aspects of the Roman imperial court and its development. Instead, following the methodology advocated above, it consists of a series of focused case studies. Some chapters take a thematic approach, analysing specific aspects of the court, such as the appointment of officials, ceremonial developments, or stories told about the court, over several centuries. Others concentrate on certain periods, individuals, and office-holders— for example, the court of Antoninus Pius or the role of the magister militum in the fifth century—while paying attention to their wider historical significance. The book concludes with a chapter placing the Roman imperial court in comparative perspective using insights drawn from scholarship on monarchical courts in other time periods and regions. We hope that our approach will contribute to our understanding of the institutional history of the Roman imperial court and the way it developed between the Principate and Late Antiquity.

¹³ On kissing the purple, see Smith 2011: 147, and on prostration, Kim 2017: 191; see further CTh. 6.23.1 on rank in the ceremony of adoratio. ¹⁴ For example, note Wallace-Hadrill 1996 and Winterling 1999, seminal studies of the early imperial court, and Smith 2007 and 2011, focusing on the late empire. This is not a criticism of these excellent works of scholarship, but an observation of the way in which Roman political and institutional history is often divided by the so-called ‘third-century crisis’ and reign of Diocletian. There are exceptions to this rule, such as the major two-volume study by Kelly and Hug 2022, which ranges from c. 30  to c.  300. ¹⁵ Duindam 2003: 25–7, 133; Elias 2006: 153–4.

4

   

2. What Is a Court? The word ‘court’ is popularly associated with kingship in any number of familiar phrases: a monarch is required to ‘hold court’, one makes the journey ‘to court’, a young lady is presented ‘at court’, a courtier ‘pays court’ to the monarch or is banished ‘from court’.¹⁶ These expressions, which primarily derive from the language of European monarchies in the Medieval and Early Modern periods, have become lodged in our consciousness through histories, novels, and costume dramas, not to mention the perennial public fascination with the modern-day activities of the British royal family, whose head, King Charles III, presides over the Court of St James’s. But it is important to be precise about what we mean when we use the word court, and in particular, whether our terminology and concepts, so influenced by later European history, can be applied to the Roman world. The court itself is best conceived, following the seminal work of Elias, as a ‘social phenomenon’, consisting of the circle of individuals that surrounds and interacts with a ruler.¹⁷ Courts are primarily composed of members of two institutions, the royal household and the government,¹⁸ but not exclusively so, since all individuals who are drawn into a monarch’s orbit, and who interact with him (or her) regularly, can be said to form part of a court. Such a definition of a royal court as a community means that the term ‘court’ does not refer to a fixed location, since it exists wherever the ruler happens to be, whether that is in a city palace, a country residence, or on campaign.¹⁹ This corresponds nicely with Marcus Aurelius’ own conception of a court, for he did not describe the court of Augustus as ‘the Palatine’, but as the emperor, his family members, household, friends, doctors, and soothsayers.²⁰ It is well-known, of course, that most European words for court, such as cour (French), corte (Italian), and Hof (German), not to mention our own English term, derive from, and are the same

¹⁶ Cf. the variety of entries recorded under OED s.v. court, v. I.1, 2, II.3a; court, n.¹ II.5, 6a, 6b, 6c, 7a, 8a, III.9. Some German equivalents are noted by Winterling 1997a: 13–14; Winterling 1999: 9–10; Schöpe 2014: 15. Note also, in the Hellenistic context, Erskine et al. 2017a: xvi. ¹⁷ Elias 2006: esp. 3–5, and 153 (‘a figuration of people’). See also OED s.v. court, n.¹ II.7a: ‘the body of courtiers collectively; the retinue (councillors, attendants, etc.) of a sovereign or high dignitary’. For this definition in the Roman context, see Wallace-Hadrill 1996: 285–6 and Kelly 2022a: 6–7 (emphasizing the need for members of the court to have ‘reasonably regular personal interaction with the emperor’); and, in the Byzantine context, Kazhdan and McCormick 1997: 167. Cf. Winterling 1997a: 13–14, 1999: 9–10, who proposes the complementary definition of ‘das erweiterte “Haus” eines Monarchen’ (‘the extended “House” of a monarch’). ¹⁸ Duindam 2003: 3, 318–19; Elias 2006: 3, 45; Smith 2011: 127. ¹⁹ See Wallace-Hadrill 1996: 286; Kelly 2022a: 7–8; and for the Medieval period, Vale 2001: 22–3, 31. See further Winterling 1997a: 13; 1999: 10, who allows for the movement of the ruler in one of his key definitions of the court: ‘in sachlicher und lokaler Hinsicht: den Aufenthaltsort, die Residenz eines Herrschers’ (‘in specific and factual terms: the whereabouts, the abode of a ruler’). For emperors conducting business at imperial villas, see George 2022: 245, 247–9, 256, 260. ²⁰ M. Aur. Med. 8.31; Winterling 1999: 2.



5

as, words for spaces and locations.²¹ But this reflects the fact that there was a spatial aspect to a ruler’s circle and entourage, in the sense that one could be ‘at court’ or travel ‘to court’, not that it was ever a fixed physical location.²² Indeed, the movement of a court from one place to another could often change the dynamics of the court society, not only between reigns, but even within them.²³ The Romans employed several different words to refer to the court, which overlapped in chronological and semantic range. The first of these are the Greek αὐλή and the Latin aula. These words both meant court in the sense of a courtyard or hall as well as the social institution of a monarchical court, just like modern European words for court.²⁴ The Latin aula derived from the Greek term, but it did not appear in Latin literature until the end of the Republic, when Cicero employed the word in a letter to refer to the court of the Cappadocian king Ariobarzanes.²⁵ By the reign of Nero, aula was a commonplace term for the Roman imperial court, though it could also be used to mean palace (in this sense it was synonymous with palatium).²⁶ The related word aulicus, or ‘courtier’, makes frequent appearances in Suetonius’ On the Lives of the Caesars, written in the early second century.²⁷ Aula and its cognates continued to be understood and used in Late Antiquity to refer to the court, as evidenced by their appearance in fourth-century legal texts, biographies, histories, panegyrics, and letters.²⁸ But they were soon joined by a new term, comitatus, of which the Greek equivalent was στρατόπεδον (literally meaning

²¹ Duindam 1995: 1 n. 1; 2016: 171; Kazhdan and McCormick 1997: 172–3; Winterling 1997a: 13; Winterling 1999: 9 (the latter two citations feature extensive literary references). ²² Kelly 2022a: 7–8. Cf. Duindam 2016: 157–8, who perhaps gives too much emphasis to the physical location when he defines the court as ‘the enclosed space as well as the social circle around the ruler’. The spatial dimension is better captured by Vale 2001: 31: ‘It was the prince’s environment, both a place, normally of unfixed location, and an assemblage of people.’ ²³ For example, when Tiberius moved from Rome to Capri (George 2022: 244–6; Hekster 2022: 482–3). There are more dramatic examples in Late Antiquity, when emperors who were previously resident in the East moved to the western provinces (Moser 2018: 278–87 on Constantius II and Matthews 1975: 223–52 on Theodosius I). For the principle in other societies, see Burke 1992: 90–1 on the effects of the transfer of Louis XIV’s court to near-permanent (but not exclusive) residence at Versailles. ²⁴ LSJ s.v. αὐλή I–III (court in the spatial sense), IV (royal court); OLD s.v. aula¹ 1–3a (court in a spatial sense), 4a (royal court); Winterling 1999: 196, 203; Kelly 2022a: 5–6. ²⁵ Cic. Fam. 15.4.6; Wallace-Hadrill 1996: 283; Winterling 1999: 197. ²⁶ Sen. Tranq. 6.2; Winterling 1999: 197–8, 201–3. Cf. Schumacher 2001: 332–3, who argues against this meaning. For palace, see OLD s.v. aula 1.3a: ‘a royal or noble residence, palace, hall’ and OLD s.v. palatium 2a. ²⁷ OLD s.v. aulicus -a -um; Winterling 1999: 112, 200. A key example is Suetonius’ reference (Calig. 19.3) to stories heard ‘from intimate courtiers’ (ab interioribus aulicis). For Late Antiquity, see Prosp. Tiro 1358, s. a. 448; Sid. Ap. Ep. 1.5.10. ²⁸ Winterling 1999: 200; Smith 2007: 168. Legal texts: C. Th. 13.3.12 (379), 6.30.12 (390). Literary sources: HA Marc. 6.3, refers to the ‘courtly dignity’ (aulico fastigio) supplied to Marcus Aurelius when he moved into the domus Tiberiana; Amm. 21.15.4 describes the reactions of the most senior people ‘in the royal court’ (in aula regia) after the death of Constantius II; Symm. Ep. 9.88 refers to the influence of a man ‘in the court’ (in aula). See further Schlinkert 1996: 461 n. 21 for a list of late Roman texts.

6

   

‘army camp’).²⁹ In the early empire, comitatus specifically referred to the emperor’s travelling retinue, usually (but not exclusively) when he was on campaign, and he was accompanied by senators and equestrians who were given the special title of comes for the duration of the expedition.³⁰ However, the increasingly peripatetic nature of imperial rule, especially in the latter decades of the third century, meant that comitatus was used to describe the court (or more precisely, courts) from the period of the First Tetrarchy onwards.³¹ Even when emperors became primarily sedentary rulers once again—as, for example, during the reigns of Arcadius (395–408) and his son Theodosius II (408–50), when the court rarely left Constantinople— comitatus could still be used alongside aula to describe the court.³² Comitatus no longer carried with it the connotation of mobility, especially since Arcadius and Theodosius II did not lead their armies on campaign, although it still effectively captured the sense of the court as a social community.³³ Sometimes the words for palace as a physical structure—palatium in Latin and παλάτιον in Greek—could indicate the court community. We find this sense of palatium in Martial’s Epigrams, as Fuscus is told that ‘the Palace praises you’ (te Palatia laudent) and Apollo ‘let the Palace cherish and love you’ (Palatia te colant amentque).³⁴ In the Constantinian era, palatium was used as a synonym for comitatus, so that some soldiers recorded that they served in sacro comitatu, while others stated that they did so in sacro palatio; administrators could be termed palatini who held office intra palatium; and senators were described as comites intra palatium.³⁵ These terms should be understood as referring to service at court or in the imperial bureaux attached to the court, rather than in a specific palace, because there was no one fixed imperial residence in the early fourth century. Indeed, many palatini would not even have worked in one of the major

²⁹ Destephen 2016: 109–11. For στρατόπεδον, see Jul. Ep. ad Ath. 278B, Ep. 19, 451D (ed. Wright). ³⁰ OLD s.v. comitatus² 1c. To this entry, one should add a baggage plaque reading ‘from the retinue of the emperor Domitian Augustus Germanicus’ (ex comitatu | Imp(eratoris) Domitiani Aug(usti) | Germanici) (CIL V 7506) and the inscribed career of a centurion which records that he was ‘enrolled in the retinue of the emperor Commodus Augustus, Pious, Fortunate’ ([adl]ecto in comitatu Imp(eratoris) | [Com][[modi]] Aug(usti) Pii Fel(icis)) (AE 1939, 48). For the concept, see also Hurlet 2018: 279. ³¹ Schlinkert 1996: 460–1; Smith 2007: 168, 188–9, 196–203. See especially two documents dated to 293: SB 18.13851 (a reference to an official in the comitatus of Galerius); CJ 7.67.1 (a legal ruling in which Diocletian and Maximian refer to their comitatus). ³² Noethlichs 1998: 15–16. For example, comitatus appears in CJ 12.40.4 (issued at Constantinople in 400), CTh. 6.15.1 (issued at Constantinople in 413). On the integration of the Byzantine court into the city of Constantinople, see Noethlichs 1998: 17–18; McCormick 2000: 136–42; Destephen 2016: 68–77. ³³ On the Theodosians and Constantinople, see Diefenbach 1996; Croke 2010; McEvoy 2020a. Arcadius and Theodosius II only made rare journeys outside Constantinople, mainly to selected cities in Asia Minor (Destephen 2016: 84–105). ³⁴ Mart. 7.28.5, 9.42.5 (we owe these references to Ben Kelly). This meaning of palatium in the sense of court is not admitted by the OLD, but seems clear from the context. ³⁵ CIL III 6194 (in sacro comitatu); AE 1922, 72 (in sacro palatio); CTh. 6.22.1 (318/326) (intra palatium), 6.35.1 (314) (palatini); CIL VI 41426 (intra palatium).



7

imperial cities; their title and attachment to the court were marks of prestige.³⁶ The use of palatium and cognate terms in the early fourth century, at a time when there were actually several emperors and courts, may initially appear counterintuitive, but it reflects the fact that wherever the emperor and his court were based was considered to be a palace.³⁷ The subsequent long-term entrenchment of the emperor and court in the city of Constantinople meant that the words for palace eventually became the preferred terms for the court, though the use of aula and comitatus did not die out.³⁸ This necessarily brief survey shows that the Latin and Greek words used by the Romans for court changed over time between the first century  and the sixth century  to reflect the different phases of the court as an institution. But these linguistic developments do not alter the fact that the court remained throughout this period a social phenomenon, a community centred on the emperor.

3. The Scholarly Context Scholarly analysis of the Roman imperial court in the Principate as an institution has not been considerable when compared with research on other aspects of Roman politics and political culture, such as the constitution, the administration, the careers of senators and equestrians, and, more recently, the dynamics of imperial representation. Any analysis of the early imperial court rests on the foundations provided by two German scholars writing in the late nineteenth century, Theodor Mommsen and Ludwig Friedländer. Mommsen’s Römisches Staatsrecht featured a short chapter on ‘Court and Household’, which examined the emperor’s amici, his formal audiences, and the role of his slaves and freedmen.³⁹ Although the court proved difficult to incorporate into Mommsen’s work because of its legal and constitutional focus, he nevertheless recognized its significance to the Roman imperial state.⁴⁰ Friedländer addressed the court in his Darstellungen aus der Sittengeschichte Roms in der Zeit von August bis zum Ausgang der Antonine, which was subsequently translated into English as Roman Life and Manners under the Early Empire.⁴¹ Like Mommsen, Friedländer ascertained the importance of the court as an institution, and engaged in a detailed examination of the interactions between the emperor and courtiers at receptions and banquets, ³⁶ MacMullen 1964: esp. 307. ³⁷ As already formulated by Cassius Dio in the third century (53.16.6). ³⁸ McCormick 2000: 136. In the sixth century, comitatus described the court of the noncampaigning Justinian in CJ 7.63.5 (520); for the continued use of αὐλή, see Joh. Lyd. de mag. 2.6.2. For the language of the Byzantine period in the ninth to thirteenth centuries, see Kazhdan and McCormick 1997: 172–5, 185–7. ³⁹ Mommsen 1887–8: II.2, 833–9. ⁴⁰ Winterling 1997b; Winterling 1999: 12–15. ⁴¹ Friedländer 1862–71 (German first edition); 1919–21 (German ninth edition); 1908–13 (English translation).

8

   

the role and influence of slaves and freedmen in the household, and the emergence of a court establishment.⁴² His analysis was broader than Mommsen’s in many ways and included a novel section on the court’s ‘influence on form and manners’, a theme which is now a familiar part of aulic studies.⁴³ While subsequent twentieth-century scholarship devoted attention to aspects of the imperial court delineated by both Mommsen and Friedländer, such as the slaves and freedman and the emperor’s amici, this rarely formed part of an examination of the court as an institution.⁴⁴ One reason for this may be that Norbert Elias’ fundamental analysis of the court as a social phenomenon, Die höfische Gesellschaft, a revised version of his 1933 Frankfurt Habilitation, was not published until 1969 (and subsequently translated into English as The Court Society in 1983).⁴⁵ Although focusing primarily on the court of Louis XIV of France, Elias’ work provided a model of a royal court that could be applied to, or tested against, other periods and cultures. But the fact that Elias’ work was not published until 1969 cannot account for the whole picture. Indeed, it would be accurate to say that Mommsen’s focus on constitutional, legal, epigraphic, and prosopographical aspects of the imperial household and administration was much more influential on subsequent scholarly generations than Friedländer’s treatment of the social world of the court.⁴⁶ Fergus Millar’s magisterial 1977 work, The Emperor in the Roman World, although it deals with many elements of government and administration that we would consider part of the court, does not examine the court itself as an institution, and there is no entry for ‘court’ in the index.⁴⁷ One important exception to this overall trend was Andreas Alföldi’s fundamental, and still unsurpassed, examination of court ceremonial in his monograph Die monarchische Repräsentation im römischen Kaiserreiche, which encompassed both the Principate and Late Antiquity.⁴⁸ Important studies of the Roman imperial court that took it seriously as an institution in its own right did not appear until later in the twentieth century.⁴⁹ Robert Turcan’s 1987 book, Vivre à la cour des Césars, followed in the tradition of Friedländer in its assiduous collection of ancient evidence to illuminate aspects of

⁴² Friedländer 1908–13: I, 30–98. ⁴³ See, for example, Duindam 2003: 287–97; Duindam 2016: 273–81. For an assessment of Friedländer’s work, see Winterling 1999: 15–18. ⁴⁴ On this phase of scholarship, see Winterling 1999: 23–32. ⁴⁵ Duindam 1995: 9–11; Mennell 2006. In this volume, we cite Elias’ The Court Society from the 2006 Dublin University Press edition of his collected works. ⁴⁶ Winterling 1999: 12–28; Wallace-Hadrill 2011: 92–5. ⁴⁷ For example, the discussion of ‘Entourage, Assistants and Advisers’ (Millar 1977: 59–151) is fundamental to all future work on the court, but it is not framed through the lens of the court as an institution. ⁴⁸ Alföldi 1970 (a republication of articles originally published in the 1930s). For an assessment of this work, see Winterling 1999: 28–32; Canepa 2009: 149–53. ⁴⁹ For a comparable survey of the impact of Elias in the Hellenistic world, see Strootman 2014: 13–15.



9

court life.⁵⁰ But it was a decidedly old-fashioned work, which did not engage with Elias and the new field of aulic studies. This was left to Andrew Wallace-Hadrill’s insightful 1996 article in the Cambridge Ancient History and Aloys Winterling’s fundamental 1999 monograph Aula Caesaris.⁵¹ In each case, the stimulus for examining the court came from quite different directions. Wallace-Hadrill, a classical scholar, was inspired by his study of the imperial Lives written by the second-century ab epistulis C. Suetonius Tranquillus, during which he uncovered the centrality of the court to Suetonius’ thinking about emperors and imperial power.⁵² Winterling, on the other hand, began his career as a historian of Early Modern Europe, writing a PhD dissertation on the court of the Electors of Cologne between 1688 and 1794, before turning to ancient history for his Habilitation, which became Aula Caesaris.⁵³ Both Wallace-Hadrill and Winterling were influenced by, and formed part of, the flourishing wave of scholarship on monarchical courts stimulated by Elias, and which engaged critically with the model of the court society that he proposed.⁵⁴ The studies of all three scholars stimulated the excellent 2007 edited volume, Court and Court Societies in Ancient Monarchies, overseen by Anthony Spawforth, and the superb 2011 special edition of the American Journal of Philology edited by David Potter and Richard Talbert on Classical Courts and Courtiers. These works tackled key questions about the concept and function of courts, but also integrated the analysis of the Roman court into an ancient cross-cultural perspective.⁵⁵ There has now appeared an important two-volume study of the Roman imperial court from Augustus to the First Tetrarchy, which consists of interpretative essays and a sourcebook, under the editorship of Benjamin Kelly and Angela Hug.⁵⁶ This project not only takes stock of the new wave of scholarship on the court which has appeared since the 1990s, but also outlines new directions for future research. The study of the Roman imperial court of Late Antiquity might initially seem to have followed a different trajectory. The survival of a large corpus of Greek and

⁵⁰ See the prefatory comments in Turcan 1987: 9–10. ⁵¹ Wallace-Hadrill 1996; Winterling 1999. ⁵² Wallace-Hadrill 1995 (second edition). The author’s reflections on his path to research can be found in Wallace-Hadrill 2011: 96–7. His work on the court in Suetonius had an impact on Roman historians prior to the publication of his 1996 article. See, for example, Jones’ biography of Domitian, which has two sections on the court and acknowledges Wallace-Hadrill’s advice and influence (Jones 1992: viii, 202). ⁵³ The earlier study is Winterling 1986. Prior to Aula Caesaris, he also published two important edited volumes (Winterling 1997d; 1998). ⁵⁴ On the post-Elias phase of aulic studies, see Duindam 1995: 6–9. For their discussion of Elias’ work, see Wallace-Hadrill 1996: 285, 294–5; Winterling 1997a. ⁵⁵ Spawforth 2007a; Potter and Talbert 2011. For the intellectual heritage, questions, and parameters of the first volume, see Spawforth 2007b: 1–10. See also the short but valuable Italian study by Pani 2014, which situates the Roman court in historical context (see especially Pani 2014: 3–16, with notes on 107–8), and two studies of the court in specific periods: the reign of Claudius (Michel 2015) and the Severan dynasty (Schöpe 2014). ⁵⁶ Kelly and Hug 2022.

10

   

Latin panegyrics from the later Roman Empire has meant that analysis of imperial ceremonial has occupied a central place in the field of Late Antiquity over the past fifty years. In works produced in the 1970s and 1980s, such as Alan Cameron’s Claudian: Poetry and Propaganda at the Court of Honorius, Sabine MacCormack’s Art and Ceremony in Late Antiquity, and Michael McCormick’s Eternal Victory, we have learnt to understand the vital role of speeches, ceremonies, and imagery in the political culture of the late Roman period.⁵⁷ There was no comparable flowering of interest in imperial ceremonial by scholars of the Principate, who were instead largely content to rely on the scholarship of Alföldi, if they addressed the issue at all. This has given the false impression that ritual and performance were less important in the Principate, when, in fact, as the work of Wallace-Hadrill has shown, they were absolutely essential to it.⁵⁸ Ceremonial, however, is but one aspect of the Roman imperial court, and one struggles, as with the Principate, to find cohesive studies of the late Roman court as an important institution in its own right before the late twentieth century. In A. H. M. Jones’ The Later Roman Empire 284–602: A Social, Economic, and Administrative Survey, published in 1964, there is no entry for ‘court’ in the index, merely ‘courts of law’, and the lemma for comitatus is glossed as ‘central government’.⁵⁹ In the body of the text itself, comitatus is described as both the ‘central administration’ and the ‘group of ministries which were attached to the emperor’s person and formed the central government’, despite Jones’ knowledge that the comitatus also included the imperial household.⁶⁰ This suggests that he only conceived of Roman institutions, in the vein of Mommsen, within a larger constitutional framework.⁶¹ This perspective received serious reconsideration in John Matthews’ brilliant study of 1975, Western Aristocracies and Imperial Court .. 364–425, in which the multiple imperial courts of Late Antiquity emerged as important institutions in their own rights. These courts, primarily located in the provinces, offered pathways to power for provincials that were quite distinct from membership of the senate.⁶² In the 1990s, the more theoretically minded wave of court scholarship prompted by Elias had an impact on both German and English-language scholarship. In Germany, this can be seen in Winterling’s edited collection Comitatus and several important articles by Dirk Schlinkert,⁶³ and in the Anglophone world, with the publication of significant chapters in the Cambridge Ancient History by Christopher Kelly and Michael McCormick.⁶⁴ These studies not only paid closer

⁵⁷ Cameron 1970; MacCormack 1981; McCormick 1986. ⁵⁸ Wallace-Hadrill 1982; 1996: 285–95. ⁵⁹ Jones 1964: 1489 (‘comitatus’), 1491 (‘courts of law’). ⁶⁰ Jones 1964: 49, 366–7. ⁶¹ For Jones’ lack of interest in the court, see Smith 2007: 163 and Whitby 2008: esp. 65–8, and for the influence of continental European scholars on Jones, see Rebenich 2008: 45–53. ⁶² See especially Matthews 1975: 41–55, 76–81 on Valentinian I. ⁶³ Winterling 1998; Schlinkert 1996; 1998; 2002. ⁶⁴ Kelly 1998; McCormick 2000.



11

attention to the ceremonial world of the court than Jones had done, but also attempted to define the court community. The work of Elias and sociological approaches to the court had greater influence on German than Anglophone scholarship in this period. However, two important studies by Rowland Smith in 2007 and 2011 have applied, tested, and engaged with Elias’ model of a court society.⁶⁵ This is not to say, of course, that the only way of approaching the Roman imperial court is through the scholarship of Elias,⁶⁶ but it cannot be doubted that the field of aulic studies which his work inspired has provided Roman historians with valuable ways of conceptualizing the imperial court.

4. Ritual and Spatial Dynamics Even Rome herself, the mistress of nations, in a transport of extravagant joy at your proximity and in an attempt to get a glimpse of you from the summits of her own mountains, the closer to sate herself with your countenances, advanced as near as she could to get a look. Indeed, she has sent the leaders of her senate, freely imparting to the city of Milan, most blessed during those days, a semblance of her own majesty, that the seat of imperial power could then appear to be the place to which each emperor had come.⁶⁷ Thus did a Gallic orator, addressing the Emperor Maximian on the occasion of his birthday in 291, describe the reaction of Rome to the arrival and meeting of the Augusti Maximian and Diocletian in Milan the previous December. The peripatetic nature of the imperial court was one of the most significant changes to take place in Late Antiquity, as we have foreshadowed above. Of course, even in the early empire, emperors had never been tied to the Palatine and the city of Rome: they sojourned in Italian villas, embarked on provincial tours, or led their armies on campaign. While abroad in the provinces, emperors and their courts were quartered in important provincial cities, such as Sirmium in Pannonia or Antioch in Syria.⁶⁸ The time spent at these and other regional cities became more pronounced in the middle decades of the third century, when emperors had to devote

⁶⁵ Smith 2007; 2011. Note especially the comments of Spawforth 2007b: 7 of the influence of German scholarship. For the Hellenistic world, the works of Strootman 2014 and Erskine et al. 2017b represent a comparable step forward. ⁶⁶ See, for example, considerations of the value of Elias’ approach in Duindam 1995; 2004. ⁶⁷ Pan. Lat. 11(3).12.1–2: ipsa etiam gentium domina Roma immodico propinquitatis uestrae elata gaudio uosque e speculis suorum montium prospicere conata, quo se uultibus uestris propius expleret, ad intuendum cominus quantum potuit accessit. lumina siquidem senatus sui misit beatissimae illi per eos dies Mediolanensium ciuitati similitudinem maiestatis suae libenter impartiens, ut ibi tunc esse sedes imperii uideretur quo uterque uenerat imperator (trans. Rodgers in Nixon and Rodgers 1994). ⁶⁸ Millar 1977: 24–53.

12

   

themselves to protecting three key frontier regions—Germany, the Balkans, and Syria—which were under constant threat from foreign invasion and often also from internal rebellion. The emperors’ presence in these areas was expected and demanded by armies and provincials alike, who valued and celebrated their connection with their imperial protectors.⁶⁹ This meant that by the time of Diocletian and Maximian, true rivals to Rome as ‘the seat of imperial power’ (sedes imperii) had emerged, as our panegyrist observed.⁷⁰ For the Tetrarchs began the transformation of these regional bases into ‘imperial cities’ which were designed to be longer-term residences for their courts, a change that came to define interaction between centre and periphery in the fourth century.⁷¹ As Verena Jaeschke and Caillan Davenport explore in Chapter 3, the development of these cities, such as Nicomedia and Thessalonica in the East and Trier and Milan in the West, signified that there were now multiple imperial cities designed to house the emperor and his comitatus between their campaigns. In these principal residences, the emperors constructed palaces, circuses, and new imperial bath complexes in emulation of the construction of those in the city of Rome.⁷² Many of these cities also had their walls extended and they were equipped with new urban infrastructure, such as warehouses. The specific placement of the palaces within the cities was largely determined by pre-existing structures. But these residences were enlarged, remodelled, and embellished with colonnades, porticoes, arches, and monumental façades which expressed imperial power, such as the Tetrapylon of the Elephants in Antioch or Galerius’ monumental arch in Thessalonica. When the emperor and members of his court were resident in these cities, their presence would be displayed to the people through formal adventus (‘arrival’) ceremonies and triumphs, festivals and religious celebrations, and their attendance at games and races.⁷³ As Hendrik Dey has remarked, processions were a ‘quintessentially urban phenomenon’, binding the emperor and court into the fabric of the city.⁷⁴ In monarchical societies, the construction of new imperial residences did not usually mark the end of the mobility of a royal court—and this certainly was not the case in the Roman

⁶⁹ Hartmann 1982: 155–69. Note, for example, that in the reign of Probus, the Balkan city of Siscia minted coins with the reverse legend SISCIA PROBI AVG(VSTI), ‘Siscia of Probus Augustus’, accompanied by the image of the personification of the city and its rivers (RIC V Probus 766). ⁷⁰ On the continuing significance of the city of Rome throughout the third century, see Davenport 2017; Machado 2019: 4–5, 46, 49–50, 65, 167. ⁷¹ For the problems inherent in calling these cities ‘capitals’, see Kelly 2003: 598; Grig and Kelly 2012: 7–8; McEvoy 2020b. We thus prefer the term ‘imperial cities’. ⁷² On the potential of monarchical residences to transform cities, see Strootman 2014: 55 (discussing the Hellenistic world, but with general applicability). ⁷³ Pan. Lat. 11(3).11.3–5 imagines the reaction of people of Milan to the arrival of Diocletian and Maximian. The classic study of the adventus is MacCormack 1981: 17–61. On third-century triumphs from the Severans to the Tetrarchy, see Haake 2017. ⁷⁴ Dey 2015: 61.



13

Empire—but it did usually signal a new phase in leadership.⁷⁵ The Tetrarchic construction boom was designed to show that Rome could be replicated throughout the empire, a message articulated in panegyrics such as that delivered in honour of Maximian in 291. The multiplication of emperors and the consolidation of the imperial college as the primary model of governance mean that we can no longer speak of a single imperial court in this period, but of multiple courts. This development did have earlier precedents, but in the third century it became much more common for there to be multiple Augusti and Caesars, such as during the joint reigns of Valerian and Gallienus (253–60) and Carus, Carinus, and Numerianus (283–5).⁷⁶ Diocletian’s Tetrarchy, with its system of two Augusti and two Caesars, although itself relatively short-lived (293–305), institutionalized and promoted collegiate rulership and unity on an unprecedented scale, through its administrative structures and text and iconography, from the famous porphyry statue groups to coins, medallions, inscriptions, paintings, and panegyrics.⁷⁷ The four emperors— Diocletian, Maximian, Constantius, and Galerius—never actually met together as an imperial college, either in a palace audience hall or in a public space. Instead, their unity had to be articulated solely through speeches, writings, or artistic depictions. Jaeschke and Davenport discuss how the audience halls in the principal imperial residences featured statues or frescoes of all four emperors, even though only one or two might be present on each occasion. When audiences were admitted to perform the adoratio ceremony before one emperor, they did so before all four rulers, ensuring that court ritual played a vital role in articulating the unity of the imperial college. The evolution of court ceremonial, as we observed above, is one of the most notable changes that can be traced across several centuries of Roman history. In the sixth century, the end of the period covered by this volume, Justinian’s magister officiorum Peter the Patrician authored the first known work on the subject, the Treatise on Court Ceremonial.⁷⁸ It is worth asking how this evolution and intensification came about. In Elias’ study of the court of Versailles, Louis XIV

⁷⁵ Duindam 2016: 159–68 discusses examples from monarchical societies across Africa, Asia, and Europe. ⁷⁶ On the creation of imperial colleges to deal with pressures on the frontiers, see Davenport 2017: 29–30. ⁷⁷ Rees 1993; Boschung and Eck 2006; Hekster 2015: 280–7. ⁷⁸ The fragments of Peter’s work can be found in the tenth-century Book of Ceremonies: Const. Porph. de cer. 1.84–95 (1.93–104), 2.51; Bury 1907: 212–13; Sode 2011: 163–4, 169–76; Moffatt and Tall 2012: xxvi–xxvii, 699; and see now the full reassessment by Feissel in Dagron and Flussin 2020: I, 64*–70*, IV.1, 471–81. The work is described by the Suda (π 1406) as On State Ceremonial (περὶ πολιτικῆς καταστάσεως) and by the scholia on the Basilika law code as The Treatise on Court Ceremonial (σύνταγμα τῆς τοῦ παλατίου καταστάσεως), the latter probably the more accurate (Laniado 1997: 406–7). Here and throughout this book, we give the numbering of chapters in the Book of Ceremonies according to Reiske’s 1829 Bonn edition first, followed by the revised numbering in Dagron and Flussin 2020 in brackets.

14

   

is portrayed as a king who used etiquette and ceremonial, and the hierarchy they created, as a way of maintaining control over his courtiers.⁷⁹ Roman emperors likewise made choices about where and how they held ceremonial events.⁸⁰ The Tetrarchs set the agenda in the way that their audience halls were designed to heighten the impact of the adoratio ritual. They sat on bejewelled thrones in the apses of the hall, framed by the veil of curtains, a ceremonial innovation discussed by Fabian Schulz in Chapter 10. But we also need to account for the fact that recent work on ritual and ceremonial in the Roman world has emphasized the considerable degree of negotiation involved between emperors and subjects.⁸¹ At court, the emperor’s friends and officials could experiment with forms of flattery in order to win the favour of the princeps. New ceremonial gestures are usually discussed in the context of the courts of ‘bad’ emperors such as Caligula, who encouraged senators to grovel and kiss him on the hands and feet, rather than on the lips as equals. However, as Christoph Michels explores in Chapter 1, experimentation could occur under other emperors, even Antoninus Pius, the paradigmatic civilis princeps of the second century. Marcus Cornelius Fronto, tutor to the emperor’s adopted sons Marcus Aurelius and Lucius Verus, not only flattered Pius by styling him dominus imperator (‘lord emperor’) and μέγας βασιλεὺς (‘great king’), but also promised to kiss the emperor on his chest and hands as an apology for missing his accession day celebrations. As Michels cogently argues, there is a real incongruity between the image of Pius as a ruler who shunned the pomp of the court and the behaviour of courtiers such as Fronto. This shows that the evolution of court ceremonial between the Principate and Late Antiquity was not solely dictated by the emperors themselves, and restores agency to the members of the court. Ceremonial innovations would not survive long-term if they were not accepted by the court community.⁸² Lucius Vitellius famously engaged in sycophantic acts of proskynesis before Caligula, but this did not become accepted court behaviour.⁸³ While Caligula’s court was, in general, hostile to the emperor’s desire to emphasize his superiority, in Late Antiquity asymmetry was widely accepted as the proper power dynamic between emperor and subject, as Christopher Kelly and Rowland Smith have shown.⁸⁴ However, Smith’s discussion has perhaps placed too much emphasis on the rise of army officers—who did not have the patience or the need to appear as a civilis ⁷⁹ Elias 2006: 90–3, 150–2. Note also Weiser 2003, a study of Charles II of England, in which, it is argued, the king used access to control his court. ⁸⁰ According to Peter the Patrician, in the sixth century, it was the emperor’s decision whether to appoint his comites of the scholae or cura palatii in the consistory or, as was more usual, in his apartments; he also recorded that it was known for candidati to be appointed without an audience (Const. Porph. de cer. 1.84, 86 [1.93, 95]). ⁸¹ For the early empire, see Wallace-Hadrill 1995: 165–6; Paterson 2007: 145–6; Davenport 2022. For Late Antiquity, see Van Nuffelen 2012; Pfeilschifter 2013: esp. 85–7, 111–17. On this theme in aulic studies in general, see Duindam 1995: 97–136. ⁸² Duindam 1995: 56, 111–12, 121. ⁸³ Suet. Vit. 2.5; Dio 59.27.5–6. ⁸⁴ Kelly 1998: 139–50; Smith 2007: 174–9, 214–20.



15

princeps—to the imperial office in the late third century as the crucial determining factor in driving change.⁸⁵ This neglects the fact that in previous decades and centuries, ceremonial experimentation at court was often driven by senators and equestrians, men like Fronto and Vitellius, rather than being imposed or directed by the emperor himself. Moreover, in recent years Matthew Canepa has rejected the idea that the ceremony of adoratio, which our ancient sources present as an innovation of Diocletian, was imported wholesale from Persia. Indeed, the Persians may well have adapted it from the Romans, where it emerged after the process of ritual experimentation described above.⁸⁶ The need to account for negotiation and adaption in the production of court ceremonial reminds us that even when we arrive in the world of Late Antiquity, we should not assume that ritual procedure was in any way static or fixed: indeed, it continued to change and develop into the Byzantine period.⁸⁷ Indeed, the persuasiveness of Elias’ model of royal control through ceremonial, and its applicability to other monarchical societies, has been questioned by Duindam, who has pointed out that ceremonial could also place very real restrictions on the kings themselves.⁸⁸ An example of Roman rulers who served mainly symbolic and ceremonial functions comes with the extraordinary series of child emperors who wore the purple from the late fourth to mid-fifth centuries. As Meaghan McEvoy has argued, these imperial minorities could not (and did not) usually command armies, but they could be presented to the court and the world at large in any number of ceremonial contexts, swaddled in an aura of ritual and majesty.⁸⁹ The acclamation of these child emperors was to the advantage of, and was often engineered by, senior generals such as Stilicho and Aetius, who exercised control over the court and the government at large. In Chapter 8, McEvoy argues that these and other magistri militum, who are usually described as shadowy ‘powers behind the throne’, were in fact extremely visible at court, to the extent that they were widely acknowledged and celebrated as sharers in the emperor’s power. This state of affairs, McEvoy points out, could only have come about with the support of members of the court, whose acquiescence to these arrangements reinforced the power of the magistri militum. The power-sharing relationship between emperors and magistri militum, established under minority regimes, became so entrenched that it even persisted under emperors who came to ⁸⁵ We of course do not wish to dismiss the importance of these soldier emperors and their impact on the court (see, for example, Chapter 5 by Benjamin Kelly), but would like to emphasize that their selection was not the only factor in driving ceremonial change. ⁸⁶ Canepa 2009: 149–53, followed by Bang 2011: 104–5; Davenport 2022: 308. ⁸⁷ Peter the Patrician notes developments in ceremonies over time, including changes in the location of imperial acclamations from the Hebdomon to the Hippodrome (Const. Porph. de cer. 1.91 [1.100]) and palace architecture resulting in modifications to rituals under Justinian (Const. Porph. de cer. 1.86 [1.95]). For changes through to the end of the Byzantine period, see the fourteenth-century work of Pseudo-Kodinos (Macrides et al. 2013: 12–14). ⁸⁸ Duindam 1995: 133–6. For the same idea in the Roman world, see Brown 1992: 56. ⁸⁹ McEvoy 2010; McEvoy 2013.

16

   

the throne as capable adults.⁹⁰ This is shown by the continued prominence of the eastern general Aspar, who even rode in Leo I’s carriage during his entrance into Constantinople in 457. The public nature of such processions meant that they gave the people at large the opportunity to observe the emperor. Pliny paints a vivid picture of men, women, and children in the city of Rome clamouring to catch sight of Trajan on the day of an imperial distribution of money (congiarium).⁹¹ Personal audiences with the emperor were always restricted and regulated, though the nature of such controls changed over time. Augustus established the practice of ‘public receptions’ (promiscuae salutationes), which could be attended by anyone, and these were a regular occurrence on holidays, such as New Year’s Day.⁹² These receptions were hardly intimate encounters with the emperor given the large crowds they attracted, but there does occasionally seem to have been the possibility of speaking to the emperor while presenting a petition.⁹³ By the reign of Diocletian, however, such audiences had been curtailed altogether.⁹⁴ Only the most privileged were admitted to the palace at Milan in 290 to gaze upon Diocletian and Maximian— everyone else had to be content with the glimpses during the adventus ceremony.⁹⁵ Emperors came to prefer large public audiences outside of the palace itself, as shown by the increasing role of the Hippodrome in Constantinople as a locus for viewing the emperor from the fifth century onwards.⁹⁶ The emperor never appeared in the kathisma, the imperial box, alone, but was accompanied by leading members of the court, including the quaestor sacri palatii and the magister officiorum.⁹⁷ All encounters with emperors, whether in city streets or audience halls, were regulated by his guards, who form the subject of Chapter 2 by Christian Rollinger. Their presence was apparent from the beginning of the Principate, as they were regarded by Tacitus as one of the key attributes of a monarchical court. He wrote that Tiberius controlled the Praetorian guard and possessed ‘watchmen, arms, and other symbols of the court; he was accompanied by the soldiery to the forum and soldiery to the senate house’ (excubiae, arma, cetera aulae; miles in forum, miles in

⁹⁰ For the influence of child emperor government on the last generation of western emperors, see McEvoy 2013: 324–9; 2017. ⁹¹ Plin. Pan. 26.1. ⁹² Suet. Aug. 53.2. ⁹³ Crowds: Fronto, Ad M. Caes. 5.45 (vdH², p. 77). Petitions: Suet. Claud. 37.1. ⁹⁴ The late Roman evidence, such as the struggles of Porphyry of Gaza and Synesius of Cyrene to obtain audiences with Arcadius, suggests that there were no longer any special days in which anyone could come and present a petition to the emperor (see Pfeilschifter 2013: 341–2). Procopius, Anecd. 13.1, 15.11, emphasizes Justinian’s accessibility, yet elsewhere in the work, it is clear that petitions were still given to the emperor through officials known as referendarii (14.11). ⁹⁵ Pan. Lat. 11(3).11.1–3; see Magdalino 2015 on the Byzantine period. ⁹⁶ Cameron 1976. For ceremonies in Constantinople providing an opportunity to view the emperor, see Pfeilschifter 2013: 333–52. ⁹⁷ Const. Porph. de cer. 1.92 (1.101); Feissel in Dagron and Flusin 2020: IV.1, 579–80; Cameron 1976: 128–9.



17

curiam comitabatur).⁹⁸ Republican magistrates had always had large retinues, but there was something about the military configuration of Tiberius’ entourage that Tacitus regarded as inherently monarchical and courtly.⁹⁹ The Praetorian guard played an important role here. From Augustus onwards, there was a Praetorian cohort stationed in the domus on the Palatine to accompany the emperor when he went out into the city.¹⁰⁰ Imperial women such as Agrippina and Julia Domna had their own retinues of German guards or Praetorians to accompany them as an outward sign of their importance within the court.¹⁰¹ Following Constantine’s disbandment of the Praetorians, these duties were taken up by the scholae palatinae and, from the reign of Leo I, the reorganized excubitores assumed an important role as a palace guard.¹⁰² However, as Rollinger shows, the appearance of the guards underwent a change that mirrored the transformation of the image of the emperors themselves. Constantine’s scholae palatinae looked quite different from the Praetorian guards, wearing armour glittering with jewels that signified their privileged monarchical status. As with the transformation of Tetrarchic imperial residences, Rollinger argues that this was a change designed and instituted by the imperial administration itself.¹⁰³ Jeroen Duindam has proposed that Early Modern court ceremonial should be divided into several different categories, two of which are particularly relevant to the Roman Empire. These are domestic ceremonial, the everyday interaction between emperors and members of his court, and dynastic state ceremonial, which in the Roman world consisted of major events, such as accessions, funerals, adventus ceremonies, and triumphs.¹⁰⁴ Dynastic state ceremonial—the focus of Rollinger’s essay on the guards and Audrey Becker’s examination of successions in Chapter 4— played an important role in marking moments of change both small and large, such as the arrival of the court into a new city, the celebration of a usurper’s defeat, the death of an emperor, and the accession of a new ruler. The transition between emperors could be a time of acute uncertainty and crisis, even when the emperor had one or more male relatives, as Becker notes in her contribution. The performance of rituals which followed a prescribed and expected pattern helped to provide a sense of reassurance to the court and to the population at large.¹⁰⁵ Up until the early ⁹⁸ Tac. Ann. 1.7.5 (trans. Davenport); Pani 2014: 18. ⁹⁹ On the Republican period, see Östenberg 2015. For the argument that Tacitus’ account indicates his disapproval of changes to established Republican practices, see O’Sullivan 2011: 67. ¹⁰⁰ Millar 1977: 61. ¹⁰¹ Tac. Ann. 13.18.3; Suet. Ner. 34.1; Dio 61(61).8.4–6. ¹⁰² Frank 1969; Destephen 2016: 233–47. For the excubitores, see Croke 2005b. ¹⁰³ On the similarly ornate uniform of the excubitores, see Croke 2005b: 146–9. ¹⁰⁴ Duindam 1995: 102–4, drawing upon Giesey 1987. Duindam also identified diplomatic ceremonial as its own distinct category. This is a topic that demands further study in the Roman context. The evidence for the Principate is relatively limited, but we are better informed about the forms of diplomatic ceremonial in Late Antiquity and Byzantium: see Gillett 2003: 251–9; Canepa 2009: 128–38; Pohl 2013; Becker 2018. ¹⁰⁵ MacCormack 1981: 12–13. See further Bell 1992: 92, 103 on the importance of formal ritual in creating order and certainty.

18

   

fourth century, Roman imperial funerals were usually entirely public occasions, with processions and eulogies in the Forum Romanum, followed by military manoeuvres around the pyre, and cremation on the Campus Martius. This procedure changed with the death of Constantine in 337.¹⁰⁶ The military still played a significant role: the palace guards were the first to show extravagant grief at Constantine’s death before escorting his body to the palace in Constantinople, and they regulated the daily homage paid to the corpse by military and civilian elites, including members of the imperial family. This, as Becker points out, provided a sense of reassurance as everyone waited for the arrival of the emperor’s sons. But over the course of Late Antiquity, the funeral changed into a Christian ceremony, taking place in a church, rather than being a full military display on the Campus Martius or its equivalent in Constantinople, the Hebdomon. Christianity also transformed the ceremonial procedure of Roman imperial accessions, though once again this was a gradual process. The Praetorians and their prefects often took the decisive role in hailing a new ruler as imperator, a practice which first occurred with Caligula in 37. This continued into Late Antiquity, when emperors, beginning with Julian in 360, were often raised aloft on their shields by the soldiers at their acclamation. This practice, which was integrated into the proclamation ceremonies at Constantinople, was first firmly attested with Anastasius I in 491.¹⁰⁷ The continued involvement of guards as acclamatory agents provided a sense of continuity and reassurance that a new emperor had the support of the army, even when based in the city on the Bosphorus and no longer leading the army in person. The consolidation of Constantinople as the pre-eminent imperial city and seat of the eastern emperor meant that both its patriarch and its population came to play increasingly important roles in legitimating imperial authority.¹⁰⁸ Phocas received both honours—crowning by the patriarch and raising on a shield in 602—as Rollinger’s chapter explores, after which the shield may have fallen into abeyance, at least until the tenth century.¹⁰⁹ The military remained an important part of imperial ceremonial, providing practical protection as well as courtly pomp

¹⁰⁶ See Price 1987 on the evolution of funerals. ¹⁰⁷ For Caligula’s acclamation by the praetorian prefect, see Tac. Ann. 6.50.4. For Julian, see MacCormack 1981: 194–6, 241; Enßlin 1942. On Anastasius I, see Const. Porph. de cer. 1.92 (1.101), and see further the description of the ceremony for Justin II in Coripp. Iust. 2.137–58. ¹⁰⁸ See, inter alia, MacCormack 1981: 243–8; Diefenbach 1996: 41–3; Dagron 2003: 80–3; Pfeilschifter 2013: 378–83; Kaldellis 2015: 102–17. The idea that the support of different groups was required to confer legitimacy on the emperor is known as Kaiserakzeptanz (‘imperial acceptance’), as articulated by Flaig 1992 (second edition, 2019). ¹⁰⁹ The continuance of the practice later in Byzantine history has been the subject of considerable debate: although there is much artistic evidence (Walter 1975), it was assumed that it fell into abeyance until the thirteenth century (Ostrogorsky 1955). However, now scholars are more likely to posit a continuous history, although the lacuna from the seventh to tenth centuries in the sources remains (Macrides et al. 2013: 418–21).



19

(to echo Tacitus’ description), but it is clear that over time, the nature of these rituals changed to accommodate a vital new component: the Christian God.¹¹⁰ Rather than interpreting these developments as a straightforward and linear development, we need to take into account the negotiation that took place in choosing an emperor and in deciding upon the ceremonial that gave him the authority to rule.¹¹¹ In this negotiation, both the imperial candidates and the ‘leaders of the court’ (archontes, to use the terminology of Peter the Patrician)¹¹² had prominent roles to play, as Becker shows in her essay. In 491, the Emperor Anastasius was the first emperor to be crowned by the patriarch in the Hippodrome, a move which Becker argues was decided by leading court officials in order to demonstrate that it was they, not the military, who controlled the accession of an emperor. This veneer of consensus would not be maintained when Anastasius himself died in 518. The court could not agree on a successor, and different archontes put forward various candidates and solicited support from a range of popular and military groups. Despite these divisions, the patriarch and the Church establishment at large were not able to play a deciding role in selecting emperors.¹¹³ Their role was largely ceremonial. When the court was sufficiently organized and united—as happened after the demise of Justinian in 565—they presented a new candidate to the patriarch to be crowned and then have him acclaimed by the people. As Gilbert Dagron has argued, the Church was important ‘because it could give a degree of credibility to the oath of loyalty sworn by the dignitaries and the representatives of the people’, not because it played a decisive part in selecting the emperors themselves.¹¹⁴ This examination of the complex interaction between place, space, and ritual shows how the transformation of the Roman imperial court between the Principate and Late Antiquity was influenced and shaped by multiple agents. Roman emperors were not the master manipulators, pulling all the strings, as in Elias’ model of the court, but there was still potential for them to exert considerable influence (a potential which obviously depended on many factors: babies and children had limited capacity). Emperors could choose where to establish their principal residences, where and how they would stage ceremonial adventus and triumphs, and the location of palaces within urban landscapes. Factors such as the style of guard uniforms, which became more ornate and jewelled in the fourth century, or the precise layout of palace corridors and audience halls, cannot necessarily be attributed to the personal design choices of the emperor. But they ¹¹⁰ Diefenbach 1996: 48–9. ¹¹¹ This is a key theme of Dagron 2003: 54–83, esp. 60, 65. See also MacCormack 1981: 254 on flexibility in the ceremonial procedures. ¹¹² For example, Const. Porph. de cer. 1.92–3 (1.101–2); Feissel in Dagron and Flusin 2020: IV.1, 479. In Latin, the archontes are styled proceres (Coripp. Iust. 2.281). ¹¹³ Pfeilschifter 2013: 383 emphasizes that no conflict between emperor and bishop ever resulted in the fall of the emperor. ¹¹⁴ Dagron 2003: 82.

20

   

were shaped by his administration and household to reflect his aims and ambitions for power. Cities, structures, ceremonies, and uniforms were not the court, but they created the environment in which emperors and courtiers interacted with each other, and with the people at large. It is within this larger context that courtiers could play a role in influencing rituals, attempting to innovate by flattering the emperor in new ways, or pushing back against new ideas, for it was only with the consensus of the court community that change could happen. It is to the composition of this community that we now turn.

5. Individual and Community at Court Otho was an affable governor of the province and, as the first to align himself with Galba’s faction, hardly lacked initiative. As long as the war lasted, he was the most distinguished among those of Galba’s circle, and the hope of adoption, which he had embraced from the very beginning, took hold of him more intensely as the days went by. He had won the favour of most of the military, and the court (aula) was also inclined towards him because of his similarity to Nero.¹¹⁵ So wrote Tacitus of the hopes and dreams of M. Salvius Otho, the thirty-five-yearold governor of Lusitania, when the suicide of his former comrade Nero in 68 generated a new, and hitherto unprecedented, race for the imperial purple. Otho’s hopes lay not in staking a claim himself, but in attaching himself to the elderly and, at this point, childless candidate Ser. Sulpicius Galba. His formative years, spent in carousing and rivalry with Nero, meant that Otho was the court’s choice for princeps. The court thus emerges from this passage as a community that could, and did, express collective feelings, opinions, and ideas. A much later account of life under the Emperor Hadrian, written by the anonymous author of the Historia Augusta, shows that the court also had norms, standards, and expectations. The praetorian prefect Septicius Clarus and the emperor’s ab epistulis Suetonius Tranquillus were dismissed from their posts, so the HA recorded, because their behaviour towards the emperor’s wife, Sabina, was more familiar ‘than the deference required at the imperial court’ (reverentia domus aulicae postulabat).¹¹⁶

¹¹⁵ Tac. Hist. 1.13.4: Otho comiter administrata provincia primus in partis transgressus nec segnis et, donec bellum fuit, inter praesentis splendidissimus, spem adoptionis statim conceptam acrius in dies rapiebat, faventibus plerisque militum, prona in eum aula Neronis ut similem (trans. Davenport). ¹¹⁶ HA Hadr. 11.3 (trans. Davenport). OLD s.v. reverentia 1, ‘a feeling of restraint in the presence of a superior, etc., awe, deference, respect, etc.’. See also HA Marc. 29.7, which describes the court possessing pride (aulicam adrogantiam), and Sid. Ap. Ep. 1.9.1, in which Sidonius considers how to enter into the court’s favour (in aulam gratiae aditus).



21

One of the themes of this book is the capacity of the court community to articulate consensus. In Chapter 5, Benjamin Kelly analyses the court in terms of an ‘emotional community’ that sets standards for how its members should behave and express their feelings to one another. He argues that under Nerva and Trajan, and again under the Antonines, there were clear expectations that the court should not be a place of fear and anger, but one in which its members (including the emperor) regarded each other with love and devotion. Although these ideals continued into Late Antiquity, there were subtle but perceptible changes we can witness, especially during the period from the 350s to the 380s, which Kelly takes as a case study. Emperors such as Constantius II, Valentinian I, and Valens engaged unchecked in ferocious displays of rage. Ostentatious displays of affection, present in Fronto’s relationship with the Antonines, were often frowned upon. Kelly persuasively argues that there was no ‘normative consensus’ about what the appropriate behaviour and etiquette at court should look like in the late fourth century, since courtiers and the emperors had diverse expectations. The reason for this probably lies, at least in part, in the rise of career officers to the position of emperor, a change that began in the late third century. We should not reduce these ‘soldier emperors’ to brutish and uncultured stereotypes, but they did have a different ‘value system’, since they were men with the ‘capacity to make a public virtue out of professional violence’, as John Matthews has memorably put it.¹¹⁷ Indeed, it was often the case that soldier emperors brought with them to court a coterie of hardened officers whom they had known long before they assumed the purple.¹¹⁸ The courts of the fourth-century emperors examined by Kelly were thus melting-pots made up of senators from Rome, administrators from provincial schools, and military men elevated through the ties of barracks brotherhood. It may be no coincidence that the fourth-century ruler who came closest to expressing the emotional norms of the Antonine period was Julian, a popular and successful general, but a man with a scholar’s education and sensibilities, not a soldier’s soldier like Valentinian I. Julian, who modelled himself on the second-century Emperor Marcus Aurelius, espoused the behavioural ideals of the civilis princeps.¹¹⁹ Indeed, we should not take the behaviour of the wrathful late fourth-century emperors as emblematic of the entire period of Late Antiquity. For these military men ensured that their sons were raised in quite different ways from themselves, providing them with the best classical education possible. Gratian, the son of

¹¹⁷ Matthews 2000b: 438–9. On the background of the Pannonian emperors, Valentinian I and Valens, see Lenski 2002: 45–55, 92–5. ¹¹⁸ Lib. Or. 1.76 complains about the senate of Constantinople being drawn from the army, rather than the educated elite. Cf. Chapter 7 by Kevin Feeney, which explores the heterogeneous appointment policy of another former military man, Jovian. ¹¹⁹ Wallace-Hadrill 1982: 48. For the continuation of the ideal of the civilis princeps in Late Antiquity, see Matthews 1989: 235–6; Kelly 1998: 147–9; Hekster 2022: 493–4.

22

   

Valentinian I, was originally brought up in the military camps with his father, but was later entrusted to the tutorship of Ausonius to be transformed into the ideal prince.¹²⁰ We can also traces change over several generations of the Theodosian dynasty, from Theodosius I—an ‘army brat’, as Neil McLynn has described him¹²¹—to his grandson Theodosius II, whose upbringing and education ensured that he would behave as a pious Christian ruler.¹²² Indeed, Late Antiquity saw successive cases of this phenomenon, as military officers, often derided as violent and uneducated, came to the throne, but tried to ensure that their heirs were trained differently.¹²³ Herein lies the fundamental change between the world of the Principate and Late Antiquity, which can be traced to the transformation of the military command structure of the mid-third century. In the early Roman Empire, the court was marked by a consistency of aristocratic values, because the military high command was drawn from among the senatorial and equestrian elites. However, in the later Roman Empire, there was a separation between the military and administrative career paths and men from the ranks of the legions were elevated to the imperial purple and high office. Sidonius Apollinaris, for example, considered the established senatorial elite to be distinct from the ‘privileged military faction’ (praerogativa partis armatae) at court.¹²⁴ This meant that there were divisions—Kelly’s ‘fault lines’—that kept being exposed, as the values of senators and soldiers interacted, sometimes clashing, sometimes influencing each other.¹²⁵ It is to these individuals who made up the court community, and whose collective views established court norms or whose different opinions created divisions, that we now turn. In both early and late sources, from Suetonius to Sidonius Apollinaris, they are described as aulici—quite simply, the ‘people of the court’ or ‘courtiers’. The English word ‘courtier’ has the potential to mislead, conjuring up as it does images of powdered lords and ladies in ruffs and wigs, dancing and romping their way through Elizabethan or Regency England. Some classicists have indeed thrown up their hands in despair when confronted with the term ‘courtier’.¹²⁶ But its meaning is clear, as the Oxford English Dictionary states: ‘One who frequents the court of a sovereign; an attendant at court’.¹²⁷ In the Roman world, there were many different expressions and circumlocutions that corresponded to this meaning of courtier.¹²⁸ To focus on merely the most obvious

¹²⁰ McEvoy 2013: 66–7, 106–7. On Valentinian I’s army career, see Lenski 2002: 20–2. ¹²¹ McLynn 2005: 102. ¹²² Soc. 7.22; Soz. 9.1. ¹²³ For example: Valentinian I and Gratian; Theodosius I and Arcadius and Honorius; Justin I and Justinian. ¹²⁴ Sid. Ap. Ep. 1.9.2. ¹²⁵ For different aspects of these transformations, see Demandt 1980; Matthews 1989: 249; Matthews 2000b; Smith 2011: 139–40; Davenport 2019: 19–21, 591–9. ¹²⁶ For example, Whitmarsh 2007: 33–5 argues that the Severan author Philostratus was not a courtier, but then states: ‘whatever that might mean’ (35). ¹²⁷ OED s.v. courtier n.¹ 1a. ¹²⁸ Smith 2007: 168.



23

example, aulicus could be used either as a substantive in phrases such as ‘intimate courtiers’ (interioribus aulicis) and ‘courtiers either hateful or stupid’ (aulici vel stulti vel detestabiles), or adjectivally to refer to ‘court attendants’ (aulicis ministris), ‘court freedmen’ (liberti aulici), and ‘court wrestlers’ (luctatoribus aulicis).¹²⁹ All of these counted as members of the court, as did—to draw on the works of Marcus Aurelius and Tacitus—philosophers, doctors, and soothsayers.¹³⁰ We should, therefore, not assume that the term courtier automatically refers to aristocratic grandees; the word encompassed all those who formed part of the court community.¹³¹ Indeed, the guard units discussed by Christian Rollinger were a crucial aspect of this community, protecting and policing its membership. This is not to say that there was no differentiation among the members of the court society. As the reference to ‘intimate courtiers’ in Suetonius shows, the Romans certainly recognized, as did their counterparts in other monarchical societies, that courtiers had different levels of familiarity and favour with the emperor, and that this could change throughout a courtier’s life. Modern scholars of royal courts often speak of ‘inner’ and ‘outer’ courts, a conceptual distinction which more often than not reflects the organization of space in palaces.¹³² This terminology is difficult to apply wholesale to the Roman imperial court of the Principate, given that the Romans themselves never had such a distinction which mapped onto the layout of imperial residences.¹³³ The vocabulary of the Roman court, and the hierarchy within it, was quite different. The word amicus (‘friend’) appears often in our sources. But it does not always indicate special intimacy with the emperor, for it was not an official title, and indeed ‘the word was used with considerable elasticity’, as Ryan Wei and Benjamin Kelly have noted.¹³⁴ The emperor relied on the advice of his consilium, although once again the term is somewhat deceptive, because it was not a body with fixed membership; the emperor could invite members of the senatorial and equestrian orders to advise him and sit in judgement on cases as he saw fit.¹³⁵ Membership of the consilium may have varied, but there were some individuals who were called to serve much more frequently than others, and these were ¹²⁹ In order, the references are to Suet. Calig. 19.3; HA Aurel. 43.1; HA Ant. Pius 6.4, 10.5; HA Comm. 7.2; Suet. Ner. 45.1. On entertainers, see Pani 2014: 22. ¹³⁰ M. Aur. Med. 8.31; Tac. Ann. 1.7.5. For modern discussions, see Jones 1992: 23–6, 50–1 (on Domitian’s court), and more widely, Paterson 2007: 140–1; Pani 2014: 17–18; Kelly 2022a: 8. ¹³¹ See Llewelyn-Jones 2013a: 4, 35–40 for similar comments with reference to the Persian court, and Erskine et al. 2017a: xvii on the Hellenistic court. Cf. Duindam 2016: 235–7, who cautions against the term courtier because it suggests a ‘deceptive equivalence’ (236) across different societies. ¹³² Spawforth 2007b: 4; Duindam 2016: 168–200, esp. 168–71. ¹³³ Kelly 2022a: 8 argues for the use of these concepts because they capture the idea of hierarchy of the Roman court. Cf. the court of the Chinese Quin and Ming dynasties, which used clear terminology to distinguish between the inner (neichao) and outer court (waichao). See Duindam 2016: 190–1. ¹³⁴ Wei and Kelly 2022: 92–3. See further, Crook 1955: 22–4; Millar 1977: 111; Eck 2006: 69–70. ¹³⁵ Crook 1955: 104–6; Davenport 2012a: 800–1; Wei and Kelly 2022: 104–6. Cf. Drinkwater 2019: 69, who refers to the consilium anachronistically as a ‘privy council’ on the dubious grounds that it is a term ‘too useful to abandon’.

24

   

recognized both inside and outside the court as the emperor’s closest advisors. When Statius delivered his now lost poem in praise of Domitian’s German campaigns, he lauded the emperor’s most prominent amici. These included Vibius Crispus, who possessed the wisdom of Homer’s Nestor and the dignity of three consulships, and the distinguished M’. Acilius Glabrio, described as ‘a close associate of Caesar’s court’ (prope Caesareae confinis . . . aulae).¹³⁶ In the absence of a fixed court-based hierarchy of offices and status, other ways to rank courtiers emerged. One of the most evident displays of a hierarchy of favour came at the imperial salutatio, since, after the reigns of Augustus and Tiberius, courtiers were organized into groups to greet the emperor in a specific order.¹³⁷ Romans also frequently spoke of familiaritas, or intimacy, with the emperor.¹³⁸ This was never based solely on social status, official position, or even service on the consilium—the equestrian patron of the arts, Maecenas, was closer to Augustus than the majority of senators.¹³⁹ The emperor’s wives and family members may not have held any administrative posts, but they could nevertheless be enormously influential.¹⁴⁰ Romans used various terms to refer to these group(s) which shared the emperor’s familiaritas. We read of the emperor’s cohors, meaning ‘circle’ or ‘entourage’, into which courtiers could be admitted or from which they could be expelled.¹⁴¹ The word contuburnium (‘fellowship’) expresses a similar idea: Vespasian was infamously removed from Nero’s contuburnium after leaving or falling asleep during the emperor’s performances.¹⁴² The correspondence of M. Cornelius Fronto, expertly analysed by Christoph Michels in Chapter 1, enables us to see first-hand the anxiety felt by courtiers who wished to gain or maintain a position within an emperor’s select group of favourites. In Late Antiquity, the fundamental principle that intimacy with the emperor was based not solely on social status, but also on the ruler’s own favour, remained.¹⁴³ We can find many complaints about the influence of eunuchs, for instance, that echo the gripes about slaves and freedmen from the Principate.¹⁴⁴ Indeed, anxiety about inappropriate influence of servants and functionaries is a ¹³⁶ Courtney 1993: 160 presents the edition of the surviving four lines. Domitian’s advisors were later mercilessly satirized in Juvenal’s Fourth Satire, in which, rather than being summoned to discuss the preparations for his German campaigns, they were instead called to debate what to do with a large turbot. ¹³⁷ Davenport 2022: 300–1, contra Winterling 1999: 127–35. ¹³⁸ OLD s.v. familiaritas a; Hurlet 2018: esp. 278–9. ¹³⁹ Wallace-Hadrill 1996: 300–1; Pani 2014: 9, 28–31. We should resist seeing three fixed groups of imperial amici, as proposed by Winterling 1999: 166–94; for refutation of this idea, see Wei and Kelly 2022: 93–4. ¹⁴⁰ Hug 2022: 67. ¹⁴¹ OLD s.v. cohors 5b; Tac. Ann. 6.9.2; Suet. Galb. 7.1; Fronto, Ad M. Caes 4.1.3 (vdH², p. 54). ¹⁴² OLD s.v. contuburnium 2a; Suet. Vesp. 4.4; Hurlet 2018: 278–9 argues that the term represents a ‘spatial translation’ of the concept of familiaritas. Note also Sid. Ap. Ep. 1.4.1 on the sodalitas, or ‘close association’, enjoyed by his friend Gaudentius in the emperor’s house (OLD s.v. sodalitas 1). ¹⁴³ Leppin 2018b: 54–5. ¹⁴⁴ Matthews 1989: 274–7.



25

perennial aspect of court life in monarchical societies.¹⁴⁵ But the organization of courtiers did certainly change in Late Antiquity. The consilium was replaced by the consistorium, the emperor’s advisory council, the name of which derived from the fact that its members stood in the emperor’s presence.¹⁴⁶ The ex officio members of the consistorium included the emperor’s key officials, such as the quaestor sacrii palatii, the magister officiorum, the comes sacrarum largitionum, and the comes rei privatae, who were collectively termed the comites consistoriani.¹⁴⁷ The staff of their bureaux were known as palatini, ‘officials of the palace’.¹⁴⁸ These were not freedmen, as had been the practice in the Principate, but freeborn administrators, who could now rise through the palatine ranks to head up the bureaux.¹⁴⁹ The consistorium also consisted of the praetorian prefect (either praesentalis or the prefect of the region in which the court was currently located), the comes domesticorum, and the magister militum praesentalis, as well as past incumbents of these offices.¹⁵⁰ Cooperation between these officials and leading courtiers in the imperial household was important not only for the functioning of the imperial regime itself, but also for ensuring a peaceful transition between individual emperors, especially when there was no clear successor, as Audrey Becker explores in Chapter 4. From the reign of Constantine onwards, the emperor frequently appointed other senior administrators or senators to the consistory, giving them the title of comes intra consistorium (‘within the consistory’), intra palatium (‘within the palace’), or domesticus (‘household’).¹⁵¹ Comes was thus transformed from an honorific term to describe senators and equites who accompanied the emperor on campaign to an official title, formally bestowed by the emperor.¹⁵² In fact, the Emperor Constantine created an entire new order of comites, to which he appointed both senators and equestrians. They were divided into three grades—the ordo primus, secundus, and tertius—with the highest rank of comes ordinis primi being granted to both ex officio and appointed members of the consistorium.¹⁵³ The distinction of being made a comes ordinis secundi or tertii gave a senator or eques

¹⁴⁵ Duindam 2016: 215–20. ¹⁴⁶ Schlinkert 1996: 475; 1998: 138–9; Smith 2007: 198. As Färber 2014: 102, 106–7 points out, in the fourth century, the consistorium did not have a fixed location, but was convened where the emperor was in residence. On the workings of the consistorium in Late Antiquity, see the detailed studies of Elton 2009 and Harries 2013 on the reign of Theodosius II. ¹⁴⁷ Weiss 1975. For Julian and his comites consistoriani, see CTh. 11.39.5. ¹⁴⁸ Jones 1964: 104. ¹⁴⁹ Winterling 1997c: 164–5 identifies this as a key change between the Principate and Late Antiquity; see further Davenport 2019: 337–41, 584–5. ¹⁵⁰ Jones 1964: 333–4, 1134–5. See further Weiss 1975; Schlinkert 1996: 463–71. Cf. Gutsfeld 1998: 87, who is sceptical (but unnecessarily so) about the place of a praetorian prefect on the consistorium. ¹⁵¹ Comes intra consistorium: Q. Flavius Maesius Egnatius Lollianus (PLRE I Lollianus 5); comes intra palatium: L. Aradius Valerius Proculus (PLRE I Proculus 11); comes domesticus: Flavius Eugenius (PLRE I Eugenius 5). There was still some flexibility in emperors calling ad hoc advisors, or acting without consulting the consistorum, of course: see Leppin 2018b: 51 on the reign of Justinian. ¹⁵² Millar 1977: 117; Smith 2007: 181–2; Destephen 2016: 187–90. ¹⁵³ Eusebius, V. Const. 4.1.2; CTh. 12.1.26 (338), with Jones 1964: 104–6, 526–9; Scharf 1994.

26

   

a formal honorific attachment to the emperor and his court, even if they were not permanently part of the comitatus.¹⁵⁴ When Q. Aurelius Symmachus travelled to Trier in 369 as a representative of the senate to mark Valentinian I’s fifth year of rule, he joined the emperor during one of his expeditions, and was therefore made a comes ordinis tertii.¹⁵⁵ Senators who no longer commanded armies often took great pride in designation as a comes of the second or third order, as it enabled them to boast of following in the footsteps of their forefathers.¹⁵⁶ Most importantly, the favour of the emperor, and association with the court, was now recognized with an official title. In 372, the emperors Valentinian I and Valens established a formal order of precedence for all court, administrative, and senatorial offices.¹⁵⁷ Orders of precedence would henceforth become a major feature of the Roman imperial government, and they were regularly revised to accommodate new titles, ranks, and officials.¹⁵⁸ The creation of new senior military positions under the Constantinian dynasty was another major innovation which differentiated the world of the Principate and Late Antiquity, introducing new dynamics to both the consistorium and the court at large.¹⁵⁹ There were originally two such offices, of magister peditum and magister equitum, who commanded the emperor’s mobile field armies (known as comitatenses). They were joined later in the fourth century by an overall commander, the magister utriusque militiae, often simply referred to as the magister militum.¹⁶⁰ The senior general who was attached to the court, the magister militum praesentalis, was a member of the imperial consistorium.¹⁶¹ One might be tempted to envision these generals as the late Roman equivalent of the praetorian prefect, who exercised great influence at court in the Principate. But this would be a mistake. Firstly, they controlled different groups and resources. While the magister militum exercised authority over the entire army, the praetorian prefect’s role as the senior field army commander did not develop until the middle decades of the third century.¹⁶² Secondly, their social origin was different. In the early empire, praetorian prefects were usually men from the municipal or provincial aristocracies, promoted after following the combined equestrian military and administrative career, but magistri militum were career soldiers and, over time, drawn increasingly from non-Roman backgrounds.¹⁶³ They thus formed

¹⁵⁴ Smith 2011: 143–5. ¹⁵⁵ CIL VI 1699; Matthews 1975: 17, 32–3. Sidonius Apollinaris also bore the title of comes when in attendance at Majorian’s court (Ep. 1.11.13). ¹⁵⁶ Davenport 2015: 284–5; 2019: 596–7. ¹⁵⁷ Jones 1964: 142–4, 528–9; Smith 2011: 142. The law is made up of CTh. 6.7.1, 6.9.1, 6.11.1, 6.14.1, 6.22.4. ¹⁵⁸ Jones 1964: 534–5. For the Byzantine court, see Kazhdan and McCormick 1997: 175–85. ¹⁵⁹ See Demandt 1970: 560–7, 781 on the origins of the new military command structure. ¹⁶⁰ Jones 1964: 97, 124–5, 608–10. The military hierarchy followed different trajectories in the East and West, as did the numbers of senior commanders. See Demandt 1970: esp. 781–6; Landelle 2014. ¹⁶¹ Jones 1964: 333, 339, 352. ¹⁶² Davenport 2019: 520–33. ¹⁶³ On the changing career structures, see Davenport 2019: 302–3, 595–6. For the non-Roman origin of some magistri militum, see Jones 1964: 135, 177–8, 552; Demandt 1970: 785–6.



27

part of the new battle-hardened cabal of soldiers at the imperial court, whose ethos and approach often clashed with aristocrats and administrators. The power and prominence of the magistri militum grew significantly with the series of child-emperors from the late fourth century onwards: Gratian, Valentinian II, Honorius, and Valentinian III in the West, and Arcadius and Theodosius II in the East. In her 2013 study of the child emperor phenomenon in the West, Meaghan McEvoy argued that this imperial system was sustained by a partnership between ‘a ceremonial and divinely endorsed emperor, accompanied and “supported” by an active, all-powerful manager’, the manager usually being the magister militum praesentalis who occupied a permanent place at court.¹⁶⁴ McEvoy extends her analysis of the role of these senior generals in Chapter 8, arguing that they should not be regarded as ‘powers behind the throne’, an expression which implies illegitimate or unconstitutional power. Most of these generals had extremely long tenures as magistri militum, sometimes spanning multiple emperors, and they served in office until they died, through either assassination or old age. Throughout these long tenures, the role of these generals as the partner of a non-campaigning emperor was not concealed, but actually celebrated through panegyrics and honours, prominent roles in imperial state ceremonial, and intermarriage with the imperial family. This state of affairs—the ‘age of the magister militum’, as McEvoy calls it—lasted until the 470s, with the deaths of Aspar in the east (471) and Ricimer in the west (472). One major difference between this period and the influence of the early Roman praetorian prefects like Sejanus and Plautianus is that there was no long-term support for a prefect to be the emperor’s equal partner. When both Sejanus and Plautianus fell from power, their successors as prefects did not achieve the same pre-eminence. McEvoy does not argue that this ‘partnership’ was the case throughout all Late Antiquity—the role of generals certainly changed in the sixth century. But elements of the late Roman world, such as the creation of new military commanders and the accession and acceptance of child emperors, provided the conditions in which this emperor–general partnership could emerge and shape politics for almost a century. The creation of the consistorium changed the role of another group of individuals at the court, the jurists, who are studied by Jill Harries in Chapter 6. The emperor and his court always attracted the Roman world’s best legal minds, but they were initially only occasional advisors, since too much time spent at court might damage their reputation as jurists and teachers. The Antonine and Severan periods witnessed the appointment of prominent jurists to high equestrian office, especially the praetorian prefecture, in concert with the increasing legal oversight inherent in that position.¹⁶⁵ Under the Severans, the prefecture was held by ¹⁶⁴ McEvoy 2013: esp. the conclusion, 305–29 (quotation from 310). ¹⁶⁵ On the legal powers of prefects, see de Blois 2001; Eich 2005: 216–22.

28

   

Aemilius Papinianus, author of the monumental thirty-seven-book Quaestiones and the seventeen-book Responsa, and Domitius Ulpianus, whose vast output included the Ad Edictum in eighty-three books, along with handbooks on official duties. These men had to straddle two worlds, publishing and earning the respect of their peers for their juristic expertise, while also dabbling in what Harries aptly dubs the ‘dark arts’ of court politics. These arts led both Papinianus and Ulpianus to bloody ends. The creation in the fourth century of the new office of the quaestor sacrii palatii, the author of imperial laws and one of the ex officio members of the consistorium, meant that the role of jurors at court became firmly institutionalized.¹⁶⁶ The new quaestores did not have independent careers as independent legal minds, but became court officials.¹⁶⁷ Indeed, an epigram lauding the sixth-century quaestor Proclus states that the ‘imperial court snatched him away’ (βασιλήϊος ἥρπασεν αὐλή) from his legal practice.¹⁶⁸ The character, actions, and behaviour of these officials reflected on the emperor positively or negatively, as shown by one of the cases examined by Harries, that of the unpopular Tribonianus under Justinian. Indeed, the formation of the consistorium gave Roman emperors a new way to articulate their own philosophy of governance through the appointment of officials to this council, as Kevin Feeney discusses in Chapter 7. During the Principate, the selection of equestrians to fill positions such as ab epistulis, a libellis, and a rationibus was not closely watched for insights into the tenor of the emperor’s regime; the presence of all these officials on the consilium is not consistently attested in our evidence, and the a rationibus in particular seems not to have travelled with the court.¹⁶⁹ These appointments were therefore primarily ‘clientelist’, because their function was to reward and promote supporters with high office. In contrast, in Late Antiquity, the selection of the four key comites consistoriani and other ex officio members of the consistory, as well as the promotion of others to the rank of comes ordinis primi, offered insights into how the emperor would balance regional, religious, and ideological interests at court. Feeney persuasively argues that in the fourth century, elites who wished to be appointed to the consistorium would try to behave in a manner that would appeal to the emperor and the tenor of his government. Feeney’s chapter examines two crucial regime changes: that which followed the accession of the pagan Emperor Julian in November 361, and then the appointment of the Christian Jovian following Julian’s death in June 363. Soon after Julian’s rise to the purple, he removed most of the members of the consistorium and the senior administrators of his predecessor Constantius II. One man who did survive the transition

¹⁶⁶ Harries 1988. ¹⁶⁷ Proc. Anecd. 30.30 may offer a satirical take on this situation when he writes that under Justinian, the law courts were empty, but the palace was a hive of activity. ¹⁶⁸ Anth. Gr. 16.48. ¹⁶⁹ Davenport and Kelly 2022.



29

(Modestus) and two new appointees (Felix and Helpidius) were former Christians who had abandoned their faith, just like Julian himself. The new emperor, Feeney argues, used his court appointments to model a way forward for Christian elites: apostatize and you will be welcomed into the fold. When Jovian succeeded Julian in turn, his choices for high court office set a very different tone. There was to be no cleaning house: instead, Jovian promoted men from a wide variety of backgrounds and religious affiliations, sending a message that his regime was to be one of conciliation. We have thus far concentrated on courtiers who occupied official positions in the military and civilian administration and who served on the consistorium in Late Antiquity. However, there is one very important group whose members played fundamental roles in court society, but held no government posts and still remained outside formal political decision-making processes: imperial women.¹⁷⁰ Monarchical courts have traditionally been established along the lines of male institutions such as the army, with the result that female members of royal families have usually been expected to act primarily in their capacities as wives and mothers, and to take on the ceremonial roles related to them.¹⁷¹ When we discuss the place of imperial women at court in both the Principate and Late Antiquity, we must remember that they had no constitutional powers, and never formed part of a consilium or sat on the consistorium.¹⁷² Nor did they formally act as ‘regent’ for a young emperor, despite the assumptions of many modern scholars who have grafted this real constitutional position, found in later European societies, onto the Roman world. Therefore, how should we conceive of the role and influence of imperial women at court? And did this change between the Principate and Late Antiquity? These questions are addressed by Anja Busch in Chapter 9, which focuses on the women of the eastern branch of the Theodosian dynasty: Aelia Flaccilla, first wife of Theodosius I; Aelia Eudoxia, wife of the Emperor Arcadius; her daughter Aelia Pulcheria, sister to the Emperor Theodosius II; and Theodosius’ wife, Aelia Eudocia. Busch draws attention to the increased public prominence given to the Theodosian women through the award of the title of Augusta and the right to wear ceremonial regalia. However, in line with recent scholarship, she carefully delineates the limits of their powers.¹⁷³ The title of Augusta carried with it no formal position or responsibilities at court, and thus it was left to individual imperial women to carve out their own role. This returns us to the theme of negotiation and adaption at court, which is equally applicable to individuals—both with and without official posts—as they attempted to establish their own place. Imperial

¹⁷⁰ ¹⁷¹ ¹⁷² ¹⁷³

For the term ‘imperial women’, rather than ‘empresses’, see Boatwright 2021: 2. Courts as male institutions: Duindam 2003: 38. The powers of imperial women are carefully delineated by Boatwright 2021: 10–46. Harries 1994a; Harries 2013; Elton 2009: 136–7.

30

   

women could exercise influence on their husbands, sons, and the court at large.¹⁷⁴ Eudoxia, the wife of the Emperor Arcadius, employed her many children as tools to persuade her husband. The emphasis on motherhood was then replaced by the pious virginity espoused by Pulcheria, Arcadia, and Marina, sisters of Theodosius II. But Pulcheria’s vow gave her enhanced religious authority, which enabled her to position herself as a figure of Christian influence at her brother’s court.¹⁷⁵ This was recognized by the many bishops, such as Cyril of Alexandria, who lobbied her to take sides in Christological disputes, as explored by Daniëlle Slootjes in Chapter 11. One could, of course, point out that women of the early Roman Empire, such as Agrippina and Julia Domna, had also been able to establish influential places at the courts of their husbands and sons.¹⁷⁶ But there are three crucial changes that took place in Late Antiquity. Firstly, the adoption of Christianity gave women new ways to define themselves within the imperial context, as pious and dutiful caregivers and providers of charity, as patronesses of churches and other ecclesiastical establishments, and as models of Christian virtue.¹⁷⁷ The second is that imperial women possessed what Busch aptly terms ‘dynastic potential’, in the sense that marriage of an outsider to a female member of the imperial family became a viable route to the imperial purple in its own right.¹⁷⁸ These two developments coalesced in the figure of Pulcheria, whose celibacy protected the Theodosian house from the interference of predatory outsiders so long as she lived, and whose eventual arranged marriage to Marcian ensured his accession as emperor following her brother’s death.¹⁷⁹ This would set a powerful precedent for the unique position of the Augusta Ariadne in 491 after the death of her husband Zeno, as highlighted both by Busch and by Audrey Becker in Chapter 4. When the leaders of the court, the members of the consistorium and the chief officials of the emperor’s household, could not come to a decision about who should succeed Zeno, they offered the choice to Ariadne. This is an example of courtiers who held official positions with defined powers delegating to their authority to an Augusta who did not, and, in doing so, helping to redefine the position of an Augusta over the longer term. The third and final change is the acceptance of a female form of co-rulership: whereas Agrippina was accused of treason because she aimed at consortium imperii (‘partnership in rule’), by the sixth century, empresses such as Ariadne could be depicted sitting on a throne that matched that of their husband, Justinian could state in a law that he had discussed matters with his most revered

¹⁷⁴ On female power as the ‘power of influence’, see Hug 2022: 70. ¹⁷⁵ Holum 1982: 91–3, 159–64; Millar 2006: 35–6, 195. ¹⁷⁶ See Hug 2022. ¹⁷⁷ McEvoy 2021. ¹⁷⁸ The ability of imperial women to transfer power to new husbands was threatened, but never came to pass in the Principate: McEvoy 2019b: 204–6; Hug 2022: 79–80. ¹⁷⁹ On the ‘anti-alliance’ policy of the eastern Theodosians, see McEvoy 2019a: 118–21.



31

wife (Theodora), and the poet Corippus could openly proclaim that Justin II’s wife Sophia was a consors imperii (‘partner in rule’).¹⁸⁰ The Roman imperial court therefore emerges as a community made up of different groups, some of whom held official civilian and military positions, and others who sat outside the formal framework of government, but were nevertheless still able to exercise influence and power. This community possessed the potential to articulate a consensus and establish normative values of behaviour, conduct, and approaches to administration, though this was not always accomplished due to conflicts of ideals and ambitions. The membership of the court society changed over time: some developments were the result of long-term systemic changes (such as the integration of jurists into the court), while others depended on the values and ideals of individual emperors (as in the courts of Julian and Jovian). We are thus able to trace the interplay of long-term change and individual circumstances in shaping the Roman imperial court society.

6. The Politics of Access Now a new wonder has appeared, the invention of the Arian heresy. The heretics and the emperor Constantius have united for this purpose: he, with the bishops as a pretext, can do whatever he wants with authority and can persecute without being called a persecutor, while they, with the power of the emperor, can plot against whomsoever they wish. For they want to plot against anyone who is not impious like them. Someone might think that this was a comedy written for these people to perform on the stage, with the so-called ‘bishops’ acting out roles and Constantius being the producer for their performance; once again he makes promises, as Herod did to the daughter of Herodias, and once again they dance their slanders to bring about the exile and death of those who are pious towards the Lord.¹⁸¹ It is to the pen of Athanasius, the exiled bishop of Alexandria, that we owe this memorable picture of his ecclesiastical opponents dancing before Constantius II,

¹⁸⁰ Tac. Ann. 14.11; Angelova 2004. For Justinian and Theodora, see Just. Nov. 8.1; for Justin II and Sophia, Coripp. Iust. pref. 23. ¹⁸¹ Athan. Hist. Ar. 52.5–6: νῦν δέ θέαμα καινὸν καὶ τοῦτο τῆς ἀρειανῆς αἱρέσεώς ἐστιν εὕρημα. συνῆλθον γὰρ αἱρετικοὶ καὶ βασιλεὺς Κωνστάντιος, ἵνα κἀκεῖνος τὴν ἐπισκόπων ἔχων πρόφασιν τῇ ἐξουσίᾳ καθ᾽ ὧν ἄν ἐθέλοι πράττῃ καὶ διώκων μὴ λέγηται διώκτης. καὶ οὗτοι δὲ τὴν βασιλέως ἔχοντες δυναστείαν ἐπιβουλεύωσιν οἶς ἂν ἐθέλωσι· θέλουσι δὲ τοἶς μὴ άσεβοῦσιν ὡς αὐτοί. τοῦτο δὲ ὡς ἐπί σκηνῆς ἄν τις ἴδοι κωμῳδούμενον παρ᾽ αύτοῖς, καὶ τοὺς μὲν λεγομένους ἐπισκόπους ὑποκρινομένους, τὸν δὲ Κωνστάντιον τὰ ἐκείνων ἐνεργοῦντα καὶ πάλιν ἐπαγγελλόμενον μὲν τοῦτον, ὡς ῾Ηρώδης τῇ Ἡρωδιάδι, τούτους δὲ πάλιν ὀρχουμένους τὰς διαβολὰς ἐπὶ ἐξορισμῷ καὶ θανάτῳ τῶν εἰς τὸν κύριον εὐσεβούντων (trans. Flower 2016).

32

   

in the same way that Salome’s movements bewitched King Herod, leading him to execute John the Baptist at her request.¹⁸² Athanasius envisaged himself as the intended victim in this reimagining of the biblical story, which expressed his anxiety and anger at the influence Arian bishops were able to wield at Constantius’ court. His anecdote vividly brings to life the politics of access which flourished both at the Roman imperial court and in other monarchical societies.¹⁸³ We could have begun this section with any number of other stories about women, freedmen, slaves, and eunuchs exerting inappropriate influence over the emperor and preventing access to his person. As Caillan Davenport explores in Chapter 13, this conception of the court as a locus of competition existed not only in the writings of those with intimate knowledge of the institution, but also in tales told by or about visitors to the court. Petitioners and embassies did not benefit from unfettered access to the emperor: the promiscuae salutationes of the Principate, as we have mentioned, were busy and crowded affairs. Visitors often needed to lobby and persuade the emperor’s courtiers, both government officials and household staff, to ensure that their requests would be favourably received.¹⁸⁴ It is in these interactions that we can observe the interconnection between the interior politics of the court and the politics of the empire at large.¹⁸⁵ If access to the monarch is a universal feature of the Roman court in both the Principate and in Late Antiquity, and indeed of all monarchical courts, what then makes it worth studying? We argue that the ubiquity of the politics of access means that it is all the more important to determine how this competition to influence the emperor manifested itself in different societies, courts, and time periods. Firstly, as several scholars of aulic studies have shown, simply being in the monarch’s presence did not translate to influence with him.¹⁸⁶ One could be permitted to watch the French king get dressed or a Habsburg monarch eat, but not be allowed to speak to him.¹⁸⁷ Even if one could talk with the king, the situation and audience might mean that the topic had to be banal and inoffensive, thus preventing any real discussion from taking place.¹⁸⁸ Sidonius Apollinaris’ account of a dinner party hosted by Majorian shows that it was the emperor who dictated the topic of conversation and the courtiers to whom the discussion was directed.¹⁸⁹ This means that the question of access is only half the story. The other is what David Starkey has aptly termed the ‘politics of intimacy’, the ability to win the monarch’s trust and confidence and to have private conversations that genuinely persuade him.¹⁹⁰ We need, therefore, to examine precisely who was granted access to a monarch and in what context. Addressing Emperor ¹⁸² ¹⁸³ ¹⁸⁴ ¹⁸⁶ ¹⁸⁷ ¹⁸⁹

The relevant biblical passages are Matthew 14:6–11; Mark 6:21–8. For the term ‘politics of access’, see Weiser 2003: 13–15; Raeymaekers and Derks 2016: 4. Millar 2006: 193. ¹⁸⁵ Spawforth 2007b: 3; Duindam 2016: 221–4. Duindam 2003: 234–42; Raeymaekers and Derks 2016: 4–5; Hengerer 2016: 147. Duindam 2003: 162; Hengerer 2016: 141. ¹⁸⁸ Weiser 2003: 14; Asch 2016: 181–2. Sid. Ap. Ep. 1.11.11–12. ¹⁹⁰ Starkey 1987a: 13; Starkey 1987b: 101.



33

Constantius II in a public audience was quite different from the intimacy enjoyed by his financial officer, the rationalis Mercurius, who sought out gossip about the incriminating dreams of courtiers and then ‘poured it into the emperor’s waiting ears’ (patulis imperatoris auribus infundebat).¹⁹¹ Secondly, the courtiers whom one needed to appease and placate in order to gain access to the ruler differed from court to court. At the Early Modern French court, the captain of the gardes de la porte played a key role in regulating access to the king.¹⁹² The guard commanders at the Roman court in Late Antiquity are not usually described as playing such a crucial part for much of our period, in contrast with eunuch chamberlains.¹⁹³ But Leo I’s reorganization of the excubitores invested this unit and their commander with greater authority to manage access to the palace in Constantinople, in particular controlling the entrances and exits, demonstrating the potential for alterations in power dynamics.¹⁹⁴ Thirdly, the spatial dynamics of access changed. At one court, the ruler’s bedchamber might be accessible, while at another, it was largely out of bounds, as detailed in Jeroen Duindam’s comparison of the courts of Versailles and Vienna.¹⁹⁵ Fourthly, the age and sex of the ruler could change the networks of power that gave access to the monarch, or indeed whether the king was actually the pivotal person to persuade. In the ‘Abbasid Caliphate, the appointment in 908 of a thirteen-year-old caliph, al-Muqtadir, saw all real authority handed to the vizier.¹⁹⁶ Finally, the parameters of access and the ceremonies that regulated it were not always instigated solely at the whim of the monarch, but depended on negotiation with the court community at large, as we have already noted.¹⁹⁷ This volume therefore presents several case studies which examine these dynamics in action, in order to determine more precisely the elements of continuity and change in the politics of access and intimacy between the Principate and Late Antiquity. In Late Antiquity, we can observe the rise of a new interest group involved in the competition for access: Christians. This included members of the Church establishment as well as holy men and women who operated outside of the mainstream ecclesiastical hierarchy.¹⁹⁸ Christians did, of course, appeal to pagan emperors as supreme authorities in matters of Roman law during the Principate.¹⁹⁹ But the conversion of Constantine to the Christian faith, espoused by all subsequent emperors except for Julian, entwined the imperial office and the

¹⁹¹ Amm. 15.3.5. ¹⁹² Murphy 2016: 40–1. ¹⁹³ On the power of court eunuchs, see Tougher 2008: 36–53. ¹⁹⁴ Croke 2005b: 146–50. Note the ability of Marcellus, comes excubitorum, to control the flow of information to Justinian (Proc. BG 7.32.23–6, 40–2). ¹⁹⁵ Duindam 2003: 163–4; Raeymaekers and Derks 2016: 9–10. ¹⁹⁶ Kennedy 2016: 161–71. ¹⁹⁷ Raeymaekers and Derks 2016: 7, 15. ¹⁹⁸ Thus McCormick 2000: 155: ‘Although it may not have changed the structures of social access to the emperors, Christianization helped transform the kind of people who used them.’ On holy men, see Brown 1971. ¹⁹⁹ See Millar 1971: 14–17 on one particular case, that of Paul of Samosata.

34

   

Church together in a new, and often tumultuous, relationship.²⁰⁰ By summoning the Council of Rome in 313, the Council of Arles in 314, and later convening the first ecumenical Council at Nicaea in 325, Constantine positioned himself as the arbitrator of disputes within the Church hierarchy and on matters of theological doctrine.²⁰¹ Even before he openly espoused his Christianity in 312, Constantine’s court in Trier had attracted bishops from the Gallic region.²⁰² The presence of bishops subsequently became commonplace at the imperial courts of Late Antiquity, but it is important to emphasize, following the important article of E. D. Hunt, that they did not hold any official court or government post.²⁰³ Therefore, bishops had to negotiate their access to the imperial court like other high-ranking elites, as Fabian Schulz and Daniëlle Slootjes examine in Chapter 10 and Chapter 11, respectively. Indeed, as Schulz discusses, in 343, the Council of Serdica ruled that bishops had to be specifically invited by the emperor to court, as an attempt to limit excessive lobbying by proponents of different Christological traditions. The regional nature of the imperial courts in the fourth century meant that a bishop of a city such as Trier, Milan, or Aquileia became more influential when the emperor was resident there.²⁰⁴ Schulz argues that bishops based in these imperial cities presumably expected to be invited to court on a regular basis, and they had the privilege of introducing bishops who had travelled from other sees to the court, thus giving them a heightened political role as patronage brokers. Once the eastern imperial court became permanently based at Constantinople from the late fourth century onwards, the patriarch assumed a new importance.²⁰⁵ In 528, Justinian decreed that no bishop could even enter Constantinople without first gaining the permission of the patriarch.²⁰⁶ The adoption of imperial Christianity enveloped both emperors and bishops in a new ceremonial calendar, that of the Church and its liturgy, in Constantinople.²⁰⁷ But the presence of the patriarch and the imperial court in the same city also led to friction and competition. In the Christian Roman Empire, the emperor could never be the sole source of religious power and authority.²⁰⁸

²⁰⁰ On the impact of imperial Christianity, see Leppin 2012: 257–76. For the emperor as a Christian priest or bishop, Dagron 2003 is fundamental; see now Lenski 2016: 76–8, specifically on Constantine, and Pfeilschifter 2013: 76–85 on early Byzantine emperors. ²⁰¹ On these councils, see Barnes 1982: 56–8; 2011: 100, 120–6; Lenski 2016: 57–8. ²⁰² Eck 2007. ²⁰³ Hunt 1989. ²⁰⁴ Note, for example, the influential position of Ambrose of Milan when the young Emperor Valentinian II was resident in the city (McLynn 1994: 158–219; McEvoy 2013: 86–99). ²⁰⁵ McCormick 2000: 155; Pfeilschifter 2013: 355–451. ²⁰⁶ Just. Nov. 6.3. ²⁰⁷ Religion was absent from Elias’ model of the court society, as Duindam 2011: 7 has noted. For these developments in Constantinople, Diefenbach 1996; 2002 are fundamental, and for specific studies of early emperors, see Croke 2010: 254–7 on Theodosius I, McEvoy 2020a: 188–90 on Arcadius, and Van Nuffelen 2012 on the Theodosian dynasty as a whole. ²⁰⁸ See Leppin 2018b: 57–61 for the situation under Justinian.



35

In Chapter 11, Daniëlle Slootjes explores the ramifications of these changes by focusing on the events leading up to the Council of Ephesus. Convened on the order of Emperor Theodosius II in 431, the council led to the bishop Nestorius being deposed from the see of Constantinople.²⁰⁹ Slootjes’ analysis moves beyond the inward-looking model of court society proposed by Norbert Elias to show how outside forces, such as the populace of Constantinople, church congregations, and bishops, could exert pressure on the court. As part of this affair, Cyril of Alexandria wrote letters to Emperor Theodosius II, his wife Eudocia, and his sisters Pulcheria, Arcadia, and Marina in an attempt to persuade them of the need to depose Nestorius. This reveals a key aspect of the politics of intimacy. Cyril targeted the emperor’s wife and sisters because they were individuals who could be expected to be able to have genuinely private conversations with the emperor.²¹⁰ Cyril also sent an extraordinary range of gifts to courtiers in order to encourage them to support his position. It would be too simplistic to dismiss these actions as mere bribery, for, as Christopher Kelly has shown, such gift-giving was an integral part of the exercise of power and patronage in Late Antiquity. In Cyril’s case, it was not the fact that he sent gifts that was considered improper, but their extravagance.²¹¹ The recipients of Cyril’s presents not only included male officeholders in Theodosius’ household such as the praepositi sacri cubiculi, but also the equivalent individuals in Pulcheria’s household, as well as leading female aristocrats, such as the wife of the praetorian prefect. This shows that petitioners did not only target male officials in their lobbying efforts, but included all male and female members of the court society who could be of assistance. Not that all individuals would be able to speak to and persuade Theodosius II personally, but they could certainly speak to other people who could then have a private audience with the emperor. For example, Pulcheria’s cubiculariae could lobby the Augusta, who would then try to persuade her brother, thus creating a chain of communication that extended to the emperor. Cyril’s ultimate goal in appealing to so many different individuals was to turn the mood of the court towards his point of view—creating a consensus, to return to this important theme—so that there would be a critical mass of people who shared intimacy with Theodosius II all lobbying the emperor to remove Nestorius. The most sought-after aim of petitioners was an audience with the emperor himself in order to convince him of the rightness of their cause. But this was not easily achieved. When Sidonius Apollinaris travelled to Rome in 467 to present a petition on behalf of the Arverni to Emperor Anthemius, he had to consider carefully which leading senator to approach in order to gain access. On the advice

²⁰⁹ For the context, see Millar 2006: 157–67. ²¹⁰ On this influence at other courts, see Duindam 2003: 239–40; Hengerer 2016: 142–5. ²¹¹ Kelly 2004: 138–85, esp. 171–5. For gifts as blessings in the fifth-century West, see Shaw 2011: 518–19.

36

   

of his chosen patron, Fl. Caecina Decius Basilius, Sidonius delivered a panegyric to Anthemius on the occasion of the emperor’s assumption of the consulship in 468 in order to ingratiate himself with the new ruler.²¹² The process was even more difficult for those petitioners who did not have such lofty contacts. This is vividly brought to life in first-hand accounts of visiting the Roman imperial court from the pen of authors such as Philo of Alexandria, who led a Jewish embassy to Caligula (examined by Caillan Davenport in Chapter 13). The eyewitness insights provided by Philo are just as valuable as those of Louis de Rouvroy, duc de SaintSimon, whose memoirs constituted a vital source for Elias’ study of Versailles under Louis XIV.²¹³ Reading Philo’s Embassy to Gaius, we get a real sense of the frustrations felt by the Jewish embassy as it followed Caligula around his different residences in Italy in the hope of achieving an audience. Moreover, when the Jews did obtain their hearing, it had to be shared with their opposing embassy, the Alexandrian Greeks, which complicated their ability to get their message across. The problem of speaking privately with the emperor was an acute one. One of the central arguments of Athanasius’ Defence Addressed to the Emperor Constantius was that he could never have been left alone with Constans to speak with the emperor in person, and in laying out his arguments, the bishop provides us with important insight into court protocol. He states that all discussion with Constans in the audience hall would have been public, and their conversation would have been heard by everyone who had also been admitted beyond the veil. Both Philo and Athanasius’ accounts are highly partisan, telling a story that puts their actions in the best light, but then so were Saint-Simon’s.²¹⁴ No narrative of court life will ever be impartial, and in fact, these accounts are all the more interesting and valuable because they represent a particular point of view, allowing scholars to assess different experiences and interpretations of court society. The desire of Roman aristocrats to have access to, and acquire intimacy with, the emperor can be found in many different aspects of elite political discourse, such as philosophical treatises, histories, and panegyrics, as Martijn Icks explores in Chapter 12. The Panegyric of Pliny memorably contrasted the open and welcoming court of Trajan with the fearful environment under Domitian, in which the emperor lurked monster-like in the shadows, protected by guards and barriers.²¹⁵ Although the basic idea that an emperor should not be a princeps clausus, closed off from his court and society at large, remained largely constant from the Principate to Late Antiquity, Icks shows how the nature of the discourse itself changed. In Late Antiquity, those who critiqued the emperor—such as

²¹² Sid. Ap. Ep. 1.9.1–8. For the context, see Harries 1994b: 143–8. ²¹³ Saint-Simon’s memoirs especially helped Elias to illuminate the workings of etiquette and ceremonial at court, as vividly demonstrated by Saint-Simon’s treatment by Louis XIV after he decided to resign his military commission (Elias 2006: 97–9). ²¹⁴ Duindam 1995: 59–62, 186–7. ²¹⁵ Plin. Pan. 48.3–5.



37

Synesius of Cyrene, philosopher and author of a famous screed directed at Arcadius—drew more heavily on the trope of Oriental despotism. This had its origins in Hellenistic perceptions of the Persian court as decadent and its monarch as despotic. The new fascination with Persian orientalism as a negative model of court society was motivated, Icks argues, by the rise of the Sasanian Persian Empire as Rome’s main rival in the East. Its applicability to the Roman world only increased as emperors themselves ceased to campaign in person from the late fourth century. This made comparisons with the Persian shahanshahs the ideal way to criticize child emperors such as Arcadius and Honorius, who, even as they grew to adulthood, did not lead their armies in person.²¹⁶ In reality, none of these emperors, even the child rulers, were really principes clausi cut off from the world around them; in fact, they were deeply involved in ceremonial performances in Constantinople.²¹⁷ However, there was a difference between access and visibility— processions and appearances made an emperor visible, but they did not often grant genuine access to his person, let alone intimacy.²¹⁸ Synesius’ misleading and inaccurate portrayal of Arcadius as a palace-bound ruler could have been prompted by the fact that it was difficult to interact genuinely with the emperor in any situation, both at court and in the city. Icks’ paper is complemented by Chapter 13, in which Caillan Davenport analyses how the court is represented in accounts of imperial audiences. He focuses on stories written by, or featuring, members of interest groups who visited the imperial court, such as Greek sophists, philosophers, Alexandrian Greeks, Jews, and Christians. Some of these stories can be found in the works of individuals who were well acquainted with the world of the emperors, like Philostratus, who counted Julia Domna as his patron. Others have more diverse origins, such as the tales found in Jewish rabbinic literature and early Christian apocrypha and hagiographies. The aim of Davenport’s chapter is to identify common concerns and interests shared by these different narratives. He shows that the elite preoccupation with gaining access to an emperor manifests itself equally in popular stories about the imperial court. In these tales, a range of stock figures, from malevolent advisors to wicked women, successfully block petitioners or poison the emperor’s mind against them. The ultimate—but often unattainable—goal of visitors was a frank and honest conversation with the emperor. It is particularly interesting that some stories feature fantastical conclusions in which outsiders are able to gain real intimacy with the emperor to the extent that they can even touch parts of his body. The folkloric nature of many of these tales mean that they shed little light on the individual circumstances of the politics of access at a particular

²¹⁶ On the military promise initially attributed to Honorius, but which he did not fulfil, see McEvoy 2013: 162–9. ²¹⁷ Pfeilschifter 2013: 86–99, 112–22. For the case of Arcadius, see McEvoy 2020a. ²¹⁸ On this principle, see Raeymaekers and Derks 2016: 10; Murphy 2016: 50–1.

38

   

court (in contrast to the case studies by Schulz and Slootjes), but they are valuable windows into perceptions of the court society. We conclude this Introduction with one particularly important aspect of Davenport’s narratives that contributes to our understanding of the court as an institution. Even though many of the stories were authored by outsiders, they were not as interested in offering detailed descriptions of the physical structures in which the court was located as we might have expected. Buildings and architecture were of much less importance than the individuals at court, and the relationships between them. Indeed, all the accounts of interactions at the court discussed in this volume, from the first-hand experiences of Philo and Athanasius to the letters and gifts sent to lobby courtiers during the Nestorian affair and the narratives of seclusion obsessed with the idea of the princeps clausus, focus on negotiations with and around other individuals to gain access to the emperor. This insight returns us to, and provides support for, the modern scholarly conception of the imperial court as a social phenomenon: the community of individuals that surrounded and interacted with the emperor.

PART I

RITUAL AND SPATIAL DYNAMICS

1 Great King or Civilis Princeps? Monarchical Ideals and Daily Interaction in the Reign of Antoninus Pius Christoph Michels

1. Introduction In August 142, Marcus Cornelius Fronto, teacher of the Caesar Marcus Aurelius, wrote a letter to Marcus’ mother, Domitia Lucilla.¹ He starts the letter, written in classicizing Attic Greek, by apologizing for not having contacted her for some time. Fronto justifies this neglect by stressing that he had to prepare his gratiarum actio for the princeps, which was due because Fronto was suffect consul at the time.² In the course of this letter, Fronto twice refers to Antoninus Pius as μέγας βασιλεὺς (‘great king’) without further qualifying or commenting on this title, which was once used by the Greeks for the Persian kings.³ Fronto adopts a Greek persona in this letter, which has several humorous elements—for example, when he makes fun of the ancient theory on similes.⁴ The title given to Pius is, however, certainly not meant as a joke. Fronto, himself stemming from Cirta in Numidia, assumes the perspective of a Greek provincial.⁵ In the Greek East, the title basileus had already established itself in the course of the first century as an acceptable alternative to autokrator, the Greek translation of the Latin imperator. We thus find basileus used in petitions addressed to Pius.⁶ For Greek inhabitants of the Roman Empire, the principes were the successors of the Hellenistic monarchs. The Alexandrian historiographer Appian, for example, remarks in the introduction to his Roman History (probably written in the 150s) that although the Romans did not call the emperors kings, they were exactly that, given their wide range of powers.⁷ The title of megas basileus, however, was not ¹ Fronto, Ad M. Caes. 2.3 (vdH² pp. 21–4); cf. also Richlin 2006b: 90–5 (no. 21). I thank the editors for their useful comments and my colleague Jörg Fündling for many conversations about Hadrian, Pius, Marcus, and Fronto. ² For the date of Fronto’s consulship, see Eck 1998; for his gratiarum actio, see Michels 2022. ³ Fronto, Ad M. Caes. 2.3 (vdH² p. 21, l. 20, p. 23, l. 17, compare p. 23, l. 7, where Pius is only referred to as βασιλεὺς). ⁴ Cf. Wenskus 2001: 228–9; Mullen 2015. ⁵ On Fronto’s self-designation as a Libyan barbarian: van den Hout 1999: 60. ⁶ Marotta 1991: 73–9. Cf. e.g. Dig. 14.2.9. ⁷ App. Pref. 6.23. Christoph Michels, Great King or Civilis Princeps? Monarchical Ideals and Daily Interaction in the Reign of Antoninus Pius In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0002

42

   

used in the time of Pius. Fronto thus even surpasses the honorific terminology of the provincials in his letter. However, when Fronto eventually did thank the emperor for his suffect consulship—the speech regrettably has not survived, apart from some possible fragments⁸—he certainly did not address him as king using the Latin word rex. This term had retained its negative connotations from republican times.⁹ He probably called him princeps or perhaps maximus princeps, as he does in a letter to Marcus when alluding to an edict of thanks issued on the occasion of the spectacles that Fronto planned to stage as consul.¹⁰ In one of only two surviving letters from Antoninus Pius in the corpus of Fronto’s letters, the princeps thanked Fronto for his original speech.¹¹ Fronto had delivered it in the senate, but in the absence of both Antoninus Pius and Marcus Aurelius. Therefore, Fronto had sent Pius a written version of the speech and a covering letter. In his reply, the princeps, in a seemingly humorous way, stresses his reluctance to praise the praise of himself because he does not want to appear arrogant. He nevertheless does so because he does not want to deprive Fronto of the deserved compliment.¹² Despite the ironic tone of the sentence, the modesty that Pius shows towards the senator Fronto is an expression of his civilitas—that is, the ideal that the princeps remained in principle an ordinary citizen (and a senator among senators) and did not exhibit superbia, or pride.¹³ He instead praises Fronto for his appropriate language, which had not broken the bounds of seemliness (nihil . . . civilius) through exaggerated deference. These divergent and contradictory images of the emperor as an absolute monarch on the one hand and as a mere citizen on the other hand existed in parallel. The princeps of the second century still basically operated within the structural framework of a contradictory combination of republic and monarchy introduced by Augustus—‘ruling without giving orders, wielding power without appearing to do so’, as Aloys Winterling has described his position.¹⁴ The Principate had, however, evolved since the days of Augustus, and aspects that had been ill-defined in the beginning had by now become increasingly institutionalized.¹⁵ ⁸ Cf. Davenport and Manley 2014 (no. 14) on potential fragments of the speech; cf. also Michels 2022. ⁹ Sigmund 2014: 338–54. This is not sufficiently considered by Rémy 2005: 129 when analysing the aforementioned letter. Fronto does, however, use the adjective regius/regia more freely than Pliny, cf. van den Hout 1999: 186 with regard to vdH² p. 67, l. 14 and p. 91, l. 21. ¹⁰ Fronto, Ad M. Caes. 2.4 (vdH² p. 24, ll. 19–21). Games hosted jointly by the two consuls are mentioned by Mart. 8.78.13f. and Plin. Pan. 92.4. ¹¹ Fronto, Ad Ant. Pium 2 (vdH² pp. 161–2 = Davenport and Manley 2014 [no. 14]). ¹² Fronto, Ad Ant. Pium 2 (vdH² p. 161, ll. 21–3). ¹³ Cf. Pliny’s insistence in Pan. 54 that Trajan was reticent about too much praise from the senate. On the image of the civilis princeps, see Wallace-Hadrill 1982. ¹⁴ Winterling 2011a: 15. On the idea of the princeps in the first and second centuries, cf. Winterling 2017. ¹⁵ There are of course very different views of what institutionalization means precisely. Compare, for example, the divergent assessments of Flaig 2019: 58, who advocated for an early institutionalization, and Gotter 2015: 230 who saw ‘astonishingly deficient institutionalization’.

     ?

43

While, for example, Caligula could do without the deification of his unpopular predecessor Tiberius, this was no longer an option for Antoninus Pius when Hadrian’s consecration was at first forestalled by the senate.¹⁶ One of the institutions that stood most clearly at odds with the ideal of the ruler as a primus inter pares and illustrates the monarchical dimension of the Roman principate was the imperial court, the social community surrounding the ruler.¹⁷ The court had become an important part of the Roman Empire, without ever evolving into a legal or constitutional institution.¹⁸ No princeps could afford to ignore the court community; an emperor who failed to meet the expectations and standards of the court could be branded a tyrant, like Caligula or Domitian. Indeed, Marcus famously warned himself in his Meditations to ‘take heed not to be transformed into a Caesar’ (Ὅρα, μὴ ἀποκαισαρωθῇς).¹⁹ In our main narrative sources, imperial histories and biographies written by authors such as Cassius Dio and Suetonius, ‘good’ emperors like Pius were remembered positively precisely because it was thought that they had not been corrupted by power and luxury. But to adopt the assessment of the narrative sources wholesale leaves us with an incomplete picture of the imperial court and prevents us from grasping the importance of the paradoxical role of the civilis princeps. In this chapter, I will begin with some introductory remarks on Pius’ role as civilis princeps with regard to the court and how it should in my view be understood. Then I will focus on Fronto as courtier and friend of the princeps, using his letters as a corrective to the image evoked by histories and biographies. Finally, I will ask whether in consideration of this context an Antonine innovation in the structure of the court—namely the establishment of private convivia, argued for by Winterling in his seminal study of the imperial court—is plausible.

2. Pius as a Civilis Princeps Antoninus Pius is often characterized in modern literature as an embodiment of the civilis princeps.²⁰ He showed the utmost respect to the senate as an institution and to the senators as his peers, being visible and available to them. He also demonstrated his civilitas to the people of Rome, though in this relationship other factors come into play, which I will leave aside in this context.²¹ At the imperial court itself, Pius was determined to reduce the level of pomp and ceremony in ¹⁶ Dio 59.3.7–8; Suet. Tib. 75.3; Joseph. AJ 18.236; cf. Price 1987: 86. On the consecration of Hadrian, see Fündling 2017: 37–40; Michels 2018: 22–78. ¹⁷ Winterling 1999; Vössing 2004; Wallace-Hadrill 2011. ¹⁸ Wallace-Hadrill 2011: 97–8. ¹⁹ M. Aur. Med. 6.30.1–5, see also 1.16.5; cf. Brunt 1974: 12; Winterling 1999: 1–3. ²⁰ Cf. e.g. Gotter 2015: 226: ‘Later . . . Antoninus moulded himself into the very incarnation of civilitas.’ ²¹ Cf. Michels 2018: 123–39; cf. also Fündling 2006: 912–13 on the comitas of Hadrian (HA Hadr. 20.1).

44

   

order to make himself as accessible as possible. This is described by the author of the Historia Augusta as follows: The imperial pomp (imperatorium fastigium) he reduced to the utmost simplicity and thereby gained the greater esteem, though the palace-attendants opposed this course, for they found that since he made no use of go-betweens, they were not able to terrorize men or take money for decisions about which there was no concealment. In his dealings with the senate, he rendered it, as imperator, the same respect that he had wished another princeps to render him when he was a privatus.²²

The imperatorium fastigium reduced by Pius seems to refer to alterations in court ceremonial, which made it easier to approach the princeps directly rather than through officials. This was a change calculated to appeal to the aristocratic members of the court, the members of the senatorial and equestrian orders, even as it isolated other sections of the court community, chiefly the emperor’s slave and freedman staff. The palace attendants were anxious that their influence would be threatened by no longer being able to control access to the princeps.²³ They were right to worry, apparently, as the HA also mentions that Pius cut their stipends.²⁴ The emperor’s firm control over his officials appears in another passage, where we learn that Pius treated his freedmen severely and prevented them from selling favours. This was another mark of the bonus princeps that appealed to the aristocracy, who resented having to propitiate slave and freedman officials.²⁵ Marcus Aurelius’ own characterization of his adoptive father in the Meditations has also significantly influenced our image of Antoninus Pius.²⁶ In the work, Marcus thanked the gods that my station in life was under a governor and a father who was to strip off all my pride and to lead me to see that it is possible to live in a palace and yet not to need a bodyguard or embroidered uniforms or candelabra and statues bearing

²² HA Ant. Pius 6.4–5: Imperatorium fastigium ad summam civilitatem deduxit, unde plus crevit, recusantibus aulicis ministris, qui illo nihil per internuntios agente nec terrere poterant homines aliquando nec ea quae occulta non erant vendere. senatui tantum detulit imperator, quantum, cum privatus esset, deferri sibi ab alio principe optavit (trans. Magie, Loeb, adapted). For the word fastigium and its meaning in this context, see OLD s.v. fastigium 7, ‘degree of eminence or importance’, here translated as ‘pomp’. ‘Imperial airs’ might be another possible rendering. ²³ Wallace-Hadrill 1982: 41 n. 64 has wondered whether passages like this in the HA might ‘reflect fourth-century tensions’. However, I see no compelling reason to consider tensions like these as anachronistic for the second century. Wallace-Hadrill himself pointed out that Marcus’ Meditations reflect the lure of luxury at court and efforts to resist it. ²⁴ HA Ant. Pius 6.4, 7.7–8. ²⁵ HA Ant. Pius 11.1. ²⁶ On Antoninus Pius in the Meditations, see Brunt 1974; Pernot 2001: 108–12; Gangloff 2019: 255–301.

     ?

45

lamps and the like accompaniments of pomp, but that one is able to contract very nearly to a private station and not on that account to lose dignity or to be more remiss in the duties that a prince must perform on behalf of the public.²⁷

In many respects, the Meditations present Antoninus Pius as an embodiment of Marcus’ Stoic ideals.²⁸ However, the image of the emperor in Marcus’ writings is consistent with the Historia Augusta, as we again find Pius behaving like a privatus, which is meant to illustrate his civilitas.²⁹ We must emphasize that Pius only behaved ‘like’ a privatus—he was not actually a private citizen, because the princeps was actually the direct opposite of a privatus.³⁰ In order to understand Pius’ conduct, it is crucial to acknowledge that a Roman princeps did not have a private life and seemingly personal practices were all actually part of imperial selfrepresentation.³¹ This is nicely illustrated by the HA’s statement that Pius’ amici sometimes saw him busy with domestic activities while visiting him.³² He knew how to represent himself to, and communicate effectively with, the aristocratic members of the court. Many aspects of Pius’ behaviour are not specific to him, but correspond to established strategies of communication that can be identified for his ‘good’ predecessors. Pius, however, who had been senator for decades before he became princeps at the age of fifty-one—a stark contrast to his grandson Commodus, for example—seems to have filled this role especially convincingly. He also presented a contrast to the eccentric Hadrian, who had alienated wide circles of the ruling elite, notably around the time of his accession and in his last years. Pius struck a more conciliatory tone that is perhaps reflected when the consul Fronto surprisingly states, in a letter addressed to Marcus already mentioned, that he could not love Hadrian but loves Antoninus all the more.³³ The fact that Pius never left Italy during his reign—again in stark contrast to the ‘restless emperor’ Hadrian, but also to his successors who were forced to spend years on campaigns—meant that he was more approachable for a Rome-based senator than Hadrian, who had his

²⁷ M. Aur. Med. 1.17.5: τὸ ἄρχοντι καὶ πατρὶ ὑποταχθῆναι, ὃς ἔμελλε πάντα τὸν τῦφον ἀφαιρήσειν μου καὶ εἰς ἔννοιαν ἄξειν τοῦ ὅτι δυνατόν ἐστιν ἐν αὐλῇ βιοῦντα μήτε δορυφορήσεων χρῄζειν μήτε ἐσθήτων σημειωδῶν μήτε λαμπάδων καὶ ἀνδριάντων τοιῶνδέ τινων καὶ τοῦ ὁμοίου κόμπου, ἀλλ̓ ἔξεστιν ἐγγυτάτω ἰδιώτου συστέλλειν ἑαυτὸν καὶ μὴ διὰ τοῦτο ταπεινότερον ἢ ῥᾳθυμότερον ἔχειν πρὸς τὰ ὑπὲρ τῶν κοινῶν ἡγεμονικῶς πραχθῆναι δέοντα (trans. Farquharson, 1944). ²⁸ Rosen 1996: 156–7. ²⁹ HA Ant. Pius 7.6, 11.1, 6. ³⁰ Winterling 2009: 58–76; for Trajan acting like a privatus as princeps in Pliny’s Panegyricus, see Rees 2001: 156–60 (with reservations). ³¹ Vössing 2004: 267; Gotter 2008: 180. See also Pflug and Wulf-Rheidt 2022: 227: ‘it is not appropriate to designate the palace as imperial private space’. ³² HA Ant. Pius 6.12. Cf. Walentowski 1998: 207. ³³ Fronto, Ad M. Caes. 2.4.1 (vdH² p. 25, ll. 3–9). Cf. Fündling 2017: 40–3. For love as an emotion at the Antonine court, see Chapter 5 by Benjamin Kelly in this volume. In my view, however, we cannot classify the letters between Marcus and Fronto as ‘private’, as the former was a member of the domus Augusta. Indeed, he read this particular letter to Pius; cf. Ad M. Caes. 2.5.3 (vdH² p. 24, ll. 17–20).

46

   

court travelling with him.³⁴ So far, we have a consistent image of the modest civilis princeps. But the picture is more complicated than this. When he became Augustus in 138, Pius did, of course, take up residence in the domus Tiberiana on the Palatine without hesitation. He also immediately made Marcus relocate to the palace, despite the latter’s reluctance on account of his philosophical scruples. The Historia Augusta mentions that Marcus Aurelius was confronted here with the ‘the pomp of the court’ (aulicum fastigium), which at least partly contradicts the image of the HA’s own Life of Pius and of the Meditations that Pius had significantly reduced imperial luxuries.³⁵ This impression is strengthened when the Life of Marcus later mentions that Marcus, too, sold expensive objects, such as furniture and jewellery, to raise funds for the Marcomannic Wars. That those auctions allegedly lasted for two months shows that the court had not been completely stripped of its luxuries under Antoninus Pius.³⁶ A comment in Marcus’ Meditations further illustrates that Pius had also not made significant changes to the court system as such. Remarkably, in order to illustrate the sentiment that everything stays the same, Marcus places the court of his adoptive father in one continuous series following Hadrian, Philip II of Macedonia, Alexander the Great, and, notably, the Lydian king Croesus, ‘for those were all like these; the actors only were different’ (πάντα γὰρ ἐκεῖνα τοιαῦτα ἦν, μόνον δἰ ἑτέρων).³⁷ This comparison reflects a monarchical, perhaps even autocratic, self-image of the princeps,³⁸ and it illustrates that the gestures of modesty by Antoninus Pius should be understood—like those of his predecessors—as a specific form of communication between the princeps and the aristocracy at court. As it is phrased by the Historia Augusta, through this ‘self-abasement’ Pius ‘gained the greater esteem’ (unde plus crevit).³⁹ This behaviour, which conformed to the imperial ideal of civilitas, can also be found in the characterization of Trajan in Pliny’s Panegyricus.⁴⁰ It should not be interpreted, however, as a complete picture of the relationship between emperor and elite (and even less so with regard to the interaction with the plebs, the provincials, or the army). Antoninus Pius conformed to the expected rules of communication by showing modesty and politeness towards his aristocratic peers, which means that the distance between senator and princeps is not stressed in our major literary sources. It does become clear, however, in the daily interaction with members of the court, which we can see in the correspondence of Fronto. ³⁴ For the changing location of the Roman imperial court, see Chapter 14. For Pius’ use of imperial villas in Italy, see Seelentag 2011: 299–300, 302, 311; cf. Halfmann 1986: 90–110. ³⁵ HA Marc. 6.3; cf. HA Ant. Pius 10.4; Dio 72(71).35.4. ³⁶ HA Marc. 17.4–5, 21.9; Winterling 1999: 81–2. ³⁷ M. Aur. Med. 10.27 (trans. Farquharson 1944). ³⁸ Vössing 2004: 369 with n. 5. ³⁹ HA Ant. Pius 6.4 (trans. Magie, Loeb). ⁴⁰ Plin. Pan. 71.4. Cf. also with regard to Hadrian, HA Hadr. 20.1 with Fündling 2006: 912–14; Wallace-Hadrill 1982: 33.

     ?

47

3. Fronto as Courtier and Friend of the Princeps As one of the teachers of Antoninus Pius’ adopted sons, Marcus Aurelius and Lucius Verus, Fronto certainly occupied a special position at court, and a certain familiarity with Pius has been deduced from his correspondence. That there are only few letters between them does not, however, speak for a very close relationship.⁴¹ Moreover, mutual amicitia is never invoked in their correspondence, even in places where we might have expected it to make an appearance. As Champlin has shown, Fronto was—although he repeatedly heaps reproach on the court in his letters—unmistakably what we would call a courtier.⁴² Because of his connection to the Caesar Marcus, he was considerably closer to the imperial family than Pliny the Younger had been.⁴³ Fronto, however, as far as we know, was never part of the consilium principis like Pliny, and he was therefore not one of the ‘inner circle’.⁴⁴ Both Pliny and Fronto were amici (‘friends’) of their respective emperor, just as all senators and equites who were admitted to the salutatio were friends of the princeps due to their social rank.⁴⁵ There were, however, closer ‘friends’ of the emperor. Winterling rightly stressed that the amicitia of the emperor was a hierarchical system based on the emperor’s favour. Winterling distinguished between three different types of friends: (i) the familiares intimi (‘the most intimate relations’) or proximi amicorum (‘closest of friends’) who were present at court on a daily basis and who enjoyed the emperor’s particular trust; (ii) a wider circle of amici, with whom the princeps interacted at a social level—we might see Fronto at this level in the time of Pius; and finally (iii) the whole of the aristocracy admitted to the salutatio in a kind of institutionalized form of amicitia.⁴⁶ Members of the aristocracy experienced differences in access to the emperor during the daily morning receptions, the salutationes. In contrast to the salutationes of the Republic, the imperial receptions primarily served as manifestations of status, and it seems that by the time of Pius, they had become increasingly institutionalized.⁴⁷ We hear from Cassius Dio that the Caesar Marcus received the ‘most noble’ (ἀξιώτατοι) of the senators in his bedchamber in the domus Tiberiana before he attended upon his adoptive father in his turn.⁴⁸ A morning reception of senators in the princeps’ bedroom is already attested for Vespasian.⁴⁹ ⁴¹ Champlin 1980: 102. There are two short letters by Pius and five by Fronto; originally there may have been more, of course (e.g. the accompanying letter Fronto sent with his published gratiarum actio is almost completely lost). ⁴² Champlin 1980: 94, 97. ⁴³ Champlin 1980: 80–1; Noreña 2007b: 256–7. ⁴⁴ Champlin 1980: 108: ‘apparent omission from any imperial consilium’; Van den Hout 1999: viii, 249. ⁴⁵ Winterling 2011b. ⁴⁶ Winterling 1999: 166–9; Winterling 2009: 90–2. ⁴⁷ Winterling 1999: 135–8; Goldbeck 2010: esp. 277. ⁴⁸ Dio 72(71).35.4. ⁴⁹ Suet. Vesp. 21; Dio 66(65).10.5; Epit. de Caes. 9.15; Plin. Ep. 3.5.9; cf. Winterling 1999: 127.

48

   

The ‘most noble’ men mentioned by Dio have, however, been interpreted as a novel development of the second century. Winterling has argued, based on short mentions in several sources, that they represent a new status group defined by social rank that had come into existence in the time of Hadrian.⁵⁰ While at first the closest imperial associates tended to be of lower rank (such as the freedmen who clustered around Claudius), he proposed that this changed during the course of the first century, so that in the second century the noblest persons also became part of the innermost circle at the imperial court. For Winterling, this elite within the aristocracy permanently surrounded the princeps and thereby took over the position previously occupied by the group of household freedmen. Because social rank, not personal familiarity, was now the crucial factor, Winterling regarded this development as part of a tendency to depersonalize familiaritas with the emperor in the early second century .⁵¹ There is certainly the risk of over-interpreting the literary sources in identifying different groups of the court. We can certainly see this tendency in another more personal circle identified by Winterling, composed of individuals who share the emperor’s interests—scholars in the case of Hadrian, philosophers in the case of Marcus.⁵² Yet that does not mean that we should dismiss the importance of Cassius Dio’s ‘most noble men’. Even if we do not wish to regard them as a separate group, their exclusive reception clearly shows differentiation within court ranks. It is telling that Dio mentions this practice in order to illustrate the civilitas of Marcus. The Caesar received the senators in simple garments in his bedchambers and not in the atrium, as was the practice with clients—a detail which was probably meant to indicate the special honour in which he held the nobles. However, the fact that these aristocrats had to pay their respects every day clearly illustrated the imbalance of power in their relationship with the Caesar. This power differential has to be kept in mind when we consider that Dio mentions that Marcus received the ‘most noble men’ wearing his ‘private’ or ordinary clothing, just as the Historia Augusta described his father Antoninus Pius doing.⁵³ As with Pius, this was a carefully calibrated performance of civilitas which was designed to ensure effective and harmonious relationships with the nobles at court. Even in the case of someone who was close to the imperial family as Fronto, who was entrusted with the education of the Caesars, there was still a considerable distance between courtier and princeps. The ambiguity in the communication between the princeps and the aristocracy—in which the emperor had to act as if every aristocrat were his amicus and the aristocrats pretended that they were all friends of the princeps⁵⁴—is manifested in an episode surrounding the will of one

⁵⁰ ⁵¹ ⁵² ⁵³

Winterling 1999: 188–91; Winterling 2009: 92; Winterling 2011b: 231. Winterling 1999: 189–90; Winterling 2009: 92–3. Winterling 1999: 190–1; but see Fündling 2006: 741–2. Dio 72(71).35.4; HA Ant. Pius 6.12. ⁵⁴ Winterling 2011b: esp. 225–6.

     ?

49

of Fronto’s friends. This incident, which took place between  146 and 158, could have led to Fronto losing the emperor’s favour.⁵⁵ Fronto came into conflict with the long-serving praetorian prefect Gavius Maximus, who was certainly one of the most influential persons of the time, even if the nature of the literary sources means that his full life and character remains obscure.⁵⁶ As is clear from the three relevant letters of Fronto to Antoninus Pius, Gavius Maximus, and Marcus Aurelius, the problems arose as a result of the death of Fronto’s friend Censorius Niger. He had bequeathed a part of his property to Fronto, but in the corresponding will he had spoken disparagingly of Gavius Maximus. This eventually led to Fronto having to explain to the princeps why he had accepted the inheritance in the first place, since Niger had incurred imperial displeasure. The episode illustrates the ‘rules of the game’ of friendship with the princeps, for the events that at first glance do not seem dramatic could well have had serious consequences for Fronto.⁵⁷ Examples from the reigns of other emperors show that the most severe form of relationship breakdown, the formal termination of friendship by the princeps, could even result in the suicide of the person concerned, as a way of protecting his family and its assets.⁵⁸ That Fronto sometimes calls Antoninus Pius dominus imperator (‘lord emperor’) in his letters is particularly telling of the way in which he conceived of this relationship.⁵⁹ The title dominus imperator had appeared in inscriptions since the time of Trajan, but Fronto is the first writer to use it as a form of address.⁶⁰ Pliny regularly employed the vocative form domine to address Trajan, although in the Panegyricus he stressed that ‘we do not speak of him as a master, but as a father’ (non de domino, sed de parente loquimur).⁶¹ The poet Martial, who had willingly hailed Domitian as ‘lord and god’ (dominus et deus),⁶² soon adopted the prevailing tone of the Trajanic period, writing that ‘here there is not a master, but an emperor’ (non est hic dominus, sed imperator).⁶³ In the polite language of Fronto’s day, dominus was a respectful term covering relationships of all kinds. Throughout Fronto’s correspondence, Marcus, for example, refers to his brother Lucius as dominus meus (‘my lord’). Nevertheless, there is a real contrast in tone when Fronto writes to Marcus, ‘I ought to give thanks to my lord, your father’ (gratias agere domino meo patri tuo debeo).⁶⁴ The use of dominus imperator as a way to address the emperor should be placed in this

⁵⁵ The affair is documented in Fronto, Ad Ant. Pium 3, 7 und 4 (vdH² pp. 162–4, 165–6) with van den Hout 1999: 199 (on the order of the letters, see Davenport and Manley 2014: 30–2). Cf. Champlin 1980: 100–2; Fleury 2006: 148–51; Wei 2013; Michels2023. ⁵⁶ On Gavius Maximus, cf. Birley 2020. ⁵⁷ Cf. Fleury 2006: 148, ‘l’implication immense’. ⁵⁸ Rogers 1959. ⁵⁹ Fronto, Ad Ant. Pium 7 (vdH² p. 165, l. 16); Anton. Pium 10 (vdH² p. 168, l. 24). ⁶⁰ van den Hout 1999: 399. ⁶¹ Plin. Pan. 2.3. Cf. Lavan 2018: 290. ⁶² Mart. 5.8.1; 8.2.6; 9.66.3. After Domitian’s death, Martial distanced himself from this (10.72.3 and 8)— ostensibly relieved. ⁶³ Mart. 10.72.8. ⁶⁴ Fronto, Ad M. Caes. 2.4 (vdH² p. 24, l. 17).

50

   

second context, as an attempt at intensifying the hierarchical relationship between courtier and emperor. Fronto’s usage is paralleled in a passage in the Digest, in which a certain Vibius Zeno, perhaps a libertus, addresses Marcus Aurelius in a hearing from 166.⁶⁵ The institutional distance is even more evident in court rituals.⁶⁶ Marcus stresses in his Meditations that Pius in most cases did not insist on senators excusing themselves when they could not take part in appointments with him, thereby conforming to an ideal formulated also by Pliny in his Panegyricus with regard to Trajan.⁶⁷ A letter from Fronto to Antoninus informs us, however, that Fronto was obliged to do exactly this when he had, due to an illness, missed Pius’ dies imperii (‘accession day’): Fronto to Antoninus Pius Augustus More dearly than with a portion of my life would I bargain to embrace you on this most happy and wished-for anniversary of your accession, a day which I count as the birthday of my own health, reputation, and safety. But severe pain in my shoulder, and much more severe in my neck, have so crippled me, that I am still scarcely able to bend, sit upright, or turn myself, so rigid must I keep my neck. But before my Lares, Penates, and household gods have I discharged and renewed my vows and prayed that next year I might embrace you twice on this anniversary, twice kiss your chest and hands, fulfilling at once the office of the past and the present year.⁶⁸

In a previous letter, Fronto had already asked Marcus to tell the princeps of his illness and inquired whether it would be necessary to write to Pius himself.⁶⁹ The Caesar had advised his teacher to do just that. Fronto’s written offer to kiss Pius’ chest and hands twice over is clearly a gesture of obeisance, fundamentally different from the kisses on the face that he and Marcus exchanged as expression ⁶⁵ Dig. 28.4.3; Lavan 2018: 290–4. On the status of Vibius Zenon, see Wankerl 2009: 81 n. 447. In CEL no. 149, a letter from an Egyptian recruit to the prefect of Egypt, Trajan is referred to as imperator dominus noster, suggesting that it may have been current outside court contexts earlier in the second century. ⁶⁶ Cf. Winterling 1999: 28–32; Winterling 2017: 428–9 on the problem of categorizing ‘court ceremonial’ as distinct from the symbolic interaction of the princeps in the public sphere of Rome. ⁶⁷ M. Aur. Med. 1.16.8; cf. Plin. Pan. 48.2. ⁶⁸ Fronto, Ad Ant. Pium 5 (vdH² p. 164, ll. 20–8): ANTONINO PIO AUG. FRONTO Vitae meae parte depicisci cupio, imp(erator), ut te complecterer felicissimo et optatissimo initi imperii die, quem ego diem natalem salutis, dignitatis, securitatis meae existimo. sed dolor umeri gravis, cervicis vero multo gravissimus ita me adflixit, ut adhuc usque vix inclinare me vel erigere vel convertere possim: ita immobili cervice utor. sed apud Lares, Penates deosque familiares meos et reddidi et suscepi vota et precatus sum, uti anno insequenti bis te compleoterer ista die, bis pectus tuum et manus exoscularer praeteriti simul et praesentis anni vicem persequens (trans. Haines, Loeb). On the date of the letter, which remains uncertain, see van den Hout 1999: 390. ⁶⁹ Fronto, Ad M. Caes. 5.35–6 (vdH² p. 75); it is not clear that the three letters which Haines 1919–20: 1.226 grouped together actually belong together; cf. van den Hout 1999: 199.

     ?

51

of their close connection.⁷⁰ But it is also different from kisses given by the emperor to greet senators. While a kiss on the cheek could be interpreted as a sign that the emperor saw the senators as peers, to kiss someone on the hand was the characteristic action of a client begging a favour, and a kiss on the chest only became a standard element of the adoratio in the fourth century.⁷¹ The importance of these gestures is illustrated by literary criticism of ‘bad emperors’ who simply held out their hand to be kissed, overemphasizing their superior status. Even to allow your feet to be kissed was the mark of the tyrant, something that Trajan, as Pliny stresses in his Panegyricus, did not do.⁷² The tone of the exchange between emperor and courtier is exemplified by Pius’ short reply to Fronto’s excuse which has, as Champlin noted, ‘an audible ring of formality, if not of the secretariat’.⁷³ A simile used by Fronto in another letter to illustrate his affection towards and close connection with Pius may likewise instead be interpreted as reinforcing the profound distance between senator and emperor. ‘Indeed, I love and I cherish Antoninus like the sun, like the day, like life, like breath, and I think that I am loved by him in turn,’ he wrote, contrasting the current emperor with Hadrian.⁷⁴ When Fronto wrote that he loves Antoninus like the sun, this could be seen as an allusion to the representation of the emperors as Helios/Sol, which is already attested in the early Roman Empire, though it would grow in importance under rulers such as Aurelian.⁷⁵ Although we lack examples of other senators communicating with Antoninus Pius, Fronto’s correspondence suggests that the picture of Pius as primus inter pares, reflected primarily in our major literary sources, should be contrasted with the daily interaction between princeps and courtiers, in which the vastly superior social standing of the emperor is much more in evidence. We shall keep this conflict between literary ideal and daily reality in mind as we turn to examine another possible Antonine court innovation proposed by Winterling.

4. A New Type of Convivium in the Reign of Antoninus Pius? This possible innovation concerns one of the central spheres of communication between the princeps and the elite, namely the convivium, or dinner party.⁷⁶

⁷⁰ Williams 2012: 255–7. ⁷¹ Alföldi 1970: 40–2; Paterson 2007: 147–8; Hartmann 2016: 71–88. ⁷² Plin. Pan. 24.2; cf. Vössing 2004: 452–5; Hartmann 2016: 79–83. ⁷³ Fronto, Ad Ant. Pium 6 (vdH² p. 165); Champlin 1980: 98. ⁷⁴ Fronto, Ad M. Caes. 2.4.1 (vdH² p. 25, ll. 9–10): Antoninum vero ut solem, ut diem, ut vitam, ut spiritum amo, diligo, amari me ab eo sentio. ⁷⁵ Beaujeu 1955: 325; Bergmann 1998: 243–6. Cf. Pius with astral nimbus on the reverse of BM Coins, Rom. Emp. IV 1666q. Cf. also Chapter 5 by Benjamin Kelly. ⁷⁶ Winterling 1999: 145–60; Vössing 2004; Schnurbusch 2011; cf. the classic flattering account of an imperial banquet in Stat. Silv. 4.2.

52

   

Aristocrats participated almost daily in these banquets, which could involve some six hundred attendees, according to a figure from the reign of Claudius.⁷⁷ Invitations to dinner parties with the emperor and the emperor’s acceptance of invitations by senators at their own houses functioned as particularly important markers for elite participation in imperial rule. Therefore, the demeanour of the princeps during this social interaction and his ability to conform to the expectations of his notional peers were very important.⁷⁸ Through gestures of modesty—for example, when Hadrian stood up to greet senators arriving at a convivium—the emperor showed respect towards the aristocracy.⁷⁹ Such communicative acts were both relevant and necessary. Transgressions of ‘bad’ emperors are often reported by the literary sources in the context of banquets. Ancient authors commented on the behaviour of the emperor during convivia given by him—not exclusively but frequently—in the palace.⁸⁰ The emperors’ responses to dinner invitations were likewise monitored. Antoninus Pius accepted them regularly, just as Hadrian had done before him. Marcus seems to have been more reserved in this regard, however, and this was, as the HA notes, accordingly registered as encouraging a courtly arrogance because he wanted to keep his friends away from dinner parties.⁸¹ Ostentatious representation of wealth and splendour, shown for example by Caligula and Nero, continued the tradition of tryphē (‘luxury’) of the Hellenistic monarchs. This was, however—in its Roman guise of luxuria—clearly regarded negatively by the Roman aristocracy. Augustus had already recognized the importance of self-restraint in this area if he did not want to appear as an ‘oriental’ monarch. Pius apparently hit the right mark in this regard, for the Historia Augusta stresses that his lifestyle showed opulence without cause for reproach and ‘frugality without meanness’ (parsimonia sine sordibus).⁸² As far as we know, Pius did arrange an extraordinarily splendid wedding feast for his daughter Faustina and his adopted son Marcus, so he recognized the importance of display on specific occasions.⁸³ Apart from the behaviour of the princeps, another crucial aspect of the banquets was the question of who was permitted to attend. The Life of Pius mentions only in passing that the emperor held both ‘public’ and ‘private’ banquets.⁸⁴ This is a surprising statement which has only rarely been questioned. Winterling, however, recognized the importance of this information.⁸⁵ The question is, however, whether it is accurate. For Winterling, this passage shows that the separation ⁷⁷ Suet. Claud. 32. The participation of leading aristocrats is attested for Vespasian, Trajan, and Hadrian, and is to be assumed in the case of Antoninus Pius. Cf. Winterling 1999: 154, 156–7, 182. ⁷⁸ Vössing 2004: 534–5. ⁷⁹ HA Hadr. 22.3. ⁸⁰ Cf. e.g. Suet. Vit. 13; HA Macrinus 1.4–5. ⁸¹ Dio 69(68).7.3–4; HA Hadr. 9.7; Ant. Pius 11.7; Marc. 29.7; cf. Vössing 2004: 328. ⁸² HA Ant. Pius 7.5; cf. M. Aur. Med. 1.16.30. ⁸³ HA Ant. Pius 10.2. ⁸⁴ HA Ant. Pius 11.4. ⁸⁵ Winterling 1999: 157. Walentowski 1998: 272 does not comment on it at all.

     ?

53

between public and private convivia represents the establishment of a new category of imperial banquets.⁸⁶ Who were the guests of these supposed private convivia of Antoninus Pius? Winterling identifies them as precisely the two elite groups of courtiers newly established in the time of Hadrian that I mentioned earlier—the ‘most noble’ members of the senate and those who shared the emperor’s scholarly interests.⁸⁷ In Winterling’s view, the relevance of this innovation goes beyond the convivia, as he concluded that in the time of Pius, a new personal or ‘private’ sphere of the princeps developed. This, he argued, formed part of a larger tendency to reintroduce the separation between public and private.⁸⁸ In contrast to earlier attempts by ‘bad’ emperors, namely Domitian, that led in this direction, this was, from Winterling’s point of view, now socially acceptable. Winterling placed the establishment of the res privata, a new imperial treasury, separate from the patrimonium, in this context.⁸⁹ He has also argued, on the basis of the Historia Augusta, that Pius employed his own personal slaves, who should be distinguished from the imperial familia Caesaris.⁹⁰ However, it is problematic to attempt to combine these two quite different changes into a major development at the court, and neither of them carry much weight in themselves. It is certain neither whether the res privata was actually established by Pius nor what its function was.⁹¹ The servants of Pius are merely mentioned in the Historia Augusta to illustrate his modesty in that he did not buy his meat on the market but had his own hunters.⁹² This returns us to the convivia themselves. Vössing has rightly stressed that the Historia Augusta is the earliest source to mention ‘private’ banquets of the emperor. In the time of its author—that is, in the late fourth century—the princeps clausus, the secluded emperor who never left his palace, was at least a perceived normality, and an institution like a private banquet had nothing strange under such conditions.⁹³ The profile of the Principate in the second century vastly differed from the world of Late Antiquity. Restricting banquets, one of the central aspects of the emperor’s communication with the senatorial aristocracy, would have been dramatically at odds with the ideal of the civilis princeps, the importance of which is especially evident for the self-representation of Antoninus Pius. During the convivia, the princeps had to present and prove himself as an approachable patronus, not as a tyrant who shut himself off from the senators.⁹⁴ Domitian, the obvious precursor for this sort of behaviour, was certainly not an emperor whom Pius would have wanted to emulate. ⁸⁶ Winterling 1999: 157; Winterling 2009: 97. ⁸⁷ Winterling 1999: 190. ⁸⁸ Winterling 1999: 157–8, 206–7; Winterling 2009: 73–4, 97. ⁸⁹ Nesselhauf 1964. ⁹⁰ HA Ant. Pius 6.4, 7.5. ⁹¹ See, for example, Millar 1977: 625–30. ⁹² Vössing 2004: 473 n. 4. In this context, Winterling 1999: 107 has also referred to HA Marc. 17.5, which actually refers to special clothing; cf. Vössing 2004: 369 with n. 4. ⁹³ Vössing 2004: 473–4; on the late antique ruler, cf. Smith 2007 and Chapter 12 by Martijn Icks in this volume. ⁹⁴ Vössing 2004: 474.

54

   

The latest archaeological research into the remains on the Palatine has demonstrated that that the second-century palace contained numerous, partly interconnected dining rooms of very different sizes that allowed more intimate dinner parties as well as huge banquets.⁹⁵ These were first laid out in the Flavian period, with subsequent modifications carried out under both Trajan and Hadrian. It seems to be equally clear, though, that it is no longer tenable to label any complex of rooms in the domus Augustana as a ‘private area’ belonging to the princeps, distinct from ‘public’ reception rooms.⁹⁶ Emperors used a wide variety of venues throughout the palace to receive and host members of the court. Thus, the differentiation made by the author of the Historia Augusta between ‘public’ and ‘private’ banquets is likely to be their own addition to accounts in their sources about the emperor holding banquets of different sizes with his amici. These were not his own innovations, however, but basically followed the tradition that had developed under his predecessors. As we have seen, Pius placed a significant premium on the effective communication of his civilitas.

5. Conclusion Antoninus Pius was an emperor who consistently conformed to the ideal of the civilis princeps. Although there are elements of the Antonine court that do show an institutionalization of hierarchy and ceremonial in comparison to the early Principate,⁹⁷ the basic challenge for the princeps—to stay on good terms with the aristocracy by treating them as his peers—remained the same. The emperor was rewarded for his performance of civilitas by a positive reception in works written by contemporaries and later authors. But despite the endurance of the idea of equality between the emperor and his peers in the Antonine period, it is clear that a distance was in fact developing between them. This becomes very clear in the correspondence of Fronto who, although he was tutor to Marcus Aurelius and Lucius Verus, cannot be seen as a member of the inner sanctum of court under Antoninus Pius. He instead represents the view of a courtier in the outer circles seeking to ingratiate himself with the emperor. In his letters, Fronto certainly does not meet the emperor at eye level as a peer, but recognizes Pius’ superior status, alternatively flattering him (within acceptable bounds) or apologizing to him depending on the circumstances. Here we see the contrast between the emperor’s exhibition of civilitas and the courtier’s own recognition that despite this performance, he was still dealing with a monarch. Indeed, the idea of the civilis ⁹⁵ Sojc and Winterling 2009; Pflug 2014; Wulf-Rheidt 2015; Pflug and Wulf-Rheidt 2022: 214–16, 225–6. ⁹⁶ Wulf-Rheidt 2012: 106–8; Wulf-Rheidt 2015: 7. ⁹⁷ Cf. Winterling 1999: 28–32.

     ?

55

princeps was a facet of Antoninus Pius’ imperial persona that he mainly employed in his communication with the aristocracy. In other circumstances and in artistic media, he could be imagined as a king ruling over land and sea or as a conquering hero. It was from this discourse that Fronto borrowed in order to express his relationship with Pius, the μέγας βασιλεὺς.⁹⁸

⁹⁸ On the different roles of the princeps in the second century, see Seelentag 2004; 2011; 2017; Michels 2018.

2 Changing the Guard Guard Units and Roman State Ceremonial from the First to the Fourth Century Christian Rollinger

1. Introduction From the bearskin-wearing Queen’s Guard of the modern British monarchy to the Pontifical Swiss Guard in their colourful early twentieth-century uniforms and medieval halberds, guard units always were, and still are, part and parcel of any monarchic court.¹ Their core purpose and role has changed very little over the millennia: in addition to providing different layers of security to the sovereign, they are also essential elements of royal and imperial pomp and circumstance, to which any number of tourists witnessing the daily ritual of the Changing of the Guard at Buckingham Palace will attest. As such, not only do guards actually mount guard within and without the monarch’s palaces, they also participate in a wide variety of ceremonial occasions intended to legitimize the monarch in the eyes of their subjects. This includes (but is not limited to) rituals such as coronations, weddings, or funerals, which, as formal state ceremonies, are certainly among the most impressive pieces of royal theatre still being staged. The same principle applies to the guards of Roman emperors. From the very beginning of their rule, the Caesars surrounded themselves with a number of guards,² the most well-known of which were the Praetorians during the first three centuries of the empire and the scholae palatinae in Late Antiquity, created at the beginning of the fourth century as a replacement for the Praetorians. Perhaps as a reflection of the wide array of duties that guards were asked to fulfil, however, there was never just one single unit detached for protecting the emperor but rather several coexisting ones, including the German bodyguard (corporis Germani custodes) and their successors, the mounted horse guards known as the equites

¹ Cf. Mansel 1984. Contrary to popular belief, the Swiss Guards uniforms were not designed by Michelangelo but are credited by the Guard itself to Commandant Jules Repond (1910–21). ² For the purposes of this chapter, I will refer to individual units by their distinct terminology only when precision is required and will instead mostly use the more general descriptor ‘guard(s)’. Christian Rollinger, Changing the Guard: Guard Units and Roman State Ceremonial from the First to the Fourth Century In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0003

  

57

singulares Augusti.³ The Praetorians themselves were likewise always more than simple ‘bodyguards’: they are better understood as a reserve army stationed in and around Rome and tasked with a variety of duties, among which was the protection of the imperial palaces on the Palatine.⁴ Their close relationship and association with the centre of imperial power and the many privileges they enjoyed over normal legionaries, however, led to their wrongly being seen as somewhat pampered, by ancient authors such as Cassius Dio, and by some modern authorities as well.⁵ By the second century, they had already become an object of ridicule to Dio.⁶ Despite their occasionally murderous meddling in imperial politics, by that author’s time, the Praetorians were perceived as primarily a parade-ground unit and their ‘ceremonialization’ has long been understood as a process of degeneration from their original status of military elite. That being said, even if we were to accept that the guards of the late empire were second-rate soldiers (which, at least for most of the time, is far from the truth), this does not mean that their ceremonial role was not important within the system of imperial government, though it has hitherto received little to no attention from scholars.⁷ Before this can change, however, some general remarks about this court and our understanding of ‘ceremonial’ are in order. The Roman imperial court was the community of individuals surrounding the emperor.⁸ This includes interaction with members of the elite and the populace both within and beyond the boundaries of an architecturally defined palace. It is not always easy to discern from our sources what exactly this entailed. What we have at our disposal are mostly short asides in the literary sources and some surviving iconographic remains. As Michael McCormick has pointed out, this is a fundamental problem in analysing imperial ceremonies: the ‘banality of ceremonies’, as he puts it, in most cases leads ancient authors to ignore them, or at

³ Bellen 1981; Speidel 1994. ⁴ For the Praetorians as such, see now Bingham 2013; De la Bédoyère 2017. Together with other (para-) military units such as the cohortes urbanae or the sailors of the imperial fleets, the Praetorians also provided security during public spectacles and a force of last resort during urban riots, occasionally assisted the vigiles in firefighting, and accompanied the emperor in his travels or as a crack military force on his campaigns. Cf. Bingham 1999; 2013: 100–13; Ricci 2004; 2017: 89–104. Their organization as heavy infantry—complete with military engineers, and detachments of cavalry and speculatores— attest to their military role, as they regularly accompanied emperors or imperial generals on campaign (Menéndez Argüín 2011). For the difference between ‘bodyguards’ and ‘guard units’, see UrriskObertyński 2004, who has elaborated on this in the case of the Hapsburg court. Wolff and Faure 2020 and Hebblewhite and Whately 2023 collect case studies on a variety of ancient bodyguard units. ⁵ Privileges included shorter terms of service (Dio 54.25.6; 55.23.1), higher pay (Dio 53.11.5; 55.23.1) and better living conditions in Rome. ⁶ Dio 74(73).16.3–4. But cf. De la Bédoyère 2017: 219. The ‘softening’ of military units garrisoned in large cities is a common trope, however, and not limited to the Praetorians; cf. Wheeler 1996. Still, this ancient criticism is not completely without merit; when confronted with the armies of provincial governors or usurpers competing for the imperial purple, the Praetorians were usually unsuccessful. ⁷ Even in Benoist 2005, a monograph devoted wholly to some of the same imperial ceremonies under consideration here (and mostly within the same period), the Praetorians scarcely figure. ⁸ See the Introduction by Caillan Davenport and Meaghan McEvoy.

58

   

least to refrain from putting much effort into describing them in any meaningful fashion.⁹ Ceremonial and imperial pomp were very much a routine aspect of life in the capital and our sources could (probably) assume that their readers were familiar with it. Thus, to some degree, we hear of ceremonies (including those involving guards) only when something unusual happened. Nevertheless, there is enough in the literary and archaeological record to shed at least some light on the ceremonial roles of guards, even if the evidence for it is scattered und unevenly distributed. During most (if not all) interactions between emperors and their subjects, palace troops and guards were present and sometimes involved in what is usually referred to as ‘court ceremonial’, though that again is a somewhat vague term that has to be further defined. Nowadays, ‘ceremonial’ has a rather broad semantic meaning, encompassing ‘the manner in which to approach, greet, and depart’ (e.g.) a ruler, as well as ‘words, facial expression, gestures, posture, clothing, and attributes’.¹⁰ In that sense, the meaning of ‘ceremonial’ is more akin to ‘etiquette’ or ‘protocol’. In that sense, too, guards were intimately involved as an ever-present background to everyday appearances of the emperor, both within and beyond the architectural boundaries of an imperial residence. For ‘ceremonial’ stricto sensu, Jeroen Duindam, in a seminal study on Early Modern courts, differentiates between three types: dynastic state, domestic, and diplomatic ceremonial. While dynastic state ceremonial includes the most obvious forms of ceremonial occasions revolving around the monarch’s family and his position as head of state, domestic ceremonial is intended to order a monarch’s household and daily interactions. Finally, diplomatic ceremonial, in the most basic terms, regulates the exchange and reception of ambassadors and envoys at court and their place relative to court hierarchy.¹¹ Aggregating the available evidence from the first to the fourth centuries, the involvement of guards is clearly discernible in all different types of ceremonial. This chapter will focus on the first of Duindam’s categories, however—that is, dynastic state ceremonial. This can further be defined as a form of ‘ritualized contact between the monarch and his subjects’, both high and low.¹² Individual ceremonies that we will discuss include accession rituals and (later) coronations, triumphs, and adventus, as well as imperial funerals, all of them occurring at different points of the evolution of Roman imperial rule. By no means should this be taken to mean that nothing could be said about the guards’ role in domestic or diplomatic ceremonial; this is emphatically not the case, though an exhaustive analysis of that role will need to be undertaken somewhere else.¹³ ⁹ McCormick 1985: 7. ¹⁰ Duindam 1995: 104. ¹¹ Duindam 1995: 102–4. Such categorizations must by no means be seen as absolute; they are modern interpretative aids that are appropriate and fitting for some court contexts but not for others, and even in the case of the Roman court, they are only partially helpful. ¹² Duindam 1995: 103. ¹³ See Rollinger 2023a.

  

59

Guards were involved in audiences granted by the emperor, where they presented an appropriately intimidating military visual background to diplomatic exchanges and negotiations; they were present during imperial banquets and at games and spectacles given by the emperor. Indeed, their very presence as the only military unit within the city, surrounding and guarding both emperor and the imperial domain on the Palatine, formed a constant backdrop of military imperial authority. The exclusion of this part of court ceremonial is thus not motivated by a lack of information. There is, however, a distinct advantage to restricting ourselves to an analysis of dynastic state ceremonial in the context of this present volume: as part of the pageantry of empire, the ceremonial appearances and duties of the guards changed along with the general understanding of the court and the imperial role itself in the slow but steady development of the imperial court from the Principate to Late Antiquity. Imperial state ceremonial was an expression of imperial ideology and, as this ideology evolved, so too did the guard.

2. Rituals of Accession In the early empire, no formal accession ritual per se existed—at least, not in any shape that we would recognize as ‘monarchic’. There could be no coronation as there was no crown.¹⁴ Ideally, the future emperor was a natural son or relative of the previous one or had been adopted into the imperial house, and had already been invested with the major imperial powers by the senate, as Tiberius had been long before the death of Augustus. Even if the new emperor had not previously been granted imperial authority and could boast of no connection to the imperial house, the senate invariably accorded him special legal powers, as it did with the lex de imperio Vespasiani and as it continued to insist on doing well into the third century.¹⁵ This apparent lack of evidence for ‘rituals associated with Roman imperial inauguration’ in the early period, however, must not be taken to mean that no such rituals existed.¹⁶ There is in fact significant evidence for this if we move beyond a strict definition of ‘inauguration’ as ‘coronation’. The closest thing to an official ceremony involving the guard was its formal acclamation of the emperor, to whom they then swore oaths of loyalty.¹⁷ In 14, during the first imperial succession, there was not even this, though the close ideological connection between Praetorians and emperor was already in evidence. Tiberius giving out the watchword to the guard after Augustus’ death is treated by Tacitus as evidence

¹⁴ For the lack of a clear-cut ritual of accession, see MacCormack 1981: 161–5. ¹⁵ Ando 2000: 33 and 152–60. ¹⁶ Pace Nelson 1976: 100. ¹⁷ Cf. Campbell 1984: 23–32.

60

   

for his taking over the mantle of his late predecessor’s rule, with him now being surrounded by ‘guards, weapons, and other courtly attributes’ (excubiae, arma, cetera aulae).¹⁸ Suetonius likewise equates command of the Praetorians with imperial authority.¹⁹ Nor was this combination specific to the early imperial period. In the fourth century, Eusebius explicitly notes that the sons of Constantine were allocated military guards and escorts upon their promotion to the rank of Caesar, and in his letter notifying Constantius II of his acclamation as Augustus, Julian explicitly reserves the right to appoint guards (stipatores) among the concessions he is prepared to make.²⁰ Given the circumstances of his own ascent to power, it was only natural for Claudius to cultivate a close relationship to the Praetorians, as expressed by his coinage. Nero, too (though for slightly different reasons), coveted a close relationship with the guards. His connection to the Praetorians was emphasized in a curious, but obviously choreographed ceremony some few hours after the death of Claudius: the gates of the palace were suddenly thrown open, and Nero, accompanied by Burrus went forth to the cohort which was on guard after military custom. There . . . he was hailed with joyful shouts, and set on a litter.²¹

The young emperor, like his predecessor before him, was then conveyed to the Praetorian Camp and officially acclaimed. The conscious decision of associating himself closely with the Praetorians might have been orchestrated by his mother Agrippina, who had paved the way for this during his boyhood: in order to further his standing and position him as potential heir and rival to Britannicus, Agrippina had had him parade together with the guards.²² A series of sestertii struck in the final phase of his rule shows Nero on horseback alongside a Praetorian soldier (mounted and unmounted). It may be an allusion to the episode of his youth or, perhaps more likely, an attempt at re-inventing the notoriously un-military Caesar as a military man (Figure 2.1). The legend DECVRSIO hints at his possible participation in that military exercise.²³ That the presence of guards was a

¹⁸ Tac. Ann. 1.7.5. In M. Aur. Med. 1.17.3, guards later appear as a natural part of court life: the example set by his predecessor Antoninus Pius, who seems to have reduced the presence of guards noticeably, is remarkable precisely because it was not the norm. ¹⁹ Suet. Tib. 24. ²⁰ Eus. V. Const. 4.18.1, 51.3. For Julian’s letter, see Amm. 20.8.5–17, esp. 14. ²¹ Tac. Ann. 12.69: fortibus palatii repente diductis, comitante Burro Nero egreditur ad cohortem, quae more militiae excubiis adest. ibi . . . faustis vocibus exceptus inditur lecticae (trans. Jackson, Loeb). ²² Tac. Ann. 12.25–6; Suet. Ner. 7.2. ²³ RIC I² (Nero) 103–8, 163–77. For the decursio, both a training exercise for young Romans and an imperial ritual, see below, and cf. Richard 1966: 314–15. The same motive is also present in provincial coinages. Nero’s coinage also heavily features the Praetorians, if we understand the soldiers on the ADLOCVT COH sesterces (e.g. RIC I² Nero 130–6) as members of that unit (and there are valid grounds for this).

  

61

Figure 2.1 Sestertius of the Emperor Nero (RIC I² 105). Münzkabinett, Staatliche Museen zu Berlin, 18,220,895. © Münzkabinett, Staatliche Museen zu Berlin. Photograph by Dirk Sonnenwald.

hallmark of monarchy and thus inherently fitting on the occasion of a new ruler ascending to the purple is confirmed by other sources. Vitellius, in one of his first measures after claiming the throne, ascended the Capitol to offer sacrifice; he was accompanied by the Praetorians.²⁴ Nerva likewise ascended that sacred hill to offer sacrifice and then, surrounded by his Praetorians, announced his adoption of— and thus the soon-to-follow accession of—Trajan.²⁵ In the late second century, by which time the evolution of imperial rituals had progressed considerably and accession rituals had solidified in their form, the ritual had changed directions, so to speak. It was now the laurel-wreathed Praetorians who escorted the new emperor to the imperial palace.²⁶ Claudius and Nero, after their acclamation by the Praetorians, had presumably been escorted back to the Palatine (as, indeed, other emperors likely also were). But Pertinax, after his acclamation at the castra Praetoria, rather than on the Palatine, was accompanied by both Praetorians and parts of the populace to the palace, as were Didius Iulianus and Pescennius Niger.²⁷ After Septimius’ dramatic adventus (see below), he likewise ascended the Palatine surrounded by his guard.²⁸ The

²⁴ Dio 65(64).5.2. ²⁵ Dio 68.3.4. Cf. Cedr. p. 433.20–434.2 (ed. Bekker). ²⁶ Hdn. 2.2.9 (Pertinax). ²⁷ Iulianus: Dio 74(73).13.1; Hdn. 2.6.13. Niger: Hdn. 2.8.6. Cf. also HA Comm. 42.7, Max. 14.4, Tres Gord. 9.6, Balb. 8.3. ²⁸ Hdn. 2.14.2.

62

   

ceremonial had acquired such normative power that it was performed even if neither palace nor Praetorians (or any guards, really) were at hand to perform it. When Pescennius Niger had been declared emperor in Antioch, soldiers and populace—in an imitation of the urban Roman ceremonial—had escorted him to his home, which they now ‘regarded as an imperial palace’ (βασίλειον αὐλὴν νομίζοντες) and decorated accordingly with the insignia of empire.²⁹ Tellingly, after Gordian I was acclaimed as emperor in Thysdrus and travelled to the provincial capital (and palace) of Carthage, garrison soldiers and the fair young men ‘in the manner of the Roman guard [i.e. the Praetorians]’ (ἐν σχήματι τῶν κατὰ τὴν Ῥώμην δορυφόρων) accompanied him, surrounding the usurper in a procession of appropriate imperial pomp.³⁰ The close connection between the presence of guards and accession rituals of the emperors remained evident between in the late third century and throughout Late Antiquity, though the guard units themselves and their specific ceremonial roles changed considerably. Under the Tetrarchy, emperors such as Maximinus Daza and Maxentius were ‘made’ by and in front of the Praetorians who acclaimed them; if possible, they were clad in the imperial purple by reigning Augusti. The purple paludamentum remained the most visible sign of rulership until the appearance of the diadem under Constantine; even then, it was supplemented by the diadem, not supplanted.³¹ Likewise, even under the changed ceremonial circumstances of later accessions, the presence of military guards remained a constant. Julian, as the first Roman emperor to be raised on a shield and crowned with a torques³²—elements that were to remain part of an increasingly regulated (though highly flexible) ceremonial until the sixth century and beyond—was surrounded by his troops, among them, one must presume, his guards.³³ Though this is of course conjecture, one may reasonably surmise that it was also the shoulders of guards that bore the shield—‘O happy shield!’ (ὡ μακαρίας ἀσπίδος) in Libanius’ turn of phrase—on which Julian was raised.³⁴ In his Ecclesiastical History, at any rate, Socrates explicitly states that it was one of the guards who ‘crowned’ Julian with his torques.³⁵ Likewise, Valentinian I received

²⁹ Hdn. 2.8.6. ³⁰ Hdn. 7.6.2, though ἐν σχήματι could also be taken to mean their physical appearance. Imperial pomp included the new emperor being clad in purple (e.g. Hdn. 2.8.6) and preceded by what Herodian (1.16.4) calls ‘the symbols of imperial power’ (τῶν βασιλικῶν συμβόλων), among which a religiously connoted fire/torch (e.g. Hdn. 1.8.4: τὸ πῦρ; cf. Dio 71.35.5: τὸ φῶς) was most prominent (Alföldi 1970: 111–18), as well as a number of now-canonical sacrifices to be offered (Hdn. 2.3.11, 2.6.12). The imperial fire (πῦρ) was also carried in front of imperial women (Hdn. 1.8.4, 1.16.4). ³¹ Cf. Lact. DMP 19 (Daza), 26 (Maxentius). For imperial insignia, see Alföldi 1970. ³² The torques is ‘a collar of twisted metal’, worn as a military decoration (OLD s.v. torques). In this case, it was allegedly gifted to Julian by the standard-bearer Maurus. ³³ Amm. 20.4.17–18; Lib. Or. 18.99; Zon. 13.10; Zos. 3.9; Jul. Ep. ad Ath. 284D all provide detailed descriptions of the ritual. On the history and enduring presence of the torques, see Enßlin 1942. ³⁴ Lib. Or. 13.34. ³⁵ Soc. 3.1.35.

  

63

the purple, torques, and diadem in front of troops that certainly included members of the scholae palatinae.³⁶ In later cases, of which we are relatively well-informed thanks to the so-called ‘coronation’ protocols preserved in the tenth-century Book of Ceremonies, the presence of guards is more difficult to pinpoint, though they were certainly involved, and military units in general played an important role. The presence of members of the scholae is explicitly attested in these ceremonies, but their specific role is uncertain. In the case of Anastasius and Justin I, for instance, the torques is placed on the emperor’s head by a campidoctor of the lanciarii, one of six legiones palatinae, instead of a member of the scholae or the excubitores. There is some evidence, however, that already in the late third century soldiers from the Praetorians had been recruited into the elite units of the imperial comitatus, some of which would then evolve into the future legiones palatinae.³⁷ This, together with the appearance of other units of the comitatus enjoying a close connection to individual emperors, may well be the historical background to the notice in Sozomen, who alludes to a tradition of officers of the Ioviani and Herculiani (which were also part of the palatine legions) acting as bodyguards to the emperors.³⁸ The absence of an explicit mention of other guards is perhaps particularly surprising in the case of Justin I, given that he was himself comes excubitorum, though it is a good indication as to the degree to which a fixed ceremonial was already established.³⁹ We do know, however, that the testudo formation that acted to shield the emperor from view during his change of dress consisted of candidati (i.e. a select corps of bodyguards from the scholae).⁴⁰ Though it is virtually certain that guards continued to take part in such ceremonies throughout Late Antiquity and well into what we now call the Byzantine period, their role steadily diminished as ‘coronation’ ceremonial changed and the military facets of the associated rituals receded from view.⁴¹ In 602, Phocas was likely the last emperor to be raised on a shield in a purely military ceremony until the thirteenth century, when this specific ritual surfaced again. He also— tellingly—was the first one to be crowned by the patriarch in a church.⁴² ³⁶ Amm. 26.2.2–3; Philost. HE 8.8; Joh. Mal. 13.28. ³⁷ Petitjean 2020, particularly 436–7 with n. 33 for selected epigraphic evidence. ³⁸ Soz. HE 6.6.4; Petitjean 2020, 440–4. Cf. Const. Porph. de cer. 1.92–3 (1.101–2) for the coronations of Anastasius and Justin. Likely, Leo I (de cer. 1.91 [1.100]) had also been ‘crowned’ by a campidoctor. ³⁹ For these late coronations, see e.g. Trampedach 2005; Sode 2009, and compare Const. Porph. de cer. 1.91–5 (1.100–4). ⁴⁰ Explicitly mentioned for Leo in de cer. 1.91 (1. 100) and probable for Justin in de cer. 1.93 (1.102). On the candidati, see Frank 1969: 127–42 and Whitby 1987: 465. ⁴¹ Cf. Trampedach 2005, esp. 280–3, with Rollinger 2021: 267–337. ⁴² That is, in the church of St. John at the Hebdomon; cf. Theoph. Sim. 8.10.6; Chron. Pasch. s.a. 602. The first ‘coronation’ in Hagia Sophia by the patriarch of Constantinople likely occurred in 610 during the successful usurpation of Heraclius, and not in 641, when Heraclius’ grandson Constans II was certainly crowned on the ambo of Hagia Sophia (Nic. Brev. 31); cf. Rollinger 2021: 300–26; Rollinger forthcoming. For the later reintroduction of shield-raising, see Mantas 2000; Teitler 2002.

64

   

3. Triumph and Adventus Mostly thought of as a quintessentially republican ritual, the Roman triumph by no means disappeared under the empire; in fact, it merged with the already not dissimilar ceremony of adventus.⁴³ As specific victories worthy of triumphs were slowly replaced by the general victoriousness of emperors and empire, both ceremonies became virtually indistinguishable. Soldiers were a traditional element of both pageants and must certainly have played a significant part in both, though our sources very seldom explicitly attest to this. The lone exception for the Principate is Claudius’ British triumph in 44, though even here the Praetorians’ presence is implied rather than explicitly stated, based on the fact that their prefect figured prominently in the procession.⁴⁴ Seven years later, the Praetorians were used to represent visually imperial power and to impress the military might of Rome on populace and visitors alike. On this occasion in 51, they set aside their usually more restrained city dress and instead appeared fully armed to ritually accentuate Claudius’ British victory when he presented the captive Caratacus to the Roman populace, surrounded by the Praetorian cohorts.⁴⁵ Tacitus chides that the populace was invited ‘as if to some spectacle of note’ (quippe ut ad insigne spectaculum) and comments that Claudius had done this to add to his own prestige.⁴⁶ As with the triumph, Praetorian participation in the imperial adventus was routine procedure. At the beginning, the involvement of the guard in this ritual may be said to have been somewhat incidental, a consequence of their escorting the emperor on the move and particularly when travelling outside of Rome. But from this simple beginning, they became an integral part of the adventus ritual, which at its most basic consisted of the inhabitants and particularly the upper classes of a city setting out to meet a high-ranking visitor such as (but by no means limited to) the emperor some miles outside the city and escorting him in.⁴⁷ The presence of different guard units during adventus soon became a marker of imperial distinction. Nor was this privilege limited to the emperor alone; it also extended to the imperial family, with Praetorians welcoming members of the imperial house in a formal adventus ceremony on the order of the emperor. Thus, on Germanicus’ return shortly after the mutinies on the Rhine in 14, two praetorian cohorts were sent to greet him by Tiberius. However, in a breakdown of discipline that must have rankled the cantankerous princeps, the entire guard

⁴³ On the subject of imperial triumphs, see Benoist 2005: 195–272 and now especially the excellent papers in Goldbeck and Wienand 2017, as well as Beard 2007 generally. ⁴⁴ Dio 60.23. ⁴⁵ For Praetorian dress, see Rankov 1994: 18–27; Bingham 2013: 75–9. ⁴⁶ Tac. Ann. 12.36. ⁴⁷ For an in-depth study of the adventus ritual during the principate, see Lehnen 1997; Benoist 2005: 25–102. For the late antique configuration of the ritual as a victory celebration, see McCormick 1986: 35–130; Rollinger 2021: 417–55.

  

65

ultimately went on the mission and, together with a large part of the Roman populace, met the general at the twentieth milestone.⁴⁸ Upon his death, Germanicus’ ashes were met in Brundisium by two praetorian cohorts and escorted back to Rome in solemn fashion, the urn being carried by officers of the guard.⁴⁹ It is worth noting that in both cases the Praetorians were present in the retinue of Germanicus anyway, as detachments already accompanied him on campaign and official duties.⁵⁰ The additional cohorts sent to greet him were thus meant to honour him further; they also acted as symbolic representatives of the princeps himself. The adventus ritual itself was highly adaptive and could be easily modified to suit particular demands. In 59, Nero entered the city of Naples in a travesty of a traditional triumph and then repeated this entry in Antium, Albanum, and finally Rome, each time accompanied by a theatrical claque, soldiers (meaning Praetorians), equestrians, and senators.⁵¹ Contrast this with the image of Trajan’s entry into Rome, as described by Pliny the Younger in his panegyric: no throng of followers [satellitum manu] attended you; you moved in the midst of the flower of the senators or equites, as the numbers of either party prevailed as they gathered round you, and your lictors quietly and courteously cleared your path. As for the soldiers [milites] present, they differed from the civilians in neither dress, propriety, nor discipline.⁵²

Pliny lauds the emperor for the pointedly unmilitary bearing of his Praetorians. Having the Praetorians don civilian dress or at least forego full armour and military equipment and thus appear in a simple tunic was a programmatic statement, a visual marker of the kind of ruler Trajan wanted to be, or rather, wanted to be seen to be. In avoiding openly showing the ultimate base of power for any Roman emperor—the loyalty of his soldiers—he chose a mode of rulership modelled on Augustus himself, the paradigmatic princeps.⁵³ Intriguingly, however, and in a telling indication of the importance of imperial pomp, the inhabitants of

⁴⁸ Suet. Calig. 4. The distance from the city at which the visitor was welcomed correlated directly with the degree of honour rendered to him. ⁴⁹ Tac. Ann. 3.2. ⁵⁰ Thus e.g. Tac. Ann. 2.16. ⁵¹ Suet. Ner. 25; Dio 63.20. ⁵² Plin. Pan. 23.3: Neque enim stipatus satellitum manu sed circumfusus undique nunc senatus, nunc equestris ordinis flore, prout alterutrum frequentiae genus invaluisset, silentes quietosque lictores tuos subsequebare; nam milites nihil a plebe habitu tranquillitate modestia differebant (trans. Radice, Loeb). ⁵³ The ostentatious statement by Pliny that Trajan was surrounded by senators and equestrians and not by a throng of attendants (satellitum manu) is more difficult to interpret. It might be nothing more than a literary trope, allowing Pliny to extol the civic virtues and affability of Trajan. On the other hand, it may also refer to the absence of ‘proper’ bodyguards (as opposed to the Praetorian milites), likely meaning members of the German guard of that time, whose presence Trajan may have deemed inappropriate for the occasion. Cf. Lact. DMP 38.7, where the bodyguards (stipatores, not soldiers) of Maximinus Daza are described as his satellitibus et protectoribus.

66

   

Rome seem to have expected and looked forward to it, much as modern-day royal pageants are widely anticipated.⁵⁴ Septimius Severus, on the other hand, entered Rome during the tumultuous period after Commodus’ murder accompanied by his army.⁵⁵ Though he himself had changed into civilian clothes and entered on foot, presumably to soften the already hyper-militaristic appearance of the whole affair, the armoured soldiers and mounted troopers in his wake—Severus had created his own guard from the troops available to him—stood out in an otherwise rapturous adventus.⁵⁶ Though such ‘military’ arrivals would have been a regular occurrence in provincial cities when the emperor was travelling or campaigning (and are occasionally depicted in imperial art⁵⁷), a military procession outside of a traditional triumph or ovation within Rome was a clear violation of the traditional rules of imperial decorum. In a way, it might be seen as portending the heightened importance of the army during the Severan dynasty and particularly during the third century. As with the empire as a whole, so too did the guards emerge from this tumultuous period in a different shape than they had entered, though very little is known about them during this time.⁵⁸ The description of Gallienus’ guards during his decennalia—if it is indeed a guard detachment that is meant by the five hundred hastae auratae (‘golden spears’)—is more a reflection of the later appearance of fourth-century guards than those of the late third century. The notorious reforms of Diocletian, who introduced imperial garments set with precious stones and a strict ceremonial of submission to imperial authority, was likely mirrored in the appearance of the guards themselves, perhaps already then decked out in shimmering gold and glittering jewels, heralding the more developed and ostentatious court ceremonial of the late antique period.⁵⁹ There is too little in the way of explicit sources, however, to be quite certain of this. Neither are the iconographic sources helpful. Recent excavations in Nicomedia have produced a truly stunning early Tetrarchic relief of an adventus scene including Diocletian and Maximianus that shows both emperors dismounting from their carruca, a ceremonial state car, and embracing with the original vibrant colouring mostly intact. Naturally, guards are also included in this depiction, but unfortunately only in a classicizing iconographic mode—including ⁵⁴ Mart. 10.6.7f. ⁵⁵ Dio 75.1.3–5; HA Sev. 7. ⁵⁶ For Severus’ reform of the equites singulares, see Speidel 1994: 42–5. ⁵⁷ Cf. for instance panels LXXXIX and XC on the column of Trajan for an adventus in solo barbarico. ⁵⁸ We know, at least, that Praetorians were actively campaigning with Alexander Severus (Hdn. 7.11.2), Gordian III (HA Tres Gord. 27.2, 27.7), Aurelianus (Zos. 1.50–3), and Maximian (CIL VIII 21021). ⁵⁹ For Gallienus’ decennalia, see SHA Gall. 8.1–6. For Diocletian’s reforms, see Rollinger 2023b. For the concomitant change in the appearance of the guard, see e.g. Lact. DMP 37.5. There is, however, evidence of an earlier predilection of the guards for expensive and precious equipment; see e.g. Hdn. 2.13.10. Laeben-Rosén 2005: 193 views daggers inlaid with silver and gold as intended ‘for ceremonial use’, but there may be other reasons.

  

67

muscle cuirasses and classically circular shields—that tells us little about their actual appearance.⁶⁰ An equally eye-catching though little-known terracotta medallion mould discovered in Sardinia nearly two decades ago shows a Tetrarchic triumphal procession consisting of no fewer than one hundred and fifty persons, with two Tetrarchs seated in the same type of ceremonial open carriage (carruca), drawn by four elephants and surrounded by the Roman senate and a plethora of guards who are, by far, the most numerous group depicted.⁶¹ Perhaps the most well-known depiction of a late-antique adventus is that of Constantine’s arrival into Rome in 312 on the Arch of Constantine. The seated figure of the emperor, motionless on his carruca, is preceded precisely by lines of infantrymen and columns of troopers with dragon standards. This matches Ammianus’ description of an adventus of Constantius II in 357, replete with all the pomp and splendour that fourth-century Rome could muster. By then, both Praetorians and equites Augusti had ceased to exist, disbanded after Constantine’s victory at the Milvian Bridge. In their place, a new guard had been established, the scholae palatinae.⁶² In both appearance and demeanour, the scholares differed from their predecessors. As a distinctive sign of their unit, the scholae carried golden spears (the hastae auratae of Gallienus’ adventus?) and were ensconced in jewel-encrusted armour, the glittering nature of which is dramatically emphasized by Ammianus.⁶³ Two things are immediately apparent when comparing this adventus to earlier forms, showing the evolution that the ceremonial had undergone some time between Severus and Constantine. Whereas Trajan’s entry in Rome was a masterpiece of understatement—an attempt, really, to hide the military foundations of Roman imperial rule and to present himself as a civilis princeps—Constantius’ entry is nakedly militaristic, even more so than Severus’ had been.⁶⁴ Ammianus himself compares the adventus, a thoroughly conventional ceremony, to a military campaign and emphasizes that Constantius approached the Tiber ‘as if he were planning to overawe the Euphrates with a show of arms, or the Rhine’.⁶⁵ There is an unmistakably disapproving and ironic undertone to Ammianus’ account, as he explicitly connects the adventus to the defeat of the usurper Magnentius and bitterly accuses Constantius of triumphing ‘over Roman ⁶⁰ Though this may yet change, as the finds have not yet been published in their entirety. See however Hacì Zeyrek and Özbay 2006, especially p. 297 and fig. 13, as well as Şare Agtürk 2018 for excellent figures (esp. fig. 8). ⁶¹ See Haake 2017: 378–82 with notes and figs. 12.1–2. For a detailed iconographic analysis of this little-known artefact, see Gualandi 2010. For adventus and other ceremonies in the Tetrarchic principal residences, cf. Guidetti 2023 and see Chapter 3 by Verena Jaeschke and Caillan Davenport in this volume. ⁶² On the scholae in general, see Frank 1969 and now Janniard 2020. ⁶³ Amm. 16.10.6–8. On the apparel of late antique guards, see Delmaire 2008b. ⁶⁴ For the inherent tension in Roman emperorship between appearing ‘civil’ and wielding autocratic power during the Principate, see also Chapter 1 by Christoph Michels in this volume. ⁶⁵ Amm. 16.10.6: et tamquam Euphraten armorum specie territurus aut Rhenum.

68

   

blood’ (ex sanguine Romano).⁶⁶ We cannot discount the possibility, therefore, that Ammianus consciously stressed the military aspect of the adventus to drive his point home; indeed, this is likely. The criticism becomes doubly apparent when contrasted with his more laconic account of Julian’s adventus in Constantinople, which only vaguely alludes to the fact that Julian had arrived in the suburbs with an army—and as a usurper. In fact, Julian is shown here in a distinctly Trajanic mould, with throngs made up in equal measure of guards and citizens surrounding and accompanying his adventus, and Ammianus subtly makes the contrast to Constantius’ arrival explicit, when he adds that Julian was escorted ‘as if by an army in line of battle’—but an army of togati, of citizens.⁶⁷ However, that is not at all the same as saying that the military aesthetics in Ammianus’ account of the Roman adventus were invented, exaggerated, or, indeed, would have been perceived as pejorative by a majority of spectators. The presence of well-turned-out and armed guards in the vicinity of the emperor was expected, particularly during a ritual that had become intermingled with the traditional triumph. It bears pointing out that even Constantine, in the famous relief on the arch which the Senate erected for him to commemorate his ‘liberation’ of Rome in 312, was surrounded by armati, not togati. That this military presence in Late Antiquity enhanced, rather than diminished, the prestige of an imperial progress, is a common trope in contemporary sources; Synesius, for one, emphasizes that guards announced the near presence of imperial majesty just as sunbeams announced the presence of the warming sun.⁶⁸ One further subtlety should be pointed out, however. The emperor in Ammianus’ description is still, immobile like a sculpture; this is explicitly phrased in these terms for Constantius and the account has become the locus classicus for describing the hieratic nature of imperial ceremonial in Late Antiquity. There is no reason to suspect that Julian, for all his later resistance to and purposeful deviation from what we might call ‘standard’ late antique ceremonial, would have deliberately chosen not to stage his first official entry into the imperial city according to the prevailing norms.⁶⁹ As guards were widely seen as extensions of the emperor, as mirror images of the imperial person, it is no surprise that we should find the same imagery applied to the guards in late antique panegyrics. Authors such as Claudian compared mounted and armoured troopers, clibanarii, which were organized into a number of army units and also made up several of the ⁶⁶ Amm. 16.10.1. Cf. Them. Or. 3.42C, who characterizes the occasion of the adventus in Rome as νικητέρια. ⁶⁷ Cf. Amm. 22.2.4: stipatusque armatorum et togatorum agminibus, velut acie ducebatur instructa. See Flower 2015; Meurer 2019: 75–7; Ross 2021: 107–13. ⁶⁸ Syn. de regno. 12.6. ⁶⁹ See Chapter 14 by Caillan Davenport and Meaghan McEvoy for Julian’s attempt to ‘dial down’ ceremonial. An important point, explicitly made in the Introduction, is taken up in Section 3: changes to ceremonial were, to a significant degree, matters of negotiation and Julian’s attempts at reform came to naught because they lacked support or even acquiescence by the court and his successors.

  

69

guard scholae, to moving statues made of iron (simulacra . . . ferrea) and no less a figure than Julian himself lauds their immobile, artificial appearance in his two panegyrics in honour of Constantius: they all sat their horses like statues, while their limbs were fitted with armour that followed closely the outline of the human form . . . The head and face are covered by a metal mask which makes the wearer look like a glittering statue . . . They ride their horses exactly like statues.⁷⁰

In reality, as in the panegyrics, the dignity of the imperial office was thought to be enhanced by a motionless emperor who appeared more as an icon (tamquam figmentus hominis, as Ammianus describes Constantius) than as a living being. The same qualities were also to be expected by his guard units in their ceremonial roles. The appearance of the scholae was deliberately designed to inspire both awe and wonder, with respect to the military prowess of the troops themselves and with respect to the state and/or the ruler capable of fielding such troops.

4. Imperial Funerals While triumphs and adventus of all imperial periods were certainly impressive, perhaps the most visually arresting ceremony that guards regularly took part in was associated with imperial obsequies, the funus imperatorum, and the rituals of divinization. The ritual itself, as we understand it, was arranged in a precise and fixed order, progressing from the funeral to the cremation and subsequent declaration of the emperor’s deification (consecratio) by the senate.⁷¹ For the whole of the Principate, (somewhat) detailed descriptions of only three such funerals survive: those of Augustus, Pertinax, and Septimius Severus.⁷² Late antique imperial burials are scarcely better attested: we have more or less extended descriptions of those of Constantine and Constantius II as well as the very detailed (albeit highly panegyrical) account of that of Justinian in Corippus’ panegyric to Justin II.⁷³ Other than that, there are only scant references and brief asides. In the case of Augustus, there are only very slight hints of guards participating in the procession or arrayed in Rome in the available accounts. Dio simply ⁷⁰ Claud. in Ruf. 2.355–65 (trans. Platnauer, Loeb); Jul. Or. 1.37C–D (ἄπειρον γὰρ ἦγες ἱππέων πλῆθος, καθάπερ ἀνδριάντες ἐπὶ τῶν ἵππων ὀχουμένους, οἷς συνήρμοστο τὰ μέλη κατὰ μίμησιν τῆς ἀνθρωπίνης φύσεως) and 2.57C–D (αὐτοὶ δὲ ἀτεχνῶς ὥσπερ ἀνδριάντες ἐπὶ τῶν ἵππων φερόμενοι) (trans. Wright, Loeb). Cf. Amm. 16.10.9–10. ⁷¹ Arce 2000: 116–17. Arce 1988 remains the standard work on imperial funerals, but see also Benoist 2005: 103–94. ⁷² Augustus: Suet. Aug. 100; Tac. Ann. 1.8.4–5; Dio 56.30.47. Pertinax: Dio 75(74).4.2. Septimius Severus: Hdn. 4.2. ⁷³ See below and cf. Rollinger 2021: 358–92 for late antique funerals.

70

   

mentions their participation and Suetonius indicates only that forty Praetorians carried the august body forth from its deathbed.⁷⁴ The presence of Praetorians at the funeral is also confirmed by Tacitus, who adds that the soldiers were arrayed ‘in guard formation’ (velut praesidio stetere) and that this was grounds for ridicule by the crowd, who derided Tiberius’ need for military protection ‘to ensure a quiet burial’ (ut sepultura eius quieta foret).⁷⁵ As part of the (belated) funeral of Pertinax, nearly two centuries later, the military was in full view. A columned shrine of ivory and gold was constructed in the Forum Romanum and a wax effigy of the emperor lay in state inside, while a large procession filed past. Senators and their families were followed by waxen images of great men (viri illustres), singing choruses, and bronze statues dressed up as representatives of peoples subjected to Rome. Behind these were detachments of cavalry and infantry in full armour. This can only mean the new Severan Praetorian cohorts and their cavalry detachments, the equites praetoriani, though these may have been accompanied by the equites singulares, and Dio’s account does not specify to which units these soldiers belonged.⁷⁶ The procession then moved to the Campus Martius, where a pyre had been built. Here, the praetorian cavalry and infantry, among other participants, performed a ritual known as the decursio, which seems to have consisted chiefly of groups and units circling and counter-circling around the pyre. Though we do have an abbreviated visual depiction of this ritual on the base of the Column of Antoninus Pius, much of the detail remains unknown. The column relief itself is heavily restored, but the identity of the soldiers is not in question: the imperial imagines included in the standards are characteristic of the Praetorians.⁷⁷ The cavalry, on the other hand, is an amalgam of riders wearing military dress and carrying vexilla, likely Praetorians and equites singulares, and a number of what appear to be civilians, perhaps members of the equestrian order.⁷⁸ It is also possible that the two prominent figures depicted on horseback in the lower register represent the new emperors Marcus Aurelius and Lucius Verus, who may personally have participated in the ritual.⁷⁹ Public displays of manoeuvres called armaturae or pyrrhicha militaris are known to have been performed by young equestrians of Rome in the second century, and they were very similar to, or maybe even identical with, the decursio. The latter was, however, strongly associated with the emperor in his function as ⁷⁴ Dio 56.42.1; Suet. Aug. 99. ⁷⁵ Tac. Ann. 1.8.6. ⁷⁶ Severus had disarmed and dismissed the existing cohorts even before his triumphal entry into Rome (Hdn. 2.13). He then later created new cohorts from among his legionaries (Hdn. 2.14.5). In contrast to the earlier corporis custodes, which were a private formation, so to speak, of the emperor, the praetorian equites were recruited from the auxilia and thus part of the ‘official’ Roman military. Cf. Speidel 1994: 61–78. Speidel’s confident assertion that the same was true for the corporis custodes is somewhat dubious; there is absolutely no proof for it. ⁷⁷ For Praetorian insignia, cf. Rankov 1994: 24–7. For the imagery in general, see Vogel 1973. ⁷⁸ Cf. Hdn. 4.2.9. ⁷⁹ Speidel 2017: 58f.; cf. HA Marc. 8.2.

  

71

commander-in-chief and thus an important part of imperial ideology: ‘the final tribute to the commander of the legions and his felicitas in the theatre of his exploits’.⁸⁰ In fact, it seems to have been performed even in cases where the emperor or a member of the imperial family died and was cremated away from Rome, as was the case with the elder Drusus and Septimius Severus.⁸¹ It is also worth pointing out that we know of a select few occasions on which imperial women were honoured with a ‘public’ (i.e. imperial) funeral, which included the involvement of the guard. One such case is the funeral of Drusilla, sister of Caius, as described by Dio. It involved both Praetorians and equestrian order circling the pyre and noble youths performing the lusus Troiae.⁸² The decursio equitum likely remained part of imperial funerals in Late Antiquity as well, though there is little in the way of explicit evidence to support this. Upon Constantius’ death, for instance, the later Emperor Jovian, then member of the protectores domestici, was tasked with transferring the body to Constantinople, ‘leading it with imperial ceremony’ (cum regia prosequi pompa).⁸³ It was accompanied by parts of the army, the scholae palatinae and the protectores, just as Germanicus’ ashes had been escorted by the Praetorians three centuries earlier. Javier Arce has taken this imperial procession together with an aside in Ammianus to mean that a decursio was held as part of the traditional pompa funebris, which indeed seems likely but is not explicitly stated.⁸⁴ Constantius, like his father before him, was then buried in a Christian ceremony in the Church of the Holy Apostles.⁸⁵ What the guard’s role was in the obsequies, properly speaking, is difficult to establish with any certainty. Upon the death of Constantine, at least, they had been the first to react to the news in a ritual way, tearing their clothes, throwing themselves on the ground, beating their heads and bewailing his demise. They then placed the purple-wrapped body in a golden coffin and laid it to rest in state in a hall of the palace, likely the Dekanneakoubita (‘Hall of the Nineteen Couches’), resting on a high pedestal and surrounded by candles mounted on golden stands.⁸⁶ There, the grandees of the empire filed past and

⁸⁰ Richard 1966: 315: ‘un ultime hommage rendu au chef des legions et à sa felicitas sur le théâtre de ses exploits’. ⁸¹ Drusus: Suet. Claud. 1.6. Septimius: Dio 77(76).15.3–4. ⁸² Dio 59.11.2. For a description of the Trojan games, see Verg. Aen. 5.545–603. ⁸³ Amm. 21.16.20. Cf. Greg. Naz. Or. 5.16f. ⁸⁴ Arce 1984. Cf. Amm. 21.16.21: ex usu crebescebant occursus. A similar procession is described for Constantius I (Eus. V. Const. 1.22) and Constantine (Eus. V. Const. 4.70). ⁸⁵ There is little solid information and much discussion about this church, its appearance, and its specific function; see Rebenich 2000: 309–10 and esp. n. 50 for the literature. The recently published collection of chapters in Mullet and Ousterhout 2020 is unlikely to be surpassed soon. Cf. Eus. V. Const. 4.71 for Constantine’s ceremony. ⁸⁶ Eus. VC 4.65–7. It is unclear what hall was meant exactly; the Nineteen Couches are named as the traditional place for it in Const. Porph. de cer. 1.60 (1.69) and in the account of the burial of Constantine VII in Theoph. Cont. 6.20.

72

   

paid obeisance.⁸⁷ This ceremonial likely remained standard procedure; the details of Constantius’ funeral which Gregory of Nazianzus records show that it closely resembled that of his father, with the added detail of a pompa funebris stretching from Mopsuestia in Cilicia to the imperial city itself, where it was greeted by his successor in a variation of a formal adventus.⁸⁸ Apart from a difference in location, then, this part of the funeral closely resembled that of other emperors and, in fact, newer Christian elements seem to have formed something of a ceremony within a ceremony during the early Constantinian dynasty.⁸⁹ The funeral mass itself, now the climax of imperial burials and in contrast to the earlier pompa, was no longer a public event in the same sense that the formal cremation of Severus had been. Since participation in the ceremony was restricted to the baptized, parts of the populace were ipso facto excluded. This included large swathes of the soldiery, guards, and officer corps (if they were not excluded from the ceremony in Holy Apostles on the basis of their social status alone), and even some emperors: Constantius II had been unable to attend the funeral mass of his father for precisely that reason.⁹⁰ Naturally, Julian would also have been absent from that of Constantius. In a way, then, these early Constantinian burials were reflective of that ‘quintessential Constantinean [sic] ambiguity’, incorporating both traditionally pagan and innovative Christian elements (including divinification) in imperial representation and ceremonial.⁹¹ Though the Constantinian model remained influential for some time, the specific ceremonial must have remained susceptible to changes and evolved in the centuries until its codification in the Book of Ceremonies.⁹² There is no mention of military exercises in other extant accounts of late antique or properly Byzantine imperial funerals, and that ritual may have lost its connection to imperial funeral rites.⁹³ Instead, we may assume, the guards were relegated to a less ostentatious display of loyalty and mourning: stationed around the catafalque, they stood guard. Likely, as was the case since Augustus, it was still the guards that carried the imperial bier in the final procession. While this is nowhere explicitly mentioned, it seems a logical conclusion and there is ⁸⁷ For the role of the court community in the funeral, see further Chapter 4 by Audrey Becker in this volume. ⁸⁸ For a detailed analysis of the funerals of Constantine and Constantius, see Rebenich 2000. ⁸⁹ Karlin-Hayter 1991: 136. ⁹⁰ As Eus. V. Const. 4.71 attests, the ceremony began only when Constantius and the military had withdrawn. ⁹¹ Arce 2000: 122; Rebenich 2000; Dagron 2003: 135–43. ⁹² There is some uncertainty as to the degree of later adherence to that model; apart from those of Constantine and Constantius II, the funerals of only three later emperors are described in any detail before the tenth century: Justinian in Corippus, Michael III in Theophanes Continuatus, and Constantine VII in the Book of Ceremonies and Theophanes Continuatus. Cf. Karlin-Hayter 1991, esp. 114–15. ⁹³ Instead, they appear as part of the festivities on the occasion of Honorius’ adventus in Rome, displayed by the scholae as a form of entertainment for populace and emperor in the Circus Maximus: Claud. de VI Cons. Hon. 612–32.

  

73

corroborating evidence from later centuries.⁹⁴ Even though it is strictly speaking outside the chronological timeframe of this chapter, a brief look at the sixth century is instructive: by then, the ceremony had evolved in a recognizably more Christian fashion. If we believe the description of Justinian’s funeral given by Corippus, the formal ceremony conveying the body of the defunct ruler to Holy Apostles resembled a church procession more than a traditional pompa. Participants carried candles in their hands and copious amounts of incense were burned both inside and outside the palace. The funeral cortège was accompanied by two choirs—one of deacons, one of young girls (‘virgins’)—singing religious hymns and psalms.⁹⁵ A similarly religious atmosphere is detectable in the scant notices about the funerals of Maurice in Theophylact Simocatta.⁹⁶ Notably, though, neither mentions the presence or participation of the patriarch.

5. Conclusion It has long been held that the early emperors, in imitation of Augustus himself, were forced to adhere publicly to a ‘persistently republican ideology of the Principate’, ‘the emperor being expected to act as a holder of republican office’.⁹⁷ Those emperors who closely adhered to this model of the civilis princeps were among the ‘good’ ones, while others, who innovated or changed aspects of this representation to reflect more accurately the nature of their rule, were ‘bad’. But as Fergus Millar already pointed out decades ago, we cannot accurately gauge the relationship between emperor and his subjects ‘without taking into account the fact that he was almost always escorted by armed soldiers’.⁹⁸ Naturally, this was above all the case with official state ceremonies of the sort we have described. Rituals and ceremonies in which guards were involved were extensions of contemporary imperial ideologemes, norms, and expectations on the part of the Roman emperor’s subjects. As such, they evolved in parallel with these expectations. Whereas military ceremonial was kept to a bare minimum during the early Principate and, presumably, under the Antonines in favour of the image of the civilis princeps, ‘civilian’ here in every sense of the word, the later Roman Empire functioned along different lines. As the case studies of this chapter have shown, however, this process by no means began only at the end of the second century. On the contrary, beginning with the very first moment of transition—Augustus’

⁹⁴ Const. Porph. de cer. 1.60 has the bier carried by τῶν βασιλικῶν, which may mean guards; Theoph. Cont. 6.53, recounting the burial of Constantine VII, explicitly mentions τὸ βασιλικὸν δορυφορούμενον. Cf. Rapp 2012 for these later Byzantine funeral rites. ⁹⁵ Coripp. Iust. 3.39–43. ⁹⁶ Theoph. Sim. 1.2.4–5. The involvement of a choir of deacons in funeral processions was usual by this point and had, in fact, been regulated by Justinian himself (Nov. 59.4). ⁹⁷ Nelson 1976: 100. ⁹⁸ Millar 1977: 61–6, quote at 61.

74

   

death and the accession of Tiberius—we can observe both a steady evolution of the role of guards in imperial ceremonial, becoming more and more prominent, and even a conscious experimentation with military/guard elements in imperial rituals. Importantly, however, the guards and the concomitant military ceremonial were present from the very beginning and even the optimus princeps Trajan, in what Pliny described as a perfect example of imperial adventus, in fact (if, perhaps, not in appearance) participated in a military procession. If (and it is rather a large ‘if ’), as is sometimes claimed, transformations to the fabric and culture of empire under the Severans prefigured the coming crises and metamorphoses of the third and fourth centuries, this is especially true of imperial bodyguard units. They were background actors in the same splendid theatre, to paraphrase Gibbon, on whose stage the emperor played both the main part and directed the production. Thus, as various emperors attempted to interpret their roles in new and different ways over the centuries, their guard was bound to adapt with them. Imperial ideology and popular expectations changed. The continuous and generalized ideology of victory that Michael McCormick has shown to have existed not only made a more pronounced military ceremonial possible, it necessitated it. Now that the military aspect of imperial rule no longer had to be— indeed, after the crucible of the third century, no longer could be—hidden, to be brought out only for the occasional blunt reminder, both the function and the attire and appearance of the guard underwent dramatic change. In fact, so did the very meaning of ‘guards’. Seen from a purely military point of view, they may have ‘degenerated’ in the sense that the private army of the Praetorians ceased to exist and was replaced by a smaller unit, the scholae. As imperial rule became more and more centred, even immobilized in Constantinople from the end of the fourth century onwards, their primary function gradually evolved and they, in turn, became a purely ceremonial outfit by the fifth century, commissions to which were expected to be bought. Their main function, as described by Agathias, the author of a sixth-century history in the tradition of the Wars of Procopius, was ‘enhancing the pomp of a royal progress’ (ἐν ταῖς προόδοις μεγαλαυχίας ἔξευρημένοι).⁹⁹ In Constantinople, the emperor did not need protection—which could be handled by the new and smaller corps of the excubitores—as much as he needed charismatic authority. Tying into a general and highly conscious tendency to conceptualize the later imperial court as a sacred space, the increasingly dramatic and theatrical appearances and choreographies of imperial ceremonial were designed to endow him with just that. The guard was an important part of that design. As the conventions of interaction with the imperial person had changed in the centuries since Augustus—indeed, had actively been changed by the innovations traditionally attributed to Diocletian and Constantine—so, too, had the guard.

⁹⁹ Agath. 5.15.2.

3 Cities, Palaces, and the Tetrarchic Imperial Courts Verena Jaeschke and Caillan Davenport

1. Introduction This chapter deals with a crucial period in the transformation of the Roman imperial court between the Principate and Late Antiquity: the age of the Tetrarchs, which lasted from Diocletian’s accession in 284 until Constantine’s defeat of Licinius and assumption of sole power in 324.¹ In this period, the multiplication of emperors, both Augusti and Caesars, required the establishment of different imperial courts, which moved between cities such as Milan, Trier, Nicomedia, and Antioch as dictated by the emperor’s campaigns. This represented a move away from what Jeroen Duindam has called a ‘court-in-capital’ model of governance, which in this case was centred on Rome, to a mobile imperial court.² Although these cities have often been described as Tetrarchic ‘capitals’, the term is not without its problems.³ The individual cities were not permanent seats of government, but one of many residential options available to the emperor and his comitatus during and between campaigns. A better term is ‘imperial cities’, because they were equipped with a range of buildings such as palaces, circuses, and baths, which were modelled on the amenities of Rome itself.⁴ It was this conspicuously imperial infrastructure that differentiated them from, and elevated them above, other cities. In this chapter, these cities are described as functioning as principal residences for members of the Tetrarchic imperial colleges. Although the emperors and their courts also stayed in other locations, the principal residences were either the seat of a member of the imperial college

¹ The use of the terms ‘Tetrarchy’ and ‘Tetrarchic’ for the period between 284 and 324 is, as Leadbetter 2009: 5 aptly puts it, ‘a scholarly shorthand for the period of collegiate rule’. For the imperial colleges, not all of which were ‘Tetrarchic’ in the sense that they had four members, see Barnes 1982: 3–8. ² See Duindam 2016: 159–68 for these different types of courts and the concept of a ‘court-incapital’. ³ Grig and Kelly 2012: 7. ⁴ For the imperial status of these cities, see Pan. Lat. 11(3).12.1, which uses the term sedes imperii (‘seat of empire’) to describe them, and Mayer 2002: 1–3. Verena Jaeschke and Caillan Davenport, Cities, Palaces, and the Tetrarchic Imperial Courts In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0004

76

   

for several consecutive years or a city to which they returned repeatedly over a long period.⁵ In this volume, the court is defined as a social phenomenon, the community of people surrounding the emperor, rather than a specific location. But spatial aspects such as palace layout, architecture, and decoration remain important to our understanding of the court because they affected how courtiers and visitors interacted with the emperor. The significance of spatial dynamics to the field of aulic studies can be traced back to Norbert Elias’ examination of the layout of Versailles and the way in which it shaped Louis XIV’s court society.⁶ In recent years, scholars such as Jeroen Duindam have demonstrated the importance of analysing how the court functioned in both a palatial and urban context.⁷ Although the building programmes which transformed the Tetrarchic principal residences did not all take place simultaneously, the cities share many similarities in their planning and layout, particularly in the provision of accommodation for the emperors and their courts, as well as amenities and avenues in which they could interact with the local elites and the population at large.⁸ The development of these regional centres into truly imperial cities also had the effect of transforming the court community. The mobile imperial comitatus consisted of the emperor, his household, and his administrative officials, but when it took up residence in one of these cities, the court expanded to include provincial elites attracted by the imperial presence.⁹ The chapter will contrast the Tetrarchs and their principal residences with the example of Maxentius, who ruled from Rome between 306 and 312, but we never officially part of the imperial college. Maxentius is an interesting case, because he controlled the sacra urbs, which was the model for all other imperial cities in its infrastructure and amenities, as well as the seat of the senate. Moreover, Maxentius and his court did not move between different cities, even within Italy itself, although his villa outside Rome on the Via Appia certainly served as an important residence. We will explore how Maxentius’ building programme in Rome, and the way it shaped the relationship between his court and the urban environment, shared some similarities with his Tetrarchic peers, but also had significant differences. The chapter will conclude with a brief discussion of Constantine’s foundation of Constantinople in 324 after his defeat of Licinius. ⁵ The term ‘principal residence’ is used by Humphrey 1986: 633–8. For the difference between these and other cities, the case of Ravenna is instructive. Diocletian assumed his eighth consulship in 303 in Ravenna (Lact. DMP 17.3), but this was never a principal residence as it was not augmented with significant new public buildings, nor did the Tetrarchic emperors ever spend extended periods there. ⁶ Elias 2006: 45–72, 79–82, 88–93. ⁷ Duindam 2016: 159–68; for a Roman example, see the fundamental analysis of Domitian’s modifications to the Palatine by Zanker 2002. ⁸ Previous discussions of the building programme of these cities include Ćurčić 1993; Brenk 1996; Duval 1997; Mayer 2002: 28–68; von Hesberg 2006; Wulf-Rheidt 2007. Mayer 2014 offers an accessible scholarly overview of Tetrarchic architecture. ⁹ Kulikowski 2015.

, ,     

77

Constantinople combined aspects of the Tetrarchic building programme with new elements designed to highlight Constantine’s prominent status as a sole ruler. It was likely that Constantine intended this ‘New Rome’ to become a key residence on the court-in-capital model. But Constantinople would not become a permanent imperial seat until the Theodosian age, when the eastern Roman Empire made the firm transition to this court-in-capital mode of governance.

2. Tetrarchic Principal Residences The principal residences used by emperors in the period 284–324 include Antioch (Antakya), Nicomedia (İzmit), Sirmium (Sremska Mitrovica), Thessalonica (Thessaloniki), and Serdica (Sofia) in the eastern provinces and Mediolanum (Milan), Augusta Treverorum (Trier), and Aquileia (ancient name retained) in the western half of the empire.¹⁰ Principal residences usually had pre-existing administrative functions (for example, acting as a provincial capital) and served as social, cultural, and economic hubs for the wider region.¹¹ They occupied strategic locations and were connected to important roads and water routes which provided access the Germanic, Balkan, or Persian frontiers. In the Tetrarchic period, these cities were equipped with new palaces suitable for housing a reigning Augustus or Caesar. The palaces were often connected to pre-existing or newly built circuses, where the emperors would appear before the people while hosting games.¹² The principal residences were equipped with new imperial bathhouses, adorned with arcades and colonnades which provided the setting for ceremonial occasions, and augmented by infrastructural works such as new or extended walls and warehouses. The Tetrarchic building programme embellished the urban framework of these cities so that they could serve as suitable stages for imperial display, hosting key moments of state ceremonial, such as adventus ceremonies, triumphs, and consular inaugurations, which were usually combined with the performance of panegyrics and the staging of games.¹³ The expectation that emperors should be seen and interact with the people while in residence is vividly demonstrated by the fact that when Diocletian did not appear in public in Nicomedia during the winter of 304/5, a rumour spread that he had actually

¹⁰ For an overview of the principal residences, see Millar 1977: 40–54; Barnes 1982: 49–82; Jaeschke 2020. ¹¹ This paragraph draws on the conclusions of Jaeschke 2020. ¹² Ćurčić 1993: 67 on the palace-circus connection. For pre-existing circuses, see Wulf-Rheidt 2007: 63 on Thessalonica and Antioch. ¹³ Mayer 2014: 107. Adventus and panegyric: MacCormack 1981: 22–33; Rees 2002: 6–19. Triumphs: McCormick 1986: 36–7. Consular inaugurations: Pan. Lat. 10(2).6–4; Davenport 2016: 393–4 (on Trier in the third century); Cameron 2013: 199–201 (the development of the tradition of holding consular games at court from c. 300).

78

   

died. This fear was only allayed when he finally made a public appearance on 1 March 305.¹⁴ At the same time, one of the defining features of the Tetrarchic period was that the emperors and their court did not base themselves in only one city, but moved between multiple principal residences, much as Hellenistic rulers had done before them and Medieval European monarchs would do in later centuries.¹⁵ This meant that that the ceremony of adventus, the arrival of the emperor and his court in a city for a new period of residence, came to represent a defining moment of Tetrarchic political culture, as Sabine MacCormack has brilliantly explored.¹⁶ We have long known about scenes of imperial arrival on the Arch of Galerius in Thessaloniki and the walls of the imperial cult chamber in Luxor.¹⁷ To these we can now add the stunning series of recently discovered reliefs from a Tetrarchic cult complex in Nicomedia, which depict an adventus and the meeting of Diocletian and Maximian.¹⁸ The entire relief programme exhibits a significant degree of Nicomedian civic pride, demonstrating the city’s celebration of its history and its new status under the Tetrarchs.¹⁹ It may be no coincidence that a new coin type, showing all four Tetrarchs sacrificing outside the walls of a city, was first produced in Nicomedia and the nearby new imperial mints in Cyzicus and Heraclea c. 294.²⁰ In 313/15, the mint of Trier struck a double solidus which depicted Constantine (or a statue of him) inside Trier, surrounded by walls and the flowing river Moselle, while subjugated barbarian captives sit forlornly outside.²¹ The presence of the emperor was thus cherished and celebrated, as it signalled his ability to protect his subjects and confer privileges on individual cities and on the region at large. As a Gallic panegyrist who praised Constantine at Trier in 310 proclaimed: ‘For in whatever places your divinity distinguishes most frequently with his visits, everything is increased—men, walls, and favours.’²² It is important at the outset to distinguish between the cities which served as Tetrarchic principal residences and the imperial retirement villas, which were intended to accommodate an emperor and his household after he had stepped down from the imperial college. The best-known examples of these retirement residences are Diocletian’s palace at Split and Galerius’ villa in Gamzigrad, to ¹⁴ Lact. DMP 17.8. ¹⁵ Duindam 2016: 161–2. ¹⁶ MacCormack 1972; MacCormack 1981: 17–33, esp. 21. ¹⁷ For Luxor, see Kalavrezou-Maxeiner 1975: 239–43; Jones and McFadden 2015; McFadden 2015a: 108–10, 117–18. ¹⁸ Şare Ağtürk 2018. ¹⁹ Şare Ağtürk 2020a; 2020b. ²⁰ For this coin type, see Elkins 2013: 286–7, 292–3; Woods 2017: 160, 165–6. ²¹ RIC VII Treveri 1. ²² Pan. Lat. 6(7).22.6: quaecumque enim loca frequentissime tuum numen inlustrat, in his omnia et hominibus et moenibus et muneribus augentur (trans. Nixon in Nixon and Rodgers 1994). Note also the Jewish parable in which a king enters a province to the benefit of the population: ‘He built a city wall for them, he brought in the water supply for them, and he fought their battles’ (Mekhilta de-Rabbi Ishmael, Tractate Bahodesh 5, trans. Lauterbach 2004). :

, ,     

79

which we should also add the site of Šarkamen, which was probably constructed by Maximinus Daza.²³ The retirement villas were equipped with fortified walls and had no circuses or other entertainment venues, which meant that they were designed to give the former emperors peace and seclusion, rather than to display them before large audiences.²⁴ The principal residences were also distinct from extra-urban villas in the provinces, which served as accommodation for an emperor during the course of his travels or campaigns. A possible example of this type of residence is the large villa near Corduba, which may have been used by Maximian during the Spanish expedition of 296.²⁵ These villas were luxurious residences, but they were only intended for short-term use. Lactantius provides a vivid description of the process of transforming the provincial centres into imperial cities in his account of Diocletian’s building programme in Nicomedia: In addition, Diocletian had a limitless passion for building, which led to an equally limitless scouring of the provinces to raise workers, craftsmen, waggons, and whatever was necessary for building operations. Here he built basilicas, there a circus, a mint, an arms-factory, here he built a house for his wife, there one for his daughter. Suddenly a great part of the city [sc. Nicomedia] was destroyed, and all the inhabitants started to migrate with their wives and children, as if the city had been captured by the enemy. And when these buildings had been completed—and the provinces ruined in the process—he would say: ‘They have not been built rightly; they must be done in another way.’ They then had to be pulled down and altered—perhaps only to come down a second time. This was the way he was always raving in his eagerness to make Nicomedia the equal of the city of Rome.²⁶

The tenor of this passage is, of course, a hostile one, which formed part of the Christian Lactantius’ denigration of the Tetrarchs as persecuting emperors. For one thing, these constructions would have been a boon for the local

²³ See Mayer 2014: 117–23 for a recent overview of Split and Gamzigrad. For Šarkamen, see Popović 2007a. For Daza, rather than Daia, as the correct form of the emperor’s name, see Mackay 1999: 207–9. ²⁴ See Chen 2016: 220–4, who suggests that the inspiration for the design of these residences may have come from the Sasanian Empire, where palaces emphasized the Shah’s seclusion. For the idea that reigning Roman emperors were supposed to be open and accessible, see Chapter 12. ²⁵ For the Spanish villa, see Haley 1994. ²⁶ Lact. DMP 44.7.8–10: huc accedebat infinita quaedam cupiditas aedificandi, non minor provinciarum exactio in exhibendis operariis et artificibus et plaustris, omnia quaecumque sint fabricandis operibus necessaria. hic basilicae, hic circus, hic moneta, hic armorum fabrica, hic uxori domus, hic filiae. repente magna pars civitatis exciditur. Migrabant omnes cum coniugibus ac liberis quasi urbe ab hostibus capta. et cum perfecta haec fuerant cum interitu provinciarum, non recte facta sunt, aiebat, alio modo fiant. rursus dirui ac mutari necesse erat iterum fortasse casura. ita semper dementabat Nicomediam studens urbi Romae coaequare (trans. Creed 1984).

80

   

building industry.²⁷ But it does concur with the archaeological evidence (reviewed below) in terms of the buildings that were regarded as necessary for an imperial residence, most notably a palace and circus.²⁸ The palace-circus combination was inspired by the model of the Palatine and Circus Maximus at Rome, where the connection between the two buildings allowed emperors to move directly from the palace to the pulvinar in the circus.²⁹ Lactantius’ remark that Diocletian wanted Nicomedia, his new residence, to be the equal of Rome, was probably therefore an accurate representation of the emperor’s ambitions. It must be emphasized that the transformation of provincial centres into imperial cities that rivalled the sacra urbs was a long process. Not all the cities named at the beginning of this section as Tetrarchic principal residences were used by emperors consistently throughout the period 284–324, nor did construction of new palaces and other public amenities occur simultaneously. In the early years of the Tetrarchy, many of the cities used as principal residences likely had no new structures. When the panegyrist of 289 addressed Maximian and his court at Trier, it would not have been in the magnificent Basilica Constantiniana that still stands today (albeit heavily restored), but perhaps in a smaller hall previously used by imperial procurators.³⁰ Diocletian was based at Sirmium for almost a year between September 293 and August 294, but once again, there was no palace in the city at the time. In fact, it was probably only during this stay that construction began on a new residence at Sirmium.³¹ The process of transformation sometimes required the resumption of preexisting structures. Lactantius’ account of construction in Nicomedia, quoted above, states that a great part of the city was destroyed to accommodate Diocletian’s new works. This could be dismissed as invective, but the archaeological evidence suggests that areas of several cities did have to be cleared out to accommodate the new Tetrarchic embellishments.³² In Thessalonica, although the region that became the palace quarter was not densely inhabited, there were certainly pre-existing buildings on the site that were torn down.³³ The dedicatory inscription for the Baths of Diocletian in Rome provides helpful comparative evidence here, because it refers to the emperors buying up buildings to acquire land for the new bathing complex.³⁴ This construction process often took several years. In Nicomedia, Diocletian was not able to dedicate the circus until late 304, ²⁷ See Wightman 1970: 101, 108, 203 on Trier. ²⁸ Pflug and Wulf-Rheidt 2022: 235–7. ²⁹ Ćurčić 1993: 67. Cf. Wulf-Rheidt 2013: 304; Pflug and Wulf-Rheidt 2022: 234. ³⁰ Rees 2002: 35. Trier had been a residence of the ‘Gallic’ Emperor Tetricus, who minted his coinage there (Knickrehm 2014), but there is little evidence for major building projects. On the hall, see Wightman 1970: 86–7. ³¹ Popović 2016: 385 connects the beginning of construction with Diocletian’s visit to Sirmium in 293/4, based on finds of porphyry statues from a polygonal structure connected with the palace. For Diocletian’s itinerary in these years, see Barnes 1982: 53. On the transformation of Sirmium into an administrative centre, see Popović 2019. ³² Mayer 2002: 30; Mayer 2014: 109. ³³ Hadijitryphonos 2011: 208–11. ³⁴ CIL VI 1130 = 31242.

, ,     

81

shortly before his retirement as Augustus.³⁵ This pattern is repeated in other cities, such as Thessalonica and Sirmium, where construction took place over a decade or more, sometimes spanning the reigns of multiple emperors.³⁶ We can therefore only establish a rough chronology of when work began in each of the principal residences. It is plausible, if unprovable, that Nicomedia was the first city to begin the transformation process, since it was the most important long-term principal residence of Diocletian.³⁷ Our main source for Nicomedia is the passage of Lactantius quoted above, since there has been, until recently, very little archaeological excavation which would enable us to reconstruct a firm chronology.³⁸ Antioch, the capital of Syria Coele, was also remodelled during the reign of Diocletian, who used the city as his base for his Persian campaign in 299–300. Once again, archaeological evidence is sorely lacking, but the testimony of John Malalas and Libanius shows that the construction of the palace in Antioch’s new city, located on an island in the river Orontes, was completed by 303.³⁹ The evidence from the western provinces suggests that Maximian sought to emulate his senior colleague by embellishing the city of Milan. Archaeological excavation dates the palace and circus complex to the general period of the late third and early fourth century; further precision is currently impossible.⁴⁰ The chronology of building in Aquileia, the other north Italian imperial residence, is likewise difficult to determine. The panegyric of 307 refers to a palace existing in the 290s, but the nature and extent of this structure are unknown.⁴¹ Aquileia did gain a circus, indicating its status as a principal residence, but this was probably a little later than Milan, based on the archaeological material.⁴² The extent of construction in Trier under Maximian and Constantius I during the 280s and 290s was probably very limited. The greatest period of renewal can be dated to the period from 306 onwards, when Trier became the principal residence of Constantine. It was then that the construction of the circus and the palace complex, including the Basilica Constantiniana, took place. They were completed by 310, since they are mentioned in a panegyric delivered in that year.⁴³

³⁵ Lact. DMP 17.4; Barnes 1982: 56. ³⁶ Thessalonica: Hadijitryphonos 2011: 211. Cf. Spieser 2015: 25–6, arguing for a more compressed chronology. Sirmium: Jeremić 2009: 476. ³⁷ Humphrey 1986: 634–6; Wulf-Rheidt 2007: 62. ³⁸ Mayer 2002: 29–31. See now Şare Ağtürk 2018; 2020a; 2020b for exciting new evidence from the city. ³⁹ Joh. Mal. 12.38; Lib. Or. 11.161, 11.205–7. See Ćurčić 1993: 68; Barnes 1982: 49 n. 18, 55. ⁴⁰ Aur. Vict. Caes. 39.45; Humphrey 1986: 634; Mayer 2002: 31–4; Caporusso et al. 2014: 158–75. ⁴¹ Pan. Lat. 7(6).6.2, with Nixon and Rodgers 1994: 198 n. 19. Mian 2012/13 places the palace in Aquileia east of the circus, based on archaeological finds. ⁴² Humphrey 1986: 621–5, 634. ⁴³ Pan. Lat. 6(7).22.4–6. See further Wightman 1970: 58–60, 102–9; von Hesberg 2006: 150; Lenski 2016: 188. The ‘imperial baths’ at Trier are slightly later (Wightman 1970: 99–102).

82

   

The best-attested site in the Balkan region is Thessalonica, which served as a principal residence of Galerius. Although significant parts of the palace complex have been excavated, scholars are divided as to whether construction began in 299, at the start of Galerius’ first period of residence, or several years later, following his elevation to the rank of Augustus in 305.⁴⁴ Nearby cities saw the most intensive construction under later members of the imperial college. In Sirmium, there is some evidence for construction of the palace in the late third century, beginning at the time of Diocletian’s residence in the city in 293–4.⁴⁵ However, the most significant expansion did not occur until Licinius became Augustus in 308 and made Sirmium his primary residence, at which point he added both the circus and the bath complex.⁴⁶ Nearby Serdica presents the most difficult case out of all the principal residences. No circus has yet been uncovered, but that does not mean that it did not have one, as archaeological excavation has been hampered by the fact that the Roman city lies under modern-day Sofia.⁴⁷ That said, its heyday as a principal residence did not occur until the reign of Constantine, who spent considerable time there between 316 and 322.⁴⁸ Therefore, the principal residences were not designed and embellished simultaneously, and certainly not in the very restricted period of the First Tetrarchy (293–305), but over several decades and under different iterations of the imperial college. However, despite the chronological differences, the evidence shows that there was a clear conception of the types of buildings and structures that a principal residence required in order to be considered a properly imperial city. We have already emphasized the importance of the palace and circus as key features, which distinguished the principal residences from other cities and the retirement villas. All the cities—with the exception of Thessalonica—also had new imperial bath complexes built in them.⁴⁹ The Herculean Baths at Milan, named after Maximian, covered a massive 14,500 square metres, while Libanius wrote that the baths at Nicomedia were like a city in themselves.⁵⁰ Magnificent new baths were also built in Rome, the first significant bathing complex in the sacra urbs since the Severan period (only the small Baths of Decius on the Aventine had

⁴⁴ See, for example, Leadbetter 2009: 100, 233–6; Spieser 2015: 24–5 (earlier chronology) and Hadijitryphonos 2011: 206–7 (later chronology). ⁴⁵ Popović 2007b: 22–3; Jeremić 2009: 476; Popović 2011; Popović 2016. ⁴⁶ Circus: Humphrey 1986: 612, 634. Baths: CIL III 10107. ⁴⁷ Humphrey 1986: 579–81; Heucke 1994: 314–16. De Cena 2014: 17 notes that there is a possibility it has yet to be discovered. ⁴⁸ Barnes 1982: 73–5. For Constantine’s affinity for Serdica, see the Anonymous Continuator of Cassius Dio (FHG IV, fr. 15.1). The palace appears in Athan. Hist. Ar. 15.5. For discussion of the possible location of the palace in Serdica, see de Cena 2014: 13–17. ⁴⁹ Mayer 2014: 110 notes the difference between pre-existing bath complexes and the new imperial baths. ⁵⁰ Milan: Aus. Ordo nob. urb. 5.41; Caporusso et al. 2014: 176–86. Nicomedia: Lib. Or. 61.17. Note also the evidence from other cities. Trier: Witschel 2004/5: 227 (emphasizing the baths’ proximity to the Palace). Antioch: John. Mal. 12.388.

, ,     

83

been erected in the intervening years).⁵¹ The construction of the ‘fortunate Diocletianic baths’ (thermae felices Diocletianae), which began in 299 after Maximian’s return to Italy from his African campaign, may have reflected his desire to spend more time in Rome. The project was a massive effort which revitalized the Roman building industry and, when completed in 305/6, constituted the largest bathing complex the city had yet seen.⁵² Indeed, it is worth posing the question as to whether Rome itself—the model for the other Tetrarchic imperial cities—constituted a ‘principal residence’. It is traditional to exclude Rome from consideration and emphasize provincial cities instead. But the fact is that Maximian is not attested in Milan in the last decade of his reign, between 21 December 295 and 1 May 305.⁵³ In contrast, Maximian did spend time in Rome during these years, as did several members of his family.⁵⁴ This suggests that the sacra urbs was not as peripheral as it has often been thought to be. In several cases, principal residences had their city walls augmented or extended and new large warehouses constructed.⁵⁵ Many acquired an imperial mint, which usually operated during those periods in which the emperors were in residence, though mints are also attested in other cities such as Arles, Siscia, Cyzicus, and Alexandria.⁵⁶ Interestingly, neither mausolea nor temple complexes dedicated to the tutelary gods, Jupiter and Hercules, were a consistent part of the Tetrarchic architectural programme in the principal residences. There are some exceptions to this, such as the mausoleum of Emperor Maximian, which was probably located in Milan.⁵⁷ But in other cases, the imperial mausolea were located in the retirement palaces instead.⁵⁸ Table 3.1 presents a schematic overview of the different Tetrarchic principal residences and their major public buildings. The evidence shows that there was no fixed configuration of palace, circus, and mausoleum, which marked out a city as a Tetrarchic principal residence.⁵⁹ In fact, imperial baths, which emulated the magnificent complexes in Rome, were a more common feature of these cities than mausolea. It has been suggested that the construction programme reflected the key Tetrarchic qualities of concordia (‘harmony’) and

⁵¹ Curran 2000: 19. ⁵² CIL VI 1130 = 31242; Coarelli 1999; Curran 2000: 44–5. Machado 2019: 67–8 emphasizes the personal agency of the emperors in the baths’ construction, in contrast with later fourth-century projects in Rome. ⁵³ Barnes 1982: 59–60. ⁵⁴ Maximian’s connections with Rome are a consistent theme in recent scholarship, for example, Potter: 2010: 29–30; Davenport 2017: 34–5; Hillner 2017: 60–2. ⁵⁵ The full discussion of these structures can be found in Jaeschke 2017a. ⁵⁶ Hendy 1972; Kuhoff 2009: 100. ⁵⁷ Johnson 2009: 70–4; Caporusso et al. 2014: 282–7. It has been suggested that Constantius I was buried at Trier after his death in the north of England (Johnson 1992: 145–6), though no mausoleum has yet been discovered. ⁵⁸ Johnson 2009: 59–70, 74–82. ⁵⁹ For the absence of mausolea, see Johnson 2009: 104–6.

Table 3.1 The buildings in the Tetrarchic principal residences. Table created by Verena Jaeschke. BUILDINGS IN TETRARCHIC PRINCIPAL RESIDENCES RESIDENCE

PALACE

CIRCUS

IMPERIAL BATHS

EXTENSIONCITY WALLS

WAREHOUSES

MAUSOLEUM

TEMPLE

Nicomedia Sirmium Antioch Thessalonica Serdica Trier Milan Aquileia

YES YES (?) YES YES YES (?) YES YES YES (?)

YES YES YES YES NO YES YES YES

YES YES YES NO YES YES YES YES

YES YES YES (?) YES YES NO YES YES

NO YES YES NO NO YES YES YES

NO NO NO YES (?) NO YES (?) YES NO

NO NO NO YES (?) NO NO NO NO

, ,     

85

similitudo (‘uniformity’) in an urban framework.⁶⁰ But one needs to bear in mind that building did not take place simultaneously in all these cities during the First Tetrarchy (293–305), when such qualities were most emphasized. Instead, the principal residences were transformed in different stages across several decades. The building work undertaken reflects the conception that there should be several cities located throughout the empire that had the necessary palaces, entertainment facilities, and infrastructure to accommodate the emperor and court in an urban environment that was considered, like Rome itself, suitably imperial.

3. Courts and Communities On 21 April 289, a Gallic panegyrist delivered a speech in honour of Maximian at court in Trier as part of the celebrations for Rome’s birthday. In the oration, he described the emperor and his retinue which he saw before him: Your triumphal robes and consular fasces, your curule thrones, this glittering crowd of courtiers, that light which surrounds your divine head with a shining orb, these are the trappings of your merits, very fine indeed, and most majestic.⁶¹

The orator’s words captured a specific ritual moment, with the emperor and court arrayed in a vivid still-life tableau.⁶² But the Tetrarchic comitatus was, above all, a mobile community, moving from city to city and province to province. The emperor, administrative officials, household staff, officers, and soldiers travelled between different principal residences and frontier posts, their movements primarily dictated by the needs of warfare. This is vividly brought to life in the description of the progress of Diocletian and Maximian in Milan by the orator of 291, who depicts the city’s inhabitants running to view the emperors as they rode through the city.⁶³ The panegyrics understandably foreground the emperors, but other sources reveal the presence of leading court officials during the arrival ceremony. A Jewish parable from the third century has the emperor riding in his carriage with his praetorian prefect, resulting in confusion among provincials as to who was who.⁶⁴

⁶⁰ Kuhoff 2009: esp. 113–16. On these imperial qualities, see Rees 2002: 55–66. ⁶¹ Pan. Lat. 10(2).3.2: trabeae vestrae triumphales et fasces consulares et sellae curules et haec obsequiorum stipatio et fulgor, et illa lux divinum verticem claro orbe complectens, vestrorum sunt ornamenta meritorum, pulcherrima quidem et augustissima (trans. Nixon in Nixon and Rodgers 1994). ⁶² For a similar depiction of the court in a later panegyric, see Pan. Lat. 5(8).2.1. ⁶³ Pan. Lat. 11(3).11.3–5. ⁶⁴ Midrash Rabbah, Genesis 8:10 (trans. Freedman and Simon 1939, but following the modification to the translation by Appelbaum 2007: 254). Note also Mekhilta de-Rabbi Ishmael, Tractate Shirata 3 on the imperial retinue: ‘A king of flesh and blood enters a province surrounded by a circle of guards; his heroes stand to the right of him and to the left of him; his soldiers are before him and behind him’

86

   

The courts of the Augusti were larger than those of the Caesars, since the senior emperors possessed praetorian prefects and all three magistri, the heads of the departments of letters (magister epistularum), responses (magister libellorum), and ‘the memory’, which may have had responsibility for edicts (magister memoriae).⁶⁵ The Caesars did have a magister epistularum to manage their correspondence, but not a praetorian prefect nor the full palatine establishment.⁶⁶ Sometimes two emperors and their courts travelled together, as when Diocletian accompanied Galerius from Sirmium to Byzantium in 293, after which the Caesar continued on to Syria separately.⁶⁷ The business of government did not stop during these travels, since the Augusti and their chief administrators continued to answer petitions and issue rescripts while on the road.⁶⁸ The officials often sprang from the specific regions in which emperors were based, as shown by the number of Gallic elites who served in the administrations of Maximian and Constantius I, such as the magister memoriae Eumenius.⁶⁹ Eumenius travelled with the Augustus Maximian between his residences in Milan and Aquileia, and probably also journeyed with him to Trier.⁷⁰ Given the backgrounds of the Tetrarchs themselves, it is not surprising that military officers constituted a significant contingent at court. They served as close confidants and advisors, and were sometimes related by marriage or birth to the emperors themselves.⁷¹ The military nature of the court community is brought to life by the frescos in the imperial cult chamber at Luxor, which depict adventus and adoratio scenes on the east and south walls, respectively. A surviving portion of the latter scene shows men gathered around Diocletian’s throne. The central figure holds a red vexillum, the banner of Roman military units, and many behind him are readily identifiable as senior military officers by their cloaks (paladumenta) (Figure 3.1).⁷² Susanna McFadden has argued that the individuality

(trans. Lauterbach 2004). The Hebrew word gibôr, here translated by Lauterbach as ‘heroes’, likely refers to elite troops and officers. We are grateful to Dr Louise Pryke and Prof. Ian Young of the University of Sydney for their advice on this point. ⁶⁵ Senior officials tasked with managing imperial letters and edicts tended to travel with the emperor (Connolly 2010: 48–9). The mobility of financial officials is less clear, as they continue to be attested in Rome, for example (Davenport 2017: 34; Davenport and Kelly 2022: 137). Later in the fourth century, the comes sacrarum largitionum certainly travelled with the emperor as part of the consistorium (Jones 1964: 333). ⁶⁶ For discussion, see Corcoran 2000: 132, 268–9, 273–4; Davenport 2021: 11–12. ⁶⁷ Connolly 2010: x, 48. ⁶⁸ See Corcoran 2000: 85–94 on imperial administrators and Connolly 2010: 47–62 for a vivid description of the court’s movements and the role of the imperial scrinium. ⁶⁹ Kulikowski 2015: 141–5; Davenport 2019: 576–9, 583–4. On Eumenius, see Rodgers 1989; Davenport 2021. ⁷⁰ Rodgers 1989: 258; Davenport 2021: 21. ⁷¹ Lact. DMP 11.5–6, 12.2, 19.1; Pan. Lat. 10(2).11.4 on marriages. Particular examples include officers who were future members of the imperial college, such as Severus and Maximinus Daza (Lact. DMP 18.12–13), Licinius (Lact. DMP 20.2–4; Aur. Vict. Caes. 40.8), and Constantine (Eus. V. Const. 1.19.1–2; Lact. DMP 24.4–7; Anon. Val. Origo 2.2–3; Protagoras FGrH 219 T1). ⁷² Kalavrezou-Maxeiner 1975: 234; McFadden 2015a: 119, 122.

, ,     

87

Figure 3.1 Fresco on the south wall of the Imperial Cult Chamber at Luxor. © Manfred Bail/imageBROKER/agefotostock.

present in the faces in this scene suggests that they are recognizable portraits of members of Diocletian’s comitatus.⁷³ We should not exclude the presence of some senators at the Tetrarchic courts, but it varied depending on the movements of individual emperors. For example, it is unlikely that many senators attended the court of Maximian in Trier during the early part of his reign, unless they were on an embassy from Rome.⁷⁴ Such a mission is attested at Milan during Diocletian and Maximian’s meeting there in 290/1.⁷⁵ This dynamic would have changed in later years, when Maximian was more often in Rome.⁷⁶ A prominent senator like T. Flavius Postumius Titianus, who had the high honour of serving as a judge in Maximian’s place, was wellconnected enough at court to secure the promotion of his protégé, T. Aelius Poemenius, to the office of procurator aquarum.⁷⁷ In the eastern regions, senators continued to govern provinces and hold posts with significant administrative responsibilities.⁷⁸ L. Aelius Helvius Dionysius, for example, was governor of

⁷³ McFadden 2015a: 126. ⁷⁴ This is certainly implied by the panegyrics delivered there, in which Rome’s senate appears rather distant, such as Pan. Lat. 10(2).13.4; Davenport 2016: 394. The visit of a senatorial embassy to Milan in the winter of 290/1 appears to be a significant expedition worthy of remark: Pan Lat. 11(3).12.2. ⁷⁵ Pan. Lat. 11(3).12.1. ⁷⁶ See Potter 2010: 29–30 on the connections between Maximian and the Italian aristocracy. ⁷⁷ CIL VI 1418 = LSA 1325 (C. Machado, dating the inscription between 295–6 and 301). ⁷⁸ Corcoran 2000: 223; Davenport 2010: 351–3.

88

   

Syria Coele and acted as a judge in the emperors’ stead throughout the entire East in the 290s.⁷⁹ Dionysius’ position suggests that he was based at Antioch while Diocletian and Galerius were campaigning against Persia, which would have required close collaboration and contact with the emperors and other courtiers. Maximinus Daza’s closest advisor in his final years, a certain Peucetius, was of senatorial rank. Although Eusebius described Peucetius as being three times consul, this is more likely a reference to an extended term as proconsul of Asia Minor.⁸⁰ One issue that is unclear is whether the emperors’ wives and children usually travelled with them. There was certainly precedent for imperial women, such as Faustina the Younger, Julia Domna, Sabina Tranquillina, and Cornelia Salonina, accompanying their ruling husbands and sons on military campaigns.⁸¹ Evidence, however, for the Tetrarchic period is conspicuously lacking, except in times of emergency, such as when Maximinus Daza’s wife and children had to flee Nicomedia with him during the civil war against Licinius. They accompanied Maximinus and his small army on his final campaign, during which they were captured by Licinius and brought to Antioch.⁸² It was more common for Tetrarchic women to travel for other reasons, such as when they moved between courts as a result of marriages to other members of the imperial college. For example, Valeria Maximilla, Galerius’ daughter, moved westwards when she married Maxentius, some time in the late 290s, and the new couple established themselves in Rome.⁸³ The presence of imperial women and children is much better attested in the principal residences.⁸⁴ Diocletian’s wife Prisca and daughter Galeria Valeria had their own palaces in Nicomedia, which may suggest that this was their primary place of residence.⁸⁵ Maximian’s wife, Eutropia, gave birth to their daughter Fausta at Rome; it is uncertain whether the emperor was himself present, but we certainly know that both mother and daughter resided in the city for some time.⁸⁶ We get a sense of the community that grew up around imperial women from Lactantius’ description of Galeria Valeria’s retinue, which included companions (presumably official attendants), eunuchs, and close female friends of senatorial rank, one of whom was described as a ‘second mother’ (matrem alteram) to Galeria.⁸⁷ The fractured family politics of the later Tetrarchic era meant that members of different bloodlines ⁷⁹ CIL VI 1673. ⁸⁰ PLRE I Peucetius; Eus. HE 9.11.3–4; Barnes 1982: 158. For senatorial families in the East under Maximinus Daza, see Lact. DMP 40.1–6. ⁸¹ Davenport 2017: 24, 28, 30. ⁸² Lact. DMP 47.5–6, 50.1–4. ⁸³ Lact. DMP 18.9; Barnes 2011: 48. ⁸⁴ Lact. DMP 15.1 (Diocletian’s wife Prisca and daughter Galeria Valeria at Nicomedia), 47.5 (Daza collects his wife and children from the palace at Nicomedia), 50.6 (Daza’s wife in residence in Antioch). In other cases, we lack firm contemporary attestation, for example, the tradition associating Constantine’s mother Helena with Trier is Medieval (Drijvers 1992a: 21–30; 2011: 131–5). ⁸⁵ Lact. DMP 7.9. Connolly 2010: 48 suggests that Diocletian’s decision to winter in Nicomedia in 294/5 was motivated by a desire to return to his wife, Prisca. ⁸⁶ Barnes 1982: 58; Hillner 2017: 60. ⁸⁷ Lact. DMP 39.1–40.5. The role of Tetrarchic imperial women is further discussed in Chapter 9.

, ,     

89

sometimes came to reside at the same court. In the years prior to his death in 313, Maximinus Daza’s court was home not only to his (unnamed) wife, his son Maximus, and his unnamed daughter, but also to two sons of deceased emperors.⁸⁸ These were Candidianus, the illegitimate offspring of Galerius, who was betrothed to Daza’s young daughter, and to Severianus, the son of Severus.⁸⁹ When an emperor was based in one of the principal residences for an extended period, the court community expanded in different ways. Major state ceremonies, such as the promotion of Galerius and Maximinus Daza to the rank of Augustus at Nicomedia on 1 May 305, meant that military officers and administrative officials in the region were summoned to court for the occasion.⁹⁰ Triumphs, which were increasingly staged in the provinces, brought together military units and their officers.⁹¹ Sometimes the imperial cities and the palaces had to accommodate two imperial courts, as was the case with Milan in the winter of 290/1, when Diocletian and Maximian were both present, and Nicomedia in the winter of 302/3, when Galerius and his court took up residence alongside Diocletian’s retinue during the discussions that preceded the Great Persecution.⁹² Governors also became members of the court community when the emperor was resident in their province, as shown by the influential presence of Sossianus Hierocles, governor of Bithynia, at Diocletian’s court at Nicomedia.⁹³ If necessary, an emperor could summon governors of nearby provinces—some of whom, as noted above, would have been senators—to consult on important matters, as is again attested in the lead-up to the Great Persecution.⁹⁴ The imperial presence could have significant benefits for local elites, members of the equestrian order and the municipal councils, who came to court to petition and propitiate the emperor.⁹⁵ Many of them represented their regions at major imperial events, like the Gallic orator who represented Autun at the triumph of Constantius I at Trier in 297 and the notables who came to the same city in 311 to celebrate Constantine’s quinquennalia.⁹⁶ But the length of an emperor’s stay, even in the principal residences, was always uncertain, as demonstrated by two Jewish

⁸⁸ In the years 311–13, Maximinus Daza moved between Nicomedia and Antioch (Barnes 1982: 66–7). But the family is only attested in Nicomedia in May 313, which makes it difficult to say whether they travelled with him at all times. ⁸⁹ Lact. DMP 50.1–6; Zon. 13.1.4. ⁹⁰ Lact. DMP 19.1. ⁹¹ For Constantius’ triumph in Trier in 297 following the reconquest of Britain, see Pan. Lat. 8(5).5.4. ⁹² Milan: Diocletian and Maximian met at Milan in December 290/January 291 (Pan. Lat. 11 (3).8.1–12.5; Barnes 1982: 52, 58). Nicomedia: Diocletian and Galerius were both in residence in the winter of 302/3 until March 303 (Barnes 1982: 56, 64). Lact. DMP 14.5 refers to the presence of Galerius’ household (familia) during this period. ⁹³ Lact. DMP 16.4; Barnes 1976: 243–6. ⁹⁴ Lact. DMP 11.5–6. ⁹⁵ For the benefits of imperial presence, see Millar 1977: 36–8; Moser 2018: 126–7. ⁹⁶ Pan. Lat. 10(2).4–5 of 289 articulates an implied Gallic audience for the speech, probably including the town council; Pan. Lat. 8(5).21.2–3 for the panegyric of 297; Pan. Lat. 5(8).2.1 for the quinquennalia in 311. For panegyrics which specifically address the concerns of the Gallic audience, see Rees 2002: 35, 47–8. On elites of Gaul benefiting from the emperor’s presence in Trier, see Rees 2002: 158–9; Davenport 2016: 393–4. See also Pan. Lat. 9(4).13.1 on the youth of Autun accompanying Constantius’ court on his return from Milan to Gaul.

90

   

parables. When the emperor is in a province, ‘people appeal to him and obtain redress’, but when he leaves, only his statue remains, which ‘cannot do what the king can do’, laments the first parable.⁹⁷ The second reflects that even when in residence, the emperor might be occupied with his next move. ‘The provinces close to him come and ask their needs of him’, it says, only for them to be told that the emperor is too busy with preparations for warfare to listen to their demands (this parable was possibly inspired by the residence of Diocletian in Antioch when campaigning for the Persian Wars).⁹⁸ And so the goal of many educated elites was to secure appointments in the imperial service which would lead to a more permanent presence at court, enabling them to advance their own careers and the causes of their communities. It was a great honour to be summoned ‘into the sanctuary of your palace’ (inter adyta palatii uestri), as one orator put it in a speech to Constantius I, the religious language equating service at the emperor’s court with the inner sanctum of a god’s temple.⁹⁹ There was considerable competition to secure these positions, the uppermost of which carried salaries of 200,000 or 300,000 sesterces per year.¹⁰⁰ Yet there could also be significant downsides to being a member of the council in a city that served as an emperor’s principal residence. This is illustrated by the revolt in 303 of Eugenius, a military commander charged with overseeing engineering works at Seleucia. Eugenius and his small force, only five hundred strong, marched to the city of Antioch and ransacked the palace, which was at that time without an emperor in residence (Diocletian and Galerius both being in Nicomedia). The insurgency was actually defeated by the people of Antioch, led by the municipal aristocracy. They remained fiercely loyal to Diocletian, who had resided in the city on many occasions. However, Diocletian nevertheless executed leading members of the council of both Antioch and Seleucia, a decision that appears inexplicable given their fidelity. It may have been the case that the emperor intended this harsh punishment to send a message to any would-be imitators, and the councillors were collateral damage. The incident certainly illustrates that living in an imperial city carried with it particular dangers, regardless of whether the emperor and court were actually in residence.¹⁰¹

4. Palaces and Ceremonial The decision to construct a palace—or palaces, in the case of Nicomedia—in the Tetrarchic principal residences can be ascribed to the need to provide the emperor

⁹⁷ ⁹⁸ ⁹⁹ ¹⁰¹

Midrash Rabbah, Song of Songs 8.9.3 (trans. Freedman and Simon 1939). Mekhilta de-Rabbi Ishmael, Tractate Shirata 4 (trans. Lauterbach 2004). Pan. Lat. 8(5).1.4. ¹⁰⁰ Davenport 2021: 19–20. Lib. Or. 1.3, 11.158–62, 19.45–6, 20.17–21.

, ,     

91

and his court with residential and administrative headquarters that were suitably imperial.¹⁰² This meant that they would not have to rely on staying in the governor’s residence or other elite houses in these cities, as they did in other towns that served as waypoints on their journeys.¹⁰³ The palaces were also designed to be ceremonial platforms for the performance of majesty, with the layout and decoration of audience halls, reception rooms, courtyards, and colonnades emphasizing the emperor’s superior dignity and divinity.¹⁰⁴ Identifying and reconstructing the Tetrarchic palaces has proved challenging on account of the disparate archaeological material, which means that we do not always have information about the precise date or function of individual buildings.¹⁰⁵ Nevertheless, it is possible to draw some general conclusions based on recurring features which can be observed across multiple principal residences.¹⁰⁶ Firstly, the palaces were usually located on the periphery of the cities, not in the centre. This was because they needed to be incorporated into the preexisting urban landscape: forums and temples which usually stood in the central areas could not simply be demolished.¹⁰⁷ Often residential structures were reused and enlarged accordingly, or they were purchased by the administration and then torn down to free up valuable space for the new residence. Secondly, the archaeological and literary evidence shows that many palaces had monumental façades towards the city, the circus, and either a river or seafront.¹⁰⁸ The approaches to the palace from within the city were distinguished by monumental features, such as the quadrifrons arch of Galerius in Thessalonica, part of which is still standing today (Figure 3.2).¹⁰⁹ In Antioch, a monumental colonnaded avenue extended from the Tetrapylon of the Elephants to the entrance of the palace, and similar routes can be detected in other principal residences.¹¹⁰ This had the effect of increasing anticipation for the arrival at the palace, as visitors were channelled in one direction towards the imperial residence.¹¹¹ In the case of the circus façades, excavations in Thessalonica have shown that the palace and ¹⁰² On these aspects as the defining features of a palace, see Wulf-Rheidt 2011: 1–2. ¹⁰³ See Millar 1977: 40–2; Connolly 2010: 49–50. For another example, the temple known as the Tripheion near Panopolis was prepared to house Diocletian’s retinue during the visit to Egypt (P. Panop. Beatty 1.259–61). ¹⁰⁴ For example, see Dey 2015: 41–2; Spieser 2015: 20–3 on the palace at Thessalonica. ¹⁰⁵ Archaeological evidence for these palaces is discussed in the following works. Trier: Fontaine 2003; Kiessel 2012/13. Milan: Haug 2012: 72–4, 92–4. Aquileia: Mian 2012/13. Thessalonica: Hadijitryphonos 2011; Spieser 2015. Sirmium: Popović 2011. Antioch: Ćurčić 1993: 68; Kuhoff 2001: 719–20. Serdica: it has recently been suggested that the Rotunda of St George was a small imperial bath complex, which served as a connection between the public forum and the private palace complex (Kirin 2000). ¹⁰⁶ The following remarks are based on Jaeschke 2017a. ¹⁰⁷ Wulf-Rheidt 2007: 60–1; Mayer 2014: 111. ¹⁰⁸ Cf. Witschel 2004/5: 228; Wulf-Rheidt 2007: 61. ¹⁰⁹ Ćurčić 1993: 70. Thessalonica: Pond Rothman 1977; Mayer 2002: 43–68; Leadbetter 2009: 93–6. ¹¹⁰ Downey 1963: 322–3; Dey 2015: 34–40, noting that the layout of Thessalonica suggests the same arrangement. See Jeremić 2009: 495 for a single street connecting the palace at Sirmium and the city. ¹¹¹ Dey 2015: 56–7.

92

   

Figure 3.2 The Arch of Galerius in Thessaloniki. Photo by Caillan Davenport.

circus were joined by an annexe which housed the pulvinar, and similar structures have been revealed at Sirmium and Milan.¹¹² The retirement complex of Diocletian in Split provides a striking example of a palace façade facing the sea, since it is still standing today. But we also know that the palace in Antioch had a colonnaded arcade facing the Orontes river, from which Emperor Licinius hurled the wife of Maximinus Daza to her death.¹¹³ The seafronts of the palaces at Nicomedia and Thessalonica may well have had similar features.¹¹⁴ The residential quarters within the palaces are especially difficult to identify and reconstruct. One point that we cannot be certain about is how many members of the court resided in them. The Palatine in Rome had never accommodated the entire court community, nor was it designed to do so. Only the emperor, his family, and some members of the slave and freedman household lived in the Palatine (even then, the location of the living quarters is uncertain).¹¹⁵ Senators

¹¹² Humphrey 1986: 612–13, 620, 631. See also Dey 2015: 41–2 on Thessalonica. ¹¹³ Lib. Or. 11.205–7; Lact. DMP 50.6, with Downey 1963: 118. ¹¹⁴ Nicomedia: Lib. Or. 61.10 refers to the shining palace on the bay. Thessalonica: the palace faced the sea, but the type of façade is uncertain (see Mayer 2002: 46; Hadijitryphonos 2011: 207; Spieser 2015: 20 for discussion). ¹¹⁵ There are some hints, such as the collegium of imperial cooks who served on the Palatine (CIL VI 7458). Davenport and Kelly 2022: 117 note the difficulties in ascertaining the living and working spaces of the slaves and freedmen.

, ,     

93

and equestrians instead maintained their own houses elsewhere in the city and came to the palace to participate in the ceremony of salutatio.¹¹⁶ This was in contrast with the arrangement at the French court of Louis XIV at Versailles, which was primarily a residential court society. The majority of courtiers had apartments in Versailles itself in addition to their houses in Paris, a situation which allowed the king to maintain close control over members of the French aristocracy.¹¹⁷ This raises the question of residential arrangements in the Tetrarchic principal residences. Certainly, the palaces were designed to house the emperor, his family, and their households. But the emperor’s travelling comitatus also included senior officials such as the praetorian prefect, leading military officers, the magister libellorum, the magister epistularum, and other equestrian administrators. Where did they live when the imperial courts established themselves in one of these cities? Several possibilities present themselves. They may have had their own house in one particular city (such as Nicomedia, if they were attached to Diocletian’s court),¹¹⁸ but were required to lodge in requisitioned housing when the comitatus came to another city.¹¹⁹ However, there is also the possibility that when the court was based in one of the principal residences, these high officials lived in the palace with the emperor and his family. Lactantius’ description of the flight of Constantine from the court of Galerius at Nicomedia implies that Constantine, then a military officer, resided in the palace.¹²⁰ If it was the case that senior administrators and military officers were housed with the emperor, this would have represented a significant change in the relationship between the emperor and his court community from the situation under the Principate. On the one level, a residential court provided a basis for greater social cohesion and unity among its members, but it also potentially allowed the emperor to control and manipulate them more closely.¹²¹ The most famous aspect of Tetrarchic palace architecture is the design of the monumental audience halls. These were genuinely innovative buildings. We can see from the examples of the Basilica Constantiniana in Trier (Figure 3.3) and the foundations of the audience hall in Thessalonica that they differed significantly

¹¹⁶ On senatorial residences in Rome, see Eck 1997. For praetorian prefects and other high equestrian officials living in their own residences in Rome, the evidence comes from water pipes and other inscriptions: CIL XV 7439 (Sex. Cornelius Repentinus, Antonine age), CIL XV 7505 (M. Opellius Macrinus, Severan period); CIL VI 41190–1 (T. Messius Extricatus, Severan period), CIL VI 41277 (Ignotus, a rationibus, Severan period), CIL VI 41281 (Aelius Firmus, mid-third century). ¹¹⁷ Elias 2006: 47–8. ¹¹⁸ The archaeological evidence is ambiguous. The house of the famous Trier fresco ceiling was long identified as part of the imperial palace (e.g. Kempf 1950; 1977), but may well have belonged to a wealthy noble (Rose 2006). ¹¹⁹ See Millar 1977: 31–5; Connolly 2010: 49–50 on preparations for imperial visits. ¹²⁰ Lact. DMP 24.4–7. ¹²¹ On the community of the travelling scrinium (a team with possibly as many as thirty-four members), see Connolly 2010: 58.

94

   

Figure 3.3 Interior of the Basilica of Constantine in Trier. Photo by Caillan Davenport.

from the known reception rooms in the Palatine complex at Rome, including the halls referred to by modern scholars as the Aula Regia and the Basilica in the domus Flavia.¹²² The audience chamber in Trier was flanked by porticoed courtyards and an entrance hall, which helped to regulate access and organize crowds hierarchically. Unlike the Aula Regia on the Palatine, which had multiple entrances, there was only one access point at the Basilica Constantiniana.¹²³ The longitudinal orientation of the hall towards the apse ensured that the room was completely focused on the emperor himself. The large number of windows, which distinguish the late antique audience halls from the imperial rooms of the domus Flavia on the Palatine, created an atmosphere flooded with light. This image was reinforced by the generous use of precious materials, such as marble floors and glass mosaics around the window arches of the apse (attested at Trier), which reflected the light around the hall. The inspiration for the apsidal shape of these rooms can be found in temples for the imperial cult.¹²⁴ It is true that there were apses in the Aula Regia and the

¹²² This paragraph is indebted to the arguments of von Hesberg 2006: 150–9. For Thessalonica, see Mayer 2014: 114 and Spieser 2015: 23, both of whom note that the basilica was of comparable size to the one in Trier. ¹²³ von Hesberg 2006: 160–1. ¹²⁴ Cf. Ziemssen 2012.

, ,     

95

Basilica in the domus Flavia. But in the Palatine complex, these halls were only two of many possible locations in which the emperor could greet courtiers and visitors. Early Roman emperors may have conducted their salutatio by moving from room to room, where assembled well-wishers would pay their respects in small groups. This style of greeting ceremony emphasized the emperor’s accessibility.¹²⁵ In the Tetrarchic period, the emperors did not move around in this fashion. Instead, they sat in the apses on thrones embossed with gems, which would have also reflected the light coming into the halls. The apse was draped with curtains, which emphasized the emperor’s distance from his subjects.¹²⁶ Anyone who wished to present a gift or other object to the emperor would have had their hands veiled so they did not accidentally touch the ruler’s sacred body.¹²⁷ The niches in the apse in the Basilica Constantiniana suggest that they originally contained statues or frescoes of members of the imperial college, similar to the painted images in the apse of the imperial cult room in Luxor.¹²⁸ This means that when courtiers and visitors entered the audience hall, they would not only pay homage to the ruler(s) actually present, but to the entire imperial college.¹²⁹ Since no temples for the tutelary gods Jupiter and Hercules were built in the Tetrarchic principal residences, it is probable that the halls were supposed to function as temples for the emperors in their Jovian and Herculean guises. Just as the Tetrarchic building programme transformed provincial towns into imperial cities that brought Rome to the regions, so the decoration and design of the palace interiors ensured that all of the Augusti and Caesars could be viewed and venerated throughout the empire. Although there were rituals and protocol which governed interaction with the emperor in the Principate, the ceremony of adoratio transformed the relationship between emperor and subject into one which emphasized hierarchy and remoteness rather than equality and accessibility.¹³⁰ The experience of meeting emperors in the Tetrarchic period is vividly described by the orator of 291 in his account of the visit of Diocletian and Maximian to Milan: What a thing that was, good gods! What a spectacle your piety created, when those who were going to adore your sacred features were admitted to the palace ¹²⁵ Pflug 2014; Davenport 2022: 302–3. ¹²⁶ The gems appear in porphyry statues of enthroned Tetrarchs: see LSA 1003 (M. Bergmann) and LSA 1118 (M. Bergmann with U. Gehn). A jewelled imperial footstool and the bottom of curtains can be seen in the fresco on the south wall of the imperial cult chamber at Luxor (McFadden 2015a: 120–1), and see Figure 3.1 in this chapter. Note also the late antique Noheda Mosaic, in which the Greek king Oenomaus sits on a throne with a jewelled footstool, an image clearly modelled on imperial aesthetics (Valero Tévar 2013: 315, 318). It should be pointed out that the use of curtains in judicial spaces predates their appearance in late antique ceremonial, on which see Färber 2014: 321–4. ¹²⁷ Kelly 2004: 19–22. The veiled hands also appear in the Luxor fresco (McFadden 2015a: 121–2). ¹²⁸ McFadden 2015a: 126–34. ¹²⁹ von Hesberg 2006: 159–60. On the reverence paid to imperial images, see Ando 2000: 232–9. ¹³⁰ On the late antique change, see Stern 1954; Alföldi 1970; Kolb 2001: 22–3.

96

    in Milan, you both were gazed upon and your twin deity suddenly confused the ceremony of single veneration! No one observed the hierarchy of deities according to the usual protocol; they all stopped still to spend more time in adoration, stubborn in their duplicate pious duty. Yet this private veneration, as if in the inner shrine, stunned the minds only of those whose public rank gave them access to you.¹³¹

The panegyrist describes the audience taking place in the palatium. It is not known whether this was the imperial palace itself, which was probably still in the construction phase, or another residential complex.¹³² The key point to emphasize, however, is that the ceremonial procedure of greeting the emperor at court was designed to evoke a religious experience, as visitors and courtiers lingered in their veneration. This is dramatically illustrated by the depiction of such an audience in the fresco on the south wall of the imperial cult chamber at Luxor, in which the courtiers closest to the emperor’s throne tilt their heads upwards; their faces are composed, their eyes wide in adoration (Figure 3.1).¹³³ We are also fortunate to have a brief eyewitness account of Rabbi Samuel ben Nahman, who was summoned to appear before Diocletian in Palestine.¹³⁴ In his : exegesis of a passage in the Book of Esther, he wrote: They were placed as in the basilica of the palace when it is full of people, where the king sits on his couch and all the people lie prostrate before him.¹³⁵

The rabbi emphasizes the veneration of the emperor, but in a slightly different manner to the panegyrist and the Luxor frescoes. His view is not so much that of awestruck adulation, but of ritual subservience, reflecting the experience of the provincial outsider rather than the court orator or artist.¹³⁶ Audiences in the palace were only accessible to those who possessed sufficient dignitas to be granted entrance, as the panegyric of 291 makes clear. There were no promiscuae salutationes in which the general public were admitted, as had been the case in the Principate.¹³⁷ Instead, the people could only gaze at the emperor

¹³¹ Pan. Lat. 11(3).11.1–3: Quid illud, di boni! Quale pietas vestra spectaculum dedit, cum in Mediolanensi palatio admissis qui sacros vultus adoraturi errant conspecti estis ambo, et consuetudinem simplicis venerationis geminato numine repente turbastis! Nemo ordinem numinum solita secutus est disciplina; omnes adorandi mora restiterunt duplicato pietatis officio contumaces. Atque haec quidem velut interioribus sacrariis operta veneratio eorum modo animos obstupefecerat quibus aditum vestri dabant ordines dignitatis (trans. Rodgers in Nixon and Rodgers 1994). ¹³² The function of the rooms in the via Brisa complex, still visible today, remain unknown. See Caporusso et al. 2014: 168–75. ¹³³ See McFadden 2015b: 147–50 on the connection between such images and the world of panegyric. ¹³⁴ Ma’oz 2006: 108–9. ¹³⁵ Midrash Rabbah, Esther 1.20 (trans. Freedman and Simon 1939). ¹³⁶ Compare, for example, the adulatory tone of Pan. Lat. 5(8).1.3. ¹³⁷ Davenport 2022: 292, 302–3.

, ,     

97

during state ceremonial occasions, such as during the adventus processions and appearances at games.¹³⁸ This meant that the interconnected palace and circus complexes in the principal residences were designed to articulate a hierarchy in encounters with imperial power: audiences at the games could only gaze at the emperor from afar, whereas visitors to the palace were granted the privilege of approaching and venerating him.¹³⁹

5. Maxentius in Rome Maxentius, the son of the Augustus Maximian, controlled Italy, Sicily, Sardinia, Corsica, and North Africa between 306 and 312, though he faced a usurpation in Africa between 308 and 309.¹⁴⁰ He was never officially recognized as a member of the imperial college, despite his parentage, and thus remained estranged from the Tetrarchic system. But as an emperor who reigned during the Tetrarchic period, his court deserves consideration in this chapter. Unlike his contemporaries, Maxentius was not a peripatetic emperor, since he used Rome as his principal residence. But he showed similar ambition to the Tetrarchic emperors in embarking on a building programme in Rome and its surroundings with the aim of providing a new architectural environment for his court and his personal performances of power.¹⁴¹ Maxentius acquired ownership of the palace on the Palatine hill, which had been used and augmented by emperors since the time of Augustus. This had both advantages and disadvantages. On the one hand, Maxentius was able to govern from the original imperial palace and he could take advantage of its connection with the Circus Maximus. He enlarged the pre-existing Severan bathing complex and built massive new substructures, giving the façade of the palace adjoining the Circus Maximus a monumental upgrade.¹⁴² On the other hand, there was restricted space on the Palatine hill to build new structures that would enable Maxentius to make his own mark.¹⁴³ He therefore decided instead to build a new monumental basilica north of the Palatine hill on the Via Sacra. The Basilica of Maxentius, which remained unfinished at his death in 312, was a bold project. It boasted vaulted ceiling construction and a large hall flanked by vaulted rooms, which reminds one of the central halls of imperial baths, such as those of Diocletian in Rome. The basilica’s longitudinal axis was created by the apse ¹³⁸ Kolb 2001: 45–6. For these processions, see Dey 2015: 57–64. ¹³⁹ In a parallel development, performances of imperial justice ceased to be held in public locations in Late Antiquity, henceforth taking place only in meetings of the consistorium. See Färber 2014: 99–107. ¹⁴⁰ Barnes 1982: 12–13; Cullhead 1994: 68–9. ¹⁴¹ For previous studies of Maxentius’ building programme, see Cullhed 1994: 49–60; Hekster 1999; Curran 2000: 54–63; Corcoran 2017: 67–9; Jaeschke 2017b. ¹⁴² Wulf-Rheidt 2017: 131–3. ¹⁴³ See Corcoran 2000: 68.

98

   

in the western end and the narthex in the eastern end.¹⁴⁴ The longitudinal orientation towards the apse resembles other reception halls, such as the Basilica Constantiniana. Therefore, the basilica could have been intended to function as a monumental audience hall in the centre of Rome, where Maxentius either sat in the apse in person or was represented by a colossal statue.¹⁴⁵ However, the basilica shows architectural features that differentiate it from the Tetrarchic halls at Trier and Thessalonica, such as the six side rooms flanking the central nave. These rooms may have been intended to gather together different groups of people for audiences with the emperor, or they could have been curtained off and used as administrative or meeting spaces. The location and size of the Basilica of Maxentius were intended to be grand manifestations of the emperor’s claim to power and the status of Rome as his residence. In contrast with the Tetrarchs, who did not construct new temples in their principal residences, Maxentius chose to embark on a major restoration of the Temple of Venus and Rome, which adjoined his new basilica, as a way of honouring his tutelary goddess Roma.¹⁴⁶ Originally square with a flat roof, the cellae were given apses, vaulted ceilings, and two porphyry columns. The interior now featured polychrome marble pavements, coffered ceilings decorated with stucco, and several niches for statues, which were also framed by porphyry columns. The western apse, which was oriented towards the basilica and overlooked the Forum Romanum, housed the cult statue of Roma. This not only illustrates the priority of the cult of Roma, but also emphasizes the connection between the goddess and Maxentius himself. Since the apsidal structure of the basilica was mirrored by the apsidal cellae of the temple, it may have also been the intention for the basilica to seem like a temple. However, it was a temple for the living Emperor Maxentius, the deus praesens. This argument is supported by Maxentius’ coinage, which commemorated the restoration of the Temple of Venus and Roma.¹⁴⁷ A rare gold medallion depicts Roma handing the globe to Maxentius, and the accompanying legend describes Roma as auctrix Augusti, ‘the creator of the emperor’.¹⁴⁸ This shows a clear parallel to the Tetrarchic practice of celebrating tutelary gods, notably Jupiter for Diocletian and Hercules for Maximian.¹⁴⁹ Maxentius’ building programme thus reveals the ambitious intent to incorporate the Palatine, the Basilica of Maxentius, and the Temple of Venus and Roma into a new complex for the veneration of his majesty.

¹⁴⁴ The initial east-west orientation was altered when another apse was added to the central room of the northern side aisle. A new entrance was added at the southern end, which enabled the basilica to be accessed directly from the Via Sacra. Some scholars suppose that the southern entrance was part of the original structure (Cullhed 1994: 51; Dumser 2005: 76–7), while others date the apse and the entrance to the end of the fourth century (Ziemssen 2006: 235–6). ¹⁴⁵ See Ziemssen 2006; Mayer 2014: 124. ¹⁴⁶ For what follows, see Jaeschke 2017b. ¹⁴⁷ Corcoran 2017: 64. ¹⁴⁸ Carson 1965. ¹⁴⁹ See further Ziemssen 2012: 139–41; Jaeschke 2017b: 183.

, ,     

99

The embellishment of Rome was accompanied by the construction of a magnificent suburban villa on the Via Appia.¹⁵⁰ This was unlike any other aristocratic villa: it included an audience hall of significant size, a small circus (which could nevertheless accommodate 10,000 spectators), and a mausoleum for Maxentius’ son Valerius Romulus, who died in 309 while still a child.¹⁵¹ The elements of the villa reflected a range of different aspects of contemporary Tetrarchic complexes. The mausoleum emulated the imperial retirement villas at Split, Gamzigrad, and Šarkamen (though only the tomb of Maximinus Daza’s mother has been found there, not that of the emperor himself ). The connection between the villa and the circus echoed the layout in the Tetrarchic principal residences and in Rome itself.¹⁵² Easy access to the circus pulvinar from the villa was provided by a corridor, which enabled Maxentius to appear before the spectators just as he would have done in the Circus Maximus. The villa’s audience hall had an anteroom and a portico, and it was flanked by courtyards which could be used to muster groups of people for audiences with the emperor. The layout of this reception zone resembles the Tetrarchic palaces at Trier and Thessalonica. How did these complexes shape Maxentius’ interaction with his court? We are hampered by the fact that most of the literary sources are pro-Constantinian, or at least reflect the victor’s point of view, and are therefore hostile to Maxentius. They present a consistent narrative of Maxentius as the enemy of the senate, a cruel tyrant who extorted funds from senators, imprisoned men of consular rank, and committed adultery with aristocratic women.¹⁵³ Maxentius is portrayed in panegyrics addressed to Constantine as a reclusive and indolent figure living in the bowels of the city. It is alleged that the closest he came to a military expedition was to leave the Palatine for the horti Sallustiani, another imperial residence in the northern part of the city.¹⁵⁴ Yet there are good grounds for modifying this picture, and suggesting that Maxentius did have support from at least some members of the senatorial order, and that he was not a recluse, but appeared at a number of public events in Rome. Maxentius’ permanent presence in Rome, where the senate was itself based, meant that the composition of his court community was different from that of the members of the Tetrarchic imperial college. It would be inaccurate to say that the courts of his Tetrarchic rivals were entirely devoid of senators, for, as we have already noted, a large proportion of the amplissimus ordo was located in Rome.¹⁵⁵ But both Maxentius and his father Maximian had had significant opportunities to develop relationships with members of the curia in the preceding years. Maximian and his wife Eutropia are both attested in Rome at different points ¹⁵⁰ For the villa, see Frazer 1966; Cullhed 1994: 57–60; Curran 2000: 62–3. ¹⁵¹ Johnson 2009: 86–93. ¹⁵² Frazer 1966: 386–7. ¹⁵³ Pan. Lat. 12(9).4.4, 20.1–2; Pan. Lat. 4(10).8.3, 31.1; Eus. HE 8.14.1–6, 16–17; V. Const. 1.33.1–36.2, 39.2; Aur. Vict. Caes. 40.24; Eutr. 10.4. ¹⁵⁴ Pan. Lat. 12(9).14.2–6; Pan. Lat. 4(10).27.5. ¹⁵⁵ Corcoran 2017: 65–6.

100

   

between c. 289/290 and Maximian’s abdication in 305. In fact, during the later years of his reign, Maximian is better attested in Rome than in Milan, as noted above. Maxentius, although not a member of the imperial college, was a senator with the rank of vir clarissimus, and he lived in Rome with his wife, Valeria Maximilla (the daughter of Galerius), their son, Romulus, and another son, whose name is now lost.¹⁵⁶ There is good prosopographical evidence to suggest that Maxentius respected traditional senatorial authority, appointing leading aristocrats to the urban prefecture, rather than choosing to elevate outsiders such as military men.¹⁵⁷ When Constantine seized Rome, he punished some senators who supported Maxentius, while retaining others in high office, notably Maxentius’ former praetorian and urban prefect C. Ceionius Rufius Volusianus.¹⁵⁸ Senators such as Volusianus would have formed the nucleus of Maxentius’ court community. When, in October 312, the people voiced their support for Constantine at the games, Maxentius gathered together some members of the senate and had the Sibylline Books consulted for an appropriate course of action.¹⁵⁹ Volusianus, who was a member of the XVviri sacris faciundis in charge of the books, would have been present at this meeting.¹⁶⁰ Of course, we cannot presume that Maxentius had good relationships with all senators: the evidence for forced exactions from members of the curia certainly speaks against this.¹⁶¹ But he certainly had some prominent supporters. Regardless of whether or not senators were personally favourable to Maxentius, they would have had to fulfil their obligations to appear at court, pay their respects to the emperor, and perform adoratio to him on a regular basis. Low-born military men who were elevated through the ranks are not attested in prominent positions under Maxentius. He himself had had no military training (unlike Constantine) and likely remained at arm’s length from the army brotherhood that clustered at the other Tetrarchic courts.¹⁶² The military men Galerius and Maximinus Daza bound their lines together through the betrothal of their children,¹⁶³ but Maxentius would have had to look to the senatorial aristocracy for marriage matches for his sons, had they grown to adulthood in Rome. There were certainly equestrian prefects and administrators who continued to be based in Rome and Italy, but these were not the same sort of military men as Galerius, Maximinus Daza, Severus, Licinius, and their ilk.¹⁶⁴ It can therefore be suggested that the composition of Maxentius’ court society bore greater similarity to the courts of the early Roman emperors than those of his Tetrarchic rivals. ¹⁵⁶ Davenport 2017: 34. ¹⁵⁷ Corcoran 2017: 66. ¹⁵⁸ PLRE I Volusianus 4; Barnes 1982: 100. Punishment of Maxentius’ supporters: Zos. 2.17.2. ¹⁵⁹ Lact. DMP 44.7–8. ¹⁶⁰ Corcoran 2017: 65. ¹⁶¹ Barnes 1981: 37; Hekster 1999: 736. ¹⁶² On the different training of Maxentius, see Corcoran 2017: 60. ¹⁶³ Mackay 1999: 206–7. ¹⁶⁴ For example, Manilius Rusticianus, praefectus annonae and praefectus praetorio (PLRE I Rusticianus 2 and 3). For the equestrian administrators in Rome, see Davenport 2017: 34–5.

, ,     

101

The existence of reception halls both on the Palatine in Rome and at the villa on the Via Appia would have given Maxentius a certain flexibility in where he could choose to hold court. The Palatine was invested with the weight and authority of Roman imperial rule over many centuries, but was not a distinctly individual residence, whereas the Via Appia complex glorified Maxentius himself and his family.¹⁶⁵ Maxentius was thus able to alter the physical location of his court to impart different ideological messages. This was important not only when he greeted members of the senatorial and equestrian elite, but also when he received embassies from his Tetrarchic rivals. We know that Galerius sent ambassadors to Maxentius when he was advancing on the city of Rome, as did Maximinus Daza when he wished to establish an alliance with Maxentius against the other Tetrarchs.¹⁶⁶ Simon Corcoran has plausibly argued that the Via Appia circus was designed to allow Maxentius to appear before ‘a smaller, more select and controllable audience’ compared with the Circus Maximus in Rome.¹⁶⁷ We can make a similar point about the use of the Via Appia villa as a venue for holding court and receiving embassies from his rivals. It was a more intimate residence that emphasized Maxentius’ personal authority and that of his nascent dynasty.¹⁶⁸ The second aspect of Maxentius’ reign in Rome that deserves highlighting is his appearance at public events and the performance of state ceremonial.¹⁶⁹ He was not a recluse who hid behind the walls of the Palatine, as the pro-Constantinian panegyrists proclaimed. Maxentius presided over games in the Circus Maximus and staged public addresses to the soldiers and the people.¹⁷⁰ The reception and dedication of imperial images were also significant public occasions.¹⁷¹ One of the most striking public displays occurred on 20 and 21 April 308, following the expulsion of Maxentius’ rebellious father Maximian from the city. Maxentius and his son Valerius Romulus assumed the consulship, the first for both the emperor and the boy, on 20 April.¹⁷² Then, on 21 April, the birthday of the city of Rome itself, Maxentius and the senator Furius Octavianus dedicated a statue of Mars (probably accompanied by images of Romulus and Remus) in the Forum Romanum, close to the senate house. The monument was joined by a second statue of Maxentius himself, the inscription on the base of which paid tribute to his pietas.¹⁷³ While we have no details about the dedication ceremonies, we can

¹⁶⁵ Frazer 1966: 389; Hekster 1999: 731. ¹⁶⁶ Galerius: Anon. Val. Origo 3.7. Maximinus Daza: Lact. DMP 43.3–5. ¹⁶⁷ Corcoran 2017: 69. ¹⁶⁸ Although Valerius Romulus died in 309, Maxentius did have a second son, whose name is unknown (Pan. Lat. 12(9).16.5). ¹⁶⁹ For the concept of state ceremonial, see the Introduction by Caillan Davenport and Meaghan McEvoy and Chapter 2 by Christian Rollinger. ¹⁷⁰ Games: Lact. DMP 44.7. Soldiers and people: Pan. Lat. 12(9).14.4; Lact. DMP 28.3–4; Eutr. 10.3. ¹⁷¹ Lact. DMP 43.3, 44.10. ¹⁷² Barnes 1981: 32. ¹⁷³ CIL VI 33856; CIL VI 31394a = 33857; Wrede 1981: 140–2; Hunsucker 2018: 103–5.

102

   

assume that this was an occasion at which the court came together in public. The gathering would have included Maxentius and his family (Valeria Maximilla, Valerius Romulus, and the second son, name unknown), the praetorian prefect (possibly Manilius Rusticianus), as well as close senatorial advisors, guards, and attendants. Maxentius was not required to undertake any new building work or urban embellishment in Rome. Unlike the members of the Tetrarchic imperial colleges, he did not need to construct palaces, circuses, and other amenities to elevate his principal residence to the status of a real imperial city: he controlled the one and only sacra urbs. He nevertheless chose to embark on a construction programme in order to emphasize his own personal pre-eminence and divinity within a Roman context. The way in which he used public space in the centre of Rome, in order to draw together the Palatine, the Basilica of Maxentius, and the Temple of Venus and Roma into a new central complex, was creative and innovative. Since he was not a peripatetic emperor, Maxentius’ regime fits into the court-in-capital model, in which emperor, court, and city are bound together in an urban framework. There were similarities with the court-in-capital environment of the Principate, in the form of the regular interaction between emperor and the senate, and the absence of the non-senatorial military elite who formed a key part of the courts of members of the Tetrarchic imperial college. But at the same time, Maxentius’ decision to erect another residence, in the form of a villa, circus, and mausoleum complex on the Via Appia, shows that he was influenced by his Tetrarchic rivals. This second seat gave Maxentius the ability to choose where he could hold court, depending on which aspects of his regime he wished to emphasize to courtiers and visitors alike.

6. Epilogue: The City of Constantine The Tetrarchic system died a long and slow death, eaten away by competition and rivalry between the members of the different imperial colleges. Its final death knell was sounded in September 324, when Constantine defeated his last remaining rival, Licinius, at the Battle of Chrysopolis. Less than two months later, on 8 November 324, Constantine founded a victory city, Constantinople, on the site of Byzantium on the Bosphorus.¹⁷⁴ In many ways, the building programme in Constantinople resembled that undertaken by his Tetrarchic predecessors when they sought to transform their principal residences into imperial cities: the new palace complex was placed adjacent to the city’s circus; a monumental entranceway led through the city to the palace; magnificent new baths were constructed;

¹⁷⁴ See Anon. Val. Origo 6.30; Dagron 1974: 25–7; Barnes 2011: 111–13; Lenski 2016: 54–6.

, ,     

103

the city walls were extended; and a new mint was opened.¹⁷⁵ But there were also innovations that revealed Constantine’s intention to elevate Constantinople to the status of a New Rome.¹⁷⁶ One of these was the decoration of the city with sculptures drawn from across the empire, thus emphasizing its centrality to the Roman world. Constantinople was distinguished with a new circular forum, surrounded by two-storey arcades and embellished with many of these statues, as well as representations of the emperor and his family. In the centre stood a porphyry column topped by a bronze statue of Constantine, who was depicted in heroic nudity and wearing a crown of rays like the god Sol.¹⁷⁷ In the north-west of the city, Constantine ordered the construction of the Church of the Holy Apostles, which was intended to serve as his mausoleum, where he would lie surrounded by representations of Christ’s disciples.¹⁷⁸ The combination of an imperial forum and mausoleum echoed Rome itself, which housed numerous fora constructed by the emperors, as well as the mausolea of Augustus and Hadrian, the final resting places of most emperors up to and including the Severan dynasty.¹⁷⁹ The porphyry column topped with a statue of Constantine was likewise significant, since such structures had previously only been erected in Rome, where columns of Trajan, Antoninus Pius, and Marcus Aurelius towered over the landscape. Constantine’s status as sole Augustus and the foundation of Constantinople as his imperial city, and as a New Rome, created the conditions for the slow transition from the Tetrarchic system of multiple, peripatetic courts to the court-in-capital model of rulership over subsequent generations. From 330 until his death in 337, Constantine used Constantinople as his primary residence and staging base for campaigns in the Danubian region. Members of Constantine’s extended family clustered at the new city, including his half-brother Iulius Constantius and his children. Constantine’s sons likewise formed part of the emperor’s court until the emperor established them in their own residences: Constantine II in Trier from 328, Constantius II in Antioch from 335, and Constans possibly in Milan from 335.¹⁸⁰ We do not know what Constantine’s ultimate plans for the succession were, but his death in 337 changed the situation dramatically, as none of his sons moved permanently to Constantinople to take up residence.¹⁸¹ The city may well have been the source of uncomfortable memories, given that it was where Constantius II had most of his male relatives murdered in

¹⁷⁵ On the city under Constantine, see Janin 1964: 21–31; Bauer 1996: 257–61. For the palace in the urban context, see Dagron 1974: 92–4; Pfeilschifter 2013: 50–1. On Constantine’s forum, see Bauer 1996: 167–87. ¹⁷⁶ On Constantinople as a second or new Rome, see Dagron 1974: 43–7; Grig and Kelly 2012: 4, 9–12; Grig 2012: 43–8. Despite the testimony of Anon. Val. Origo 6.30, Constantine did not found a new senate in Constantinople (Moser 2018: 57–72). ¹⁷⁷ Bauer 1996: 177–9. ¹⁷⁸ Johnson 2009: 119–29. ¹⁷⁹ Davenport 2017: 36. ¹⁸⁰ See the itineraries Barnes 1982: 76–80, 84–7. For the retinues that Constantine allocated to his sons, see Eus. V. Const. 4.51.3, with Davenport 2020b: 224–6. ¹⁸¹ Dagron 1974: 86; Grig and Kelly 2012: 13–14.

104

   

June 337.¹⁸² Constantius II, who was based in the East until the murder of Constans and the usurpation of Magnentius in 350, only made occasional visits to Constantinople, first in 337 (after the conference with his brothers in September at which they were acclaimed Augusti), then again in 342, 343, and 349, and finally in the winter of 359/60 when he returned to the East.¹⁸³ Instead, Antioch was Constantius II’s preferred residence in the eastern provinces, as it provided him with an excellent campaigning base for the Persian wars.¹⁸⁴ Even though Constantius II did not reside in Constantinople, he did not neglect the city. New public buildings were constructed during his reign, including granaries, colonnades, and a library, and it became a centre for intellectual, cultural, and religious life. On 15 February 360, the emperor personally presided over the dedication of the new cathedral of Hagia Sophia.¹⁸⁵ This attention can be ascribed to the fact that Constantinople retained enormous symbolic importance as the city of Constantine, and Constantius II benefited from his association with the city throughout his reign.¹⁸⁶ He is said to have been overcome with anger when his Caesar Gallus staged games in the city in 354, probably because he was usurping the Augustus’ own position as a patron and benefactor of Constantinople.¹⁸⁷ The continuing ideological significance of Constantinople as a key imperial centre in the East is acutely demonstrated by Valentinian I’s proclamation of his brother Valens as co-Augustus at the Hebdomon in 364, and Procopius’ decision to launch his usurpation against Valens in the city in 365.¹⁸⁸ The fact remains, however, that itinerant imperial courts, rather than a court based in a single city, remained the norm during the reigns of the sons of Constantine and their successors, the Valentinians. The pattern of collegiate rulership and movement between different principal residences, as established during the Tetrarchic period, was an essential way to ensure that emperors could adequately tackle internal and external conflicts and make their presence felt in different regions of the empire. The ideological resonance of Constantinople as Constantine’s own city could not immediately change this mode of governance. It was only under the Theodosian dynasty, particularly during the reigns of the child emperors Arcadius and Theodosius II, that the Roman Empire returned to a court-in-capital style of governance, now centred on Constantinople.¹⁸⁹

¹⁸² For the massacre, see Burgess 2008. ¹⁸³ This draws on the itinerary of Constantius II in Barnes 1993: 219–24. ¹⁸⁴ Moser 2018: 121–31; McEvoy 2020b: 281–4. ¹⁸⁵ Moser 2018: 131–41; McEvoy 2020b: 290–1, 294–5. ¹⁸⁶ Moser 2018: 156–66. ¹⁸⁷ Amm. 14.11.13; McEvoy 2020b: 300–1. ¹⁸⁸ Grig and Kelly 2012: 14; Pfeilschifter 2013: 41–2. See Lenski 2002: 24–5, 68–115 and Croke 2010: 241 on Procopius. ¹⁸⁹ This transition is discussed in Chapter 14.

4 The Court in Constantinople Facing the Death of the Emperor Audrey Becker

1. Introduction In his seminal work treating court society as a sociological phenomenon, Norbert Elias argued that the court should be understood as a mechanism created by the ruler in order to manage and control—through court ceremonial, with its protocol and pomp—the elite court society whose dependence on the ruler consequently increased more and more.¹ The influence of Elias’ work on court society studies is unquestionable. Elias’ work has given historians a framework or at least a starting point to explore the court society,² but the application of this model to the late antique court raises as many problems as it solves. For instance, it could hardly be argued that the court of the late Roman Empire only fulfilled ‘the victim’s role’ that Elias’ model needs in order to work.³ In fact, this model assumes that the ruler was strong and clever enough to use ceremonial or protocol as an instrument of power over the court, in order to manipulate it. However, it was certainly possible for a princeps clausus to be manipulated by his courtiers instead, as literary sources at the end of the fourth century imply through their strong criticism of such rulers.⁴ This was due to at least two reasons. The first one was situational: from the reign of Emperor Gratian onwards, children frequently became emperor.⁵ The second one was structural: regardless of how a new emperor achieved power (through an elective process, thanks to a military usurpation, or even by heredity), the issue of his legitimacy arose because there was not any official or formal system to choose the emperor. In other words, emperors needed people who recognized their legitimacy.⁶ This is perhaps one of the most important differences between Louis XIV, whose court was studied by Elias, and many Roman emperors: as the ¹ Elias 2006. See Winterling 1986 and Duindam 1995 for a critical analysis of Elias’ work. ² See Smith 2007 who used Elias’ model as a framework, even if he admitted that it did not fully correspond to the late Roman court. ³ Duindam 1995: 181. ⁴ See Chapter 12, as well as Stroheker1970; Rohrbacher 2016: 151–3. ⁵ McEvoy 2013. ⁶ Kaldellis 2015: 6–7 raises the paradoxical issue of the people of Constantinople regularly taking part in riots, mutilating, or killing the emperors whatever their Christian legitimacy. See also Diefenbach 1996; Pfeilschifter 2013. Audrey Becker, The Court in Constantinople Facing the Death of the Emperor In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0005

106

   

son of Louis XIII, Louis XIV was legitimate by birth and he did not need to prove it or to gain any popular support to be seen as legitimate. From a methodological point of view, I will consider here the court as a space where social interactions between actors—both the courtiers and the emperor— aimed to negotiate their respective power(s) in order to achieve legitimization. In this context, the court has a twofold dimension, being simultaneously a space and the people interacting in this space. The protocol is the frame, what both Elias and Bourdieu called a ‘structure’,⁷ in which social interactions were performed, showing very concretely how each group built and reinforced its power dialectically, which was performed and displayed through imperial protocol.⁸ We can envision a bureaucratic group headed by the magister officiorum, the praetorian prefect, and the quaestor; a military group headed by the magistri militum; the imperial household, the domus Augusta, in which Augustae had a prominent role; the group headed by eunuch chamberlains who staffed the sacred bedchamber; and the group headed by the Patriarch of Constantinople. There were also more informal groups organized around individuals who had an intimate connection with the emperor, whatever their status. All these groups interacted with each other, resulting in a very unstable social web. The lack of any official or constitutional process for electing the next emperor meant that this fragile balance was lost when the emperor—as the main political actor, at least from a symbolic point of view—died. If the heir(s) of the deceased emperor had been previously associated with the imperial power, the power balance at the court could sometimes remain stable, the leading groups being assured of keeping their power. But many other cases could undermine the political influence of other groups at the court and cause political disorder. Indeed, the court had to deal with another critical parameter, namely the role played by the people of Constantinople in the legitimizing process, especially since the beginning of the fifth century, when the emperors took up permanent residence in the New Rome.⁹ The aim of this chapter will be to explore, on the basis of individual case studies, how the court’s leaders acted during the very first moments after the death of the emperor. When there was no designated heir, the court had to manage the election of the new emperor. But whether there was a designated heir or not, the court also had to deal with the citizenry’s reactions during this specific period between the death of the previous emperor and the coronation ceremony of the new one.¹⁰

⁷ Bourdieu 1982. For a comparison between Elias and Bourdieu, see Déchaux 1993; Schumacher 2013. ⁸ On the human aspects of the court, see McCormick 2000: 145–56; Smith 2007: 196–209; WallaceHadrill 2011; Destephen 2016: 109–81. ⁹ On this aspect, see Kaldellis 2015: 102–17. See also Diefenbach 1996; Pfeilschifter 2013. ¹⁰ Rapp 2012 deals with the death of the Byzantine emperors at court in a different perspective by focusing on the boundary between private and public aspects and between secular and sacral

      

107

Furthermore, during this whole process the future emperor could be vulnerable, sometimes even more than any other member of the court, because it was essential for his power to be recognized by the court’s leaders and by the people of Constantinople. This chapter will argue that the court’s leaders manipulated and modified the protocol to suit their own political agenda. Indeed, they simultaneously re-established and assured their own prominent place within the court by securing for themselves the right to take part in electing the next emperor when it was required by the circumstances. This chapter will discuss the period between the death of one emperor and the coronation of the next one in four significant cases: the emperors Constantine, Zeno, Anastasius, and finally Justinian.

2. The Court Facing the Death of Constantine: Waiting for the Heir(s) The defeat of Licinius in 324 meant that for the first time in forty years, the Roman Empire was ruled by only one Augustus.¹¹ From 330 onwards, the Augustus and his court settled in Constantinople, which not only bore his name but was also explicitly founded as a New Rome. Although Constantine’s sons were made Caesars, assigned officials, and set up residence in different cities throughout the empire, there could be no doubt that it was the Augustus who ruled supreme from the city on the Bosphorus. And so it was that the emperor was buried in Constantinople following his death near Nicomedia on 22 May 337. For the circumstances surrounding the death of Constantine and his burial, we are entirely dependent on the narrative of the only contemporary source, Eusebius of Caesarea. He reports that Constantine’s corpse (likely embalmed) was transferred to Constantinople, after the conclamatio, the ritual expression of grief and sorrow, was performed by his bodyguards and the population of Nicomedia in a traditional way.¹² Once the translatio cadaueri (‘transfer of the deceased’) to Constantinople had been completed—the choice of the capital he had built as his last resting place was in itself an innovation¹³—a viewing of the corpse took place in the Great Palace. Eusebius describes it as follows: The military took up the remains and laid them in a golden coffin. They wrapped this in imperial purple, and bore it into the city named after the emperor; then in

dimensions in rituals. For an analysis of the coronation ceremonies, see Becker 2016. Kaldellis 2015: 102–17 examines the role played by the people in the imperial accession. For a comparison of the interregna in the western and eastern halves of the empire, see Feeney 2019. ¹¹ This was the brief period in which Diocletian ruled alone as Augustus prior to the elevation of Maximian. ¹² Rapp 2012: 268–70. For the role of the guards, see Chapter 2. ¹³ See the final section of Chapter 3.

108

   

the most superb of all the imperial halls they laid it on a high pedestal, and by kindling lights all round on golden stands they provided a wonderful spectacle for the onlookers of a kind never seen on earth by anyone under the light of the sun from the first creation of the world. Within the palace itself, in the central imperial quarters, the emperor’s remains, adorned with imperial ornaments, with purple and crown, was guarded day and night by a huge circle of people keeping vigil. The commanders of the whole army, the comites and all the ruling class, who were bound by law to pay homage to the emperor first, making no change in their usual routine, filed past at the required times and saluted the emperor on the bier with genuflections after his death in the same way as when he was alive. After these chief persons, the members of the senate, and all those of official rank came and did the same, and after them crowds of people of all classes with their wives and children came to look.¹⁴

It has already been demonstrated that Eusebius managed to describe the whole funeral process in an acceptable way for Christians while reshaping and Christianizing the Greco-Roman concept of apotheosis.¹⁵ But what remains less clear is the meaning of this post-mortem daily homage. For Eusebius, it was proof that there was no transition crisis because, even while dead, the emperor was still ruling over the empire. But this is an ideological interpretation of what happened, and it leaves aside at least two related questions which have never been fully examined: who decided on such a performance before a dead emperor and what this performance meant. Constantine’s living sons—Constantine II, Constantius II, and Constans—all held the rank of Caesar at the emperor’s death in 337, as did their cousin, the Caesar Dalmatius. No explicit record of Constantine’s intention for the composition of the imperial college after his death survives. It has been posited that he planned for a new Tetrarchy, composed of two Augusti and two Caesars, but this must remain speculative.¹⁶ Yet the death of any Roman emperor was a time of anxiety and tension, because well-laid plans, even if known to his closest advisors, ¹⁴ Eus. V. Const. 4.66.1–67.2: Ἄραντες δ’ οἱ στρατιωτικοὶ τὸ σκῆνος χρυσῇ κατετίθεντο λάρνακι, ταύτην θ’ ἁλουργίδι βασιλικῇ περιέβαλλον ἐκόμιζόν τ’ εἰς τὴν βασιλέως ἐπώνυμον πόλιν, κἄπειτα ἐν αὐτῷ τοῦ παντὸς προφέροντι τῶν βασιλείων οἴκων βάθρον ἐφ’ ὑψηλὸν κατετίθεντο, φῶτά τ’ ἐξάψαντες κύκλῳ ἐπὶ σκευῶν χρυσῶν θαυμαστὸν θέαμα τοῖς ὁρῶσι παρεῖχον, οἷον ἐπ’ οὐδενὸς πώποτ’ ὑφ’ ἡλίου αὐγαῖς ἐκ πρώτης αἰῶνος συστάσεως ἐπὶ γῆς ὤφθη. ἔνδον γάρ τοι ἐν αὐτῷ παλατίῳ κατὰ τὸ μεσαίτατον τῶν βασιλείων ἐφ’ ὑψηλῆς κείμενον χρυσῆς λάρνακος τὸ βασιλέως σκῆνος, βασιλικοῖς τε κόσμοις πορφύρᾳ τε καὶ διαδήματι τετιμημένον, πλεῖστοι περιστοιχισάμενοι ἐπαγρύπνως δι’ ἡμέρας καὶ νυκτὸς ἐφρούρουν. Οἱ δέ γε τοῦ παντὸς στρατοῦ καθηγεμόνες κόμητές τε καὶ πᾶν τὸ τῶν ἀρχόντων τάγμα, οἷς τὸν βασιλέα καὶ νόμος πρότερον ἦν προσκυνεῖν, μηδὲν τοῦ συνήθους ὑπαλλαξάμενοι τρόπου τοῖς δέουσι καιροῖς εἴσω παριόντες τὸν ἐπὶ τῆς λάρνακος βασιλέα οἷά περ ζῶντα καὶ μετὰ θάνατον γονυκλινεῖς ἠσπάζοντο. μετὰ δὲ τοὺς πρώτους ταῦτ’ ἔπραττον παριόντες οἵ τ’ ἐξ αὐτῆς συγκλήτου βουλῆς οἵ τ’ ἐπ’ ἀξίας πάντες, μεθ’ οὓς ὄχλοι παντοίων δήμων γυναιξὶν ἅμα καὶ παισὶν ἐπὶ τὴν θέαν παρῄεσαν (trans. Cameron and Hall 1999). ¹⁵ Among a large bibliography, see MacCormack 1981; Price 1987; Dagron 1996: 148–54; Rebenich 2000; Rapp 2012: 268–72; Becker 2022: 277–80. ¹⁶ See Burgess 2008: 8–9 for the Tetrarchy proposal.

      

109

could dissolve among the ambition and rivalry of legitimate heirs and opportunists. Eusebius’ account shows that the customary daily homage (προσκύνησις) to Constantine continued to be performed while everyone waited for his sons to arrive in Constantinople.¹⁷ Such a ritual was an outward expression of an agreement that had clearly been reached among all the different groups, military and civil, at court to preserve the status quo for the time being.¹⁸ This was especially important for the relatives of Constantine who were present in Constantinople, notably Iulius Constantius, his half-brother, who may have hoped to play a role in upcoming negotiations, but did not wish to show their hands publicly.¹⁹ By participating in these rituals, Iulius Constantius and his allies indicated their continuing loyalty to Constantine and to the idea of dynasty to the assembled military. Moreover, the daily ritual demonstrated to the citizenry of Constantinople that no one at court called into question the legitimacy of the emperor: even if he was dead, he carried on ruling.²⁰ Although at first sight what Eusebius describes appears to be very similar to Herodian’s account of imperial funeral practice, especially the employment of an effigy, the meaning was quite different.²¹ The effigy was used to stand in for an emperor, who was supposed to be alive until his funeral. For Eusebius, there is no doubt that the daily routine was performed before a dead emperor. The room in the Great Palace where Constantine lay became the spatial framework for this performance, the perfect organization of which would clearly show the permanence of the imperial τάξις (‘formation’), consisting of the army, the court, and the emperor.²² Furthermore, contrary to appearances, if performing προσκύνησις during the display of the imperial remains was the most concrete way to deny any vacancy in the office of the Augustus, it did not refer to the famous Medieval doctrine of the king’s two bodies.²³ As a matter of fact, Eusebius gave an ideological framework to this ceremonial innovation: the ritual προσκύνησις before the emperor’s dead body was reshaped as a way to pay homage to his immortal Christian soul which ruled on earth from heaven. Thus, the βασιλεία μετα θανατου (‘rule after death’), which sidestepped the current power vacuum, was linked with the immortal soul of the emperor who had left his human form. This explanation was given by Eusebius in an apologetic fashion, one of the goals of which was to give Constantine’s sons a kind of miroir du prince, urging them to maintain harmony and continuity in a Christian framework.²⁴ ¹⁷ On the court ceremonial, see Bravo 1997; Tantillo 2015. ¹⁸ Eus. V. Const. 4.67.1: ‘the commanders of the whole army, the comites and all the ruling class’ (oἱ δέ γε τοῦ παντὸς στρατοῦ καθηγεμόνες κόμητές τε καὶ πᾶν τὸ τῶν ἀρχόντων τάγμα). ¹⁹ PLRE I Constantius 7; Burgess 2008: 9 on Constantine’s half-brothers as ‘senior statesmen’. ²⁰ Price 1987: 101. ²¹ Hdn. 4.2.2–4. On the conduct of imperial funerals in the Principate, see Davies 2000. ²² On the taxis, see Carile 1998. ²³ Kantorowicz 1957. Cf. Arce 2000: 123–4, who uses this theory as a valid framework. ²⁴ Cameron and Hall 1999: 9–12.

110

   

It remains difficult to assess at what point people taking part in the daily routine endorsed this Eusebian framework. It is even debatable whether the object of this ceremony was the emperor’s remains themselves, as Eusebius described, or the insignia of the imperial office on his body. This would have evoked the ideology of the Roman res publica in which the power was embodied in an office and its symbols, not in a person. But how people taking part in this ritual understood it at the time was probably not as significant as the fact that the action itself allowed everyone to perform an act of political consensus by bringing all the different interest groups together. Eventually this new ritual achieved its goal: by managing to make the people of Constantinople wait quietly for one of Constantine’s sons, the court hoped to ensure a peaceful transition of power and to maintain the power balance between all the groups at court. The funeral itself was held some time in early June, after Constantius II had arrived in Constantinople.²⁵ Paradoxically, the best proof that this innovation worked was the violence of Constantius II’s reaction following the funeral. The pretence that Constantine still ruled had allowed Constantius II to play for time to solve the issue of political tensions and uneasiness about the succession.²⁶ When Constantius II arrived in Constantinople, he was clearly aware of the existence of networks at the court built by and around his relatives and the leading courtiers chosen by his father. He could see them paying their daily respects to the remains of his father. In order to take control of the court, he immediately dismissed Flavius Ablabius, the praetorian prefect, who had been one of the leading officials at Constantine’s death. In June 337, Constantius II had his relatives present in Constantinople killed, eliminating all but two of the male descendants of Constantius I and his second wife Theodora (Constantius Gallus and Julian were spared). The victims included Iulius Constantius, Flavius Dalmatius and his two sons, the Caesar Dalmatius, and Hannibalianus, who bore the title ‘king of kings and the Pontic peoples’ (rex regum et gentium Ponticarum).²⁷ Constantius II also executed Flavius Optatus, the first man to be made a patricius by Constantine and, according to Jerome, ‘many of the nobles’ (multo nobelium), among whom were probably the rhetor Aemilius Magnus Arborius, Virius Nepotianus, and Flavius Felicianus.²⁸ Even though he had retired to Bithynia, Ablabius was likewise put to death. On 9 September 337, the three brothers—Constantine II, Constantius II, and Constans—were all acclaimed Augusti by the soldiers in Sirmium. The public display of unity by

²⁵ Burgess 2008: 38. ²⁶ The pretence that Constantine was still ruling after his death appears in contemporary sources: Burgess 2008: 29–30; Barnes 2011: 167. ²⁷ Burgess 2008: 10, 40–1; Moser 2018: 151–2. ²⁸ Jer. Chron. s.a. 338. See also Zos. 2.40.3; Eun. V. Soph. 6.3.9–13; Jul. Ep. ad Ath. 270C–D.

      

111

the court following Constantine’s death forestalled open dissent and civil war, but by not taking action, many of its members had unwittingly set the stage for their own removal.

3. The Court Facing the Death of Zeno: Trying to Avoid Chaos When Zeno died on 9 April 491, the situation was complicated in a different way. His legitimate sons had predeceased him, nor had he a designated Caesar, though there was a family candidate in his brother Longinus.²⁹ In this situation, Peter the Patrician (upon whom the early material in the tenthcentury Book of Ceremonies depends) makes it clear that the leading officials at court—to whom he refers collectively as ἄρχοντες (archontes)—took charge of the next emperor’s election.³⁰ The first step of this process was for the archontes to convince Constantinople’s citizenry gathered at the hippodrome that they had a right to play a role in selecting the next emperor. The absence of any designated heir meant that the appointment to the purple was solely a matter for the court to decide. They displayed their consensus by publicly coming together to support the Augusta Ariadne, Zeno’s widow and the daughter of Leo I.³¹ The leaders of the court instructed her to go into the hippodrome dressed in a military cloak (χλαμύς). Surrounded by the principal officials, Ariadne presented herself to the populace as the legitimate embodiment of the βασιλεία, or imperial power. According to Peter the Patrician, the court office-holders present included the praepositus sacri cubiculi (who was Urbicius at that time), the magister officiorum (who since 484 had been Longinus of Cardala), the castrensis sacri palatii and quaestor sacri palatii (whose names are unknown), as well as the archbishop of Constantinople (Euphemios).³² Ariadne addressed the crowd, emphasizing that she possessed the loyalty of the population of Constantinople. Her message was quite clear. She had asked ‘the highly esteemed archons, and the sacred senate, with the common consent of the most noble’³³ to convene in secret and choose the new emperor.³⁴ In Ariadne’s address, she proclaimed that the new emperor would be selected by a divinely inspired court:

²⁹ On Zeno’s efforts to secure the succession during his lifetime, see McEvoy 2019b: 202–4, and on Longinus, 206. An illegitimate son is attested in Theoph. AM 5964. ³⁰ Const. Porph. de cer. 1.91–2 (1.100–1). ³¹ On Ariadne, see Meier 2010; Croke 2015a; McEvoy 2019b; and Chapter 9. ³² Const. Porph. de cer. 1.92 (1.101). ³³ Const. Porph. de cer. 1.92 (1.101): τοῖς ἐνδοξοτάτοις ἄρχουσι καὶ τῇ ἱερᾷ συγκλήτῳ μετὰ κοινῆς τῶν γενναιοτάτων δοκιμασίας ἄνδρα (trans. Moffatt and Tall 2012). ³⁴ On the silentium cum conventu, see Delmaire 1995: 33–4.

112

   

So that the decision is untainted and pleasing to God the ruler, with the concurrent support of the most noble armies, and with the Holy Gospels set before us and in the presence of the most holy and saintly Patriarch of this imperial city, we have given orders to the highly esteemed archons and the sacred senate, and with the holy Scriptures set before us, as has been said, for the choice to be made.³⁵

Hence, the ideology used by Aurelian and Constantine—both of them chosen by their armies, divinely inspired by pagan gods³⁶—was now expressed in a Christian civilian framework in Constantinople. The court leaders emphasized that they, and not the troops, were the medium of communication of God’s will. And it worked: people at the hippodrome, the demes (divisions of the people based on circus factions), and the military all agreed to wait until the funeral of Zeno took place, thus giving the court more time to make a decision about the next emperor. But Peter the Patrician’s account of the events shows clearly that this unity was a pretence. As soon as the first public ceremony was over, tensions arose at court. The leading candidate was Zeno’s brother Longinus, twice consul and magister militum, who had the support of the magister officiorum (confusingly, also called Longinus).³⁷ But he was passed over by the court in their deliberations—probably because of his unpopularity and their reluctance to have another Isaurian wear the purple.³⁸ Unable to reach a decision by agreeing on a name for the next emperor, the archontes gave Ariadne the responsibility of making the choice. It seems entirely possible that it was the praepositus Urbicius who suggested that they invest this power in Ariadne. Urbicius had been very close to Ariadne’s mother, Verina.³⁹ A few years earlier, in 484, thanks to her authority as Augusta, Verina had made the coronation of the usurper Leontius possible.⁴⁰ It is important to underline how politically expedient this solution was for the leading men at court. The involvement of the Augusta Ariadne—the only person who could not be suspected of wanting to become Augustus, no matter her ³⁵ Const. Porph. de cer. 1.92 (1.101): ὥστε δὲ καθαρὰν καὶ τῷ δεσπότῃ Φεῷ ἀρέσκουσαν τὴν κρίσιν γενέσθαι, ἐκελεύσαμεν τοὺς ἐνδοξοτάτους ἄρχοντας καὶ τὴν ἱερὰν σύγκλητον, συντρεχούσης καὶ τῆς τῶν γενναιοτάτων ἔξερκίτων ψ’φου, προκειμένων καὶ τῶν ἁγίων εὐαγγελίων, παρόντος τοῦ ὁσιωτάτου καὶ ἁγιωτάτου τῆς βασιλίδος ταύτης πόλεως πατριάρχου, καὶ προκειμένων, ὡς εἲρηται, τῶν ἅγίων λογίων, γενέσθαι τὴν ἐπιλογὴν (trans. Moffatt and Tall 2012). ³⁶ The Anonymous Continuator of Cassius Dio (FHG IV fr. 10.6) states that the troops of Aurelian were divinely inspired when they chose him as emperor; the same claim is made for Constantine in Pan. Lat. 7(6).7.3–5, delivered at court in 307. ³⁷ PLRE II Longinus 7; Evagr. HE 3.29; Theoph. AM 5983. Longinus’ ambitions are illustrated by the fact that he had made an imperial match by betrothing his daughter Longina to the house of Anthemius (Joh. Ant. fr. 239.3, ed. Mariev 2008; this is discussed further in McEvoy 2022). Soon after Anastasius’ succession, Longinus would be exiled to Egypt and the homonymous magister officiorum dismissed. ³⁸ For the tensions with Isaurians, see Brooks 1893: 231–2. ³⁹ In 481, Urbicius arranged an assassination attempt on Illus on behalf of Verina because he refused to release Verina’s mother (Theoph. AM 5972; Marcell. com. s.a. 484). ⁴⁰ Joh. Mal. 15.13.

      

113

popularity—avoided escalating the crisis.⁴¹ Ariadne exercised influence because she was married to the former emperor, but because of her gender she was not allowed to have any constitutional political power. By involving Ariadne in the process of selecting the new emperor, the leading courtiers showed her that she could still play a role in politics. This was something that she was undoubtedly willing to embrace, given that all her influence derived from her husband Zeno and her father Leo, both now dead. The nomination of the successor offered Ariadne a way to assert her relevance. The personal relationship that she established with the crowd at the hippodrome was probably part of a strategy to be seen by the people as legitimately carrying a part of the βασιλεία. Ariadne chose Anastasius, who, as a silentiarius under the jurisdiction of the praepositus sacri cubiculi, was a court official. This suggests that she was advised by Urbicius, the praepositus, in making her choice.⁴² Ariadne stamped her authority on the situation by selecting a man who owed his rise solely to her, and she chose to marry him to formalize the arrangement. The selection of Anastasius, who was formally acclaimed on 11 April, did not mean the end of the tensions. The very weakness of his position gave Anastasius no other choice but to agree, before his coronation ceremony, to swear an oath before the senators and the leading members of the court ‘that he would not harbour a grudge against anyone with whom he had had dealings, and that he would administer the state with a strict conscience’.⁴³ Furthermore, even before this oath, the Patriarch Euphemios had required Anastasius to issue a written declaration that he would respect the definition of the faith decided by the synod at Chalcedon.⁴⁴ It is very clear that, at that time, Anastasius did not control the course of events, but was controlled by the court through these ceremonies.⁴⁵ It had long been a crucial component of imperial ideology that God, or the gods, inspired and guided the army’s choice of the emperor. But with the death of Zeno, the leading officials of the court clearly tried to control the whole process, from Anastasius’ election to his coronation ceremony. To this purpose, they modified the protocol. According to the Patria of Constantinople, there was no public display of Zeno’s corpse (as there had been with Constantine). Access to the body was restricted to members of the court, and it is likely that the funeral was closed as well. Then, court officials decided that the coronation ceremony would be held not at the Hebdomon, but at the hippodrome instead. This was a telling decision and a very concrete way to show that the court, rather than the army, was the

⁴¹ On female rulership in Byzantium and Medieval Europe, see Chapter 14. ⁴² Haarer 2006: 1–5. See also Meier 2009. ⁴³ Const. Porph. de cer. 1.92 (1.101): ὡς οὐδενὶ, πρὸς ὃν ἔσχεν πρᾶγμα, φυλάττει λύπην, καὶ ὅτι μετὰ ὀρθοῦ συνειδότος τῇ πολιτείᾳ χρήσεται (trans. Moffatt and Tall 2012). ⁴⁴ Theoph. AM 5983. On the document signed by Anastasius, see Evagr. HE 3.32. ⁴⁵ On this issue at the court of Louis XIV, see Duindam 2016: 202–3.

114

   

medium of communication of God’s will. The troops still took part in the coronation ceremony, but only to accept and legitimate the election decided by the court.

4. The Court Facing the Death of Anastasius: Unrest Breaks Out When Anastasius himself died on 9 July 518, the chaos which had been avoided on account of Ariadne’s mediating role twenty-seven years earlier could no longer be circumvented. As soon as the news of Anastasius’ death was known, the scholae and the excubitores became nervous. In the absence of any designated heirs, the uncertainty grew even greater since the senators, the leading court officials, and the Patriarch did not manage to agree quickly on a successor. Anastasius had three nephews—Pompeius, Probus, and Hypatius—but the emperor did not designate any of them his successor by elevating them to the rank of Caesar.⁴⁶ Peter the Patrician does not even mention them. Among the three, only Hypatius seems to have been prominent enough to be a serious candidate, but he was overseas, far away from Constantinople when Anastasius died.⁴⁷ Moreover, Hypatius’ lack of success as a general made him unpopular.⁴⁸ In these circumstances, the leading figures of the imperial court took the initiative. Celer, who was the magister officiorum appointed by Anastasius, convened his colleagues, and urged them as follows: We should consult and act as soon as we can, for if we name quickly the person who ought to be appointed, everyone will follow us and be content. If after a little while we are not masters of the deliberations, we shall have to follow the rest.⁴⁹

Despite the foresight of Celer, the assembled courtiers failed to agree on a name. Faced with the procrastination of the court and without any Augusta able to maintain order by convincing the demes and the military gathered at the hippodrome to wait (Ariadne had passed away a few years before), the situation got out of hand and riots broke out. There were several different candidates for the purple: John, proclaimed emperor by the excubitores; Patrikios, by the scholarioi and the Blues; Theokritos, by Amantius the praepositus sacri cubiculi of Anastasius; and ⁴⁶ Cameron 1978; Haarer 2006: 246–9; McEvoy 2018. ⁴⁷ At that time, he was magister utriusque militiae. See Greatrex 1996. ⁴⁸ Hypatius was defeated in 503 by Cavades, which the sources ascribed to his cowardice (Proc. BP 1.8.13–19; Marcell. com. s.a. 503; Joh. Lyd. de mag. 3.53). In 513, he was even taken as prisoner by Vitalian, and Anastasius had no other choice except to pay the ransom (Marcell. com. s.a. 515; Evagr. HE 3.43; Joh. Mal. 16.16; Theoph. AM 6006). ⁴⁹ Const. Porph. de cer. 1.93 (1.102): ἐν ὅσῳ ἔξεστιν ἡμῖν, βουλευσόμεθα καὶ πράξομεν.ἐὰν γὰρ ταχέως ὀνομάσωμεν τὸν ὀφείλοντα γενέσθαι, πάντες ἡμῖν ἀκολουθήσωσιν καὶ ἡσυχάζουσιν. ἐπεὶ μετ᾽ὀλίγον οὐ γινόμεθα κύριοι τῆς βουλῆς, ᾽σλλ᾽ἡμεις ἑτέροις ἔχομεν ἀκολουθεῖν (trans. Moffatt and Tall 2012).

      

115

finally, Justin, by the senators.⁵⁰ Justin emerged as the consensus figure, and he was accepted by the military, the demes, as well as by Amantius the praepositus sacri cubiculi. This turn of events calls for comment. The ability of the court to select the emperor depended on the military and the demes allowing them this privilege.⁵¹ In order for this to happen, the different factions at the imperial court had to be able to overcome their internal divisions to reach a consensus (even if it meant a delegation of responsibility to Ariadne after the death of Zeno), in order to avoid political disorders caused by popular riots. If there was no consensus, then the court as a whole lost its ability to select an emperor. Instead, individuals and groups of courtiers championed their own candidates. After the death of Anastasius, the praepositus sacri cubiculi Amantius was in a position to block the elevation of two candidates—John and Patrikios—in order to increase the chances of his preferred emperor, Theokritos. The supporters of John and Patrikios performed their own acclamation ceremonies, with John being raised on a shield at the hippodrome and Patrikios on a couch in the Hall of the Nineteen Couches in the Great Palace.⁵² But Amantius, as praepositus sacri cubiculi, had control over the imperial apparel (τὸ σχῆμα τοῦ βασιλέως), which he refused to give to either John or Patrikios. This undermined their legitimacy, and it was enough to remove them from contention for the throne. But at the same time, Amantius himself failed to obtain the necessary backing for his candidate Theokritos because the latter lacked popularity among the citizenry and the military. Thus, the court settled on the compromise candidate, Justin, the comes excubitorum, who was granted the imperial insignia, and acclaimed in the hippodrome. There were, however, allegations that this had been achieved by bribery, since Amantius had given Justin money to secure support for Theokritos, but Justin had used it to advance his own cause.⁵³ Indeed, after Justin was crowned emperor, Amantius paid for his support of a rival—he was accused of plotting against the emperor and eliminated, along with another cubicularius called Andrew and the former praefectus praetorio Orientis Marinus.⁵⁴ Since Justin was not the unanimous choice of the imperial court, but a candidate who only emerged after dissent and negotiation, he needed to remove those officials who had supported his rivals and staff key positions with his own new appointees.⁵⁵

⁵⁰ Const. Porph. de cer. 1.93 (1.102); Evagr. HE 4.2. ⁵¹ On the continuity of the republican idea of popular sovereignty, see Kaldellis 2015. ⁵² Const. Porph. de cer. 1.93 (1.102). ⁵³ For example, Chron. Pasch. s.a. 519; Theoph. AM 6011. On later attempts to explain the succession of Justin over Anastasius’ nephews, see McEvoy 2018: 99–107. ⁵⁴ Theoph. AM 6011; Evagr. HE 4.2; Proc. Anecd. 6.26. For an analysis of these events, see Greatrex 2007, who suggests that the stories about Amantius were concocted to justify his execution. ⁵⁵ According to Zach. HE 8.1, the execution was a consequence of the opposition to Justin’s religious policies. But the speed of this purge, within the first ten days of Justin’s reign according to Procopius, suggests that there was a real urgency that was more likely political than religious.

116

   

5. The Court Facing the Death of Justinian: Learning from Previous Mistakes When Justinian died in 565, he had no co-emperor nor a designated heir. But in contrast to what happened after Anastasius’ death, the leading court officials acted very quickly in choosing between two nephews of Justinian, proclaiming Justin the son of Vigilantia, rather than Justin the son of Germanus, as emperor. Our principal source of evidence is Corippus’ ‘In praise of the younger Justin Augustus’ (In Laudem Iustini Augusti Minoris), which, despite its panegyrical nature, shows that chaos was avoided by careful preparation. Callinicus, the praepositus sacri cubiculi of Justinian and the only witness to the latter’s last moments, claimed that the emperor had designated Justin (son of Vigilantia) as his heir just before he died.⁵⁶ Then the senators acted very quickly: before the news of the emperor’s death circulated all over Constantinople,⁵⁷ they brought Justinian’s body to the Great Palace, which was secured by the comes excubitorum to avoid any troubles inside.⁵⁸ The coronation of Justin II took place only a few hours after the death of Justinian. This was an unusual development, as members of the court decided to crown the new emperor while the previous ruler was not even buried. The purpose of this innovation was to avoid any official power vacancy or at least to shorten it as much as possible, in order to make Justin II the only legitimate successor. As a result, after Justin II had been designated by Justinian on his deathbed and crowned, both emperors wore the imperial insignia at the same time in the Great Palace—one deceased, while waiting for his funeral, and the other alive after his coronation. The location of the coronation ceremony had been changed by the court itself to fit the circumstances. It took place in the Great Palace, probably in the consistory itself, behind closed doors. This meant that the idea of universal consensus for the new emperor was merely symbolic, as in reality Emperor Justin II was chosen and crowned by the court.⁵⁹ In order to make this turn of events acceptable to the people of Constantinople and to avoid any rivals or riots,⁶⁰ Justin II was crowned by the Patriarch, like his predecessors.⁶¹ By involving the Patriarch, the court could still present Justin to ⁵⁶ Evans 2001: 264. ⁵⁷ Coripp. Iust. 1.176–8. ⁵⁸ Coripp. Iust. 1.202–8. ⁵⁹ In 527, the coronation ceremony of Justinian by Justin I also took place within the Great Palace, but at the Delphax in very different circumstances (Const. Porph. de cer. 1.95 [1.104]). Justin, at that time, was very ill: the Chronicon Paschale mentions an ulcer on his foot, while John of Nikiu refers to an old arrow wound in his head which had reopened (Chron. Pasch. s. a. 527; Joh. Nik. 90.47). In both cases, the physical injury he suffered would probably have been very serious. Furthermore, the ceremony took place before the court and the troops. ⁶⁰ The last riots had taken place less than three years before, not to mention the earlier Nika riots and the coronation in 532 of the nephew of Anastasius, Hypatius, at the hippodrome. On the riots in 562–3, see Theoph. AM 6054, 6055; Joh. Mal. 18.132, 135, 136, 138. On the coronation of Hypatius, see Chron. Pasch. s. a. 531; Theoph. AM 6024; Joh. Mal. 18.71. ⁶¹ Coripp. Iust. 2.159–74; Becker 2016: 151–5.

      

117

the people as the choice of God, whose decision the court alone had witnessed. As soon as the coronation ceremony was over Justin II went to the kathisma, the imperial box in the hippodrome, where he could be publicly acclaimed by the demes. The people always played a key role in the legitimization of the emperor even if, stricto sensu, they did not take part in the coronation ceremony itself any more. Therefore, Justin II was successfully recognized as emperor because he was supported by a consensus among the leading officials at court. They managed to present his elevation to the people as a fait accompli, as he had already been crowned by the Patriarch as God’s choice. The people could not argue with that.

6. Conclusion In Late Antiquity, the death of each emperor initiated a period of uncertainty which ended only with the acclamation and coronation of his successor. During this critical period, the leading court officials found themselves in a precarious position, as the sudden disappearance of the emperor destabilized the court community. They could either choose to work together to select a successor, or they could splinter into different factions, each with its own imperial candidate. The court in Constantinople also had to seek the consent of the people gathered at the hippodrome, since their support was regarded as necessary for the legitimization of the new emperor. There was no way of escaping the fact that popular acclamations were ‘a matter of principle’,⁶² forming a concrete way of expressing the consensus on which imperial legitimacy relied. However, the court leaders could manipulate, modify, shift, and even create new ceremonies to control events in the interregna between emperors. This is exactly what the imperial court did in 337, while waiting for the arrival of Constantine’s sons. By allowing the emperor’s body to be displayed and venerated on a daily basis, they emphasized the continuity of imperial power. The situation was not quite as straightforward when there was no son to succeed the emperor, as happened after the deaths of Zeno, Anastasius, and Justinian. Leading court officials tried to control the nomination of a successor by reshaping the process of acclamation. There were various ways to do this. The court could decide to hand over the decision to a suitably legitimate individual, such as the Augusta Ariadne after the passing of Zeno. Another method was to make the imperial funeral a court event, as occurred after the death of Justinian, and then emphasize the choice of the successor as the decision of a divinely inspired Christian court, rather than handing over power to the military. The

⁶² Kaldellis 2015: 110.

118

   

Christian framework of this process for choosing an emperor gave religious legitimacy to what were in essence pragmatic formal innovations. The success of these measures, however, depended on members of the court working together to decide on how events would unfold. If there were no consensus at court, as occurred after the death of Anastasius, then there could be dissent and riots among different factions and rivals. The eventual successful candidate for the purple would then be forced to reshape the membership of the court, in order to eliminate potential threats and re-establish consensus.

PART II

IN DI VI D U AL A ND COM M UNIT Y AT COURT

5 Was the Roman Imperial Court an ‘Emotional Community’? Benjamin Kelly

1. Introduction In Pliny’s Panegyricus, one point of contrast between the principate of Trajan and the unsavoury regime of Domitian concerns the emotional atmosphere allegedly prevailing at these emperors’ respective salutationes.¹ ‘We do not gather,’ Pliny claims, ‘pale and stupefied with fear, nor delaying our approach as if in danger of our lives, but when it is convenient for us, free of fears and cheerful.’² The speech continues in a similar vein, expatiating on the terrors for the visitor that lurked within Domitian’s palace, and the inner chaos of anger, fear, and shamelessness expressed by the emperor’s very mien. We shall never know whether Pliny gives a fair account of how Trajan, Domitian, and their subjects actually felt during the salutatio. Nevertheless, the passage highlights two important points: monarchical courts are often imagined (by contemporaries and by posterity) as having characteristic forms of emotional expression, and the existence and outward manifestations of emotions at court are often the target of strong normative evaluation. An interest in emotions at court was present at the dawn of court studies: Norbert Elias saw the exigencies of life in an Early Modern court society as encouraging the suppression of strong displays of feelings.³ More recently, Barbara Rosenwein’s study of what she called ‘emotional communities’ in the Early Middle Ages has added an important analytical tool for the study of courts and emotions. By ‘emotional communities’, Rosenwein has in mind groups of people who ‘adhere to the same norms of emotional expression and value—or devalue—the same or related emotions’.⁴ Such communities are not necessarily very large, and many can coexist at any one time within a nation or kingdom; their

¹ This research was supported by the Social Sciences and Humanities Research Council of Canada. I am grateful for constructive advice on the chapter from the Editors and the Press’s reader, as well as for the comments from seminar audiences in Toronto and London, Ontario. Thanks are also due to Chris Dawson, who served as a research assistant on this project. ² Plin. Pan. 48.1, cf. 48.4: itaque non albi et attoniti, nec ut periculum capitis adituri tarditate, sed securi et hilares cum commodum est convenimus. ³ Elias 2006: 120–6. ⁴ Rosenwein 2006: 2. Benjamin Kelly, Was the Roman Imperial Court an ‘Emotional Community’? In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0006

122

   

members are generally united not just by a common set of values and behaviours relating to emotional expression, but also more generally by the fact that they have ‘a common stake, interests, values, and goals’.⁵ ‘Emotional communities’ are generally not unique in their emotional norms and values, but more often overlap to some degree with other emotional communities, and share some of their affective traits.⁶ A monarchical court has the potential to be such a community, given that its members are in frequent face-to-face interaction and share many traits and goals—above all the goal of winning and maintaining the monarch’s favour. Among Rosenwein’s case studies are the courts of the Merovingian kingdom of Austrasia (561–613) and the Neustrian kingdom, which absorbed Austrasia in 613. For Rosenwein, the comparison of the two shows how the emotional regimes of courts can vary even over short timespans. In Austrasian court circles, she detects an emphasis on familial love and a proclivity for the expression of affection, especially using dulcedo (‘sweetness’) and its cognates to describe feelings and their targets.⁷ In the Neustrian court, on the other hand, she detects an aversion to effusive expressions of emotion, together with a privileging of bonds between males and of status hierarchies, which had an impact on emotional life.⁸ This approach has recently been adopted by Sally Holloway and Lucy Worsley in an analysis of the court of early Georgian England, which they see as an ‘affective space’ where there was a shared ‘grammar of emotions’.⁹ In this chapter, my aim is to bring the history of the Roman imperial court into conjunction with the history of Roman emotion.¹⁰ The focus is on how people at the court expressed and thought about emotion. In other words, I engage in a detailed study of a specific aspect of court culture: the presence in court circles of shared opinions on emotions and their expression at court.¹¹ To study such a specific dimension of court culture, we must avoid isolated texts that potentially offer idiosyncratic perspectives or repeat standard literary images or philosophical doctrines rather than reflecting court culture.¹² Instead, we need clusters of roughly contemporaneous texts (preferably of different genres) written by authors

⁵ Rosenwein 2006: 24. ⁶ The concept of the ‘emotional community’ has been adopted by historians of emotion in a variety of times and places. For use of the concept in relation to Antiquity, see Chaniotis 2011; Chaniotis 2016. The concept is also used at points by Hagen (2017: 52–4, 62, 109, 115, 122, 133, 163, 223), although it is not at the forefront of her analysis. ⁷ Rosenwein 2006: 100–29. ⁸ Rosenwein 2006: 130–62. ⁹ Holloway and Worsley 2017. ¹⁰ For a critical survey of research on Graeco-Roman emotion, see Hagen 2017: 27–37. ¹¹ On the court as a centre for the generation of shared opinion, see Wallace-Hadrill 1996: 293–5; cf. Winterling 1999: 76–82 for the court as a centre for the generation of taste. ¹² For this reason, I do not examine Cassius Dio, for example. He clearly had close connections with the late Antonine and Severan courts, but the absence of other relevant writers of that era with court connections turns him into an isolated source whose representativeness of court values is open to doubt.

      ‘   ’ ? 123 with court connections, who are likely to reflect court mores. I propose to focus on three such clusters. Firstly, Suetonius, Tacitus, and Pliny all had connections of various kinds with the courts of Domitian, Trajan, and Hadrian, and wrote under one or both of the last two of these emperors. Secondly, the surviving letters of Marcus Fronto and his correspondents (who include Marcus Aurelius, Antoninus Pius, and various other important court figures) illustrate richly the performance and evaluation of emotion within the inner circle of the Antonine court. Thirdly, there is an important cluster of texts produced by people with various kinds of court connections covering the period from the late 350s to the 380s; these include the letters and orations of Julian, the Res Gestae of Ammianus Marcellinus, and the speeches of Claudius Mamertinus, Ausonius, and Themistius. As is usual in the study of court history in all periods, our texts come from members of an elite, so we must remember throughout that, for the most part, they only give us access to the thought-world of this echelon of the court. Moreover, this is (with the exception of Ausonius) a non-Christian elite, so if Christian courtiers had subtly different views concerning emotions at court, we should not expect to see them in these clusters of sources. My roughly diachronic analysis of these textual clusters focuses on the questions of whether one can see characteristic forms of emotional expression at court in particular periods, and on whether there are signs of shared norms relating to the emotional lives of emperors and courtiers. What I suggest is that in the second-century court, one can see a community whose members evaluated and expressed emotional norms along very similar lines. In the fourth century, however, fault lines appear, and one can find cases of dissonance between the norms of courtiers and the emotional performances of some emperors in court contexts.

2. The Emotionally Dysfunctional Court: Tacitus, Suetonius, and Pliny on the First Century In discussing the Julio-Claudian and Flavian courts, Pliny, Suetonius, and Tacitus certainly present an image of a place of high emotion and one where feelings were broadly shared across the court.¹³ But sometimes there is no hint of how these emotions were expressed outwardly, raising questions about how our authors and their sources knew the inner feelings of court members. One suspects that authors and biographers added such claims as splashes of colour on the basis of nothing more than intuition—or prejudice. There are also doubts in cases in which firstcentury courtiers are said to have shown outward signs of their emotions, as Domitian’s pallid morning callers do in Pliny.¹⁴ These doubts stem from the

¹³ Suet. Dom. 14.1; Tac. Agr. 45, Ann. 6.50, 11.28.

¹⁴ Tac. Ann. 11.28; Arr. Epict. diss. 4.1.48.

124

   

sheer negativity of the ‘emotional palette’ (to use another of Rosenwein’s terms) attributed by Tacitus and Suetonius to the courts of later Julio-Claudians and Domitian—emperors of whom these authors generally disapproved. Fear¹⁵ and hatred¹⁶ are the most frequent emotions ascribed to later Julio-Claudian emperors and their courtiers in court contexts. Prominent too is invidia. As Robert Kaster has shown, invidia in Roman culture could refer to two things: either to the emotion that we would call jealousy (and the Greeks φθόνος), or to something approaching the feeling one has when one feels righteous indignation or looks askance at a person’s breach of social norms.¹⁷ Emperors are portrayed as feeling invidia (in the sense of jealousy) towards courtiers,¹⁸ and courtiers towards other courtiers;¹⁹ courtiers also feel invidia (in the sense of righteous indignation) against misbehaving emperors.²⁰ Emperors and courtiers are also portrayed as tailoring their actions to try to avoid invidia²¹ or consciously (or unconsciously) stirring up invidia against others.²² On the other hand, more positive emotions are largely absent, and when they appear, the environment of the court perverts them. For example, hilaris (‘cheerful’) and cognates never appear in a court context in Tacitus’ Annals. In Suetonius, some emperors display the emotion in inappropriate contexts that reflect poorly on them,²³ but it is not an emotion felt by courtiers. Love at the courts of the late Julio-Claudians and Domitian is generally sexual attraction, inappropriate either because it is adulterous,²⁴ or directed at people of low social station,²⁵ or abused by courtiers to manipulate and entrap.²⁶ Even what should be normal conjugal love comes across as tainted, such as when Suetonius’ Caligula threatens to torture his wife to discover why he loves her so.²⁷ Gaudium and laetitia (‘joy’) and their cognates in Tacitus are experienced by emperors individually,²⁸ and also by some people who could be called courtiers, but almost always in contexts outside the court.²⁹ In short, then, the emotional palette ascribed to the courts of the later JulioClaudians and of Domitian is so unremittingly negative that we cannot make

¹⁵ Tac. Ann. 4.7, 6.50, 6.51, 11.28–32, 12.64, 12.47, 13.13, 13.20 14.61, 14.65; cf. Mart. 12.5 (= 12.2 + 12.6.1–6); Juv. 4.74–5. ¹⁶ Tac. Ann. 1.69, 4.7, 12.22, 13.15, 14.1, 14.3, 14.56, 14.61, 16.20. ¹⁷ Kaster 2005: 84–103. ¹⁸ Plin. Pan. 45.3, 95.4; Tac. Ann. 13.15. ¹⁹ Tac. Ann. 13.6, 14.3, 14.54, 16.18. ²⁰ Tac. Ann. 4.53 (with Hagen 2017: 170–2); Suet. Dom. 14. ²¹ Emperors: Tac. Ann. 4.30; cf. 4.70; 15.64; Suet. Dom. 1.2–3. Courtiers: Tac. Ann. 15.52. ²² Emperors: Suet. Calig. 56.1. Courtiers: Tac. Ann. 12.3. ²³ Suet. Calig. 27.4, Ner. 34.2, Dom. 11.1. Outside court contexts: Calig. 18.2; Claud. 21.5. When Suetonius’ Augustus experiences hilaritas in court contexts, there is no such negative connotation: Aug. 98.2, 98.3; Tib. 21.2. ²⁴ Suet. Dom. 3; Tac. Ann. 11.12. ²⁵ Suet. Calig. 36; Tac. Ann. 13.12–13. ²⁶ Suet. Claud. 26, 36, Otho 2; Tac. Ann. 6.45, 13.46. ²⁷ Suet. Calig. 33. Tiberius’ initial love for Julia also turns sour: Tib. 7. ²⁸ Tac. Ann. 1.52, 2.84, 11.25, 15.23, 15.60. ²⁹ Tac. Ann. 1.76 (Drusus); 2.75 (Piso); 13.46 (Otho); 14.4 (Agrippina the Younger), 14.5 (Agrippina’s slave, Acerronia). The senate also experiences gaudium twice in the Annals: 6.3, 14.46.

      ‘   ’ ? 125 confident claims about how emotion was felt or performed. With this said, this palette is still an interesting reflection of how these courts were remembered in the early second century, something that is of historical interest in itself. I would suggest, though, that one can go further than this. Tacitus, Suetonius, and Pliny at various points in their careers moved in court circles; moreover, they were of a social level that would have given them constant contact with courtiers. Inevitably, then, their accounts would have been conditioned by contemporary court discourses on emotion and its display—the ‘grammar of emotion’, to use Holloway and Worsley’s phrase. Most basically, our authors’ discussions of earlier courts show us which emotions were perceived in the early second century as unbecoming of an emotionally healthy court. This sort of discourse is explicit in Pliny’s (rather misogynistic) claims about Trajan’s wife, Plotina, and his sister, Ulpia Marciana: they have avoided the feminine pattern of close proximity leading to rivalry (aemulatio), thence to jealousy (invidia), and finally to hatred (odium).³⁰ Ultimately, this absence of invidia and odium at court is meant to reflect positively on Trajan. One can also see from our cluster of early second-century texts that norms relating to anger control circulated in court circles in that period.³¹ These norms had their origins in Greek philosophical discussions about the need for monarchs to restrain their rage, and were then absorbed into the culture of the Roman elite, steeped as it was in Greek philosophy.³² Under Trajan, claims Pliny, parents only need worry about the natural fragility of human life; the anger of the emperor (principis ira) is now off the list of diseases that might kill their offspring.³³ That anger in an emperor was frowned upon is also suggested by the cases in Pliny and Tacitus where anger is ascribed to first-century emperors of whom these authors disapproved, but where the anger is said to have been detectable only by subtle giveaways³⁴—or completely undetectable.³⁵ The absence of any real evidence in these cases shows how the ascription of anger was a standard reproach against a stereotypically bad emperor that could be made independent of historical fact. The ideology of anger control in Roman culture did not just relate to rulers, but was

³⁰ Plin. Pan. 84.2. See Tac. Ann. 2.43 for aemulatio muliebris between Livia and the Agrippina the Elder at the court of Tiberius. ³¹ Norms of anger control were not, of course, exclusive to elite culture. For anger control in Roman popular morality, see Morgan 2007: 101–2, 104, 174, 333. ³² Harris 2001: 229–63. Knight has recently attempted to nuance Harris’s position, arguing that in the early Principate, anger was sometimes viewed as an acceptable way to communicate with an emperor’s subjects ‘provided he followed certain unwritten performative or “display” rules’ (2015: 210), and that early emperors were aware of the utility of anger as a political tool (2016). However, the shortage of relevant contemporary evidence for the period makes it challenging to substantiate the theory. ³³ Plin. Pan. 27.1, cf. 40.5. ³⁴ Tac. Ann. 1.13–14, 1.74; Plin. Pan. 48.4; cf. Tac. Agr. 42. ³⁵ Tac. Ann. 2.28, 4.21, 6.50, 14.49. Note too Tac. Ann. 14.56, where Nero hides his odium in an interaction with Seneca.

126

   

more broadly directed;³⁶ courtiers were also expected to contain their rage, an idea visible already in the first century in Seneca’s De ira.³⁷ This point is suggested by the fact that in Tacitus and Suetonius, it is mainly courtiers of whom the author disapproves who are said to experience rage in court contexts, including Sejanus, Messalina, and Agrippina the Younger.³⁸ Fear also attracted normative judgements in second-century court circles. When the sources report the fear that pervaded earlier emperors’ courts, this is a clear indictment on those emperors. Rather more interesting is the existence of norms of fear management—that is, a notion that emperors should not let themselves be overtaken by a fear that those in their immediate circle will kill them. Part of the rationale for this is that fear can lead a ruler to behave cruelly and unjustly, a rationale implicit in tales of first-century emperors who executed people out of a misplaced fear that they were conspiring to kill them.³⁹ In Suetonius’ account of the Libo Drusus affair of 16, an additional rationale is hinted at: that fear causes an emperor to treat his courtly elite with a lack of civilitas. For it was only once Tiberius was liberated from fear (metus) by the death of Libo, Suetonius claims, that he was able to behave in a way that was civilis, eschewing excessive honours.⁴⁰ The idea that emperors should not be overtaken by a fear of those in their own retinues is also implicit in the praise lavished on emperors who faced the prospect of conspiracies with nonchalance.⁴¹ Closely associated with this approval of the fearless emperor is condemnation for devious courtiers who stir up emperors’ fears (and, indeed, other emotions) for their own nefarious ends. Thus, one dimension of the hostile traditions about Claudius’ manipulation by his wives and freedmen is that he allowed his emotions to be controlled. For instance, in Tacitus’ version of the Messalina and Silius episode, the freedman Narcissus uses his wiles to make Claudius furious (incensus) against his wife and her paramour,⁴² and to prevent Claudius’ amici from stirring up any regret (paenitentia) in the emperor for the punishment that Messalina is about to suffer.⁴³ Later in the narrative, the freedmen Narcissus, Callistus, and Pallas also work to overcome the pity (misericordia) of the emperor towards the pantomimist Mnester, who is duly executed.⁴⁴ In relating Agrippina’s scheme to push aside Octavia’s fiancé, L. Junius Silanus, in favour of Nero, Tacitus comments that this was easily achieved, since this was an emperor ‘who made no decisions and felt no hatred (odium) unless they were imposed upon him and ordered’.⁴⁵ It is not just Claudius who is portrayed as being manipulated in this

³⁶ Harris 2001: 201–28. ³⁷ Sen. Ira 3.15.2–3. ³⁸ Sejanus: Tac. Ann. 4.3, 4.19; 5.4; Suet. Tib. 61.1. Tac. Ann. 4.52. Messalina: 11.37. Agrippina the Younger: Tac. Ann. 12.22, 13.13, 13.18, 14.4. For other possible cases of courtiers’ anger, see Tac. Ann. 4.53, 6.26, 16.22. ³⁹ Suet. Dom. 14.4–15. ⁴⁰ Suet. Tib. 26.1, cf. Claud. 35. ⁴¹ Suet. Tit. 9.1–2; cf. Suet. Vesp. 25. ⁴² Tac. Ann. 11.35. ⁴³ Tac. Ann. 11.33. ⁴⁴ Tac. Ann. 11.36. ⁴⁵ Tac. Ann. 12.3; cf. Suet. Claud. 35–7.

      ‘   ’ ? 127 way: Tacitus repeatedly makes similar claims about the emotions of Tiberius,⁴⁶ Nero,⁴⁷ and Vitellius⁴⁸ being cynically swayed by those close to him. In some of these narratives, devious courtiers are said to have faked their own emotions as part of the process of manipulating those of the emperor. This draws our attention to a delicate area of the grammar of emotion existing in secondcentury court circles: the ethics of counterfeiting emotions.⁴⁹ On the one hand, fake emotion is portrayed as being used by wicked courtiers to manipulate the emperor.⁵⁰ On the other, blanket condemnation of counterfeit emotions at court was not viable, as those who had lived through the reign of Domitian knew full well. In the Panegyricus, we can see Pliny easing his way around this problem by suggesting that the public honours and other apparent expressions of love by courtiers and senators for Domitian were really the products of servitude and fear—and therefore their falsity was the fault of the emperor.⁵¹ In the narratives of the first-century court, there are episodes of feigned emotion that likewise seem to reflect negatively on the emperor who provokes them rather than the courtiers forced to produce them—for instance, when Britannicus is poisoned by Nero in the Annals, Octavia (whom Tacitus views positively) manages to remain impassive, since she has long since learnt to hide grief (dolor), affection (caritas), and other feeling (adfectus).⁵²

3. Expressing Love at the Second-Century Court The second-century writers who provide accounts of the later Julio-Claudians and Domitian and their courts mostly illuminate the negative emotional norms prevailing in these authors’ own day—that is, ideas about what the emotional life of a court should not be like. But what of the positive side of things? What sort of emotions and emotional performances were valued in second-century court circles? Pliny’s discussions of court life under Trajan are of great value here precisely because they are overheated panegyric: they imagine an ideal emotional atmosphere for a court. At Trajan’s salutatio, all are cheerful (hilares) and free from fear (securi);⁵³ joy (gaudium, laetitia) is felt throughout the court, and beyond.⁵⁴ Above all, for Pliny it is (appropriate) love that supposedly prevails at Trajan’s court: both Trajan’s wife and sister love him most extravagantly ⁴⁶ Tac. Ann. 1.69, 1.74. ⁴⁷ Tac. Ann. 13.21, 13.47, 14.57–9, 14.61–2. ⁴⁸ Tac. Hist. 3.38–9; cf. Levene 1997: 136–7. ⁴⁹ This theme is also visible earlier in Sen. Ira 3.14–15. On feigned emotion in Roman culture more generally, see Hagen 2017: 202–12, 230–3, 304–5. ⁵⁰ Tac. Ann. 11.3 (analysed by Hagen 2017: 78–82), 13.20, cf. 12.68. ⁵¹ Plin. Pan. 55.3, 72.6–7, 74.3–4, 85. ⁵² Tac. Ann. 13.16. For other cases where self-preservation demands feigned emotion, see Tac. Ann. 1.7, 6.50, 15.71. ⁵³ Plin. Pan. 48.1. ⁵⁴ Plin. Pan. 46.8, 68.2, 72.2, 83.6, 95.4.

128

   

(effusissime diligat), and he loves them in return;⁵⁵ he loves his amici, feels affection (caritas) for them,⁵⁶ and longs for them (desiderare) in their absence;⁵⁷ and they love him in turn—unlike Domitian, for whom amicitia was an empty name in the absence of real love.⁵⁸ And as he reaches the climax of his panegyric, Pliny declares that he loves (diligo) Trajan, just as much as he was the target of Domitian’s jealousy (invisus).⁵⁹ Discourse about the ideal of love at court had matching norms relating to its physical expression. In the Panegyricus, Pliny recounts an anecdote about an unnamed amicus of Trajan who went into retirement. The emperor, we are told approvingly, did not hold back his emotions, but kissed and embraced his friend as he departed on his retirement.⁶⁰ Also revealing are the stories of first-century emperors who offered their hands or feet to be kissed, or gave chilly and brief kisses, even when interacting with important members of the court.⁶¹ There is similar disapproval in Epictetus’ references to people desperate for imperial favour kissing the hands of imperial slaves.⁶² Such kissing is seen as an act of calculated abasement⁶³ replacing the proper performance of warm emotion appropriate to a well-ordered court. Unsurprisingly, Pliny’s Trajan reverts to the appropriate performance of emotion: ‘You do not force citizens to embrace your feet or return a kiss with your hand; the same human feeling (humanitas) when kissing remains as before, the same restraint with your right hand.’⁶⁴ It is very clear from the surviving correspondence from and to Fronto that amor continued to be emphasized in court circles under Antoninus Pius and Marcus Aurelius (and his co-emperor, Lucius Verus). Most famously, the letters exchanged between Marcus and Fronto are bursting with the language of love. To take one example of many, in a letter from the early years of his principate, Marcus writes to Fronto: I beg you, my master, love me as you do love me; love me just as you love those young children of ours; I have not yet said everything that I want to say: love me in the same way as you have loved me.⁶⁵

⁵⁵ Plin. Pan. 84.4, 87.4. ⁵⁶ Plin. Pan. 86.1, 87.5. ⁵⁷ Plin. Pan. 86.5. ⁵⁸ Plin. Pan. 85. ⁵⁹ Plin. Pan. 95.5, cf. 68.7. See too Plin. Ep. 2.1 for Domitian’s alleged invidia towards Verginius Rufus. ⁶⁰ Plin. Pan. 86.1–4. ⁶¹ Caligula: Suet. Calig. 56, cf. Dio 59.27.1, 59.29.5, Sen. Ben. 2.12. Nero: Tac. Ann. 15.71. Domitian: Suet. Dom. 12; Tac. Agr. 40; cf. Mart. 10.72. ⁶² Arr. Epict. diss. 4.1.148, cf. 1.19.24, 3.7.29–31, 4.10.20–1. ⁶³ Cf. Paterson 2007: 147–8. ⁶⁴ Plin. Pan. 24.2: non tu civium amplexus ad pedes tuos deprimis, nec osculum manu reddis; manet imperatori quae prior oris humanitas, dexterae verecundia. Cf. Plin. Pan. 23.1. For the text and interpretation of Pan. 24.2, see Kühn 1987. ⁶⁵ Fronto, Ad Ant. Imp. 1.4.1 (vdH² p. 92): oro te, mi magister, ama me ut amas; ama me sic etiam, quomodo istos parvolos nostros amas; nondum omne dixi quod volo: ama me quomodo amasti.

      ‘   ’ ? 129 The quantity and intensity of the language of love in the letters between Marcus and Fronto has prompted the controversial suggestion that the pair were at some point involved in a sexual or romantic relationship.⁶⁶ Whatever the details of Marcus’ sex life, the important point here is that this same basic language is used to express the emotional content of other court relationships, albeit without the same degree of rhetorical elaboration. Fronto writes to Marcus that he loves (amo) Antoninus Pius ‘like the sun, like the day, like life, like breath’ (ut solem, ut diem, ut vitam, ut spiritum) and feels that he is loved by the emperor in return (amari me ab eo sentio).⁶⁷ In one of the Greek letters of the corpus, Fronto puts Marcus’ birth mother, Domitia Lucilla, on the same emotional plane as Marcus when he says that he mentioned both in an encomium for Antoninus Pius ‘just as lovers name the objects of their affection over every wine-cup’ (ὥσπερ οἱ ἐρασταὶ τοὺς φιλτάτους ὀνομάζουσιν ἐπί πάσῃ κύλικι).⁶⁸ Various other members of the court are said by Fronto to be loved by or to love him: Claudius Severus (cos. ord. 146);⁶⁹ Sulpicius Cornelianus (probably later an ab epistulis Graecis under Marcus Aurelius);⁷⁰ and Gavius Clarus (a senator and former praetor).⁷¹ He also readily ascribes loving feelings to others:⁷² Marcus and Lucius Verus, he claims, shower amor on their visitors during the salutatio;⁷³ Lucius Verus, Fronto believes, loves him;⁷⁴ and in a letter to Marcus at the New Year, Fronto sends best wishes to Marcus’ father (i.e. Antoninus Pius), mother, wife, daughter (Lucilla), ‘and all the other people you deservedly love’ (ceterisque omnibus, quos merito diligis).⁷⁵ Nor is this language just the product of an overwrought emotional style on Fronto’s part; other letter writers in the corpus use it too. In an early letter, written apparently from an imperial villa, Marcus writes that last year when staying at the same place, he was ‘inflamed with longing’ (desiderio peruri) for his mother—in much the same way, he writes, that he now has a longing for Fronto.⁷⁶ Elsewhere Marcus declares his love for other members of the court, including Herodes Atticus⁷⁷

⁶⁶ Richlin 2006a; 2006b; 2011, followed by Taoka 2013a; 2013b; contra: Laes 2009; Davenport and Manley 2014: 9–11, 97. ⁶⁷ Fronto, Ad M. Caes. 2.4.1 (vdH² p. 25); cf. Fronto, Ad Ant. Pium 10.3 (vdH² p. 68). On the relationship between Fronto and Antoninus Pius, see Chapter 1 by Christoph Michels. ⁶⁸ Fronto, Ad M. Caes. 2.3.4 (vdH² p. 23). For the wine-cup simile, see Richlin 2011: 175. ⁶⁹ Fronto, Ad amicos 1.1.3 (vdH² p. 171), cf. M. Aur. Med. 1.14. ⁷⁰ Fronto, Ad amicos 1.2 (vdH² p. 171), cf. Eck 1992: 239–40. ⁷¹ Fronto, Ad Verum Imp. 1.6.2 (vdH² p. 110), cf. PIR² G 97. ⁷² Fronto’s enthusiasm for talking about affection is reflected in his borrowing of the Greek φιλόστοργος (‘warm hearted’) to describe others’ capacity for affection: Fronto, Ad amicos 1.3.4 (vdH² p. 173), Ad Verum Imp. 1.6.7 (vdH² p. 111). Marcus himself came to connect the borrowing with Fronto, and apply it to him: M. Aur. Med. 1.11; M. Aur. ap. Fronto, De fer. Als. 4.2 (vdH² p. 234). ⁷³ Fronto, Ad Verum Imp. 1.12.2 (vdH² p. 115). ⁷⁴ Fronto, Ad Verum Imp. 1.12.2 (vdH² p. 115). ⁷⁵ Fronto, Ad M. Caes. 5.45 (vdH² p. 77), cf. 5.48 (vdH² p. 78). ⁷⁶ M. Aur. ap. Fronto, Ad M. Caes. 3.9.4 (vdH² p. 42). ⁷⁷ Fronto, Ad M. Caes. 3.2.1 (vdH² p. 36), cf. 2.1.3 (vdH² p. 17).

130

   

and Fronto’s daughter, Cratia.⁷⁸ Thus, there was an ideal that love should be the paramount emotion shared by all members of the inner circle of the court; with Marcus there may or may not have been a sexual dimension, but he and his teacher were clearly using the language prevailing in their particular emotional community. In one intriguing letter, Fronto grapples with the tension between this ideal and the emotional realities of the late Hadrianic court. He writes (somewhat apologetically) to Marcus that while he loves Antoninus Pius, with Hadrian his feelings were different: he wanted to ensure that Hadrian was propitious and appeased, but did not love him, partly because love requires some degree of intimacy, but partly because he lacked confidence (fiducia).⁷⁹ As Edward Champlin has pointed out, this last comment seems to reflect the general fact that in his last years, Hadrian turned erratically against some of his courtiers, as well as the specific fact that Fronto felt that he was in a dangerous position in this period, to judge from hints in other letters.⁸⁰ What Fronto reveals, at least in the privacy of his letter to Marcus, is that memories of actual emotional states cannot always be completely sanitized to fit present political needs. In the second-century court, amor was not just expressed discursively: it had a performative element as well. The Fronto correspondence is filled with references to actual, expected, or imagined physical expressions of love, and not just between Marcus and Fronto.⁸¹ A letter from Lucius Verus as emperor jokingly feigns outrage at Fronto for coming to the palace but not seeking him out to embrace and talk to him.⁸² Fronto mentions a recent occasion on which Marcus was setting out on a journey with Antoninus Pius, and while the emperor waited in the carriage, Marcus was delayed by ‘a crowd of people greeting you and kissing you’ (te salutantium et exosculantium turba).⁸³ During his consulship, when he was obliged to remain in Rome, Fronto sent his wife, Cratia, to join Marcus and his mother, who were apparently with the court at Pius’ Campanian villa; he imagines Cratia and Domitia Lucilla exchanging many kisses.⁸⁴ Then there are the cases in which Fronto encourages others to engage in displays of affection on his or their own behalf. He asks Marcus to give a kiss to his mother—Domitia Lucilla is probably meant⁸⁵—as he delivers her a letter,⁸⁶ and sends another to Marcus’ daughter, presumably Domitia Faustina.⁸⁷ ⁷⁸ M. Aur. ap. Fronto, Ad M. Caes. 5.68 (vdH² p. 84). ⁷⁹ Fronto, Ad M. Caes. 2.4.1 (vdH² p. 25). ⁸⁰ Fronto, Ad Ant. Pium 5 (vdH² p. 164), De nep. Am. 2.8 (vdH² p. 238), with Champlin 1980: 95–6. ⁸¹ For expected and imagined physical expressions of love between Fronto and Marcus: M. Aur. ap. Fronto, Ad M. Caes. 2.6.3 (vdH² p. 27); Fronto, Ad M. Caes. 2.13.2 (vdH² p. 32), 3.14.3 (vdH² p. 46), 4.12.4 (vdH² p. 66). ⁸² Fronto, Ad Verum Imp. 1.11.1 (vdH² p. 114). For other references to physical affection between Verus and Fronto, see Ad Verum Imp. 1.4 (vdH² p. 109), 1.7.1 (vdH² p. 111–12), 1.12.3 (vdH² p. 116). ⁸³ Fronto, Ad M. Caes. 3.14.3 (vdH² p. 46). ⁸⁴ Fronto, Ad M. Caes. 2.13.1 (vdH² p. 32). ⁸⁵ van den Hout 1999: 56. ⁸⁶ Fronto, Ad M. Caes. 2.2.9 (vdH² p. 21). ⁸⁷ Fronto, Ad M. Caes. 5.48.2 (vdH² p. 78); cf. van den Hout 1999: 209; PIR² D 177. See too Ad M. Caes. 4.12.7 (vdH² p. 67).

      ‘   ’ ? 131 To say that the Antonine court had a shared ideology that prized amor and its physical expression is not to imply that everyone at court spoke about and demonstrated their feelings for everyone else with equal intensity. In any community—and especially in one as steeply hierarchical as a court— relationships vary greatly in their degrees of intimacy. A delicate ‘grammar of emotions’ determines the intensity of emotional performance appropriate to a given relationship. Rosenwein’s category of the ‘emotional community’ does not imply emotional uniformity in intra-group relationships, but rather a shared set of values and rules, and a common basic vocabulary to discuss feelings that are prized or avoided. In the case of Marcus and Fronto, the intimacy of their relationship naturally meant that the basic language of amor was intensified with the full battery of metaphors, similes, and other rhetorical devices;⁸⁸ the physical performance of emotion was also frequent, to judge from the letters. To take another example, the relationship between Fronto’s wife, Cratia, and Domitia Lucilla was evidently close,⁸⁹ so the reference to the pair exchanging frequent kisses is unsurprising. On the other hand, less intimate court relationships evidently made less intense protestations of love seem appropriate. The two letters from Antoninus Pius to Fronto are perfectly pleasant, with Fronto being called carissime in both, but there is nothing like the intensity of language used by Marcus and Fronto for each other, or indeed for certain other protagonists in the Frontonian corpus.⁹⁰ This does not mean that Pius was a kind of emotional princeps clausus who articulated and performed emotion differently from that of his courtiers: his statement that he would rather live on the grim island of Gyara with Faustina than in the palace without her expresses great depth of feeling;⁹¹ and Fronto also assumes that Marcus will kiss and embrace his adoptive father on occasion.⁹² To take a second example from an earlier court, while the letters exchanged between the Pliny the Younger and Trajan are warm, they contain nothing like the intensity that one sees in letters between Fronto and Marcus Aurelius or Lucius Verus. Nor do Pliny’s letters suggest that Trajan directed to Pliny anything resembling the intense physical demonstration of feeling for the unnamed friend and praetorian prefect described in the Panegyricus.⁹³ For Trajan, Pliny was always just carissime⁹⁴ and there is no sign of imperial kisses beyond the routine—as was appropriate for a member of the ‘outer court’. Another point to stress is that the language of love and lavish demonstrations of it by kisses and embraces was not unique to the court. The letters of Pliny are filled

⁸⁸ On metaphor and expressions of love in the correspondence of Marcus and Fronto, see Taoka 2013b. ⁸⁹ On this relationship, see Richlin 2011: 175–7. ⁹⁰ Antoninus Pius ap. Fronto, Ad Ant. Pium 2 (vdH² pp. 161–2), 6 (vdH² p. 165). ⁹¹ Antoninus Pius ap. Fronto, Ad Ant. Pium 2.2 (vdH² p. 162). ⁹² Fronto, Ad M. Caes. 2.4.2 (vdH² p. 25). ⁹³ Plin. Pan. 86.1–4. ⁹⁴ Plin. Ep. 10.16, 10.20, 10.44, 10.60, 10.62, 10.82, 10.91, 10.115, 10.121.

132

   

with declarations of love not just for people who were directly associated with the court⁹⁵ but also for other friends and acquaintances.⁹⁶ Fronto uses love language in relation to people with no known court connections, such as Licinius Montanus, a man from Cirta on whose behalf Fronto writes a commendatio letter to the governor of Africa.⁹⁷ It is well documented that in Roman aristocratic culture more broadly, displays of physical affection between family members and friends were routine.⁹⁸ But this was probably precisely the point: to ostentatiously avoid the dysfunctional emotional atmosphere that was (fairly or unfairly) imagined for the court in the first century, and instead to embrace an emotional language and a performative style that was typical of the aristocracy at large. This was essential to the image of civilitas that Trajan and the Antonines sought to cultivate. To use Rosenwein’s image, then, the smaller emotional circle of the court overlapped quite strongly with the larger circle of the Roman aristocracy. The overlap was not necessarily complete, however, since court life had the potential to accentuate certain emotional problems, and these correspondingly received special emphasis in court discourse. In the Antonine period, just as earlier in the century, invidia seems to have been recognized as a particular challenge at court because of the paramount importance of the favour of the emperor (and members of the imperial family) and the fact that favour was necessarily a finite resource. In one letter, Fronto urges Marcus as Caesar to work to banish invidia within his circle (cohors) of friends—a feeling that Fronto sees as especially likely when Marcus shows some special favour to an individual.⁹⁹ Lucius Verus was evidently aware of this issue as well, since we hear that he admitted Fronto into his bedroom so that he could kiss him in private without inciting anyone’s jealousy (invidia).¹⁰⁰ Fronto must also have impressed on the young Marcus the undesirability of jealousy in a ruler, since the emperor states in the Meditations that from Fronto he learnt about the jealousy characteristic of the tyrant (ἡ τυραννικὴ βασκανία).¹⁰¹

4. The Fourth-Century Court as an Emotional Community In his speech praising Emperor Gratian on the occasion of his consulship of 379, Ausonius discusses the emotional atmosphere of the court (palatium) at Trier. ⁹⁵ Plin. Ep. 1.22 (Titius Aristo, legal advisor to Trajan); 2.1, 6.10 (Verginius Rufus, cos. ter. 97); 3.3, 4.17 (Corellius Rufus, cos. suff. 78, influential in the court of Nerva). ⁹⁶ Plin. Ep. 1.5, 2.13 (Voconius Romanus, an equestrian from Spain); 3.15 (the otherwise unknown poet Silius Proculus); 4.4 (Varisidius Nepos, an aspiring military tribune). ⁹⁷ Fronto, Ad amicos 1.3.5 (vdH² p. 175). ⁹⁸ Kühn 1987: 263; Timpanaro 1987; Paterson 2007: 147; Thraede 2008: cols. 545–52. ⁹⁹ Fronto, Ad M. Caes. 4.1.3 (vdH² p. 54). ¹⁰⁰ Fronto, Ad Verum Imp. 1.7.1 (vdH² p. 111–23); for the letter generally, see Timpanaro 1987. See too Fronto, Ad M. Caes. 1.3.2 (vdH² p. 3), in which Domitia Lucilla jokes about her invidia towards Fronto. ¹⁰¹ M. Aur. Med. 1.11.

      ‘   ’ ? 133 Under Valentinian I, he says, the court was terrifying (terribile), but now under Gratian, it has become lovable (amabile).¹⁰² The statement illustrates the fact that in the fourth century, as in the second, the language of fear and love was fundamental when it came to stigmatizing or praising the emotional atmospheres of courts. The early years of Gratian’s reign saw an attempt to distinguish his regime from that of his father, especially given the string of trials that had terrorized the senatorial elite towards the end of Valentinian’s life.¹⁰³ Ausonius therefore furthers this agenda by claiming that fear was the dominant emotion at Valentinian’s court.¹⁰⁴ At the same time, Ausonius grapples with an inconvenient fact: that late antique court spaces and the ceremonies that transpired within them were liable to inspire fear regardless of the emperor involved, since they emphasized the gulf between emperor and subject. Through the alchemy of panegyric, this awkward reality is transmuted into praise. Ausonius claims that the setting for the speech— likely the imperial audience hall (the Basilica Constantiniana) at Trier¹⁰⁵—under Gratian has become a place of ‘tranquil dread and reverent terror’ (horrore tranquillo et pavore venerabili) of the sort one might feel in an oracle shrine.¹⁰⁶ Themistius deals with a similar problem in his speech to Theodosius I from 381 by likewise assimilating emotions felt in court space with those experienced in sacred space: Nobody enters the palace with their heart throbbing, their teeth chattering, or ‘pallid because of fear’ but with cheerful and lofty thoughts, as though entering the sanctuary of a temple.¹⁰⁷

This results from the serenity of Theodosius’ face, so the orator claims, and the fact that he (unlike Valens) has an aversion to imposing death sentences.¹⁰⁸ That fear at court was considered to reflect badly on the late antique emperor is also quite clear from how frequently the emotion appears in the courts of those emperors of whom Ammianus disapproves. As well as ascribing terror and

¹⁰² Aus. Grat. Act. 1. ¹⁰³ Matthews 1975: 65; McEvoy 2013: 60–1. ¹⁰⁴ On comparison with previous rulers as a basic feature of imperial panegyric, see, for example, Rees 2018. ¹⁰⁵ On the audience hall at Trier and its atmosphere, see Chapter 3 by Verena Jaeschke and Caillan Davenport in this volume. ¹⁰⁶ Aus. Grat. Act. 1. To be sure, Gratian was not present on the occasion of the speech, but in Sirmium: Aus. Grat. Act. 11, with McEvoy 2013: 80; McEvoy 2016b: 165–7. However, Ausonius is not speaking about a particular occasion, but generally about the supposed emotional atmosphere of the audience hall, which was used by Gratian as a court space when in Trier. Most crucial for our purposes are the lexical choices that the courtier Ausonius makes in discussing emotion in this court space. ¹⁰⁷ Them. Or. 15.190c: εἰσφοιτᾷ δὲ εἰς τὰ βασίλεια πατάσσων οὐδεὶς τὴν καρδίαν οὐδὲ τοὺς ὀδόντας ἀράσσων οὐδὲ ῾χλωρὸς ὑπὸ δέους᾽, ἀλλ᾽ εὐέλπιδι τῇ γνώμῃ καὶ ἀνεστηκυίᾳ ὥσπερ εἰς τὰ ἄσυλα τῶν ἱερῶν; cf. Hom. Il. 10.376, 15.4. ¹⁰⁸ Them. Or. 15.190b–c. There is an interesting contrast in Them. Or. 9.125c, where the orator, when praising Gratian, implies that Celts and Germans who encounter him enthroned in his audience hall feel fear. Terrorizing foreigners when they visited court was evidently not negative.

134

   

anxiety to individual emperors (and Caesars),¹⁰⁹ and to courtiers¹¹⁰ in particular contexts, the historian presents it as an emotion shared by groups of courtiers, such as when news of an impending Persian attack in 359 causes great fear in all.¹¹¹ There is a clear sense that fear in an emperor is unbecoming.¹¹² Part of the concern is that it can cause emperors to take inappropriately harsh repressive measures when conspiracies are alleged.¹¹³ Valens also attracts the explicit condemnation of Ammianus for reacting with fear to the attempted usurpation by Procopius, which rendered the emperor ineffectual at dealing with the crisis.¹¹⁴ On that occasion, Valens’ intimates (proximi) managed to galvanize him into action; more commonly, though, Ammianus imagines devious courtiers manipulating emperors’ fear for their own gain, and clearly disapproves of this.¹¹⁵ There are also similarities between the second and the fourth centuries in relation to normative evaluations of invidia at court. Jealousy is part of the bleak picture Julian paints of the court of Constantius II in the Letter to the Athenians: he reports that, when he first came to the court of Constantius II, a group of unnamed courtiers initially mocked him, then grew suspicious of him, and finally felt jealousy (φθόνος) towards him.¹¹⁶ He also frowned upon jealousy in an emperor, to judge from his charges that Constantius II began to feel jealous (φθονεῖν) of Gallus as soon as he came to court, and that he perhaps later felt envy (ζηλοτυπία) regarding Julian’s own military successes, leaving him open to be persuaded by court sycophants to strip the Caesar of his Gallic command.¹¹⁷ The (equally prejudiced) account that Ammianus gives of the court of Constantius II has the same underlying normative assumptions. For the historian, invidia is an emotion shared by groups of deplorable courtiers: in one of his memorable animal metaphors, he reports that at the court of Constantius II, invidia kept barking at Arbetio;¹¹⁸ members of the comitatus are driven by invidia to denigrate Julian’s military achievements;¹¹⁹ and eunuch chamberlains driven by invidia try to turn the emperor against Ursicinus.¹²⁰ Along with fear, jealousy is an emotion that Constantius II’s unscrupulous courtiers would stir up in the emperor;¹²¹ it is also an emotion that the historian attributes to Valentinian I and Valens.¹²²

¹⁰⁹ Amm. 14.10.2 (Constantius II), 14.11.7–9 (Gallus), 14.11.21–2 (Gallus), 16.8.10 (Constantius II), 20.8.2 (Julian), 22.3.9 (Julian), 26.7.13 (Valens), 30.8.11 (Valentinian I), 31.10.8 (Valens). See too Jul. Ep. ad Ath. 273 D (Constantius II). ¹¹⁰ Amm. 14.9.1 (Ursicinus), 15.8.3 (Eusebia), 20.9.2 (envoys of Julian to Constantius II). ¹¹¹ Amm. 18.4.2. ¹¹² Cf. Pan. Lat. 3(11).26.3. ¹¹³ Amm. 15.3.9. Note too Them. Or. 11.153c, where Valens is praised for being unperturbed by false allegations that a certain (unnamed) associate was plotting against him; cf. n. 41 above. ¹¹⁴ Amm. 26.7.13. ¹¹⁵ Amm. 14.11.3–4, 15.3.9, 16.8.10, 18.4.2, cf. 30.8.12. ¹¹⁶ Jul. Ep. ad Ath. 274 D. ¹¹⁷ Jul. Ep. ad Ath. 272 A, 282 C. Note too Ausonius’ attack on the invidia of Domitian (Grat. Act. 6). ¹¹⁸ Amm. 16.6.1. ¹¹⁹ Amm. 17.11.1–2. ¹²⁰ Amm. 18.4.4. ¹²¹ Amm. 14.11.3, cf. 26.4.4 (rumours that Valentinian I and Valens sought to stir up invidia against Julian and his supporters). ¹²² Amm. 30.8.10, 31.12.1.

      ‘   ’ ? 135 The more general suspicion of feigned emotion at court is also visible in the fourth century. Most explicit here is Julian’s letter of 362 to St Basil, an acquaintance from his student days, in which the emperor affirms his friendship, and declares: We interact with each other without the play-acting of the court, which is the only thing you have hitherto experienced, I think—a play-acting according to which people at court, while singing praises, hate with a hatred greater than what they feel for their worst enemies. But we, although we refute and rebuke one another with appropriate freedom whenever necessary, love each other no less than the greatest friends.¹²³

Also firmly condemned are emperors and courtiers who would counterfeit or distort positive emotions to cloak more sinister agendas. This is clear in the panegyric for Julian by Claudius Mamertinus, who was comes sacrarum largitionum and praetorian prefect in 361, and was entering the consulship when he gave the speech in 362. Mamertinus praises the sincerity that Julian shows in his friendships. The emperor does not participate in insincere flattery, or (unlike some emperors) engage in ‘the cheerful frightfulness, or the guffawing savagery, by which ingrained cruelty was concealed with a charade of joy’ (hilarem diritatem cachinnantemque saevitiam, a quibus ingenita crudelitas figmento laetitiae tegebatur).¹²⁴ While both the ‘palette’ of emotions ascribed to members of the fourth-century court and their normative evaluation are generally similar to what we can see in the second-century sources, when it comes to anger, there are subtle differences. Just as in the second-century reports discussed above, so too in the fourth-century texts, emperors experience wrath, being sometimes goaded into anger by devious courtiers.¹²⁵ But in Ammianus, fourth-century emperors fly into rages much more frequently than in Tacitus or Suetonius,¹²⁶ as do prominent courtiers.¹²⁷ Outbreaks of ‘courtly theatrical rage’ were a distinctive feature of the period.¹²⁸

¹²³ Jul. Ep. 26.381 B–C: σύνεσμεν γὰρ ἀλλήλοις οὐ μετὰ τῆς αὐλικῆς ὑποκρίσεως, ἧς μόνης οἶμαί σε μέχρι τοῦ δεῦρο πεπειρᾶσθαι, καθ᾿ ἣν ἐπαινοῦντες μισοῦσι τηλικοῦτον μῖσος ἡλίκον οὐδὲ τοὺς πολεμιωτάτους, ἀλλὰ μετὰ τῆς προσηκούσης ἀλλήλοις ἐλευθερίας ἐξελέγχοντές τε ὅταν δέῃ καὶ ἐπιτιμῶντες οὐκ ἔλαττον φιλοῦμεν ἀλλήλους τῶν σφόδρα ἑταίρων. ¹²⁴ Pan. Lat. 3(11).26.2; cf. Amm. 14.7.9, 14.11.6, 16.10.18, 22.14.4. On the composition of Julian’s court community, see Chapter 7 by Kevin Feeney. ¹²⁵ For emotional manipulation and anger, see Amm. 14.1.10, 14.5.4, 30.5.10. ¹²⁶ Diocletian: Amm. 14.11.10. Constantinus II: Amm. 14.5.4, 14.11.13, 14.11.23, 19.12.5, 20.2.5, 20.9.2, 20.9.3, 21.16.9. Gallus: Amm. 14.1.10, 14.7.1, 14.7.13, 14.9.8. Julian: Amm. 22.14.2–3 (suppressed anger), 23.2.4, 24.5.5, 24.5.7, 24.5.10. Valentinian I: Amm. 27.7.4, 27.7.7, 28.1.11, 28.1.23, 29.3.2, 30.5.10, 30.6.2, 30.8.1, 30.8.12. Valens: Amm. 29.1.18, 29.1.20, 29.1.27, 29.2.18, 31.14.6. ¹²⁷ Amm. 14.7.11 (Domitianus, pr. pr.), 19.12.7 (Paulus ‘Catena’), 26.5.7 (Ursatius), 29.1.5 (Fortunatianus, comes rei privatae). ¹²⁸ Malone 2010: 76–7.

136

   

However, norms of anger control also persisted in court circles in the fourth century.¹²⁹ This is clear from several explicit condemnations in Ammianus,¹³⁰ from Themistius’ approval of rulers keeping their rage in check,¹³¹ and from Julian’s special pleading to justify an outbreak of wrath by his brother,¹³² Gallus. One might be tempted to suspect that reports of raging emperors are mere inventions of a hostile tradition. But it is hardly likely that Julian would invent an episode of Gallus flying into a rage, call it unbefitting of a ruler (ἥκιστα βασιλικῶς), and then try to rationalize it. It is also striking that in his panegyric for Constantius, he admits that the emperor felt anger, but tries to minimize it by saying he was merciful when angry.¹³³ Themistius engages in similar rationalizations. He presents imperial rage as acceptable if directed at deserving targets such as barbarians¹³⁴ and (in the case of Valens) flatterers at court.¹³⁵ Sometimes too, the orator (unconvincingly) claims, Valens’ irascible verbal outbursts benefit those who provoke them, and are not a sign of the emperor’s inner rage.¹³⁶ Such desperate rationalizations would not have been necessary unless some emperors really did suffer from episodes of wrath and everybody around them knew it. Moreover, many of the reported episodes of wrath by emperors and courtiers took place in court spaces where they would have been witnessed by many, making crude invention by our authors less likely. Constantius’ praetorian prefect, Domitianus, is reported as storming out of Gallus’ consistorium in a rage in 354, which in turn prompted a fit of anger from Gallus. When his quaestor suggested overthrowing the statues of Constantius and murdering Domitianus, Gallus then delivered an enraged harangue of his troops, in which the Caesar bared and gnashed his teeth.¹³⁷ Equally spectacular is the performance reported for Valens at the treason trials at Antioch in 371–2, to which Ammianus was an eyewitness.¹³⁸ He recounts that Valens ‘was raging to the utmost point of frenzy, in the fashion of a beast in the arena’ (eruditior ad laedendum, in modum harenariae ferae).¹³⁹ Most famously, in 375 Valentinian I is reported to have flown into a rage when giving an audience to Quadian ambassadors in his ¹²⁹ On the general persistence in Late Antiquity of the ideal of anger control by rulers, see Brown 1992: 58–61; Harris 2001: 257–62; and Malone 2010, who aptly notices the tension between these ideals and bouts of performative anger reported for fourth-century emperors. ¹³⁰ Amm. 27.7.4, 29.1.18, 29.1.20, 29.2.18; cf. 22.14.2–3. ¹³¹ Them. Or. 1.7a–d, 2.30c, 7.87d, 7.93b, 7.98b, 8.110c, 11.144c, 18.217d. ¹³² Jul. Ep. ad Ath. 272 C; note too the moral that Julian draws from the quarrel between Achilles and Agamemnon: that kings should not be carried away by anger (Jul. Or. 2.50 B). He also urges his uncle Julian to avoid anger (Jul. Ep. 29 [ed. Wright]). Themistius (Or. 8.111c–112a, 13.172c–d) likewise condemns the anger of Agamemnon. ¹³³ Jul. Or. 1.48 A. ¹³⁴ Them. Or. 9.121d; cf. Or. 32.360a on the general usefulness of anger in warding off that which is harmful. ¹³⁵ Them. Or. 8.106c. ¹³⁶ Them. Or. 8.111b. ¹³⁷ Amm. 14.7.11–13. ¹³⁸ Amm. 29.1.24. ¹³⁹ Amm. 29.1.27; cf. Vanderspoel (1995: 178) for the suggestion that the emphasis on philanthropy in Them. Or. 11 is an attempt to still Valens’ rage in 371–2.

      ‘   ’ ? 137 consistorium and apparently suffered some kind of stroke or cerebral aneurism, from which he died.¹⁴⁰ A similar dissonance between court norms and emotional performance is visible in relation to the other half of Ausonius’ dichotomy, namely love. In one sense, the cluster of fourth-century texts with which we are concerned shows that love was a highly valued emotion in court relationships. Claudius Mamertinus wants to see his own relationship with Julian as one of amicitia in which amor is the key emotion.¹⁴¹ He also claims that like other emperors, Julian has ‘a retinue that is devoted and loving to him’ (devota et amans sui cohors), but unlike other emperors he does not inspire such feeling by putting star-struck lightweights in his consilium, but has an inner circle of genuine worthies.¹⁴² Love is also valued as a courtly emotion in the extant writings of Julian. The panegyric for Eusebia claims that the empress feels love for her husband, and Constantius returns the feeling.¹⁴³ The consolatio to himself following the recall of his esteemed lieutenant, Saturninus Secundus Salutius, from Gaul talks about the Caesar’s love (φιλία, ἔρως) for his friend (φίλος), and uses several verbs of love to describe the relationship (ἀγαπᾶν, στέργειν).¹⁴⁴ In his letters, Julian regularly declares his love for his correspondents and assumes they love him, as he does in the letter to St Basil, quoted above;¹⁴⁵ and he writes to one correspondent that he loves his very name, just like people madly in love (δυσέρωτες).¹⁴⁶ Themistius likewise idealizes love at court. He emphasizes the love shared between the ideal ruler and his friends,¹⁴⁷ as well as claiming that Valens loves those in his inner circle (οἱ οἰκεῖοι)¹⁴⁸ and praising his brotherly love for Valentinian.¹⁴⁹ In the thirteenth oration, the Erōtikos, delivered before the Roman senate in 376 or 377, Themistius speaks at length of his own love for the young Gratian (who was not present for the speech).¹⁵⁰ Themistius relates how he initially began an ‘amorous journey’ (ἐρωτικὴ πρόοδος)¹⁵¹ by haunting the gymnasia and palaestrae, searching for a youth with a beautiful soul in a beautiful body whom he could love.¹⁵² Unsuccessful there, he went to court, where he met Constantius II and Julian, but their beauty was too overripe for Themistius’ tastes, so did not excite his erotic love (ἔρως).¹⁵³ Finally, he claims, he found what he was

¹⁴⁰ Amm. 30.6.2–6, cf. Zos. 4.17; Soc. 4.31; Soz. 6.36. ¹⁴¹ Pan. Lat. 3(11).18.4. ¹⁴² Pan. Lat. 3(11).25.1–4, cf. 2(12).16.1–2. Note too Ausonius’ claim about the affection (caritas) Gratian feels for his friends (Grat. Act. 17). ¹⁴³ Jul. Or. 3.106 A, 112 D, 114 B–C. ¹⁴⁴ Jul. Or. 8.243 C–D, 247 C. For the nomenclature of Salutius, wrongly called Sal(l)ustius in some MSS, see PLRE I Secundus 3. ¹⁴⁵ Jul. Ep. 3.441 D, 16, 20.452 A, 26.381 C, 65.442 D, cf. 50.446 B (ed. Wright). ¹⁴⁶ Jul. Ep. 5.425 B (ed. Wright). ¹⁴⁷ Them. Or. 1.17b. ¹⁴⁸ Them. Or. 8.111b. ¹⁴⁹ Them. Or. 6.74d, 81d–82a, 9.127a. ¹⁵⁰ On the speech and its circumstances, see Vanderspoel 1995: 179–84; Konstan 2014: 128–34. For Gratian’s absence, see especially Them. Or. 13.179b, d, with Barnes 1975: 329. ¹⁵¹ Them. Or. 13.177d, cf. 177c. ¹⁵² Them. Or. 13.163d–165a. ¹⁵³ Them. Or. 13.165b–c.

138

   

looking for in Gratian, who united physical beauty and moral excellence, thus exciting Themistius’ erotic love.¹⁵⁴ His many attractions include his eyes, which are full of cheerfulness (εὐθυμία); indeed, lamentations haven been expelled from his residence.¹⁵⁵ Themistius casts himself in the role of the adult lover (ἐραστής) smitten by his beloved boy, whom he calls τὰ παιδικά and ὁ ἐρώμενος,¹⁵⁶ thus deploying the standard language of Greek pederasty. The speech is a forest of literary allusions, especially to Homer and Hesiod, and to the Platonic dialogues that consider love. It could be seen as having a ludic element—transgressive as learned jokes about pederasty may seem to modern readers. It certainly cannot be taken as evidence of a genuine emotional connection between Themistius and Gratian; indeed, there is no firm evidence that the pair had ever met.¹⁵⁷ In any case, the oration shows how the relationship between an emperor and a courtier in the fourth century could still be negotiated using the language of erotic love, just as it could in the second century, to judge from the letters of Marcus and Fronto (which the speech obviously calls to mind).¹⁵⁸ However, while the rhetoric of love and affection in fourth-century court circles echoes what we have seen in the second century, a change is detectable in relation to the performance of affection by the emperor. This is visible in a series of anecdotes about Julian and in Ammianus’ reaction to them. One of these establishes the norm for the fourth century. In 361, the praetorian prefect of the Gauls, Nebridius, who had stayed loyal to Constantius when Julian was acclaimed Augustus, came to Julian as a suppliant, and asked to touch his hand. This Julian refused, asking rhetorically: ‘What special act will be reserved for my friends, if you touch my hand?’ (Ecquid . . . praecipuum amicis servabitur, si tu manum tetigeris meam?).¹⁵⁹ This the historian reports without comment.¹⁶⁰ Quite different is a subsequent incident when Julian, holding court in the senate house at Constantinople, saw the philosopher Maximus, ‘jumped up indecorously’ (exiluit indecore), ran to his visitor, and kissed him. In spite of his general approval of Julian, Ammianus offers extended critical commentary of the emperor’s behaviour, which he sees as inappropriate ostentation and an attempt to burnish his reputation as a philosopher.¹⁶¹ Another performance of affection by Julian is narrated by Claudius Mamertinus: at dawn on the day of his oration, Mamertinus ¹⁵⁴ Them. Or. 13.165c–166c, 168c–169a, 170d–171a. ¹⁵⁵ Them. Or. 13.173a, 175c. ¹⁵⁶ Them. Or. 13.166c, 168c, 171a–b, 175d, 179a, 180a. On the ‘fully erotic’ resonances of this language, see Konstan 2014: 134. ¹⁵⁷ Vanderspoel 1995: 179; cf. Konstan 2014: 133. Note too the fact that occasionally, Themistius uses similar language to refer to other court figures: he calls a philanthropic ruler the ἐραστής of his friends (Or. 1.17b) and speaks of his ἔρως for Valens (Or. 13.177b). ¹⁵⁸ Konstan 2014: 133. ¹⁵⁹ Amm. 21.5.12. ¹⁶⁰ Note too the astonishment of the armies when the usurper Procopius allows a soldier to kiss his hand: Amm. 26.7.15. ¹⁶¹ Amm. 22.7.3–4. Note too 22.9.13, where Julian kisses Celsus, governor of Cilicia, albeit at the Cilician Gates rather than in a court context; Ammianus nevertheless hastens to add that the emperor had known him since his student days.

      ‘   ’ ? 139 and his co-consul had visited the palace for Julian’s salutatio; the emperor had jumped up to meet them and had kissed Mamertinus on the mouth. In the hands of the panegyricist, the incident is given a positive interpretation—the crowd at the salutatio goes wild with joy. But Mamertinus’ elaborate treatment of the episode makes it clear that was not a routine greeting, as it would have been in the second century, and that rationalization was required.¹⁶² When it came to the performance of both love and anger, therefore, the court in the period from Constantius II to Theodosius I did not have the same strong normative consensus that we have observed in the second century. Does this make it impossible to assert that the fourth-century court was an ‘emotional community’? In a study on emotions in nineteenth-century Italy, Mark Seymour has sought to add nuance to Rosenwein’s concept of ‘emotional communities’ by suggesting that these need not be perfectly uniform, but ‘are likely to be shot through with a wide range of potential fault lines’. He suggests that in the case of nineteenth-century Italy, these fault lines might be the product of variables including ‘gender, class, profession, and geographical area’, and also fluid factors like individuals’ propensity to certain emotions. To try to conceptualize the fact that emotional norms and expressive styles can sometimes be in tension within an otherwise coherent group, Seymour suggests a revised category: the ‘emotional arena’.¹⁶³ Such a category would seem to better capture the more complex situation of the court from Constantius II to Theodosius I. In this period, the court lacked the geographical and social uniformity of the second-century court, which centred on the palace at Rome and a handful of imperial villas in central Italy, and in which the Roman senatorial aristocracy was still the dominant element. The courts of the middle decades of the fourth century were highly mobile, as emperors moved between palaces in multiple imperial cities, and further mobility was compelled by the military needs of civil and foreign wars. The court was in Rome very rarely in this period, and men of the city’s ancient senatorial families, although not entirely absent, formed a minority within the courtly elite.¹⁶⁴ Instead, courtiers came from a variety of backgrounds and geographical locations. Ausonius was a schoolmaster from Bordeaux, whose father was a physician and whose mother apparently was from an impoverished Gallic noble family.¹⁶⁵ Themistius came from a Paphlagonian family which was not exactly wealthy, but which was able to give him a thorough classical education.¹⁶⁶ Ammianus had a military career, and his appointment as a protector domesticus at an early age suggests a family from an eastern city (perhaps Antioch) distinguished by military (or possibly administrative) service.¹⁶⁷ Other upwardly mobile courtiers had less

¹⁶² Pan. Lat. 3(11).28–30. ¹⁶³ Seymour 2012: 179. ¹⁶⁴ Matthews 1975: 16–17; Schlinkert 1998: 143, 152–5; Smith 2007: 179–86. ¹⁶⁵ Sivan 1993: 49–59. ¹⁶⁶ Vanderspoel 1995: 31–49. ¹⁶⁷ Kelly 2008: 104–32.

140

   

exalted family backgrounds, having worked their way up through the ranks of the army or the civil service.¹⁶⁸ Emperors too came from diverse backgrounds.¹⁶⁹ Constantius II and Julian were from a family that was originally Illyrian, but had somewhat peripatetic upbringings and spent much of their adult lives on campaign.¹⁷⁰ Valentinian I and his brother Valens were military men of Pannonian origin. Gratian, although obviously sharing these roots, spent his formative years in Trier attaining a high level of literary culture.¹⁷¹ In view of this greater diversity of geographical settings and personnel, one would not expect the court culture of the period from Constantius to Theodosius I to be as unified as was the case with the court from Trajan to Marcus Aurelius. One would expect ‘fault lines’ in Seymour’s sense. The dissonances that we have observed in relation to the expression of certain emotions confirm this expectation. Men like Julian and Ammianus, steeped as they were in classical paideia, still subscribed to the same norms of anger control as members of the senatorial aristocracy of the High Empire, with whom they shared a literary culture.¹⁷² For a soldier of Illyrian ancestry like Constantius II, or a military man of Pannonian upbringing like Valentinian I, prescriptions against performative anger in rulers would have been less ingrained. To them, such performances perhaps seemed to be a useful tactic to project authority at court,¹⁷³ and there is unlikely to have been sufficient normative coherence among the courtly elite to ‘domesticate’ such emperors by inculcating ideals of restraint. When it came to demonstrations of love, Julian evidently thought it appropriate to his philosophical persona to engage in traditional kissing. However, to someone like Ammianus, who as a protector domesticus at the court of Constantius II had evidently come to see a greater ceremonial distance between emperor and subject as normal and proper, this seemed inappropriate.

5. Conclusion Was the Roman imperial court an ‘emotional community’ in Rosenwein’s sense? For the period from Trajan to Marcus Aurelius, I think that we can answer in the affirmative. In this period, we can see emperors (Antoninus Pius, Marcus, Verus), members of the imperial family (the young Marcus, Domitia Lucilla), members of the inner court circle (Fronto, Cratia), and more peripheral courtiers like Pliny subscribing to a consistent set of norms that valued love and approved of the ¹⁶⁸ Schlinkert 1998: 152–4; Smith 2007: 180–1. ¹⁶⁹ Cf. Vanderspoel 1995: 14. ¹⁷⁰ Bowersock 1978: 21–32. ¹⁷¹ McEvoy 2013: 106–7. ¹⁷² Although Julian, it would appear, did not always live up to his own ideals of anger restraint: Malone 2010: 72–5, with n. 129 above. ¹⁷³ See Malone 2010: 75–7 on the utility of theatrical outbursts of rage in projecting the emperor’s power and dominance.

      ‘   ’ ? 141 warm expression of love by all. In the writings of three members of the ‘outer court’—Suetonius, Tacitus, and Pliny—we can also see a consistent set of ideas about what the emotional atmosphere at court should not be like. We can sometimes see these same ideas reflected at the heart of the court in the Frontonian corpus—for instance, in concerns about invidia at court. To argue for a consistent set of emotional norms in the court of this period does not amount to a claim that these norms were exclusive to court circles—unique emotional expression and norms are not what Rosenwein has in mind. Indeed, the expressive norms of the Antonine court look quite similar to those of the aristocracy at large, and one suspects that this was deliberate. When we come to our cluster of authors with court connections from the fourth century, we can see some consistencies between authors (and similarities between the second and fourth centuries), especially in relation to the disapproval of fear, anger, and invidia at court, the valorization of amor in court relationships, and the stigmatization of courtiers who manipulate emperors’ emotions. There seem to have been shifts, however, in relation to emperors’ performance of emotion and the evaluation of these performances. Fourth-century emperors indulge in spectacular shows of rage in court spaces that are simply not paralleled for the first and second centuries; at least some of these reports seem credible. This happened in spite of continued ideals of anger control. Likewise, we see Julian warmly expressing love in the manner of a second-century emperor, but being criticized for it. For this reason, Seymour’s notion of an ‘emotional arena’, in which there is some degree of conflict in relation to emotional norms and expressive styles, better describes the court from the late 350s to the 380s.

6 Jurists as Courtiers from Augustus to Justinian Jill Harries

1. Introduction ‘Courtiers are either stupid or hateful’ (aulici vel stulti vel detestabiles).¹ Such was the unfavourable view of courtiers held in the late fourth century by the ingenious author (hereafter Scriptor) of the imperial biographies known collectively as the Historia Augusta, and this view encapsulated the reputational dangers for legal experts seeking a career at court. To maintain a reputation for integrity, ‘men skilled in [the interpretation of] law’ (iuris periti) had to project themselves as impartial, incorruptible—and effective. Failure could invite criticism like that of Juvenal directed as Domitian’s legal advisor and prefect of the city, Pegasus, ‘most excellent and revered exponent of laws’ (optimus atque interpres legum sanctissimus) during whose feeble term of office, ‘justice was disarmed’ (inermi iustitia).² While the stereotypical aulicus played the system, benefited from patronage, subverted his rivals, or flattered the sovereign, the legal advisor, like the ideal philosopher, should speak truth to power, as the quaestor Eupraxius did to the irate Valentinian I in 370, when warning him of the rights of senators under the treason law.³ Court societies were (and are), in Jeremy Paterson’s words, ‘a negotiation between the ruler and the subject’, and the conventions and customs of courtly behaviour and discourse ‘provide an element of reassurance and confidence in a relationship which is inherently unstable and fraught with uncertainty’.⁴ Jurists’ ‘courtly behaviour’ was distinctive in that, through their mastery of a discipline essential for good governance, they could expect to engage with the emperor to some extent on their own terms. While projecting themselves as ‘friends’ (amici) of the emperor, with access to him as legislator or as judge, they also operated ¹ HA Aurel. 43.1. For his possible identity as a lawyer at Rome see Honoré 1987. ² Juv. 4.77–81. Pegasus is shown as a member of Domitian’s corrupt council of advisors, afflicted, like the rest, with the pallor amicitiae (4.75), the fear engendered by the emperor’s ‘friendship’. ³ Cf. Ulpian at Dig. 1.1.1.1 for jurisprudence as the ‘true philosophy’ (vera philosophia). For Eupraxius, see Amm. 28.1.25. ⁴ Paterson 2007: 122–3. For full discussion of the definition of a court, see the Introduction. Jill Harries, Jurists as Courtiers from Augustus to Justinian In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0007

      

143

within a long-standing autonomous and specialized tradition, with its own discourse and ways of thinking. For them, therefore, as for others, who, like Juvenal, were fearful of potential abuse of power by the autocrat, the emperor’s demonstration of respect for the law, and therefore for those who interpreted it, provided the necessary ‘element of reassurance’. The emperor made the law but the law’s content was determined by juristic expertise. Within these limits, lawyers were also creators of law and, from the second century on, they advertised their privileged access to imperial decisionmaking through their citation of imperial rescripts (which they themselves may have drafted). This reflected the increased concentration of legislative power in the emperor’s hands; in particular, the establishment of a fixed text of the Praetorian Edict by Hadrian’s jurists c. 130, transferred to the emperor and his lawyers all power to modify in future the ius honorarium (the law created by the praetor, through his Edict, which we would term one branch of private law).⁵ This increased workload in turn created a hierarchy of courtiers and legal administrators. Many ‘replies’ (responsa) to legal questions, referrals, or appeals did not emanate directly from the emperor or his council (consilium), but were the work of the secretaries for letters (epistulae) or petitions (libelli) and their staffs. The route to the top became more complex and more challenging. By the fourth century, with a few exceptions, the individual identities of the imperial jurists had been entirely subsumed by the bureaucracy. The emperors’ lawyers inevitably subscribed to the prevailing culture of competition between courtiers. Debates among imperial advisors, who also acted as assessors at trials before the emperor, were about not only legal principle but also the exertion of visible influence, enhancing reputations within the court and beyond; Paulus, the legal advisor to the Severan emperors, advertised his privileged access to inside information by publishing a six-book work entitled Imperiales sententiae in cognitionibus prolatarum (Imperial Decisions Issued in Judicial Proceedings).⁶ But access came at a price. Initially, the involvement of legal experts with the court functioned on an ad hoc basis, as a matter of negotiation. As the administration evolved and both the apparatus and the culture of autocracy strengthened, the emperors’ jurists cooperated, because it was expected of them, and over time became assimilated into the administrative machine. As late as the Severans, in the early third century, jurists close to the centre of power had not forfeited their independent identities, yet the violent deaths of Papinian and Ulpian, perhaps the two most influential of all legal writers, showed the risks as well as the rewards of incorporation.

⁵ This refers to modification at the centre. The Roman Empire was multi-legal, and law in the provinces was affected by judicial decisions and the creation of local precedents. ⁶ Lenel, Pal. I, 1111–2 (fragments 877–80), a rework of Paul’s earlier Decreta (Lenel, Pal. I, 959–65, fragments 56–80). See also Honoré 1994: 15–21.

144

   

2. Negotiation: The Early Principate Under Augustus, lawyers, like senators and other fellow members of the elite, could choose the extent of their involvement with the princeps. Some individual jurists, such as Antistius Labeo, demonstrated that a reputation for independence and the preservation of distance from the princeps’ entourage could enhance a lawyer’s reputation; Labeo’s father, Pacuvius, had supported Brutus and Cassius, and the son had been held aloof—for which reason (or so Tacitus alleged) his career advanced no further than the praetorship. By contrast, Labeo’s main rival, Ateius Capito (consul in 5) was alleged to have achieved both access and office by flattery; although this perception damaged his posthumous reputation in Tacitus’ eyes, his legal opinions were treated with respect and he was later credited (erroneously) with being the founder of one of the two law schools at Rome. And, like Labeo, C. Cassius Longinus, suffect consul in 30, governor of Syria under Claudius in the 40s and doyen of the schola Cassiana, exploited his family connection to the assassin of Caesar, his ‘Republican’ identity reinforced in Tacitus’ narrative by his well-advertised adherence to the austere—even harsh— values of the maiores.⁷ Courtiers, including lawyers, influenced which cases were brought to the emperor’s attention. In the sixth century, the lawyers of the Byzantine autocrat Justinian assumed, retrospectively, that ‘favourites’ always affected the legal agenda. Thus Augustus, for a variety of motives, allegedly insisted that the consuls intervene to enforce payment of legacies left in trust. This generated further discussion when another courtier, Lucius Lentulus, died in Africa, having added codicils to his will at a time when codicils, like trusts, were not legally enforceable. He begged Augustus to do something for him ‘as a trust’, Augustus did so, and others followed the precedent of voluntarily paying legacies, which could not have been claimed under formal law. At this point, Augustus summoned an ad hoc advisory consilium of jurists, including Trebatius, previously associated with Caesar in Gaul and with Cicero, to ascertain whether the practice was consistent with received law. Trebatius was happy to provide the required reassurance, observing ‘how convenient and even necessary the practice was for citizens’ (utilissimum et necessarium hoc civibus esse).⁸ Although the trusts and codicils narrative, being late, may also be apocryphal, it accurately reflects the ad hoc nature of juristic involvement in early imperial decision-making. While power became increasingly concentrated in Augustus’ household, the familia Caesaris, and in the hands of non-senatorial administrators ⁷ Tac. Ann. 3.70, 3.75. On Labeo’s conservatism, see Gell. NA 13.12.1. For the law schools, see Pomponius’ Enchiridion at Dig. 1.2.2.47. On Cassius’ advocacy of the harsh provisions of the SC Silianianum in 61, see Tac. Ann. 14.43–4. ⁸ Just. Inst. 2.23.1 (consular intervention); 2.25.pr. (Trebatius). Trebatius was also to advise on the divorce of Augustus’ friend Maecenas from his wife Terentia (Dig. 24.1.64).

      

145

appointed by him, Augustus, not being officially a monarch, had no ‘court’ as such.⁹ The senate, therefore, remained a reservoir of legal expertise independent of the emperor, while Trebatius, and others like him drawn from the non-senatorial elites of Rome and Italy, continued the practice inherited from the late Republic of attaching themselves to the retinue of the dynast. But while assisting the powerful, Trebatius and his contemporary lawyers on the make were not integrated into anything recognizable as a ‘court society’; they did not reside in the same buildings as their emperor and the focus of their everyday activities lay elsewhere, in private consultancy and teaching.

3. Consilium Principis Late in his reign, Augustus established an inner council of advisors, the membership of which was fluid and who were known, with others, as the emperor’s friends (amici).¹⁰ As the consilium evolved, it acquired more formal functions and a membership, which combined designated high officials with trusted advisors.¹¹ Members of the consilium assisted the emperor at trials as assessors; they debated responses to legal questions raised by petitioners (whose own access was doubtless facilitated by interested parties); and they made other policy decisions. Jurists’ mastery of the law was both a support and a potential threat to an emperor’s personal power. An emperor who showed respect for the law—and therefore lawyers—provided the necessary ‘element of reassurance’ to the anxious members of the elite.¹² Because the support of jurists, as guardians of the law, was conditional on the emperor acknowledging the importance of their discipline, emperors represented as ‘good’ by their elite historians ran their judicial sessions with due respect for the opinions of their legal counsellors. According to the Scriptor, Marcus Aurelius declared that his own opinion counted as merely one vote and Septimius Severus was praised by the senatorial historian Cassius Dio, a member of his consilium, for his practice of giving ‘to his fellow judges’ (συνδικάζουσιν) full freedom of speech (παρρησία) in stating their opinions.¹³ While most narratives of imperial legal hearings focus on the business at hand, a non-jurist demonstrated, for the benefit of the less favoured, that acting as an assessor and member of the consilium also conferred privileged social access. Pliny the Younger knew law, but handled it in the free-wheeling manner of ⁹ On this point, my view differs from that of the Editors as expressed in the Introduction. ¹⁰ For a basic account, see Crook 1955. ¹¹ See Suet. Tib. 55 (council of twenty appointed by Tiberius); Suet. Tit. 7 (council formed from amici and much consulted); HA Hadr. 8.9 and 18.1 (Hadrian praised for appointing senators, equites, and legal experts as assessors). For Hadrian as the ideal emperor-as-judge, see Tuori 2016: 196–240. ¹² Contrast Caligula (Suet. Calig. 34.2), who alone was permitted to give legal opinions. ¹³ HA Marc. 22.4, cf. Dig. 28.4.3 from Marcellus, Dig. 29 (Marcus); Dio 77(76).17.1–2 (freedom of speech).

146

   

an advocate, not a jurist. In a series of trials at Centum Cellae (modern Civitavecchia), Pliny assisted Trajan, firstly in clearing a citizen of Ephesus of a false charge; secondly in a case of adultery, in which Augustus’ Julian law on the suppression of adultery (lex Julia de adulteriis coercendis) was, despite resistance from the deceived husband, fully enforced; and thirdly in a dispute concerning a partially forged will, further complicated by the reluctance of some of the heirexecutors (who had involved the emperor in the first place) to pursue the case.¹⁴ Each hearing lasted one day and at the end of each, the advisors were invited to a simple dinner with the emperor and spent the evening listening to recitations or in civilized conversation. Finally, each advisor received a parting gift. Pliny’s letter made the most of it: ‘I took great pleasure in the importance of the cases, the honour of being an assessor and the charm of our social life’ (sed mihi ut gravitas cognitionum, consilii honor, suavitas simplicitasque convictus, ita locus ipse periucundus fuit). Unlike Pliny, the austere legal specialists avoided references to feasting, however moderate, and other social activities. Although the assertion by the Scriptor that Ulpian enjoyed cosy social evenings with Severus Alexander can be discounted as evidence,¹⁵ such jurists as Papinian, Paulus, and Ulpian under Septimius Severus and his successors would have benefited, discreetly, from social as well as formal professional contact. The jurists’ narratives of themselves, however, exploited the austerity expected of their discipline to prove that they were not, in a negative sense, courtiers. Their justification for their status as amici lay in their access to a specialized body of knowledge, not in their talents (if any) as schemers for personal advancement. The point was reinforced by an anecdote told, by an academic lawyer, of Emperor Hadrian who, when petitioned by a group of men of praetorian rank for the privilege of giving legal opinions backed by imperial authority, stated that such a grant should depend not on rank or petition, but on proven expertise.¹⁶ It helped the prestige of lawyers to be in the know. When Marcus Aurelius tried a case concerning the validity of a will, in which the names of the heredes (heirexecutors) had been deleted but the legacies, including one to a certain Leo, left intact, the jurist Ulpius Marcellus advertised his access to inside information by putting on record not only the decision but also the arguments advanced by the advocates of the different parties.¹⁷ The room, he recalled, was crowded and the presence of the treasury advocate (advocatus fisci), Calpurnius Longinus, gave the occasion a political frisson, as, if the will fell, the property would revert to public funds. Calpurnius therefore argued forcefully that it was impossible for a will to stand, if there were no heirs named. After several exchanges between him and Leo’s advocate, involving trenchant assertions of position, rather than refined ¹⁴ Plin. Ep. 6.31. ¹⁵ HA Alex. Sev. 34.6. ¹⁶ Pomponius at Dig. 1.2.2.49. ¹⁷ Dig. 28.4.3.

      

147

legal argument, Marcus’ ‘more humane’ (humaniorem interpretationem) verdict, that only those provisions deleted should be void, vindicated the rights of the beneficiaries over those of the treasury. Despite their professions of altruism, lawyers on the make were not above resorting to flattery.¹⁸ Writing under Domitian or Trajan, the holder of no less than three provincial governorships, Javolenus Priscus, commented that imperial beneficia should be given the most generous possible interpretation, as stemming from the ‘all but divine indulgence’ (a divina scilicet . . . indulgentia) of the princeps.¹⁹ Some decades later, L. Volusius Maecianus’ rise to prominence involved some clever self-promotion. He began his career as an eques and under Antoninus Pius held a series of equestrian administrative jobs, including work in the petitions (a libellis) and finance departments (a censibus), rising to the heights of praefectus annonae, in charge of the grain supply and, a natural progression, prefect of Egypt, a post reserved for the most trusted imperial administrators, in 160/1.²⁰ Under Marcus Aurelius, he was adlected to the senate, was a member of the consilium, and held a suffect consulship. For the benefit of other lawyers, he had established his credentials by a weighty work of sixteen books on trusts (fideicommissa) and a little treatise in Greek on the Rhodian Sea Law. With an eye to the future he also brought himself to imperial attention, writing, for the benefit of the young Marcus Aurelius, a treatise on coinage, specifically the as; this may explain the Scriptor’s assertion, not corroborated elsewhere, that Marcus was Maecianus’ pupil (auditor) in legal studies.²¹ Maecianus’ services were rewarded with advertisement of his merits in an imperial ruling, preserved, not by Maecianus but by Ulpian, on the rights of a grandson to benefit from the estate of a freedman accused by his father and grandfather of a capital offence.²² After some hesitation, Marcus had consulted Maecianus as ‘our friend’ (amicus noster) who, in addition to his ‘long-standing and well-founded skill’ (veterem et bene fundatum peritiam) was now ‘with extreme care’ (anxie diligens) to revisit the original decision, which rejected the grandson’s request. Having done so, Maecianus stuck by his original opinion and said so ‘in our presence’ (coram nobis). Still not satisfied, Marcus held further direct consultations with Maecianus, again by name, and ‘other friends’ (aliis amicis). This resulted in the overturning of the original decision, which they agreed was not supported by the wording of any relevant source of law, yet the

¹⁸ On the flattery of courtiers, see the discussion of Fronto by Christoph Michels in Chapter 1. ¹⁹ Dig. 1.4.3. Javolenus was governor of Germania Superior, Syria, and Africa Proconsularis. ²⁰ HA Marc. 25.4 refers to the murder of one ‘Maecianus’ as prefect of Egypt fifteen years later, in 175, because of his support of the revolt of Avidius Cassius; either this is another Maecianus, or the HA used the jurist’s prefecture of 160/1 as a basis for a further fiction. For his career, see Millar 1977: 103, and for the prosopographical conundrum, see Jarvis 2015. ²¹ HA Marc. 3.6. ²² Dig. 37.14.17, on which see Honoré 1994: 13–14.

148

   

fulsome praise accorded to Maecianus ensured that his reputation would be in no way damaged by the emperor’s rejection of his learned opinion.

4. Cooperation: Papinian and Ulpian As the imperial administration developed a separate identity in the course of the second century, lawyers were employed in the new bureaux for the drafting of imperial letters (ab epistulis, which could include legal content) and the answering of petitions (a libellis).²³ For those who made it to the top, the temptations of the courtier’s role were more insidious, and the risks more extreme. The accession of Septimius Severus after a series of civil wars (193–7) marked a profound, even sinister, change in the role of lawyers in government. Severus, African by origin and with extensive connections in the Roman East, relied on an inner circle, notably the praetorian prefects, firstly Fulvius Plautianus, not a jurist but allegedly a childhood friend; then, from 205, Papinian, arguably the greatest of all the Roman jurists, who came from a Syrian or African background; and, in a more junior role, Ulpian from Tyre. All were recruited from the countries best known to the emperor. Yet the prefecture would prove fatal for many of its holders. In 185, Commodus promoted then destroyed Perennis; in 205, Plautianus was brought down by Caracalla, Severus’ elder son;²⁴ in 212, Papinian was killed in the course of the struggle for the succession; and in the 220s, Ulpian too, having killed his two colleagues, also fell victim to the politics of the unstable court of Severus Alexander. The fate of Papinian fascinated the legally minded Scriptor.²⁵ Papinian is said to have prospered through his Syrian connection—the Scriptor claims he was related to Severus’ empress Julia Domna—but he had also acquired a reputation independently as a jurist through his public teaching in discussion seminars and his extensive writings, which, like those of most jurists, consisted mainly in the exposition of individual cases. Having administered justice as the civilian arm of the praetorian prefecture down to Severus’ death in 211, Papinian supported the claims of Severus’ younger son, Geta, and then fell victim to the struggle for the succession. From the Scriptor, more was required: he could not be ‘silent on the death of so great a man’ (de tanti viri caedi reticere). Caracalla, consistent with the stereotype of the deceitful tyrant, pretended friendship with Papinian, then, in the emperor’s presence, an assassin killed Papinian with an axe, while Caracalla commented that he should have used a sword. Later, the Scriptor reverted to the ²³ For these positions, and the emergence of the bureaux, see Davenport 2019: 307–10; Davenport and Kelly 2022. ²⁴ Dio 77(76).3–4; the official version, that Plautianus threatened the imperial house, appears at Hdn. 3.11.1–3; 3.12.12. ²⁵ HA M. Ant. 3.2 (friendship); 4.1 (murder); 8.1–9 (variant versions); Geta 6.3.

      

149

atrocity, depicting his hero as conciliatory, principled, and prophetic: Papinian had tried to reconcile the two feuding sons; he had refused to compose a justification for Caracalla’s murder of Geta: ‘it is easier to commit fratricide than justify it’ (non tam facile parricidium excusari posse quam fieri); he would not attack Geta posthumously, counting such an action as a ‘second murder’ (parricidium aliud); and he had foretold Caracalla’s own death at the hands of a future praetorian prefect (Macrinus). Much of this was probably invention and was certainly deployed for literary effect. The murder, in its bloody context, also signalled a change; association with princes was now dangerous for court lawyers too. Domitius Ulpianus, whose writings comprise some forty per cent of Justinian’s Digest, had a still more turbulent court career, the details of which are uncertain. Both his rise and fall resulted from his immersion in the court politics of Elagabalus and Severus Alexander. He may have served Septimius Severus as master of petitions from mid-202 for seven years, and may then, as a favoured courtier, have accompanied his emperor to Britain, where Severus died, at York, in 211. In the view of Honoré, Ulpian survived the purge, which destroyed Papinian, by joining Caracalla in time. An alternative hypothesis is that Ulpian’s relative obscurity protected him from physical harm but not from exclusion from the imperial presence. Exclusion—an early parallel is Augustus’ ‘withdrawal of his friendship’ (renuntio amicitiae)²⁶—meant the end of a career, the loss of office, honours, and prestige, and, in extreme cases, life itself. Out of office for the next five years, down to Caracalla’s assassination in 217, Ulpian, who had hitherto written virtually nothing, devoted himself to writing voluminous treatises on Roman law. His motives were mixed. One, surely, was to keep himself in the public eye—or at least in the eyes of those who mattered. A second, probably, was to provide an education in Roman law for the new citizens created by the Constitutio Antoniniana, which granted citizenship to virtually all the inhabitants of the Roman Empire. This was itself an indirect form of flattery; he was furthering Caracalla’s policy agenda. However, if his intention was to work his way back into Caracalla’s favour and office at his court, he failed. From the accession of Elagabalus in 218, boudoir court politics came to the fore as two imperial mothers, Soaemias, the mother of Elagabalus, and Julia Mamaea, mother of Alexander, jostled to prevail over the other as kingmaker. Ulpian’s exact connection with either is speculative; whatever the intrigues involved in the fall of Elagabalus in February 222, Ulpian profited from them. He is attested as prefect of the grain supply (praefectus annonae), in March 222, a few weeks into the reign of Severus Alexander.²⁷ By the end of the year, he was praetorian prefect.²⁸

²⁶ Ov. Tr. 1.121–40; Tuori 2016: 74–80. ²⁷ CJ 8.37.4. ²⁸ CJ 4.65.4, dated 1 December 222, addressed ‘to Domitius Ulpianus, praetorian prefect, my parent’ (ad Domitium Ulpianum praefectum praetorio parentem meum).

150

   

But his rise had been murderous. Initially he had a supervisory role over two other prefects, Julius Flavianus, formerly prefect under Elagabalus, and Geminius Chrestus, previously governor of Egypt, but by the end of the year, he had killed both of them.²⁹ Such activities, not penalized at the time, reflect both the ruthless character of the perpetrator and the weakness of the emperor whom he served. Severus Alexander could not control his own court or the factions within it. Thus Alexander and Mammaea also could not save Ulpian from the resentment of his own soldiers, who attacked him at night and murdered him ‘soon after’ (οὐ πολλῷ ὕστερον), despite his attempt to take refuge with his imperial protectors.³⁰ Ulpian’s posthumous reputation, like that of Papinian, rests on his legal achievement, but the courtier is also present in the tradition about him represented by the Scriptor in the late fourth century. According to the imaginative biographer, the lawyers put themselves in charge of the whole government; they and the chief bureaucrats were to vet state business before presenting it to the emperor. Ulpian, courtier par excellence, had uniquely privileged access: only he could see the emperor privately and he insisted on overseeing the emperor’s activities as judge.³¹ Like other successful courtiers (compare Pliny’s relationship with Trajan noted above), he was permitted to mix business with pleasure, dining with Alexander, who, we are told, enjoyed his conversation. This is the favourable perspective on Ulpian the courtier, recognized, less kindly, by Syme as responding to the ‘promptings of an eager ambition’.³² For Honoré, Syme’s verdict was ‘perverse’: Ulpian was ‘a lawyer and intellectual in government’³³ who evolved a cosmopolitan and egalitarian approach to law and human rights. The contradiction, which reflects the perspective of the sources, is more apparent than real. Court jurists lived paradoxical lives, because they could not openly acknowledge that part of their lives spent practising the dark arts of the aulicus without damaging their profiles as independent lawyers. Ulpian the jurist from Tyre was an extreme example; he brought a distinctive, universalist perspective, but his written work—also his lasting achievement—was confined to a period when he was out of office and power. He may have used his writing as a means of facilitating his return, but, once back in office, he wrote nothing. In the early 220s, as a power-player on the highest level, he was effective, but short-lived. Ulpian’s career, therefore, like that of Papinian, reflects both the dilemma of the thinker under autocracy and the dangers confronting the courtiers of weak emperors. ²⁹ Dio 80(80).2.2; Zos. 1.11.3. For the turbulent politics of Severus Alexander’s regime, see Davenport 2011: 287–8. ³⁰ Honoré 2002: 30–3. The date of Ulpian’s death is disputed, because Dio’s text was excerpted and perhaps distorted by the Byzantine monk Xiphilinus. Dio’s ‘soon after’ suggests 223 or at the latest early 224; but Dio’s reference to complaints made against him as governor of Pannonia, directed to Ulpian (Dio 80[80].4.2) but rejected by Severus Alexander, who appointed him to a second consulship in 229, places Ulpian’s death as late as 228. ³¹ HA Alex. Sev. 16.1 (vetting); 67.2 (private access); 31.1–3 (judicial activity). ³² Syme 1972: 406–9. ³³ Honoré 2002: 35.

      

151

Papinian survived in high office longer, because he served a strong emperor; both men, perhaps the two most influential of all Roman lawyers, succumbed to the politics of faction in a court out of control.

5. Coming from the Outside Around the court jurists, the world was changing. In the first century, the focus was on Rome and the perspective of the lawyers was Rome-centred. Rome-based pedagogic ‘dynasties’ coexisted with families of jurists; Emperor Nerva was the son and grandson of jurists and his grandfather, M. Cocceius Nerva, had been a close friend of Tiberius.³⁴ By the time of Hadrian, however, emperors, senators, and lawyers originated from outside Rome and the composition of the ruling elite had become more diffuse, reflecting the diversity of empire. Lawyers, therefore, who pursued a career at the centre, also prided themselves on their local eminence, duly reinforced by reference to their status as amicus (φίλος in Greek) of the emperor. The career of M. Cn. Licinius Rufus, from Thyatira in Lydia, as analysed by Fergus Millar, illustrates one such from the first half of the third century.³⁵ Licinius’ thirty-year career (c. 210–c. 240) began with five equestrian secretarial posts, mostly requiring legal expertise; he was then adlected to the senate as praetor, held the governorship of Noricum, and ended his career with the rank of consular, the status of amicus Caesaris, and the reputation of one ‘most expert in the laws’ (ἐν|πειρότατον | νόμων).³⁶ All this is known from six inscriptions set up in his honour—four in his native city, which celebrate his work as representative of his city and as builder, the other two from northern Greece, for services unspecified. This local lad made good was also a courtier, who used his expertise to gain access to high places. Early in his career (as he mentions the emperor ‘Antoninus’), he compiled a substantial work of twelve or thirteen books, entitled Regulae (Legal Rules).³⁷ Moreover, he consulted the jurist Paulus, apparently by letter,³⁸ on a knotty legal problem concerning a man given freedom under a trust who, after reaching the years of discretion (at twenty) voluntarily sells himself into slavery; could he now proclaim himself as free? Neither exercise may be purely ‘academic’. By composing a wide-ranging dossier of ‘Rules’ (the style is prescriptive, not argumentative), Licinius advertised his range of expertise as a potential legal administrator. By contrast, his approach to Paulus demonstrates the lawyer’s ability to canvass both sides of a question: ‘I am swayed by x’ (movet me) he writes of one option and then ‘on the other hand, I am swayed by y’ (e contrario movet me), ³⁴ Plin. Ep. 1.22 (Aristo); Dig. 1.2.2.48 (Nerva). ³⁵ Millar 1999. ³⁶ IG 10.2(1).142. ³⁷ Extracts at Lenel, Pal. I, 559–62. ³⁸ Dig. 40.13.4, Paulus, Quaestiones, with letter-style greeting (Licinio Paulo [salutem]).

152

   

before leaving Paulus to decide between them. Licinius would not have presented his work as self-advertisement, still less a job application. Yet he would have expected that the effect of both would be to bring his name to the attention of the powerful.

6. Assimilation: The Imperial Quaestor in Late Antiquity In the course of the fourth century, a new type of lawyer took centre stage in court life. The imperial quaestor, the draftsman of imperial laws and consistory decisions, was originally the emperor’s spokesman and retained elements of that role. The first individual described as quaestor of the sacred palace (quaestor sacri palatii), Fl. Taurus, made his mark not as a legal innovator but as Constantius II’s representative in negotiations in Armenia.³⁹ A few decades later, the poet Ausonius was Gratian’s quaestor in the late 370s, not on account of his legal knowledge, but because of his gift with language and the fact that he had been the emperor’s tutor. Moreover, although most quaestors, unlike Ausonius, seem to have been genuine experts in law as well as legal draftsmen, their situation differed from that of Papinian and Ulpian, whose legal expertise had been exercised through the praetorian prefecture, in fundamental respects. Firstly, although quaestors had acquired the highest rank, that of illustris, by the fifth century, the office lacked the clout of the praetorian prefectures, the overall heads of the provincial administrations, who acted as quasi-emperors with their power to judge in the ruler’s place (vice sacra iudicans). Secondly, the quaestor had no office staff and had to draw on the services of the palace secretariats in charge of records, correspondence, and petitions.⁴⁰ As the court of the eastern empire fixed its residence at Constantinople from the early fifth century, quaestors acquired an influence on the evolution of law, which had lasting consequences. In 429, a committee headed by the praetorian prefect Antiochus, accompanied by his son, the quaestor Antiochus Chuzon, and others, persuaded Theodosius II to inaugurate a project of collection, selection, and (minor) editing of imperial constitutions from Constantine down to his own day, along with planned further works on juristic writings and the basics of law.⁴¹ The outcome of this, by late 437, was the Codex Theodosianus; the other two enterprises were quietly (though never officially) abandoned. The elder Antiochus died or retired before the completion of the Codex; the younger became praetorian prefect and took an active role in the ecclesiastical politics of the Council of Ephesus in 431.⁴² A century later, perhaps the most learned of all quaestors, Tribonian, guided to completion Justinian’s Corpus Iuris Civilis (Body of Civil Law), ³⁹ Amm. 14.11.14. ⁴⁰ For a general survey of quaestors, see Harries 1988. ⁴¹ CTh. 1.1.5; Matthews 2000a. ⁴² PLRE II Antiochus Chuzon 1.

      

153

consisting of two editions of Justinian’s own Codex of imperial constitutions, from Hadrian to his own day (529 and 534); the Digest of juristic interpretation of law (533); and the Institutes.⁴³ How much of the credit for the achievement of so much in so short a time rests with Tribonian and his team and how much with Justinian is a matter of speculation; certainly, despite a political setback in 532, Tribonian could rely entirely on his emperor’s backing. All involved in both projects received fulsome acknowledgement, by name, in the enabling constitutions for both codification projects—and mere names without personality or identity they have, largely, remained. The achievement and reputation of most late antique quaestors (with the exception of men distinguished for other reasons, such as the poet Ausonius⁴⁴ or the pagan champion, senator and scholar Nicomachus Flavianus) was dependent on, and perpetuated by, the emperors whom they served; no imperial quaestor authored a major independent work of jurisprudence. They also shared the dangers of proximity to emperors. Under Leo, the quaestor of c. 465, Isocasius, fell victim to a purge of ‘Hellenes’ (pagans), to be saved by the intervention of a friend.⁴⁵ Then, in 529, the ex-quaestor Thomas was exiled by Justinian, again on suspicion of pagan leanings; he was, however, also associated with the former praetorian prefect, Menas, who may have been the author of a subversive dialogue between a main speaker, ‘Menodorus’ and one ‘Thomasios’, which had the temerity to advocate a less autocratic constitution (politeia).⁴⁶ In Late Antiquity, the potential reputational damage attendant on involvement with the court became an acknowledged reality. Corruption in the legal system was allegedly rife and emperors themselves routinely denounced the ‘illegal impetration’ of rescripts. Other interested parties weighed in. Late in the fourth century, the Scriptor voiced his objections to the practice of ‘selling smoke’, referring to what happened when advisors promised to intervene on behalf of petitioners to get a favourable outcome in return for a bribe. To this end, while Antoninus Pius was praised for his insistence on transparency, the annoyance of his ‘court functionaries’ (aulici ministri) at his reforms was also gleefully noted; and Severus Alexander allegedly publicly suffocated in smoke a notoriously corrupt official, and crucified another who had taken a bribe of a hundred aurei.⁴⁷ In the sixth century, Procopius the historian was quick to criticize individual lawyer-courtiers for putting the legislative process itself up for sale. Procopius’ ideal quaestor was a man of wide experience, skilled in matters legal and entirely ⁴³ For Tribonian and other early sixth-century quaestors, see Honoré 1978. ⁴⁴ PLRE 1 Decimus Magnus Ausonius 7. ⁴⁵ PLRE II Isocasius. ⁴⁶ Bell 2009 includes the dialogue On Political Science (De Politica Scientia). For Thomas’ fate, see Mal. 18.42. ⁴⁷ HA Ant. Pius 6.4; cf. 11.1 amici prevented from colluding with freedmen; HA Alex. Sev. 36.2–3 (suffocation); 23.8 (crucifixion).

154

   

incorruptible—unlike the quaestors of Justinian.⁴⁸ Tribonian himself, according to Procopius, was supremely well educated but far too fond of money, always willing to sell justice for his own profit and even ignoring laws he had written himself.⁴⁹ He and John of Cappadocia, praetorian prefect in the early 530s, were said to have harmed the politeia (state or public interest) with their behaviour, incurred the wrath of the mob during the Nika riots in 532, and were sacked by Justinian but later restored.⁵⁰ Tribonian, a genial character, also engaged in flattery, the main vice of the typical courtier, allegedly telling Justinian that he was excessively afraid that the emperor was so very pious that one day he would be snatched away and spirited off up to heaven.⁵¹ Procopius’ Tribonian and his successors were implicated in the general dysfunction of Justinian’s court. By contrast, Justinian’s illiterate but comparatively innocuous predecessor, Justin, had benefited from the wisdom of the quaestor Proclus, whose influence extended well beyond the making of laws in an elegant and rhythmical style not only to domestic policy but also to foreign relations, as when he successfully blocked the ill-judged attempt by Justin to adopt the son of the Persian king Cabades.⁵² But court politics nonetheless carried risks and entailed compromise. When the Blue faction, favoured by Justinian, who was ill at the time, rioted, some ringleaders were arrested and punished by Theodotus, prefect of the city. Justinian, now recovered, promptly tried to frame Theodotus on a charge of witchcraft, but Proclus intervened to show that the charge was groundless. To take on the heir-apparent showed both courage and confidence, but his success was only partial; Theodotus’ life was saved but he was exiled to Jerusalem.⁵³ Thanks to Procopius, we see quaestors more clearly as courtiers in the early sixth century than ever before—and in often highly unflattering terms. The Tribonian of the Corpus Iuris Civilis—and of the constitutions composed by him in Justinian’s name—with his sonorous language, respect for the classical past, and love of legal disputation for its own sake is far removed from the moneygrubbing flatterer found in the pages of Procopius. The historian’s representations were influenced by his broader philosophy. He deeply distrusted the innovations, as he saw them, favoured by Justinian, and his ministers; all concerned were by definition guilty of greed, corruption, and incompetence. Conversely, Proclus, being opposed to innovation in general, was a safe pair of hands. This preconception allowed Procopius to ignore the implications of an ingenious little device which Proclus invented, ostensibly to allow the illiterate Justin to trace personally his endorsement of imperial documents; the potential for forgery, to anyone but Procopius, should have been obvious.⁵⁴

⁴⁸ ⁵⁰ ⁵² ⁵⁴

Proc. Anecd. 20.15. ⁴⁹ Proc. BP 1.24.16; Anecd. 13.21, 23. Proc. BP 1.24.18; 1.25.1. ⁵¹ Proc. Anecd. 13.12. Proc. BP 1.11. For the style, see Honoré 1978: 226. ⁵³ Proc. Anecd. 9.35–42. Proc. Anecd. 6.13–16.

      

155

7. Conclusion The fate of jurists as courtiers mirrors that of the wider Roman elite as, over the centuries, they became accustomed to, and conditioned by, the operation of an ever more overt autocracy. Initially they acted independently, as occasional, ad hoc advisors, the extent of their involvement a matter of negotiation. Their reputation, then as now, was based on their writings and their independent work as teachers and consultants. By the second century, cooperation had become the expected mode of conduct for emperor and lawyer alike. Papinian, perhaps, and Ulpian for certain, though best known now for their writings, accorded their literary work less importance than their careers in the service of emperors. By Late Antiquity, the legal author, with his independent expertise, had been entirely subsumed by the courtier; there is no named jurist in the Digest later than the Constantinian writer Arcadius Charisius. As lawyers who drafted laws became merely ‘the emperor’s mouth’, the reputation of lawyers at court in general and some individuals in particular fell victim to the late antique culture of criticism, as the emperors, themselves unaccountable, shifted responsibility for failure to court officials and by their example encouraged others to do so. But the contrasting characters and motivations ascribed to both Ulpian and Tribonian reflect a more durable dilemma for the jurist-courtier, resulting from the inbuilt tension in the life of the court lawyer between the behaviour—devious, sycophantic, ruthless, and unprincipled—of the aulicus and the obligations imposed by the need to safeguard the integrity, if not the autonomy, of law. That Justinian’s lawyers, despite external pressures and perhaps their own moral failings, succeeded in the creation of their ‘temple of justice’, the Corpus Iuris Civilis, is testament to the fact that even at the court of an extreme theocrat, the rule(s) of law could still prevail.

7 Court Construction and Regime Change in the Mid-Fourth Century Kevin Feeney

1. Introduction The fourth century was a pivotal time for the Roman government, witnessing both the dramatic expansion of the bureaucracy and the creation of a new system of graded honours.¹ Although it was obviously impractical for any individual to exercise total control over this vast administrative apparatus, the emperor could nonetheless determine the membership of its uppermost ranks.² In this chapter, I argue that the changed political context of Late Antiquity allowed imperial patronage to be utilized not merely in a traditionally clientelist fashion to advance well-connected supporters, but also to promote a coherent political message. Successive emperors were able to exploit the premium that contemporary elites placed upon attaining high office to influence their behaviour. The opportunity for this was especially acute at the beginning of a new reign. Here I focus on two particularly well-documented consecutive accessions in the mid-fourth century, those of Julian in 361 and Jovian in 363. Although very different men operating in very different political contexts, both would deliberately harness the power of court appointments to advance their respective agendas.

2. Traditions of Court Construction The Roman court’s transition to Late Antiquity was marked by the formalization of the loose hierarchy of elites and amici Caesaris into a better defined order of imperial comites under Constantine I.³ Although the imperial administrative bureaux had traditionally been headed by members of the equestrian order, the mid-fourth century saw the most prestigious positions confer upon their holders the rank of vir clarissimus, the same honorific held by senators.⁴ These positions included on the civilian side the magister officiorum, quaestor sacri palatii, comes ¹ Smith 2011. All translations throughout are my own. ³ Jones 1964: 526–8; Scharf 1994; Smith 2011: 143–4.

² Chastagnol 1992. ⁴ Millar 1977: 83–101.

Kevin Feeney, Court Construction and Regime Change in the Mid-Fourth Century In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0008

    

157

sacrarum largitionum, and the comes rerum privatarum, collectively referred to as the comites consistoriani, as well as the praetorian and urban prefects.⁵ The magistri militum and the comes domesticorum made up the top military offices.⁶ While consistorial meetings would regularly include ex officio members and personal friends of the emperor without formal positions, the dignitas conferred by high office remained the pinnacle of the standard cursus honorum for late Roman officials.⁷ These offices thus constituted the nucleus of the official structure of the late Roman court. The prestige bestowed by the new comitival system enhanced the desirability of such offices for ambitious fourth-century elites. The court functioned as a ‘busy channel of social mobility’ which could confer elevated social status upon parvenu officials whose ancestors had been excluded from the dignity of the old senatorial class.⁸ Men such as Flavius Maximus, son of a provincial tabularius, could now aspire to reach the heights of a praetorian prefecture.⁹ The existing aristocracy themselves remained enthusiastic participants in paideia, the ‘common culture that was held to be the distinguishing mark of the diffused governing class of the empire’, which stressed the need to discharge the negotium of high office to preserve familial eminence.¹⁰ Thus even the scion of one of Rome’s noblest families, the infamous Sextus Claudius Petronius Probus, would serve as quaestor, urban prefect, and praetorian prefect, the latter on four occasions.¹¹ Contemporary vocabulary duly presented office as a personal benefaction from the emperor—a dignitas, honor, or τιμή—rather than a public duty.¹² Since senior positions at the court now represented the apex of both civilian and military careers, competition was fierce, and the process through which appointments were made was heavily scrutinized. Contemporary writers were in no doubt that personal characteristics were the chief qualification which determined public appointments. The ideal government official would obtain office through a mix of personal connections, familial status, and wealth. The court was distinguished from lesser posts in public administration chiefly in the degree of these qualities which membership required, so that the most senior office-holders should consist of no less than the ‘best men from all the provinces’.¹³ For men without the self-evidently outstanding pedigree of an aristocratic clan like the Anicii, promotion could be obtained by building a ⁵ All these posts are attested as clarissimate before 361. MO: CTh. 12.1.38; 8.5.8. CSL: CTh. 11.16.7, 12.1.38. CRP: P. Oxy. 20.2267. The first explicit confirmation of the quaestor’s status comes only in 362 (CTh. 11.39.5); however, that the office held this grade beforehand can be seen in the earlier appointment of men who already possessed senatorial status: PLRE I Nebridius 1, Secundus 3. On the quaestor, see Chapter 6 by Jill Harries in this volume. Praetorian Prefects: CTh. 8.4.5; 8.7.4. As a traditionally senatorial position, the prefecture of Rome carried clarissimate rank well before other court positions: see e.g. CTh. 13.5.2. The prefecture of Constantinople was created at clarissimate rank in 359 (Dagron 1974: 226–9). ⁶ CTh. 12.1.38. ⁷ Jones 1964: 333–4. ⁸ Matthews 1975: 43. ⁹ PLRE I Maximinus 7. ¹⁰ Brown 1992: 36. ¹¹ PLRE I Probus 5. ¹² Lendon 1997. ¹³ Pan. Lat. 4(10).35.1–2: ex omnibus provinciis optimates viros.

158

   

network of personal and familial recommendations, the process of suffragium.¹⁴ This is most visible for us in the vast quantity of recommendation letters that have survived, including more than five hundred from Libanius alone.¹⁵ These were not intended to attest to their subject’s curriculum vitae but to his character. The most frequent adjectives which they deploy are generic accolades such as ‘worthy’ (dignus, ἄξιος), ‘suitable’ (idoneus, ἐπιτήδειος), and ‘effective’ (utilis, χρήσιμος).¹⁶ Contemporaries were often more resentful of the power of wealth, with unverifiable allegations of corruption and a ‘market for offices’.¹⁷ Emperors could not be expected to be personally acquainted with the vast majority of their appointees, and for lower offices they seem to have been happy to let the suffragium system work itself out. For the most senior posts, however, a far greater degree of personal involvement was normal. While always subject to the arbitrary wishes of the emperor, higher-ranking officials enjoyed considerably longer tenures than lower office-holders, with praetorian prefects, for example, lasting for an average of three to four years, and magistri militum a good deal longer.¹⁸ Given the immense pressure imposed by fierce competition for office, this suggests that emperors were careful about selecting the men who would fill the uppermost ranks of their court. Although promotions might be based on the usual combination of status and networking, demonstrable loyalty to an emperor could also play a role in unlocking the top offices. Consequently, both members of the imperial family and men who shared the emperor’s geographical origins could expect to be the beneficiaries of considerable patronage upon his accession. This was especially true under Augusti who approached their new empires as outsiders to the local power structure, such as the Spaniard Theodosius I in the East after 379. Theodosius appointed his relative Nebridius as comes rei privatae in 382, and later as prefect of Constantinople in 386.¹⁹ He also brought to Constantinople a broader posse of Spanish officials such as the notorious Maternus Cynegius, who served consecutively as comes sacrarum largitionum, quaestor, and eastern praetorian prefect.²⁰ So significant was this group that one influential study has claimed that the governance of the East after 379 could not be understood without appreciating the culture which had given rise to these men in Spain.²¹ The overall effect was to create ‘regional groupings of elites that played an active role as regional cliques in imperial politics’.²² The attractions of this approach were obvious; regional appointees were men with whom an emperor was both personally familiar and culturally comfortable, who knew that their careers were entirely dependent upon him alone. In an era in ¹⁴ Jones 1964: 383–96. ¹⁵ Cribiore 2007: 215. ¹⁶ Pedersen 1976: 30. ¹⁷ Them. Or. 8.117a. See further MacMullen 1988: 148–67; Jones 1964: 391–6; cf. Kelly 2004. ¹⁸ Jones 1964: 380–1. For the long tenures of magistri militum, especially in the fifth century, see Chapter 8 by Meaghan McEvoy. ¹⁹ PLRE I Nebridius 2. ²⁰ Matthews 1975: 109–15; PLRE I Cynegius 3. ²¹ Matthews 1975: 145. ²² Kulikowski 2015: 145.

    

159

which most emperors could expect to control only a part of the entire empire, maintaining a strong geographic base was an eminently sensible precaution. Few observers of the late Roman court, ancient or modern, have seen experience or qualifications as particularly relevant to appointment prospects. Longevity of service was sometimes praiseworthy, but typically only for lower postings.²³ It did not matter whether an appointee had experience relevant to his position, nor was there any set of objective criteria to judge how well he performed his duty.²⁴ With the exception of the largely respected post-Diocletianic civilian/military divide, there was little sense that individual offices might require different skills. Felix, for example, would serve as comes sacrarum largitionum under Julian after previously being a local dignitary and then a notarius; there is no indication that he had any prior financial background.²⁵ Even the emperor’s quaestor would not initially be selected for legal expertise, although eloquence may have been taken into consideration.²⁶ One study has found that neither competence nor specialized training influenced government appointments, and indeed a candidate’s prior history would not necessarily even be known in detail.²⁷ While military promotion could be more flexible, here too advancement through influence remained commonplace, particularly given the tradition of sons following their fathers into service.²⁸ It is surely likely that the future Theodosius I’s rapid advancement to dux was related to his father’s current service as magister militum.²⁹ Both ancient and modern observers have thus traditionally emphasized the clientelist elements of court construction, in which ‘ties of patronage and favours exchanged lie behind appointments to high and low office’.³⁰ Yet while it is clear that appointments to senior court offices were heavily shaped by this web of connections and ‘rough personal judgements’, these were not the only factors that an emperor could consider.³¹ The emphasis on clientelist rather than programmatic aspects of the appointment process reflects the established model of the ‘reactive emperor’ developed for the Principate, with a sovereign who had neither the desire nor the ability to develop and implement a proactive political agenda.³² While some historians have sought to extend this model into Late Antiquity, however, a recent wave of scholarship has pointed in a very different direction.³³ Historians such as Schmidt-Hofner have argued that later emperors were far more capable than their predecessors of formulating and disseminating coherent political agendas.³⁴ Constantine in particular has been heralded as marking a watershed in the transformation the emperorship into a ‘relatively proactive,

²³ Pederson 1976: 33–4, 40. ²⁴ Amm. 21.16.3; Them. Or. 8, 117a–b. ²⁵ PLRE I Felix 3. Mamertinus (Pan. Lat. 3[11].1.3) vaguely suggests that he was appointed comes sacrarum largitionum because he could be trusted not to steal, but this is hardly a tangible qualification, and he was later dismissed from another office for embezzlement (Amm. 27.7.1–2). ²⁶ Harries 1988: 158. ²⁷ Pederson 1976. ²⁸ Jones 1964: 637–42; Matthews 1989: 270. ²⁹ See McLynn 2005. ³⁰ Salzman 2002: 115. ³¹ Pederson 1976:40. ³² Most associated with Millar 1977. ³³ Errington 2006. ³⁴ Schmidt-Hofner 2008.

160

   

popularizing autocracy’.³⁵ Successive studies have demonstrated the ways in which his regime manipulated the rescript system both to promote an image of the emperor as the personal wellspring of justice and to encourage cities to move towards adopting his Christian faith.³⁶ Other emperors utilized legislation as a tool of communication rather than as a way of solve legal problems, particularly at the outset of their reigns.³⁷ A growing ability and willingness to utilize all tools at the emperor’s disposal to achieve proactive political ends was thus a new hallmark of imperial government from the early fourth century. It was only natural that the power of patronage would soon be used for the same ends. Communication between any emperor and the imperial elites was fundamentally a dialogue. The competition induced by the desirability of public positions altered its terms in the emperor’s favour; if the prestige of office was a commodity, then the fourth century was a seller’s market. Far from reducing the court to a patronage machine, this allowed emperors to incentivize changes in elite behaviour. Barnes has demonstrated that Constantine displayed a marked preference for appointing Christians, and subsequent work has shown that he also acted as a consistent patron to a set of predominantly Christian individuals such as Junius Bassus and Flavius Ablabius.³⁸ Such favouritism should be seen not merely as a reward for these particular men, but also as a clear signal to other current or aspiring office-holders about the beliefs and behaviour that allowed one to succeed in Constantine’s court, and a presentation of new exempla to emulate. Elites thus did not need to settle for being passive recipients of capricious imperial favour whose rise and fall were based solely on suffragium. Instead, they possessed agency, which they could exercise by behaving in ways that would please the Augustus and elevate them above their peers in the fierce competition for office. In order to explore this argument, I will examine two successive regime changes in the early 360s: Julian’s succession of his cousin Constantius II in the East in November 361 and Jovian’s succession of Julian in June 363. Both were unusually eventful accessions, following an abortive civil war and a major military debacle, respectively. If neither can be considered ‘typical’ for a late Roman emperor, they nonetheless demonstrate the very different ways that an emperor could utilize court appointments as part of a broader political agenda.

3. Julian: A Study in Change Of the multiple imperial accessions in the mid-fourth century, Julian’s assumption of rule in the East in November 361 was perhaps the unlikeliest candidate for

³⁵ Dillon 2012: 6. ³⁶ Dillon 2012; Lenski 2016. ³⁷ Schmidt-Hofner 2015. ³⁸ See Barnes 1981; 1989; Porena 2012. Salzman 2002 argues that religion was not a consistently determinative factor in appointments, but concedes that it did play some role on occasion.

    

161

radical change. Julian was, after all, the cousin of Constantius II and his succession could conceivably be treated as a standard dynastic one. To do so, however, would be to downplay the fraught circumstances that surrounded it; not just his famous religious break with Constantius but also their eighteen-month civil war and a bloody family history.³⁹ All of this had left Julian in an ambiguous position. His blood ties to his predecessors and his support from elements of Constantius’ court meant that an outright disavowal was politically untenable, yet his personal enmity and heartfelt religious convictions strongly inclined him towards breaking with his cousin’s regime. Within these limits, there is no question that Julian opted for the greatest break possible. Standard gestures were made to highlight his dynastic legitimacy, such as accompanying his cousin’s funerary procession.⁴⁰ Yet in key and very public respects, Julian made it entirely clear that he did not simply represent the continuation of his family’s regime. The relative informality of his public image, such as his moving around Constantinople on foot and his dismissal of much of the palace staff, could hardly have been a greater contrast with his cousin’s elaborate public decorum.⁴¹ The emphasis which Mamertinus laid upon this projection of civilitas in his consular panegyric of 362 suggests that it may well have been a calculated decision to break with the recent trend towards elaborate ceremonial around the emperor.⁴² Julian’s active financial and literary promotion of his faith hardly needs restating, while he also deliberately entered into a public philosophical dialogue intended to repudiate the worldview put forth by his predecessor.⁴³ His rejection of his family extended even to his legislative output, which displayed such an enthusiasm to repeal Constantinian legislation that it has been suggested—not unfairly—that he actively delighted in doing so.⁴⁴ Considered cumulatively, the rhetoric and actions that surrounded Julian’s assumption of the purple in late 361 conveyed two central messages: the era of Constantius was firmly over, and the new emperor would aggressively promote his pagan faith. These same twin themes are visible in the construction of his court. In the winter of 361/2, the Augustus used the court to send one of the most unambiguous signals yet that a new order was entirely displacing the old. Having immediately dismissed the entire senior echelon of Constantius’ consistorium and replaced them with his own appointees, he then convened a tribunal at Chalcedon to put the leading officials of the ancien régime on trial.⁴⁵ The dismissal of the prior court in itself was hardly unusual. Not only was some level of turnover a natural part of even dynastic accessions, but, as we shall see, Julian came to power in the East with an established court of western supporters, and every appointee of ³⁹ Jul. Ep. ad Ath. 270C–D. ⁴⁰ Lib. Or. 18.118. ⁴¹ Amm. 22.7.1, 22.4.9–10; Lib. Or. 18.130. On the contrasting atmosphere at court under Constantius II and Julian, see Chapter 5. ⁴² Rees 2012: 214–15. ⁴³ Elm 2012: 60–87. ⁴⁴ Harries 1999: 43–5. ⁴⁵ Amm. 22.3.1–2.

162

   

his cousin whom he retained meant one less place for his own followers. Yet while the ousting of Constantius’ courtiers was well within the bounds of normality, their sham trials and subsequent bloody eliminations were decidedly not.⁴⁶ The purge was brutal and, at the senior levels, total. Three of the four comites consistoriani—the magister officiorum Florentius, the comes rerum privatarum Evagrius, and the comes sacrarum largitionum Ursulus—were condemned and either exiled or executed.⁴⁷ The quaestor Leonas is last mentioned in our sources in the summer of 360, and it is unclear if he suffered a similar fate, but he was certainly removed from office and replaced with Julian’s choice, Jovius.⁴⁸ The praetorian and urban prefects were also dismissed to a man, and two, Taurus and another Florentius, the consuls for 361, were convicted at Chalcedon.⁴⁹ This behaviour was characteristic of emperors who had emerged victorious in civil war rather than those peacefully succeeding a relative.⁵⁰ Dramatic changes in the ranks of the senior military officials were likewise deeply uncommon in non-civil war regime changes, in part because new emperors in precarious positions could not afford to alienate seasoned generals.⁵¹ In 361–2, however, every one of Constantius’ leading commanders was replaced.⁵² The long-time magister militum Arbitio was too powerful to be pushed out and was likely involved in the consistorium’s decision to accept Julian after the death of Constantius, but he was already old and his disappearance from public life might reasonably be attributed to voluntary retirement.⁵³ The same cannot be said for his fellow magistri Gomoarius and Agilo, neither of whom had been in position for more than two years when they were dismissed.⁵⁴ Both would snatch an opportunity to return offered by the usurper Procopius in 364, suggesting that their departure was not voluntary; indeed Gomoarius was known for his hatred of Julian.⁵⁵ A certain Lucillianus may also have been a magister but if not then he was still a high-ranking general, and he too was unceremoniously dismissed.⁵⁶ Finally, the post of comes domesticorum makes its first appearance in Julian’s reign with a new incumbent, Dagalaifus, completing the sweep of senior offices on both civilian and military sides.⁵⁷ ⁴⁶ Dynastic successions could occasionally lead to upheaval but prosecutions were usually limited, as in 375: Matthews 1975: 64–6. ⁴⁷ Amm. 22.3.6–7; PLRE I Florentius 3, Evagrius 5, Vrsulus 1. ⁴⁸ PLRE I Jovius 2; Amm. 20.9.3, 21.8.1. ⁴⁹ PLRE I Taurus 3, Florentius 10, Nebridius 1, Helpidius 4, Honoratus 2, Tertullus 2; Amm. 22.3.4. ⁵⁰ See the aftermath of the usurpations of Magnentius: Amm. 14.5.1–9; Silvanus: 15.6.1–4; Procopius: 26.10.1–14. On tactical amnesties after civil war, see Leppin 2015a. For a similar case in the third century, see Davenport 2012b. ⁵¹ Though purges could accompany other forms of regime change under non-campaigning child emperors such as in 408 under Honorius; see Chapter 8 by Meaghan McEvoy in this volume. ⁵² Our record of comites rei militaris is incomplete, but no known occupant of this post continued from Constantius to Julian. ⁵³ PLRE I Arbitio 2; Amm. 26.9.5. ⁵⁴ PLRE I Agilo, Gomoarius; Amm. 20.2.5, 20.9.3. ⁵⁵ Amm. 26.7.4, 21.15.13. Agilo was on the Chalcedon tribunal: Amm. 22.3.1. ⁵⁶ Amm. 21.6.10–11. ⁵⁷ PLRE I Dagalaifus.

    

163

While this purge of the military has received little scholarly comment, perhaps because it did not result in bloodshed, the condemnation of the civilian courtiers at Chalcedon has often been highlighted as a form of personal score settling by either the new emperor or the military establishment.⁵⁸ Certainly there was some of that, especially in cases where we know grudges to have existed. Julian is said to have harboured a personal vendetta against the praepositus sacri cubiculi Eusebius, for instance, while Ammianus alleges that the comes sacrarum largitionum Ursulus had become deeply unpopular with the army.⁵⁹ However, the number of men targeted is simply too great and includes too many men such as the consuls Taurus and Florentius for whom well-informed sources like Ammianus knew of no personal motivation. The scale of this action makes far more sense when read in conjunction with Julian’s legislation and public behaviour as a warning to the empire’s elite that although a member of the Constantinian family sat on the throne, Constantius and all that he represented were gone. Association with the old regime was no longer a virtue, but rather a mark of suspicion. If elites wished to know the path to power, they would need to look to the empire’s new rulers. In lieu of a regional following from his birthplace, Julian had assembled a court of his most devoted Gallic followers following his Paris usurpation in 360, replacing the appointees of Constantius which he had been obliged to accept while still a Caesar.⁶⁰ This put him in the unusual position of already having an established court in place before he formally took up rule of the entire empire. Unsurprisingly, the vast majority of his appointees continued to hold their offices after he took possession of the East following the death of Constantius. His magister officiorum Anatolius, quaestor Jovius, comes domesticorum Dagalaifus, and both magistri militum, Nevitta and Jovinus, were all first appointed while Julian was a usurper.⁶¹ The decision to maintain their appointments clearly reflected an emperor’s typical desire to reward long-time followers and surround himself with loyalists; the same can be said for his consular nominations for 362, Nevitta and Mamertinus.⁶² Yet it was the few senior officials whom Julian did retain from Constantius and the few outsiders who were soon promoted into his consistorium that truly signalled to imperial elites how they too could thrive in future. By far the most senior civilian official to remain in place was the comes Orientis Modestus, in post since 358.⁶³ While this position did not carry the same distinction as the most senior offices, it held more prestige than the vicariates for which it

⁵⁸ Thompson 1947: 73–9, argues that it was military revenge, while Browning 1976: 124 ascribes it to Julian settling personal scores. Both are still repeated, e.g. Boeft et al. 1995: 17; Harries 2012: 303–4. ⁵⁹ Amm. 21.15.4; 22.3.12; 22.3.7. ⁶⁰ Blockley 1972. ⁶¹ PLRE I Anatolius 5: Amm. 20.9.8. Jovius 2, Dagalaifus, Nevitta: Amm. 21.8.1. Jovinus 6: Amm. 21.8.3. ⁶² Amm. 21.8.1. Mamertinus 2 had been CSL but would be moved to a praetorian prefecture. ⁶³ PLRE I Modestus 2; Lib. Ep. 364, 686.

164

   

was the Eastern replacement, not least due to the court’s frequent presence in Antioch, where the comes resided.⁶⁴ The position often served as a stepping stone towards a prefecture, as it did for Modestus’ two immediate predecessors and then indeed for Modestus himself.⁶⁵ While his fellow senior officials of Constantius were dismissed or worse, Modestus not only survived but thrived, and would be promoted by Julian to the prefecture of Constantinople in 362. To understand why Modestus endured, we should look at those easterners whom Julian promoted to high office outside of his western circle. Two of these in particular were within the highest echelons of the consistorium. Felix was made comes sacrarum largitionum after Mamertinus in early 362, and Helpidius was brought in around the same time as comes rerum privatarum.⁶⁶ Why these men? Helpidius had met Julian in Gaul, but this can hardly have been sufficient to award him such a senior position.⁶⁷ During the standoff with Constantius, the latter had proposed Felix as Julian’s magister officiorum and the usurper had rejected the appointment—not exactly a promising sign.⁶⁸ Perhaps it was indeed simply that they had suitable recommendations, but the circumstances of their appointments suggest instead an intriguing alternative. Both men had a single crucial characteristic in common with the survivor Modestus: as Theodoret lamented, they ‘were Christians who had abandoned the faith, obliging the unholy emperor’.⁶⁹ As is so often the case with Julian, it is impossible to escape the problem of religion. It is obviously unsurprising that a pagan emperor would favour pagan officials. Yet there were many prominent pagans who would not enjoy the new emperor’s patronage, including that unofficial court stalwart Themistius whose relations with Julian were famously complex.⁷⁰ What unites these appointees was that they were not just pagans but former Christians like the emperor himself. It is telling that this fact was well enough advertised that it would have been known to prominent pagan and Christian writers alike. We should not dismiss their conversions as opportunistic flattery alone. Although that doubtless played a role, we have just seen that pagan status alone was not enough to win favour. Rather, we should understand both sides, Julian and these senior officials, to be actors participating in a public dialogue. This dialogue was intended as an unambiguous message to elites, especially in the East. The accession of a new ruler who had played no role in the eastern empire’s political life and brought with him a core of loyal outsiders would inevitably cause disquiet among aristocrats and social climbers alike. This could only have been exacerbated by the unambiguous signal that Chalcedon ⁶⁴ Liebeschuetz 1972: 110–11. ⁶⁵ PLRE I Honoratus 2, Nebridius 1. ⁶⁶ Philost. HE 7.10; Theod. HE 3.8; PLRE I Felix 3, Helpidius 4. ⁶⁷ Lib. Ep. 35. ⁶⁸ Amm. 20.9.5. ⁶⁹ Theod. HE 3.8: Χριστιανοὺς ὄντας ἀποστῆναι τῆς εὐσεβείας χαριζομένους τῷ δυσσεβεῖ βασιλεῖ; Lib. Ep. 791. ⁷⁰ Elm 2012: 60–143.

    

165

immediately sent out, that association with the old regime was no longer a benefit. The appointments of Helpidius and Felix were a demonstration of the new emperor’s willingness to admit outsiders and a display of the criteria he wished to see met, while the retention of Modestus showed those who already held positions how to maintain them. For Felix, this meant claiming that the emperor had personally converted the official to his faith, while for Modestus, it apparently meant publicly suggesting that he had never really been a proper Christian anyway.⁷¹ Even the emperor’s own uncle and namesake would be appointed comes Orientis after Modestus only when he too had publicly renounced Christianity.⁷² The ecclesiastical writers are quite right to ascribe such conversions to a desire to oblige the emperor. What mattered was that ambitious elites now knew how to do just that. Julian’s court construction then was not merely clientelist but also political, and designed to send two messages: that the regime of Constantius was dead and discredited, and that not just paganism but the renunciation of Christianity were secure routes to promotion or retention. For the ambitious elite looking to move up in the world or the anxious bureaucrats hoping to maintain their position alike, this young emperor, who seemed likely to be in place for a long time to come, was sending a clear signal of the new order. Patronage was yet another tool in Julian’s ceaseless effort to undo his uncle’s religious revolution, and his appointments in 361 are thus a perfect example of how emperors could exploit the competition for office to promote a coherent political agenda and influence the behaviour of imperial elites.

4. Jovian: A Study in Conciliation Just a year and a half after Julian used the court to signal this radical shift, an equally abrupt change came with his death during his failed Persian invasion. Deep behind enemy lines and in urgent need of an emperor, his generals came together to select a replacement but became deadlocked, leading the impatient imperial guards to elevate one of their own to the purple. Confronted with this choice and low on time and options, the council accepted the Illyrian primicerius domesticorum Jovian, who hurriedly negotiated a humiliating peace treaty.⁷³ Against this inauspicious background, the new emperor was forced to set a course for his own reign that would take it in a very different direction from his predecessor. Not only did Jovian lack anything like Julian’s ideological worldview, but he also began in a dramatically weaker position as the son of a well-regarded but ⁷¹ Felix: Lib. Or. 14.36. Modestus: Lib. Ep. 804. ⁷³ Amm. 25.7.1–12; Lenski 2000. Cf. Heather 1999.

⁷² Philost. HE 7.10; Passio Artemii 23.

166

   

distinctly non-imperial comes domesticorum.⁷⁴ Elevated in what was essentially a coup against the consistorium, he was unknown to the empire’s elite and lacked any sort of broad power base to match Julian’s western following. Moreover, his reign had begun with a capitulation to Persia so devastating that even the heroic efforts of Themistius struggled to portray it as a victory.⁷⁵ Yet the precariousness of his position did not mean that he had no choices at all. The easiest way to absolve himself of the disastrous treaty would have been to blame his predecessor, an approach with the dual merits of being accurate and having a large and vocal Christian constituency eager to embrace it.⁷⁶ Likewise, the Christian Jovian might have followed his predecessor’s militant pagan turn with an equally emphatic restatement of the empire’s new orthodoxy. Total rejection of Julian, presumably accompanied by the replacement of his courtiers with an Illyrian and Christian cohort, was certainly an option. Despite the fantasies of later ecclesiastical historians, however, this was not the path that Jovian followed.⁷⁷ In his treatment of both the invasion and religious affairs, the new emperor elected to a chart a far more middling course. Instead of blaming Julian for the Mesopotamian debacle, coins were minted with legends such as RESTITVTOR REI P(VBLICAE) (‘restorer of the state’) and VICTORIA ROMANORVM (‘victory of the Romans’), while court rhetoric also presented the campaign as a triumph.⁷⁸ There was, obviously, a heavy degree of self-interest in presenting the invasion as a success, yet the decision not to blame Julian remains notable, particularly since it would become fashionable under Valens to criticize Jovian himself for the catastrophe.⁷⁹ On the domestic front, there was no legislative flurry repealing prior statutes such as that which marked early 362. The main Jovianic legislation about which we know was a measure designed to promote religious toleration, as much a rebuke of the Constantinians’ preference for Christianity as Julian’s militant paganism.⁸⁰ The clearest political statement produced by the new regime, Themistius’ panegyric of 364, likewise downplayed any direct conflict with Julian and focused on the theme of religious pluralism.⁸¹ Where it exists, then, the sparse evidence for Jovian’s reign consistently suggests a conciliatory posture. In the same oration, Themistius also chose to single out Jovian’s appointments policy for praise within this framework of conciliation. He did this firstly by ⁷⁴ This was duly trumped up: Amm. 25.5.4; Zos. 3.30.1; Zon. 13.14. ⁷⁵ Heather and Moncur 2001: 152–4. ⁷⁶ See Matthews 1989: 130–7 for the best account of the campaign. On Christians ready to blame Julian: Greg. Naz. Or. 4.48, 5.12; Soz. 5.1.9, 6.3.2; Ephrem, Hymns Against Julian 3.2. ⁷⁷ Soc. 3.22; Soz. 6.3; Theod. HE 4.1; Joh. Mal. 13.26. ⁷⁸ RIC VIII Arelate 328, 335; Thessalonica 234–8; Heraclea 107; Constantinople 175–7; Antioch 228–9; Them. Or. 5.66a–c. ⁷⁹ Eutr. 10.16; Festus, Breviarium 29. Repeated with characteristic shamelessness by Themistius in Or. 16.213a. ⁸⁰ Heather and Moncur 2001: 154–6. ⁸¹ Them. Or. 5.67c–70b.

    

167

highlighting that Jovian had spilt no blood upon his accession, unlike his dynastic predecessors—certainly a direct allusion to Chalcedon as well as the chaos of 337.⁸² This was explained as a reflection of the new emperor’s surpassing virtue and confidence, which were in turn linked with his choices of courtiers: Understanding that the foundation of certainty for an emperor is the justice of his supporters, you brought back some of the best men from all sides, and selected others, and you dismissed the rest.⁸³

Once again, the emphasis upon quality mattering more than factionalism and the restoration of the best ‘from all sides’ (ἁπανταχόθεν) formed an implicit contrast with Julian. We can see here the composition of the court being incorporated into a broader political agenda that presented the emperor as a unifying and conciliatory figure, which could ironically be presented as a break with his predecessors. How far was this aspiration translated into practice in Jovian’s court? The identity of the speaker itself was a clue to this. After a difficult period under Julian, Constantius’ favoured orator had now returned to his former prominence, where he would remain for the duration of Jovian’s reign and into that of the Valentiniani.⁸⁴ Yet Themistius and other allies of Constantius would not enjoy the limelight alone among that nebulous circle of men who enjoyed imperial favour without the formal status of consistorial office. Under Jovian, they were joined by less traditional courtiers such as the philosophers Maximus of Ephesus and Priscus, who had been elevated to the innermost circles of power by Julian and continued to find favour and honours under his successor.⁸⁵ This strange mix of restored Constantinians and Julianic legacy figures would prove characteristic of how the new emperor dispensed the offices that made up the consistorium’s formal structure too. Three distinctive groups may be identified among Jovian’s senior appointees, and their identity is a striking contrast with Julian’s blend of Gallic loyalists and pagan converts.⁸⁶ The largest such group is one which did not exist in Julian’s court—the holdovers from the emperor’s immediate predecessor. Not only were none of Julian’s senior courtiers prosecuted, but one comes consistorianus was actively retained, the comes rei privatae Helpidius.⁸⁷ Two of Julian’s most highprofile supporters, Saturninius Secundus Salutius and Claudius Mamertinus, were retained in their capacities as praetorian prefects, as was the praetorian

⁸² Them. Or. 5.66d. ⁸³ Them. Or. 5.67b: Αἰσθόμενος δὲ ὅτι τῷ βασιλεῖ κρηπίς ἀσφαλείας ἡ τῶν συνόντων δικαιοσύνη, τοὺς ἁπανταχόθεν ἀριστοὺς τοὺς μὲν ἐπανήγαγες, τοὺς δὲ προσείλου, τοὺς δὲ ἀπήλλαξας. ⁸⁴ Jones 2010: 501–6. ⁸⁵ Eun. V. Soph. 478. ⁸⁶ Olariu 2005: 353. ⁸⁷ Lib. Ep. 1120.

168

   

prefect of Gaul, Germanianus.⁸⁸ Jovian’s choice of comes rei privatae was one Caesarius whose first known public posting, as vicarius Asiae, was made under his predecessor.⁸⁹ In three of these cases, the individual concerned was not only a Julianic appointee but also a very public pagan; Helpidius, it will be remembered, had apostatized, and there is no evidence that he converted back.⁹⁰ Only a single appointee of Julian stood trial, the deeply unpopular consularis Syriae Alexander— and he was acquitted.⁹¹ Themistius’ contrast was justified; Chalcedon this was not. The civilian situation was mirrored in the senior hierarchy of the military. It is true that one of Julian’s two known magistri equitum, Nevitta, disappears from our sources and so was surely dismissed.⁹² However, his counterpart Jovinus remained in post and would even survive another regime change in 364.⁹³ Two other magistri promoted by the new Augustus had previously been awarded senior ranks under Julian, the comes domesticorum Dagalaifus and the comes rei militaris Victor, while the comes rei militaris Arinthaeus kept his office.⁹⁴ The Julianic faction’s influence, then, was not contained to a few token representatives within the court. Appointees of Julian occupied most of the top positions in the military and a number of high-ranking civilian posts. The second most important faction in Jovian’s court was the now-customary group of regional and familial appointees, in this case Illyrians whom the new emperor clearly hoped would provide him with a power base. He likely replaced Julian’s deceased magister officiorum Anatolius with the Dalmatian Ursacius and appointed the Pannonian Viventius as quaestor.⁹⁵ His father-in-law Lucillianus became magister equitum and another relative, Januarius, was appointed to an unclear senior military role, probably as comes rei militaris.⁹⁶ Beyond these few exceptions, however, the men from Jovian’s homeland had to content themselves with largely secondary positions such as the tribuneships of the scutarii awarded to Equitius and Valentinian or the post of numerarius awarded to Leo.⁹⁷ It is probable that if Jovian had lived then he would have continued to promote these co-regionalists, but in the event we can say only that unlike his predecessor, he had ⁸⁸ Secundus: Lib. Ep. 1429, Or. 24.20; Amm. 25.7.7; Zos. 31.1, CTh. 7.4.9. Mamertinus: Amm. 26.5.5, CTh. 13.3.6. ⁸⁹ PLRE I Caesarius 1. ⁹⁰ Of the other converts from Christianity, Felix and the comes Julian both died in 362, spurring Christian glee: Amm. 23.1.5; Theod. HE 3.11–13; Philost. HE 7.10, 12; Soz. 5.8.4; Greg. Naz. Or. 5.2. Modestus reappears as praetorian prefect for the East in 367: Zos. 4.2.4. ⁹¹ PLRE I Alexander 5. Unpopularity: Amm. 23.2.3; Lib. Ep. 838, 1351, 1411. Arrest: Lib. Ep. 1256, 1294, 1456. ⁹² He last appears debating Julian’s succession at Amm. 25.5.2. ⁹³ Amm. 25.10.9, 27.2.1. ⁹⁴ Amm. 26.5.2; PLRE I Victor 4, Flavius Arinthaeus. Zos. 3.13.3 falsely suggests that Julian had promoted Victor to magister peditum, and Arinthaeus to magister equitum. ⁹⁵ PLRE I Anatolius 5; Amm. 25.3.14; Zos. 3.29.3. For the probable, but not certain appointments of PLRE I Vrsacius 3 and Viventius, see Lenski 2002: 56. ⁹⁶ PLRE I Lucullianus 3; Amm. 25.8.9–11, 25.10.6–7; Zos. 3.35.1–2; Symm. Or. 1.4; PLRE I Ianuarius 5, Amm. 26.1.4–5. ⁹⁷ PLRE I Equitius 2, Valentinianus 7, Leo 1.

    

169

no established personal following waiting in the wings to sweep the uppermost ranks of the consistorium, and modern claims of Pannonian domination of the court in his reign seem implausible.⁹⁸ The third and by far the smallest set of Jovian’s appointees may be those whom Themistius had in mind when he praised Jovian specifically for bringing former officials back.⁹⁹ These were the officials who had enjoyed favour under Constantius II before being unceremoniously dismissed by his cousin, as we have seen in the previous section. The most prominent among this group to be offered formal office were the magistri equitum Lupicinus and Malarichus, although the latter declined to return.¹⁰⁰ A further group of Constantinian honorati would return to prominence within the court, if not to actual appointment. We have seen already that this number included Themistius, and we know that the influential patricius Datianus was likewise restored to his former favour.¹⁰¹ The lack of any formal appointment to a senior civilian role among this group presumably reflects the fact that many candidates had been wiped out at Chalcedon as much as any reluctance by Jovian to restore them to office. Jovian’s initial court, then, was a far more diverse body than his predecessor’s had been, and Julian’s appointees remained the largest single faction within it. It is surely probable that this reflected weakness as much as intention. As a primicerius from a military family, the new Augustus is hardly likely to have been intimately acquainted with the civilian elites upon whose support he now depended. Similarly, his initial elevation as Augustus had been possible only because the military council meeting to determine the succession had acquiesced to the decision of the domestici. Removing many of the officials who had participated in that acquiescence was unlikely to prove a winning strategy. Given time to establish himself more securely, Jovian might well have made more sweeping changes. In the short term, the limited removal of a handful of officials and the promotion of a mix of regional loyalists and the occasional experienced old hand was a conservative but entirely logical choice. Nevertheless, however, constrained by circumstance it might have been, Themistius’ linking of Jovian’s personnel choices with the theme of conciliation suggests that the fledgling regime was actively attempting to make a virtue out of necessity. Paradoxically, the retention of so much of Julian’s court could be presented as an implicit break with his practice, and continuity could be used to suggest a changed attitude. Unlike Julian or Constantius, the new emperor would supposedly choose only the best men, regardless of their factional or religious orientation. The fact that Jovian was in no position to alienate the men who

⁹⁸ Piganiol 1947: 147. ⁹⁹ Olariu 2005 includes Arbitio and Agilo among this group, but there is no evidence for either playing any active role before they resurfaced during Procopius’ revolt in 365. ¹⁰⁰ Amm. 26.5.2; PLRE I Lupicinus 6, Malarichus. ¹⁰¹ PLRE I Datianus 1.

170

   

had served under either of his predecessors could be presented as conciliation. His inability to enforce any religious litmus test even if he had so desired could be spun as toleration. No ambitious official should feel that they were locked out— and thus, they should have no incentive to oppose him. In appointments as in rhetoric and legislation, Jovian’s regime consistently pursued the creation of consensus over the promotion of factional interests. Even with limited room to manoeuvre, the emperor could thus use the process of court construction to communicate a broader political message. Where Julian had sought to change elite behaviour, the weaker Jovian sought instead only to maintain elite approval. Although Jovian’s early demise prevents us from knowing whether his efforts would have been successful in stabilizing his regime, it is worth noting that his civilian appointments would be far more enduring than those of Julian. Within a few months of succeeding him, the Valentiniani would begin to clear the decks of Julianic appointees preserved by Jovian, except in the military.¹⁰² Similarly, and predictably, under the Pannonian brothers ‘the new tide from the Balkans continued to rise’.¹⁰³ Jovian’s senior Illyrian officials such as Ursacius and Vivacius would move west with Valentinian, and his mid-ranking regional appointees such as Leo and Equitius would go on to enjoy glittering careers.¹⁰⁴ Even the policy of restoring Constantinians would pick up pace.¹⁰⁵ Constantius’ magister officiorum Florentius was restored from the exile imposed at Chalcedon to the role of comes sacrarum largitionum, while the Nebridius dismissed by Julian would go on to be praetorian prefect of the East.¹⁰⁶ Despite a shorter reign and a far more constrained agenda, it was thus Jovian rather than Julian who would set the tone for the composition of the late Roman court in the following decades.

5. Conclusion A detailed study of the senior appointments made by Julian and Jovian shows how Roman emperors could construct their courts programmatically to promote a political agenda as well as in a clientelist manner to reward well-connected elites and supporters. For Julian with his developed ideological perspective, this meant the promotion of apostasy from Christianity and a rejection of the Constantinians, in a deliberate reversal of the appointments policy of his dynastic predecessors. For Jovian, it meant presenting weakness as principle by bolstering a largely status quo court with a few loyalists and former officials, and portraying this as part of a broader programme of tolerance and conciliation. For all their differences in

¹⁰² Lenski 2002: 105–7. ¹⁰³ Lenski 2002: 57. ¹⁰⁴ PLRE I Leo 1 would become magister officiorum, Equitius 2 would become magister militum and consul for 374. ¹⁰⁵ Tritle 1994. ¹⁰⁶ PLRE I Florentius 3 = Florentius 5, Nebridius 1.

    

171

worldview and strategy, both men thus utilized the power of patronage in tandem with other imperial tools such as rhetoric and legislation to influence the behaviour of imperial elites and advance a coherent message. The desirability of high office and a newly proactive understanding of emperorship thus allowed emperors of the fourth century to shape their courts in a manner that would have been unimaginable to their predecessors in the Principate.

8 Sharing the Imperial Limelight The Age of the Magister Militum Meaghan McEvoy

1. Introduction In the summer of 472, civil war raged in and around the city of Rome between the western Emperor Anthemius and his magister militum and son-in-law Ricimer. It was at this time that the eastern Emperor Leo I sent the aristocrat Olybrius from Constantinople to his fellow emperor in the West, bearing the following message (according to Malalas): I put to death Aspar and Ardaburius, so that nobody should oppose my orders. You too should execute your son-in-law Ricimer, to prevent him giving you orders . . . reign as one who gives orders rather than as one who takes them.¹

The remarkably long-serving magister militum Fl. Ardaburius Aspar and his son Ardaburius Iunior (a former magister militum himself ) had indeed been executed in a palace plot launched by Leo I the previous year. Anthemius, Leo urged, should do the same to remove his own overbearing magister militum, Ricimer. These men belonged to a series of generals who dominated the Roman imperial courts in the fifth century, including Stilicho, Constantius, Aetius, and Ricimer in the West, and most prominently Aspar in the East. These generals have often been characterized as ‘the power behind the throne’ in modern scholarship on the fifth century, including recently by Goffart, McCormick, Heather, Mathisen, and Lee, to name but a few.² And this is, indeed, a description which I have used myself, writing of the early years of Valentinian III’s rule, that: The fight for the real power in the realm was between the western generals, a struggle conditioned by the institutionalisation of child-rule which had taken

¹ Joh. Mal. 14.45: Ἐγὼ ἐφόνευσα Ἄσπαρα καὶ Ἀρδαβούριον, ἵνα μηδείς μοι ἐναντιοῦται κελεύοντι· ἀλλὰ καὶ σὺ φόνευσον τὸν γαμβρόν σου Ῥεκίμερ, ἵνα μὴ ἐπάνω σου κελεύῃ . . . βασίλευσον κελεύων καὶ μὴ κελευόμενος. (trans. Jeffreys et al. 1986). ² Goffart 1989: 79; McCormick 2000: 146; Heather 2006: 462; Mathisen 2009: 322; Lee 2013b: 96. Meaghan McEvoy, Sharing the Imperial Limelight: The Age of the Magister Militum In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0009

   

173

place up to this point, and which created the expectation that a single military leader behind the throne would be the true wielder of power.³

In the course of more recent research, however, I have come to question just what we mean by this term ‘power behind the throne’, and what it implies. The connotations of this phrase in itself are inherently negative, denoting some sort of concealed, illegitimate, and generally malignant exercise of power through a weak or incompetent ruler. This understanding is confirmed by the Oxford English Dictionary, which defines the phrase as ‘a person without constitutional status who covertly exercises power by personal influence over a ruler or leader’.⁴ But is ‘covert’ influence really the sort of power that the magistri militum of the fifth century were exercising? In this chapter, I will address three key questions. Firstly, I will examine why it was that generals came to be the dominant powers at Roman imperial courts in the fifth century, especially in the West but also to some degree in the East. Secondly, I will analyse how their power and place at court and in the regime at large was articulated, paying particular attention to their involvement in ceremonial occasions, the erection of statues in their honour, and public statements of their importance such as marriages into the imperial family. These acknowledgements suggest that in fact, their power was far from hidden behind the throne. The third question to be examined, therefore, is how it changes our conceptions of this period if we regard these magistri militum as ‘power-sharers’ rather than as ‘powers behind the throne’.

2. Why Were Generals So Dominant? The dominance of magistri militum (sometimes termed ‘generalissimos’) at the late Roman imperial court has been frequently noted by scholars, usually with a particular emphasis on the fifth-century West.⁵ Yet the question of why this dominance came about is often overlooked. One major factor was child emperor rule, which I have discussed extensively in my previous work. For the purposes of this chapter, it is sufficient to emphasize that the repeated accessions of child emperors in both the East (Arcadius, Theodosius II) and the West (Gratian, Valentinian II, Honorius, Valentinian III) from the late fourth century onwards resulted in the delegation of the military leadership role of emperors to their dominant generals. This represented a major change in the type of imperial leadership from that which had prevailed across much of the fourth century, when emperors such as Constantine I, Constantius II, Valentinian I, and Theodosius ³ McEvoy 2013: 312 (emphasis added). See also McEvoy 2016a: 500, n. 96. ⁴ OED s.v. power, n¹ P13. ⁵ Note, for example, O’Flynn 1983 and Wijnendaele 2017a.

174

   

I had all personally campaigned with their armies. And, as the role and function of the emperor changed in the late fourth and fifth centuries, so did the roles and functions of his key officials.⁶ The position of magister militum, the highest command in the Roman military, is thought to have originated towards the end of the reign of Constantine I, at the same time that the military responsibilities of praetorian prefects were removed.⁷ While both the eastern and western Roman armies were headed by magistri militum from the late fourth century onwards, their numbers differed in each half of the empire. According to the Notitia Dignitatum, there were two positions of magister militum in the West (which were often subsumed into one overarching command post, that of magister utriusque militiae). There were five in the East, of whom the most influential were the two generals who were in attendance upon the imperial court, who were known either as the magistri militum in praesenti or magistri militum praesentales.⁸ In the past, scholars such as A. H. M. Jones have suggested that the greater number of magistri militum in the East prevented the emergence of a single dominant general there (in contrast to the West),⁹ but, as we shall see, this was not always the case. In assessing the reasons for the dominance of magistri militum at court, we should take into account the length of their tenures. Over the course of the fourth century, magistri militum tended to serve longer than their highest-ranking civilian counterparts, such as praetorian prefects or magistri officiorum.¹⁰ Magistri militum frequently served for up to five years, while praetorian prefects tended to serve three years or less, and magistri officiorum between one and two years.¹¹ The differences between the length of tenure of civilian and military offices became more marked across the course of the fifth century. While there were always exceptions—such as the nine-year term of the praetorian prefect of the East, Anthemius (405–14), and the thirteen-year tenure of Helion as eastern magister officiorum (414–27)¹²—in general the turnover of high-ranking bureaucrats became more rapid in the fifth century, with terms of between twelve and eighteen months often the norm. Yet at the same time that civilian tenures were shortening, the average time in office of magistri militum was extending considerably, in both East and West. The western Roman Empire was to see a succession of long-serving magistri militum across the fifth century. The general Stilicho served under Honorius for thirteen years (395–408, having first received the rank from Honorius’ father Theodosius I in 392–3), followed by Constantius for approximately ten years with the same emperor (c. 411–21), then Aetius under Valentinian III for nineteen ⁶ See McEvoy 2013 for the full discussion. ⁷ Zos. 2.33; Joh. Lyd. de mag. 2.10 (=3.40). For the development of the role, see Boak 1915: 117–22; Jones 1964: 174–5, 178, 608–11; Demandt 1970: 560–612, 702–26; Lee 2014: 103–4; Bileta 2019: 75. ⁸ Not. Dig. Occ. 5–7; Not. Dig. Or. 5–9. ⁹ Jones 1964: 174, 342. ¹⁰ For these officials, see Jones 1964: 368–72. ¹¹ See the tables in PLRE I: 1047–52, 1059–61. ¹² PLRE II Anthemius 1, Helion 1.

   

175

years (432–54), and Ricimer under multiple emperors for sixteen years (456–72).¹³ These generals managed to assert their dominance in their capacity as the sole overall military commander, the magister utriusque militiae. In the East, we see similarly lengthy tenures taking hold. The general Plinta, for example, described by the ecclesiastical historian Sozomen as ‘the most powerful man at court at that time’ (δυνατώτατος τότε τῶν ἐν τοῖς βασιλείοις γεγονώς), was a magister militum for nineteen years (418–38).¹⁴ His contemporary, the general Areobindus, held the rank of magister militum for fifteen years (434–49).¹⁵ Fl. Ardaburius Aspar, who appears to have been Plinta’s son-in-law, is attested as a magister militum for a staggering forty-seven years, from c. 424 until his murder in 471. As the career of Aspar attests, even allowing for the different military structure of the East and the multiplication of magistri militum, it remained possible for one general to become the dominant figure of the court at Constantinople, just as in the West.¹⁶ And indeed, the longterm nature of appointment to this office is acknowledged in our sources: when Emperor Zeno was staging his come-back following the usurpation of Basiliscus, he won over the general Armatus, then serving under the usurper, with the promise of the rank of magister militum for life if he changed sides.¹⁷ These remarkably lengthy tenures of office far outstrip records of service we have for any civilian high office-holders during this period. They are crucial to explaining why it was the magistri militum who became the dominant figures of court and government. In some respects, it is no great surprise that generals should serve longer terms in office than bureaucrats: their careers and successes depended upon winning the loyalties of their soldiers (and indeed their wars) in a way that a praetorian prefect’s tenure of office did not. As noted above, the long service of magistri militum in the fifth century also coincides with an era of largely non-military emperors (such as Honorius, Valentinian III, and Theodosius II), whose role in the active waging of war was delegated to such generals. Steady military leadership was all the more necessary under these emperors, and longterm service of generals could bring a measure of stability and security to relations between the non-campaigning emperor and his armies.¹⁸ The issue of the generals’ age also needs to be taken into account. Given our evidence for military career paths,¹⁹ we would not expect a man to be appointed ¹³ PLRE I Stilicho; PLRE II Constantius 17, Aetius 7, Ricimer 2. There are also some long-serving magistri militum in the fourth-century West whose careers can be seen as precedents to this scenario, such as Bauto, who was magister militum from 380–5 (PLRE I Bauto), and Arbogast, who served as magister militum from 388 to 394 (PLRE I Arbogastes). ¹⁴ Soz. 7.17.4; PLRE II Plinta. See further Elton 2009: 136. ¹⁵ PLRE II Ariobindus 2. ¹⁶ For the long career of Aspar, see PLRE II Aspar; Croke 2005a; McEvoy 2016a. For Aspar as ‘the power behind the throne’, see Heather 2006: 462. ¹⁷ Joh. Mal. 15.5; Chron. Pasch. s.a. 478. ¹⁸ McEvoy 2013: 321–5. ¹⁹ For the age of recruitment as nineteen or older, see Jones 1964: 616–17; Lee 2007: 77–8. The lengths and trajectories of army careers are difficult to estimate, but the sons of officers benefited from accelerated promotion, and it was rare for senior generals to rise from the ranks (see Jones 1964: 637–8, 643).

176

   

magister militum until between the age of thirty and thirty-five, while promotion to the very highest post of magister militum praesentalis is likely to have come still later in life. If we assume that Stilicho was thirty-five at the time of his initial appointment as magister militum in 394, he will have been at least forty-nine at the time of his death in 408.²⁰ The later western magister militum Aetius, elevated in 433 and killed in office twenty-one years later in 454, would, on this same basis, have been fifty-six or older when he died.²¹ The general Ricimer, whose birth likely occurred c. 418, would have been around thirty-eight years of age when he commanded troops as a magister militum in 456, and fifty-four when he died in 472.²² Similarly in the East, if we assume that Plinta was aged thirty-five at the time of his promotion to the rank of magister militum praesentalis in 418, then by the time of his apparent (natural) death in office, in 438, he would have been at least fifty-five years of age. And finally, Aspar, who as the son of a magister militum, probably benefited from accelerated promotion, was elevated to the position of magister militum c. 424 and killed in 471. He was probably a little younger than the other cases described above, possibly thirty at the time of his first appointment as magister, making him seventy-seven years old at his death. These are only approximate estimates, but they suggest that leading generals frequently served into their fifties and sixties. Of course, civilian officials could be appointed in their fifties or later as well, but, with a few exceptions, such as the long-serving praetorian prefect Anthemius, who probably died in office in 414, they tended not to serve long enough to grow old in one specific position. The ages to which the magistri militum lived have twofold significance. Firstly, they suggest that these generals did not tend to be replaced on a regular basis as part of a natural turnover in appointments, as was generally the case with other offices at court, such as the comites consistoriani, but continued to serve until their death (natural or otherwise). There are few attested examples of fifth-century magistri militum stepping down from office, either willingly or unwillingly.²³ Given the obvious dangers of a profession based on the waging of war, we might also expect that many magistri militum would have died on campaign, ²⁰ Claud. Epithal. 323–5 refers to Stilicho’s white hair when describing the marriage of Honorius and Maria in 398, and again in his poem on the Gothic wars (de bello Get. 450–68), when he relates the story of Stilicho’s relief of Milan during its siege by Alaric in 402. Of course, we cannot assume his white hair was necessarily a sign of old age. ²¹ My estimates are similar to those of other scholars. Stickler 2002: 25 and Wijnendaele 2017b: 469 propose that Aetius was born in the early 390s. Clover 1971: 30 argues, on the basis of the language of the court poet Merobaudes in his composition for the first birthday of Gaudentius, Aetius’ son, that Aetius was already forty-six or older in 441/2, which would make him aged at least sixty at his death in 454. ²² For discussion, see Gillett 1995: 383. ²³ One example, however, is that of Aspar’s son Fl. Areobindus, who was removed from office as magister militum per Orientem in 466 as the result of allegations of treasonous correspondence with Persia (V. Dan. Styl. 55; Croke 2005a: 160–1). The magister militum Merobaudes was recalled by Valentinian III from a campaign in Spain in 443, but it is not clear whether he was dismissed from office (PLRE II Merobaudes). For fourth-century cases, see Lee 2014: 114–15; Bileta 2019: 88.

   

177

but this is less often the case than one might assume.²⁴ The most senior magistri militum, those traditionally described as ‘powers behind the throne’ and who operated in partnership with the emperor, often met violent deaths while still in office. This is shown by the coup that ousted Stilicho and led to his death in 408, the murder of Aetius by Valentinian III in 454, and the killing of Aspar at the behest of Leo I in 471.²⁵ Their long tenures gave these generals the opportunity to build up extensive patronage networks within the court and administration, and, of course, within the military itself.²⁶ When they fell from power, it was usually the case that their associates and supporters did so as well.²⁷ Secondly, the advanced ages of these long-serving generals raises the question of whether, as they grew older, they made a transition from being primarily campaigners to being primarily courtiers. For surely the question must be asked: were the active campaigning duties of an elderly magister militum ever delegated? This is not an easy question to answer, since our sources provide little in the way of evidence on this point. One potential indication may be found in the known activities of Aetius in the West from the 440s onwards: as far as our limited sources allow us to conclude, from this point he seems to have spent more time at court than on campaign.²⁸ Yet he nevertheless later led the Roman forces and their confederates personally at the Battle of the Catalaunian Fields against Attila in 452: he therefore remained the active commander of the western Roman army in a time of major crisis.²⁹ A further indication may survive in the fragmentary history of Priscus, who reports on what is generally seen as the last active campaign of the eastern general Aspar, against the Goths and Huns in 467. However, as I have argued elsewhere, the fragment, while mentioning Aspar several times, does not specify that he himself was present, and when referring to actions on the campaign, it attributes them to officers ‘on Aspar’s staff ’ or to ‘Aspar’s men’, rather than to the general himself.³⁰ It is not possible to state conclusively that the active duties of older magistri militum were delegated while they themselves still maintained their rank, but it is a distinct possibility. This may have meant that magistri militum were more present and active in court life in their later years.

²⁴ One exception is the magister militum per Gallias, Litorius, who was killed in or soon after a battle with Visigothic forces in Gaul in 439 (PLRE II Litorius). ²⁵ On the fall of Stilicho, see Matthews 1975: 275–83; O’Flynn 1983: 56–62; McEvoy 2013: 180–6; for Aetius, see O’Flynn 1983: 88–103; McEvoy 2013: 295–7; for Aspar, see Croke 2005a: 195–200; McEvoy 2016a: 491–4. ²⁶ As we see, for example, in the patronage networks of Aspar (Croke 2005a: 149–57; Lee 2013a: 100–2). ²⁷ This is discussed in Section 5, below. ²⁸ See further McEvoy 2013: 271–2, where it is also suggested that Aetius knew it was important to remain close to Valentinian III as the young emperor reached adulthood if he wished to maintain his influence. ²⁹ On the Battle of the Catalaunian Fields, see Oost 1968: 296–8; Täckholm 1969: 259–76; Clover 1973: 114–15; Heather 2006: 338–40. ³⁰ Prisc. fr. 49; McEvoy 2016a: 490, n. 34.

178

   

Overall, magistri militum were rare figures of continuity in imperial administration, serving for very lengthy periods, and often under successive emperors. The magister militum Aspar held office under the emperors Theodosius II, Marcian, and Leo I, for example, while Ricimer served under multiple emperors from Majorian to Olybrius (although admittedly, he was also responsible for the fall of a number of them and the installation of their replacements).³¹ Magistri militum were members of the imperial consistorium, the emperor’s chief advisory body, and their long-term service will have meant that they were among a very small number of key advisors who remained part of the consistory for years or even decades.³² While other office-holders, like eunuch chamberlains or female members of the imperial family, might similarly exercise influence over the emperor for years at a time, none of these individuals had official roles in the decision-making process of government in the way that the magistri militum did.³³ Senior generals also possessed a high level of access to the emperor and could speak to him privately, a privilege which was not available to all military or civilian office-holders.³⁴ For example, when Valentinian III personally murdered Aetius in 454, the general had come to the palace for a private meeting with the emperor, which was presumably a routine occurrence.³⁵ The generals’ long tenures in office, patronage networks, and positions at court made them the natural partners for the non-campaigning emperors of the fifth century.

3. How Was Their Power Articulated? Magistri Militum at Court With this picture of the long-term office-holding of generals (and its implications) in mind, we can turn to our second question: how was the power of magistri militum articulated in this period? One way to assess this is to examine the records of fifth-century state ceremonial which involved members of the court, which shows the many public honours and distinctions that the generals received.³⁶

³¹ For Aspar’s service, see Croke 2005a: 149–57; McEvoy 2016a: 484–92. For Ricimer’s role in the rise and fall of emperors, see O’Flynn 1983: 104–28, MacGeorge 2002: 178–261; Halsall 2007: 266–78; McEvoy 2017: 97–102. ³² On membership of the consistory, see Jones 1964: 333, and on the role of the consistory in the fifth century, Elton 2009: 138; Harries 2013: 73–86. Kevin Feeney examines the politics of appointments to the consistorium in Chapter 7. ³³ On the roles of different members of the court of Theodosius II competing for influence, see Elton 2009; Harries 2013. ³⁴ For the different grades of access and intimacy at monarchical courts, see the discussion in the Introduction of this book. ³⁵ Prisc. fr. 30.1.13–27. Elton 2009: 135 points out that the emphasis on reaching the emperor in his apartments assumes that he did not often come out. The privilege of a private meeting (as distinct from a session of the senate or consistory) was granted to very few individuals. ³⁶ For state ceremonial, see the Introduction.

   

179

(i) Panegyrics One of the clearest indications of the prominence of generals can be found in panegyrics. Although most of the surviving panegyrics of the late fourth and fifth centuries are devoted to emperors, many also contain references to the magistri militum.³⁷ These speeches were delivered on major occasions of state ceremonial, such as consular inaugurations, imperial adventus, and triumphs, either in the city in which the imperial court was resident or in the senates of Rome or Constantinople.³⁸ Among the most famous of these panegyrics are those written by the poet Claudian who, in the course of celebrating the consulships of Emperor Honorius, constantly lauded the talents and achievements of Honorius’ general Stilicho. Indeed, Claudian has long been understood as primarily dedicated to the service of Stilicho, thanks to the detailed studies of Alan Cameron.³⁹ We know that Claudian’s panegyrics for the third and fourth consulships of Honorius were delivered at the young emperor’s court at Milan in 396 and 398, respectively, while that for Honorius’ sixth consulship was recited at Rome in 404.⁴⁰ This delivery of these speeches to a court audience in the first two cases and to the court and senate in the third is crucial to our understanding of the public prominence accorded to Stilicho.⁴¹ The language used to describe Stilicho frequently emphasizes that the general was working in concert with the emperor, or even as a substitute for him.⁴² In Claudian’s panegyric for the third consulship of Honorius, for example, the poet describes a scene in which foreign peoples approach Stilicho seeking alliance with the Roman Empire: Germany swears allegiance to the absent Honorius and addresses her suppliant prayers to him . . . That which others were enabled to win by long wars—this, Honorius, Stilicho’s mere march gives to you.⁴³

A still more explicit statement comes in the panegyric of 404. Claudian conjures up a scene in which Roma asks Honorius to travel to the sacra urbs to celebrate his

³⁷ On the function and significance of panegyric in Late Antiquity, see McCormack 1981; Rees 2002; Omissi and Ross 2020. ³⁸ See Cameron 2013: 204–5. ³⁹ Especially Cameron 1970: 42, 46–62. ⁴⁰ For the date and location of the delivery of Claudian’s panegyrics, see Cameron 1970: xv–xvi; Gillett 2012: 290–2. For a detailed discussion of the events surrounding the delivery of Claudian’s panegyric for Honorius’ sixth consulship, see Kelly 2016. ⁴¹ Cameron 1970: 229 observes that of Claudian’s ten major political poems, eight were delivered at the imperial court in Milan. ⁴² On the presentation of the rule of Honorius, accompanied by his magister militum Stilicho, as a partnership, see McEvoy 2013: 162–9. ⁴³ Claud. De IV Cons. Hon. 448–59: iuratur Honorius absens imploratque tuum supplex Alamannia nomen . . . quod longis alii bellis potuere mereri hoc tibi dat Stilichonis iter (trans. Platnauer, Loeb). See also another of Claudian’s political poems, In Eutrop. 1.377–80.

180

   

sixth consulship. The emperor exhorts the citizens of Rome not to complain that he had not yet appeared in the city himself, since he had previously despatched Stilicho as his representative: I sent you Stilicho to sit in the curule chair to take my place, a consul instead of an emperor, a father instead of a father-in-law. In him your citizens also saw myself.⁴⁴

Similarly, in the 450s and the 460s, the panegyrics of Sidonius Apollinaris, though also dedicated to the emperors, made admiring references to the dominant general of the period, Ricimer. In the 467 panegyric for Anthemius, the general is described as ‘unconquerable Ricimer’ (invictus Ricimer). It is asserted that: If the Norican is restraining the Ostrogoth, it is that Ricimer is feared; if Gaul ties down the armed might of the Rhine, it is he that inspires the dread.⁴⁵

Indeed, panegyrists went beyond simply including magistri militum in their poems praising emperors. For not only was Stilicho fulsomely praised in Claudian’s poetry in honour of Honorius, but also he was the focus of a threebook panegyric—notably surpassing in length any of the panegyrics Claudian wrote for the emperor—for the occasion of his first consulship in 400. As Cameron has discussed, the first two books were delivered at the emperor’s court at Milan, while the third was delivered at Rome.⁴⁶ Celebration of Stilicho’s dedication to the military and his achievements as a commander is a focus of the panegyric. Claudian writes: He was always with the army, seldom in Rome, and then only when the young emperor’s anxious love summoned him thither. Scarce had he greeted the gods of his house, scarce seen his wife when, still stained with the blood of his enemies, he hastened back to the battle.⁴⁷

But in this particular scenario, in which the magister militum served a very young emperor, Claudian also focuses on the guidance Stilicho provided to Honorius:

⁴⁴ Claud. De VI Cons. Hon. 431–4: advectae misso Stilichone curules, ut nostras tibi, Roma, vices pro principe consul impleret generoque socer. vidistis in illo me quoque (trans. Platnauer, Loeb). For discussion, see McEvoy 2013: 168. Celebrations for Stilicho’s first consulship in 400 had been held in both Milan and Rome (Cameron 2013: 204–5). ⁴⁵ Sid. Ap. Carm. 2.377–8: Noricus Ostrogothum quod continet, iste timetur; Gallia quod Rheni Martem ligat, iste pavori est (trans. Anderson, Loeb). ⁴⁶ Cameron 1970: xv. ⁴⁷ Claud. De Cons. Stil. 1.116–19: Adsiduus castris aderat, rarissimus urbi, si quando trepida princeps pietate vocaret; vixque salutatis Laribus, vix coniuge visa, deterso necdum repetebat sanguine campum (trans. Platnauer, Loeb).

   

181

as a youth you teach him in secret a king’s lesson—his duty to his people; as a reverend senior you pay him honour and govern the empire at a father’s bidding; to your lord you give humble worship; you guide your master with obedience, your sire with love.⁴⁸

Claudian also emphasizes Stilicho’s fame beyond the Roman frontier, declaring to the general: ‘Embassies arrive from every quarter and in the presence of your sonin-law pray for a hundred voices to herald your renown.’⁴⁹ Honorius is duly honoured by the panegyrist, but in a way that nevertheless magnifies the importance of Stilicho as a warrior, diplomat, and guardian. The fact that Claudian could so openly celebrate Stilicho’s authority shows the prominence of the magister militum in the emperor’s regime and within the court community.⁵⁰ Later generals of the fifth century also had their own panegyrists to compose celebrations of their exploits, as an important recent study of Andrew Gillett has explored.⁵¹ In the 440s, the poet Merobaudes composed works in honour of both Valentinian III and the general Aetius. Merobaudes’ first poem celebrating Aetius, which took the form of a gratiarum actio, was delivered to the senate at Rome between 443 and 446.⁵² This was followed by a further panegyric for the magister militum on the assumption of his third consulship on 1 January 446.⁵³ Merobaudes’ praise of his subject is noteworthy for its emphasis on Aetius’ military prowess, his skills as a statesman, as well as the assertions of universal affection for him: But aside from distinction in battle, who is there who exhibits so great a celerity in planning, a strictness in judgement, a gentleness in conversation, a serenity of expression, a brevity of anger, and an enduring love? Oh most auspicious occasion of my oration! All men admit that I speak the truth; thus far they complain that I have left out certain things regarding your deeds.⁵⁴

⁴⁸ Claud. De Cons. Stil. 2.69–73: secreto consona regno ceu iuvenem doceas, moles quid publica poscat: ceu sanctum venerere senem patriisque gubernes imperium monitis; dominum summissus adores; obsequiis moderere ducem, pietate parentem (trans. Platnauer, Loeb). For further discussion, see McEvoy 2013: 168. See also Claud. De Cons. Stil. 3.118–23. ⁴⁹ Claudian, De Cons. Stil. 2.184–5: Undique legati properant generique sub ore in tua centenas optant praeconia voces (trans. Platnauer, Loeb, slightly modified). ⁵⁰ Claudian’s panegyric for the consulship of the civilian Manlius Theodorus in 399 was also presented at the court in Milan in January 399 (Cameron 1970: xv). We know of other panegyrics for the consulships of generals in the later fourth century, such as those by Libanius for Richomeres in 384 (Lib. Or. 1.219–20) and by Augustine for Bauto in c. 385 (Aug. c. litteras Petil. 3.25, 30), but neither of these survives. Sidonius Apollinaris (Ep. 8.6.5–6) also informs us of a panegyric delivered for the magister militum Astyrius in Gaul in 449. ⁵¹ Gillett 2012: esp. 271–80. See also Clover 1971: 34. ⁵² Clover 1971: 38. ⁵³ Clover 1971: 11. For the fragmentary panegyric of Merobaudes on Aetius, see Pan. II, at 13–15 (translation) and 64–8 (text), in Clover 1971. For the date and place of Merobaudes’ panegyrics, see Gillett 2012: 290–2. ⁵⁴ Merobaud. Pan. I. fr. IB, 9–16: iam vero praeter Martias laudes cuius tanta in consiliis alacritas, in iudiciis severitas, in conloquiis mansuetudo, in vultu aequalitas, in ira brevitas, in amore diuturnitas? o fortunatissimum orationis meae tempus: vera me dicere omnes fatentur, queruntur hactenus me de actibus tuis aliquanta omisisse (trans. Clover 1971). See also Pan. I. fr. IIB, 1–7, and Pan. II. 105.

182

   

Aetius’ rival generals in North Africa, Bonifatius and Sebastianus, also had their own dedicated panegyrists in their trains, according to Sidonius.⁵⁵ The orations celebrating the exploits of Bonifatius and Sebastianus have not survived, and they are not known to have been presented to the imperial court like those celebrating Stilicho or Aetius; nevertheless, their compositions reflect the trend of powerful generals having such men in their entourages to sing their praises. As Gillett has pointed out, this is a remarkable development known only from the late fourth century onwards, originating with the special relationship between Stilicho and Claudian. The practice of glorifying generals in a space previously reserved for the praise of emperors alone demonstrated that the attention of the court and senate had shifted from the emperor to the emperor and his magister militum.⁵⁶ It was a dramatic statement of the partnership rule which prevailed in the western Roman Empire during the reigns of Honorius and Valentinian III.⁵⁷ Unfortunately, we do not have similar surviving evidence from the East in the same era. However, given the frequency with which generals held the fasces in the fifth century (discussed in (iii), below), we should assume panegyrics celebrating the consulships of magistri militum were also delivered at the eastern court in Constantinople.

(ii) Adventus and Accession Ceremonial We have few extensive accounts of either adventus or accession ceremonial in the fifth century; however, those that do survive indicate that these powerful generals could be prominently involved in these state ceremonial occasions.⁵⁸ In the West, we find Claudian describing Emperor Honorius’ adventus into the city of Rome in 403 with his magister militum Stilicho actually riding alongside him in the imperial carriage. According to Claudian: Then it was, Stilicho, that Fortune repaid you for the labour of so many years when, mounted in the same chariot, you saw your son-in-law [i.e. Honorius] in his prime pass in triumph through the streets of Rome.⁵⁹ ⁵⁵ Sid. Ap. Carm. 9. 274–301; Gillett 2012: esp. 271–80. See Wijnendaele 2015: 71 on Bonifatius’ panegyrist. ⁵⁶ Statius’ now-lost De Bello Germanico had included mentions of Domitian’s advisors, but the presentation of the emperor’s own generalship is uncertain, given that only a few lines survive. On the tendency of Tetrarchic panegyric to glorify emperors rather than generals, see Davenport 2016: 394–5. ⁵⁷ See McEvoy 2013: 162–71, 268–72. ⁵⁸ On the adventus ceremony in Late Antiquity, see MacCormack 1981: esp. 17–89. ⁵⁹ Claud. De VI Cons. Hon. 578–83: Tunc tibi magnorum mercem Fortuna laborum persolvit, Stilicho, curru cum vectus eodem urbe triumphantem generum florente iuventa conspiceres illumque diem sub corde referres, quo tibi confusa dubiis formidine rebus infantem genitor moriens commisit alendum (trans. Platnauer, Loeb). On the Stilicho’s visibility in Honorius’ regime, see Cameron 1970: 42–4; Gillett 2012: 269–71; McEvoy 2013: 165–9.

   

183

The paternal presence of Stilicho alongside the young emperor sent a clear message, both to the assembled crowds in Rome and to the members of the court community who participated in these events, that the general was no ordinary courtier. Instead, he appeared as the emperor’s partner in imperial rule.⁶⁰ In the East, a similar scenario is reported on the occasion of the accession of Emperor Leo I in 457. This is described in a section of Peter the Patrician’s treatise on court ceremonial preserved in the tenth-century Book of Ceremonies, which, in this section, draws upon much earlier sources. The emperor’s ceremonial entry to Constantinople following his acclamation at the Hebdomon took place as follows: he [sc. Leo] sits in the carriage, preceded by the cross and the sceptres. The foremost patrician sits with him, or whomever he orders, kissing his hands, while the other archons precede them. When he comes to the Forum of Constantine, he gets down from the carriage and receives the eparch of the city and the senate. The leading senator, with the eparch of the city, offers him a gold crown.⁶¹

The identity of the ‘foremost patrician’ and the ‘leading senator’ is not here disclosed, but as Brian Croke has pointed out, in 457 the magister militum Aspar was both of these, and his prominent involvement in Leo’s accession must have sent a powerful message about his role in the new regime.⁶² The power and influence of Stilicho and Aspar was not restricted to domestic court ceremonial, where it could be hidden from the public at large, but was openly displayed in dynastic state ceremonial.

(iii) Consular Celebrations In the fifth century, magistri militum were frequently rewarded with the consulship. This remained the most prestigious honour in an emperor’s gift—despite the onerous financial burden of staging consular games—since consuls continued to give their names to the year.⁶³ Emperors usually dominated the consular fasti themselves, with the Emperor Honorius holding the consulship thirteen times, ⁶⁰ This may have been intended to evoke memories of Honorius’ adventus to Milan in 394, when he rode in the same chariot as his father Theodosius I, as described by Claudian at De III Cons. Hon. 126–41. See McEvoy 2013: 168–9. ⁶¹ Const. Porph. de cer. 1.91 (1.100): καὶ κάθηται εἰς τὴν καρούχαν, προηγουμένου τοῦ σταυροῦ καὶ τῶν περσικίων. συγκάθηται δὲ αὐτῷ καὶ ὁ πρῶτος πατρίκιος, ἢ ὃν ἂν ἐπιτρέψῃ, φιλῶν αὐτοῦ τὰς χεῖρας. οἱ γὰρ ἄλλοι ἄρχοντες προλαμβάνουσι. καὶ ἐλθὼν εἰς τὸν φόρον Κωνσταντίνου, κατέρχεται ἐκ τῆς καρούχας, καὶ δέχεται τὸν ὕπαρχον τῆς πόλεως καὶ τὴν σύγκλητον. προσφέρει δὲ αὐτῷ ὁ πρῶτος τῶν συγκλητικῶν μετὰ τοῦ ἐπάρχου τῆς πόλεως μοδίολον χρυσοῦν. (trans. Moffatt and Tall 2012). For the ‘archons’ as leading members of the court community, see Chapter 4. ⁶² Croke 2005a: 151–2. Such ceremonies developed what appears to have been the at least occasional Tetrarchic practice of praetorian prefects riding with the emperor, as noted in Chapter 3. ⁶³ Cameron 2013: 204–7; Lee 2013a: 104.

184

   

and his nephews Theodosius II and Valentinian III wielding the fasces on eighteen and eight occasions, respectively.⁶⁴ Such imperial domination severely restricted the opportunities for private individuals to receive the honour, and, as A. D. Lee has noted, this makes the number of magistri militum achieving the consulship in this period all the more significant.⁶⁵ In the West, between the full accession of Honorius in 395 and the deposition of Romulus Augustulus in 476, a total of thirteen consulships were awarded to magistri militum.⁶⁶ In the East, the number is even higher, with magistri militum awarded twenty-four consulships between 395 and 500, while three were also bestowed on the sons of magistri militum who had not yet achieved that rank themselves but may have been on that path.⁶⁷ Although private individuals without military careers did still achieve the consulship, the opportunities to do so became increasingly limited. Generals were the largest constituency beyond the imperial family to hold the highest honour in the empire on a regular basis. This is a clear reflection of the dominance of magistri militum in the political and social networks of the later Roman Empire. Furthermore, in the West in particular, there was also a pattern of magistri militum holding multiple consulships during their long periods in office. Stilicho was western consul in 400 and 405; Constantius in 414, 417, and 420; and Aetius in 432, 437, and 446.⁶⁸ The same trend cannot be traced in the East in the fifth century. However, the court at Constantinople did witness the emergence of military dynasties. These comprised multiple generations of the same family who rose to the rank of magister militum and were awarded the consulship. For example, the elder Fl. Ardaburius held the fasces in 427, followed by his son Aspar ⁶⁴ Imperial relatives also were awarded consulships in this period, such as Fl. Bassus Herculanus, the brother-in-law of Valentinian III and husband of the Augusta Honoria, in 452 (PLRE II Herculanus 2). ⁶⁵ Lee 2013a: 105. ⁶⁶ The western consuls were: Fl. Stilicho (400, 405), Fl. Constantius (414, 417, 420), Fl. Felix (428), Fl. Aetius (432, 437, 446), Fl. Ardaburius Aspar (434—though an eastern general, he was awarded a western consulship for his campaign in North Africa), Fl. Sigisvult (437), Fl. Astyrius (449, who received a panegyric, probably at Arles in Gaul, by Fl. Nicetius: Sid. Apoll. Ep. 8.6.5), and Fl. Ricimer (459). A further two consulships were also awarded to the comes Africae Heraclianus in 413, and Fl. Marcianus, the son of the emperor (and former magister militum) Anthemius and later a magister militum himself in the East, in 469. ⁶⁷ The eastern consuls were: Fl. Fravitta (401), Varanes (410), Fl. Constans (414), Fl. Plinta (419), Fl. Ardaburius (427), Fl. Dionysius (429), Fl. Ariobindus (434), Fl. Anatolius (440), Fl. Ardaburius Iunior (447), Fl. Zeno (448), Fl. Anthemius (455), Fl. Rusticius (464), Fl. Basiliscus (465), Fl. Zeno (469), Fl. Iordanes (470), Fl. Marcianus (472), Armatus (476), Illus (478), Trocundes (482), Fl. Theodericus (484), Fl. Longinus (486, 490), Ioannes Scytha (498), Fl. Ioannes qui et Gibbus (499), Fl. Patricius (500), Fl. Hypatius (500). The sons of magistri militum who were awarded consulships, but who had not at that point reached the rank of magister, were Iulius Patricius, the second son of Fl. Aspar (459), and Herminericus, the third son of Aspar (465). Fl. Dagalaiphus, the consul of 461, was also not a magister but came from a military family. Fl. Marcianus, the son of the magister militum and later western emperor Anthemius, had not been a magister at the time of his first (western) consulship in 469, but did hold that rank by the time of his second (eastern) consulship in 472. Two men holding the rank of comes domesticorum were also awarded the consulship in the East in this period: Fl. Sporacius in 452 and Fl. Aetius in 454. ⁶⁸ On occasion, civilians were also accorded the honour of multiple consulships, such as the senator Petronius Maximus in 433 and 443 (PLRE II Maximus 25), but this remained rare.

   

185

in 434 and each of Aspar’s three sons in the course of the 440s, 450s, and 460s. This made this military dynasty the most decorated non-imperial family in the eastern empire.⁶⁹ This consulship brought with it major public celebrations which further highlighted the generals’ prominence. Panegyrics were delivered to honour the consuls when they assumed the fasces, and examples of such orations for magistri militum have already been discussed above. Expensive commemorative items were also commissioned to mark the occasion. A number of consular diptychs survive for magistri militum, including one likely to depict Fl. Constantius on the occasion of either his 414 or 417 consulship, one certainly representing Fl. Felix in 428, and another the magister militum Astyrius in 449.⁷⁰ A similarly extravagant commemoration of the 434 consulship of the magister militum Fl. Ardaburius Aspar is found in the form of the silver Missorium of Aspar.⁷¹ In the East, the assumption of the consulship took place at Constantinople, which was also the seat of the court. In the West, as Cameron has argued, the weight of evidence points to consular inaugurations and games usually occurring in the city where the court was resident, rather than in Rome.⁷² Merobaudes provides us with an indication of the pomp of these ceremonies in his description of Aetius’ inauguration: Thus, while the leader regains the peaceful rewards of the toga and orders the consular chair, now at peace, to abandon war trumpets, these very wars have given way everywhere in admiration of his triumphal attire.⁷³

This passage is significant for the way in which it openly celebrates Aetius’ achievements domi militiaeque, in both civilian and military capacities. This was a major change from the situation that prevailed in the late third and fourth centuries, in which panegyrists tended to minimize or pass over the achievements

⁶⁹ On the military dynasties of the East, see McEvoy 2019a. Fl. Plinta, the consul of 419, is also believed to have been Aspar’s father-in-law (Croke 2005a: 152–3). Both Ardaburius’ consulship of 427 and Aspar’s of 434 were western consulships, following eastern military campaigns in western territories. ⁷⁰ On the diptych believed to depict Fl. Constantius, see Cameron 1998 and 2015, although he identifies the honorand as the eastern consul for 414, Constans, the magister militum per Thracias. For arguments in favour of Fl. Constantius, see Engemann 1999; Bühl 2001; Olovsdotter 2005: 20–3, 99–100. On the diptych of Felix, see Olovsdotter 2005: 23–4 and Cameron 2015: 251–8, and on that of Astyrius, Olovsdotter 2005: 25–6. ⁷¹ Zaccagnino et al. 2012; Cameron 2015: 275–80. ⁷² Cameron 2013: 204–5. There must have been some exceptions, however: Sidonius (Ep. 8.6.5–6) writes of witnessing the consular inauguration of the magister militum Astyrius, consul in 449. His inauguration (including the delivery of a panegyric in his honour) took place in Gaul, probably at Arles, which was not the seat of the court at this time. Cameron 2013: 206 discusses the inauguration, but does not note that it could not have been at court. ⁷³ Merobaud. Pan. II.30–3: sic tranquilla togae recipit dum praemia ductor pacatamque iubet lituos nescire curulem, ipsa triumphales habitus mirantia passim bella dedere locum (trans. Clover 1971).

186

   

of generals and attributed all military success to the emperor, who alone was able to excel domi militiaeque.⁷⁴ The consulship was an office of state, not of the court, but the ceremonies and panegyrics that celebrated Aetius and other generals did usually take place at court and in the city in which the court was resident. This highlighted their prominence as the most important non-imperial individuals in the court community.

(iv) Other Ceremonial Occasions: Marriages and Birthdays Weddings of emperors or members of the imperial family represented important dynastic state ceremonial occasions, not least because it was hoped that such marriages would lead to the birth of children, including future emperors. The marriage of magistri militum (or members of their families) into the imperial house is a key feature of their dominance of the court in this period, which will be discussed further in Section 4 of this chapter. Here, I wish to highlight how the wedding celebrations publicly articulated their position and prominence within the court. Claudian’s description of the marriage of Honorius to his first wife Maria emphasizes the part played by Stilicho, Maria’s father, in the proceedings: Meanwhile the army has laid aside its words: the soldiers are dressed in white and throng around Stilicho, the bride’s father. No standard-bearer nor common soldier fails to scatter flowers like rain and to drench their leader in a mist of purple blossoms.⁷⁵

The poet highlights the military followers whom Stilicho brought with him into the court community.⁷⁶ According to Claudian, the soldiers went on to sing of Stilicho’s nobility, majesty, modesty, and justice, declaring that they owed the emperor a firmer allegiance since the general was now the father of his bride. They concluded by urging Stilicho to ‘Crown your head with a garland, lay aside your rank for a moment and join our dances’.⁷⁷

⁷⁴ See Davenport 2016: 384–9 on the Tetrarchic panegyrics. Note particularly Pan. Lat. 10(2).6.4, a description of Emperor Maximian’s achievements domi militiaeque which exhibits marked similarities to Merobaudes’ description of Aetius. ⁷⁵ Claud. Epithal. 295–8: Candidus interea positis exercitus armis exultat socerum circa; nec signifer ullus nec miles pluviae flores dispergere ritu cessat purpureoque ducem perfundere nimbo (trans. Platnauer, Loeb). ⁷⁶ On military men as part of the court community, see Chapter 5 and Chapter 14. ⁷⁷ Claud. Epithal. 300–37: vincire corona; insere te nostris contempto iure choreis (ll. 336–7 trans. Platnauer, Loeb, adapted). See similarly Claud. Fescennine Verses 3.1–12. Stilicho must also have been present for Ambrose of Milan’s funeral oration for Emperor Theodosius in 395, an important state ceremonial occasion. See Amb. de ob. Theod. 5; McLynn 1994: 357–8, n. 3; Liebeschuetz 2005: 180, n. 3; McEvoy 2013: 144.

   

187

Still more central to the imperial marriage celebrations at Rome in 467 was the general Ricimer, who wedded Alypia, the daughter of the new emperor, Anthemius. In one of his letters, Sidonius reports how he arrived in Rome to petition the emperor amidst these festivities: Up until now I have not presented myself at the bustling doors of the emperor and his courtiers, for I arrived here at the moment of the marriage of Ricimer the patrician, whose union with the daughter of the immortal Augustus is a hopeful guarantee of the safety of the state . . . And now the bride has been given away, the bridegroom has put off his garland, the consular his embroidered robe, the brideswoman her gay mantle, the man of rank his toga, and the undistinguished citizen his cloak; nevertheless, the full pomp of the bridal ceremony has not yet subsided, for the bride has not yet passed to her husband’s home.⁷⁸

Sidonius’ account indicates that the whole court had been tied up with the celebrations, which meant that he had to delay his attempts to find a senator who could smooth his way to an audience with the emperor. Such an event publicly confirmed the prominence of Ricimer as the most important figure in the court community and in the empire itself after—or alongside—the emperor. That Sidonius recognized the general’s prominence is suggested by his description of marriage being undertaken ‘in hopeful guarantee of the safety of the state’ (in spem publicae securitatis). Beyond the realm of dynastic state ceremonial lay the daily life and routine of the Roman imperial court. We usually conceive of this in terms of imperial audiences, meetings of the consistorium and so forth, since these are the events best attested in our literary sources. However, sometimes we can catch a glimpse of the social life of the court in occasions that did not constitute dynastic state ceremonial in the same way as accessions and weddings, but were nevertheless significant events which brought the court together as a community. In 441/2, the poet Merobaudes composed a work to mark the first birthday of Gaudentius, the son of the general Aetius. The delivery of the poem and the birthday celebrations apparently took place at Rome, with Merobaudes urging Aetius to ‘ease his warlike heart (he is worthy of the staff of retirement); and let him embrace his milk-white offspring with muscular arms’.⁷⁹ Merobaudes rejoices that the baby, born in Rome, can see the royal dwelling place of Quirinus and the hearth of Mars,

⁷⁸ Sid. Ap. Ep. 1.5.10–11: necque adhuc principis aulicorumque tumultuosis foribus obversor. interveni etenim nuptiis patricii Ricimeris, cui filia perennis Augusti in spem publicae securitatis copulabatur . . . iam quidem virgo tradita est, iam coronam sponsus, iam palmatam consularis, iam cycladem pronuba, iam togam [senator] honoratus, iam paenulam deponit inglorius, et nondum tamen cuncta thalamorum pompa defremuit, quia necdum ad mariti domum nova nupta migravit (trans. Anderson, Loeb). ⁷⁹ Merobaud. Carm. IV.12–14: laxet pectora bellicosa ductor (est dignus rude) lacteamque prolem nodosis ferus ambiat lacertis (trans. Clover 1971).

188

   

since he himself is ‘an offspring of Mars’ (proles Martia).⁸⁰ That the infant son of the magister militum should merit a birthday poem by the same poet who composed orations for the emperor, and that these celebrations should take place at Rome with the emperor likely in residence there,⁸¹ indicates the centrality of Aetius and his family to the court community, as well as the extent to which the panegyrist was first and foremost part of the retinue of the general, rather than that of the emperor.

4. Public Acknowledgements (i) Statues to Generals From the late fourth century onwards, the statues of generals began to be erected in the Forum Romanum at Rome. As Robert Chenault has explored, before this period the Forum remained very much an imperial space, adorned by monuments of emperors, in contrast to the Forum of Trajan, where statues of senators were frequently erected during the Principate and in Late Antiquity.⁸² The new statues represented a landmark in the public acknowledgement of the importance of Rome’s generals to the state. Senators had ceased to command armies in battle in the late third century, which means that it had taken more than a century for the status and prestige of the new military elite, and its senior representatives, the magistri militum, to reach a level that demanded public commemoration at Rome.⁸³ The general Stilicho alone may have had up to four statues in the Forum Romanum, including a gilded statue placed on the western rostra, a remarkably prominent position for a man who was not the emperor. The inscription on the base declares that: The Roman people, on account of his unique love and providence in its regard, have decreed that a statue of bronze and silver be placed on the rostra as a memory of [Stilicho’s] everlasting glory.⁸⁴

⁸⁰ Merobaud. Carm. IV.38–40. On the likelihood of both Aetius and Gaudentius being in Rome for the celebrations, see Clover 1971: 30. ⁸¹ Valentinian III was at Rome in August 442 (Nov. Val. 2.2) and in Spoleto in September 442 (Nov. Val. 7.2); the only other records for this period indicate his presence at Ravenna in late January–midFebruary 441 (Nov. Val. 8.2; 10). See further Gillett 2001: 143. ⁸² Chenault 2012: 124. As Chenault goes on to point out (pp. 125–6), it is misleading to class dedications to these generals as senatorial as some scholars have done (e.g. Bauer 1996: 405–8), since their positions were more complex, and more akin to that of a chief minister. Niquet 2000: 86, 262–4 likewise presents a more nuanced position. ⁸³ Davenport 2015: 285; 2019: 598. ⁸⁴ CIL VI 1731 = 1195 = LSA 1437 (ed. C. Machado): Populus Romanus, | pro singulari eius | circa se amore | adque providentia, | statuam ex aere argentoque | in rostris ad memoriam | gloriae sempiteranae | conlocandam decrevit, | exequente Fl(avio) Pisidio Romulo, v(iro) c(larissimo), | praef(ecto) urb(i). As

   

189

Such monuments are also attested by comments made by Claudian in his panegyric for the consulship of Stilicho, where he refers to Stilicho’s form in gilded statues, and exhorts the Roman populace to ‘look with joy upon the sacred face you worship cast in bronze and marvel at in gold’.⁸⁵ Although we have little information about statue dedication ceremonies in Late Antiquity, we should assume they were important occasions, bringing together officials and the public (including members of the court community, if in residence at the time) in celebration of the honour paid to the recipient.⁸⁶ While no other general of the era is known to have been awarded as many statues as Stilicho, bases for statues of the western generals Constantius and Aetius have also been found in the Forum Romanum.⁸⁷ Two inscribed bases (now lost) record the statues to Constantius erected in 420, the year before his elevation as Augustus. One describes him as ‘the restorer of the state, and parent of the most unconquered princes’.⁸⁸ Similarly, a surviving base preserves a partial inscription honouring the general Aetius. The statue was awarded to the magister militum by the emperors Theodosius II and Valentinian III and the senate and people of Rome. This statue also appears to have stood in a highly prominent position, an area behind the senate house known as the Atrium Libertatis.⁸⁹ As Chenault observed, the influence of these senior generals extended far beyond their military offices, often reaching into the imperial family itself; as such, they were more comparable to emperors than to senators, whose statues continued to appear in the Forum of Trajan.⁹⁰ In the East, the paucity of inscriptions and statue bases from Constantinople means that there are fewer documented examples of statues for generals. Instead, we depend largely on testimony in later literary evidence, such as the Patria, which records that:

Chenault 2012: 126 notes, the location on the rostra was ‘of the highest symbolic importance and usually reserved for emperors alone’. For further statues of Stilicho, see LSA 1363 (erected by the senate and people of Rome; the identity of the honorand remains debated, but the inscription certainly referenced Stilicho, his name being obliterated following his fall in 408) and LSA 1436 (awarded by the senate). For further discussion, see Bauer 1996: 20–1; Messerschmidt 2004: 555–68 at 559, no. 4. LSA 1490 (awarder unknown) is also likely to be a dedication to Stilicho. ⁸⁵ Claud. De Cons. Stil. 3.11–12: os sacrum, quod in aere colis, miraris in auro, cerne libens (trans. Platnauer, slightly adapted). See also Claud. De Cons. Stil. 2.173–81. ⁸⁶ Some statue inscriptions do refer to dedicatory celebrations—see e.g. CIL XI 3303 (Italy), CIL VIII 1548 (Africa). John Chrysostom famously objected to the noisy celebrations surrounding the dedication of a statue to the Augusta Eudoxia in Constantinople in the early fifth century (Soc. 6.18; Soz. 8.20). ⁸⁷ On the statues for Stilicho establishing a pattern for later generals, see Chenault 2012: 127–8. ⁸⁸ CIL VI 1719 = LSA 1423 (awarded by the urban prefect): Reparatori rei publicae [et] | parenti invictissimo[rum] | principum. For the other lost base for a statue to Constantius, see LSA 1424. On the title of parens being used to describe individuals such as Stilicho, Constantius, and Aetius, see Boak 1915: 139; Mazzarino 1942: 106–13; Straub 1952: 94–112; O’Flynn 1983: 5. ⁸⁹ CIL VI 41389 = LSA 1434; Machado 2006: 161–2; Chenault 2012: 127–8. ⁹⁰ Chenault 2012: 125–6.

190

   

A statue of Aspar is preserved up to the present day in the region of the Tauros, mounted on a strong powerful warhorse, as can be seen.⁹¹

An equestrian statue of a late Roman general is a conspicuous and unmistakeable image of his importance and may even be interpreted as a statement of equality with the emperor: no other individuals below the rank of emperor are reported to have been honoured with statues on horseback in this era.⁹² Public statues in the cities of Rome and Constantinople could only be erected with the direct approval of the emperor. As John Weisweiler has observed, we should not therefore view statues of generals as symptomatic of imperial weakness.⁹³ These statues were rather an acknowledgement of the very public prominence of magistri militum, and, in that sense, a statement of the imperial and community consensus that legitimated that prominence.

(ii) Official and Semi-Official Titles, Public Acclamations, and Acknowledgements Across the course of the fifth century, generals were often accorded the high honour of patrician status, although, as with the consulship, magistri militum were not alone in being accorded such a distinction.⁹⁴ Yet as A. D. Lee recently and before him A. E. R. Boak have pointed out, legislation from the period indicates ever more exalted epithets being used to refer to magistri militum.⁹⁵ In a law of 441, for example, Emperor Theodosius II refers to his general Areobindus as ‘the magnificent man, master of the soldiers’ (virum magnificum magistrum militum).⁹⁶ In 440, when the Vandals were threatening the coast of Italy, Valentinian III issued a law referring to ‘the most excellent man, the patrician, our Aetius’ (excellentissimum virum patricium nostrum Aetium). In a law of 445 ⁹¹ Patr. Const. 2.99 = LSA 353 (ed. U. Gehn): Ἄσπαρος δὲ στήλη ἐν τοῖς τοῦ Ταύρου μέρεσιν σώζεται ἕως τῆς δεῦρο ἐν ἵππῳ ἐρρωμένῳ δεξιολαβεῖ καθεζόμενος ὁρᾶται (trans. Berger 2013). A base for a statue of Aspar is also recorded at Ulpia Augusta Traiana in Thrace (LSA 10) and one for his contemporary general Basiliscus in nearby Philoppopolis (LSA 367). Both men campaigned in the region, and the erection of statues to them reflects an awareness of their military and political prominence. ⁹² Jord. Get. 289 records that Emperor Zeno had an equestrian statue erected to Theoderic the Ostrogoth at the entrance to the palace in Constantinople. Theoderic did hold the rank of magister militum praesentalis under Emperor Zeno from 481, but it is unlikely the statue dates from this period; it is more likely that it belongs to the years after he established his rule in Italy in 493. See further the discussion in LSA 510 (ed. U. Gehn). Even if the Patria were mistaken (given its later date) in identifying an equestrian statue of Aspar at Constantinople, it testifies to the understanding of the degree of the general’s prominence in the politics of his day. ⁹³ Weisweiler 2012: 326–7, 335. ⁹⁴ The aristocrat Petronius Maximus, for example, was a patricius by 445 according to Nov. Val. 19 (see further PLRE II Maximus 25), as was the aristocrat Fl. Albinus by 446, according to Nov. Val. 21.1 (PLRE II Albinus 10). ⁹⁵ Boak 1915: 136–8; Lee 2013a: 104. ⁹⁶ Nov. Theod. 7.4.3 (441). See also O’Flynn 1983: 86–7.

   

191

he was named as ‘Aetius, dearest and most beloved father’ (Aeti parens karissime atque amantissime), while in 452 he was termed ‘the magnificent man, our father and patrician’ (magnifici viri parentis patriciique nostri).⁹⁷ As noted above, the use of the term parens to refer to men such as Aetius is also a significant mark of prominence and prestige, although this was not limited to magistri militum.⁹⁸ Acclamations also reflected the general awareness of the influence of magistri militum. In December 438, the Codex Theodosianus, which had been brought back from Constantinople by the delegation that had accompanied Emperor Valentinian III to the East for his marriage to Licinia Eudoxia, was promulgated at a meeting of the Roman senate. Famously, the senators hailed the new legal code with a lengthy series of acclamations, including acclamations in support of the magister militum Aetius.⁹⁹ The assembled senators are recorded as shouting: Hail Aetius! [repeated fifteen times] A third term for you in the consulship! [repeated thirteen times] Through your vigilance we are safe and secure! [repeated twelve times] Through your vigilance, through your labours! [repeated fifteen times]¹⁰⁰

As John Matthews has noted, these acclamations did not represent spontaneous actions but were closely planned and controlled.¹⁰¹ We have similar accounts in the eastern Roman Empire of generals being accorded acclamations at church councils, such as the general Zeno at the Council of Edessa in the 440s.¹⁰² The prominence of magistri militum is also reflected in the letters they received from bishops and churchmen, such as those sent by the eastern bishop Theodoret of Cyrrhus to the magistri militum Anatolius and Aspar.¹⁰³ These letters were acknowledgements of the influence of these generals at court, especially the fact that they had regular private access to, and influence with, the emperor.

⁹⁷ Nov. Val. 9 (440); 17.3 (445); 36.1 (452); McEvoy 2013: 282, 291–2. ⁹⁸ The aristocrats Fl. Albinus and Petronius Maximus were also referred to as parens in surviving legislation: see Nov. Val. 2.3 (443), 4 (440). ⁹⁹ Matthews 2000a: 39–49; see also Harries 1999: 65–9; Heather 2006: 128–32. ¹⁰⁰ Gesta of the Codex Theodosianus, 6: Aeti aveas. Dictum XV. Ter consulem te. Dictum XIII. Excubiis tuis salvi et securi sumus. Dictum XII. Excubiis tuis, laboribus tuis. Dictum XV (trans. Pharr 1952). As Matthews 2000a: 42 has observed, the acclamations of Aetius indicate that he represents military authority. ¹⁰¹ Matthews 2000a: 44. ¹⁰² AGWG, NF XV, pp. 17, 19, 23, 25. For the magister militum Zeno, see PLRE II Zenon 6. ¹⁰³ Theodoret to Anatolius: Ep. 79, 92, 111, 119, 121, 139; Millar 2006: 222–4; Lee 2013a: 93–4. Theodoret to Aspar: Ep. 139; McEvoy 2016a: 495–6. More generally on correspondence between bishops and generals in this era, see Lee 2007: 153–63. Generals might also engage in benefactions to churches themselves. For example, the gift of Ricimer of a mosaic at the church of Sant’Agata dei Goti in Rome (see Mathisen 2009) or that of Aspar to the Anastasius church in Constantinople (see Snee 1998: esp. 176–7; McEvoy 2016a: 496–7). Civic works might also be undertaken by magistri, such as Aspar’s building of a great cistern in Constantinople (see Janin 1964: 316; McEvoy 2016a: 498).

192

   

Contemporary legislation attests to the activities of generals in many spheres of government and administration beyond strictly military affairs. The novellae of Valentinian III, for example, offer us a picture of the involvement of the magister militum Aetius in a wide range of legal rulings in the 440s and 450s, including meat supplies to the city of Rome and the rescuing of impoverished children.¹⁰⁴ Generals were also seen as guarantors of weights and measures, as shown by a bronze weight from Rome, which records that it was authorized ‘by the blessings of our lords [sc. the emperors] and the patrician Ricimer’ (salvis dd. nn. et patricio Ricimere).¹⁰⁵ A novella issued by Emperor Majorian in 458 and addressed to the senate offers a vivid glimpse of the centrality—and visibility—of the magister militum Ricimer in the emperor’s regime and, in particular, how their partnership was publicly presented: Military matters will be the watchful concern of both ourself and our parent and patrician Ricimer. We shall, by the grace of God, protect the position of the Roman world, which we liberated, by our joint vigilance, from the foreign enemy and from internal disaster.¹⁰⁶

Such a partnership is all the more striking when we note that Majorian, unlike emperors such as Valentinian III and Honorius before him, was an adult at the time of his accession, and an active military man as well. Yet this system of a partnership rule had become ingrained in the West after years of child-emperorship, much as emperors like Majorian might have tried to fight against it.¹⁰⁷ The partnership between emperors and generals was understood by peoples and communities beyond the Roman frontiers. In the fifth century, we find instances of foreign peoples approaching magistri militum (rather than emperors) to establish alliances, or even viewing the alliances that they had forged as having been with the magistri militum themselves, rather than with the emperors or the Roman state. After the murder of Aetius, for example, Hydatius reports that Valentinian III took great care to send envoys to the many barbarian groups with whom agreements had been made by his deceased general, in order to assure them that the terms still stood.¹⁰⁸ ¹⁰⁴ See Nov. Val. 33 (32 Jan. 451) and Nov. Val. 36 (29 June 452). For discussion of Aetius’ involvement in civilian matters, see Delmaire 2008a: 291–4; O’Flynn 1983: 86; McEvoy 2013: 287–8. ¹⁰⁵ CIL X 8072. The precise date of this weight is unclear, but it must belong to the period 456–72. It has also been suggested on the basis of RIC X 188–92 that coinage issued under Emperor Libius Severus bore the monogram of Ricimer, but this remains disputed. See Woods 2002. ¹⁰⁶ Nov. Maior. 1: Erit apud nos cum parente patricioque nostro Ricimere rei militaris pervigil cura. Romani orbis statum, quem communibus excubiis et ab externo hoste et a domestica clade liberavimus, propitia divinitate servemus (trans. Pharr 1952). ¹⁰⁷ McEvoy 2013: 321–5. On this specific novella, see also MacGeorge 2002: 200–1. ¹⁰⁸ Hyd. Lem. 153 (161). Priscus (fr. 11.1.1–5) also refers to a treaty that Attila ‘made with Aetius, the general of the western Romans’.

   

193

(iii) Intermarriage One final, crucial aspect of the very public prominence of magistri militum in this era is their intermarriage with the imperial family. It is important to note that very few marriages between imperial women and non-imperial citizens occurred at all in the fifth century, and that those that did were almost all with military men. This is a significant reflection of their status and prominence, and an indication of the degree to which the loyalty of magistri militum was valued by the emperors, for there could be no greater confirmation of their standing than an alliance with the imperial house. These marriages changed the configuration of the court community, moving military men into the centre of the court, and pushing senatorial families who were not granted the privilege of such unions to its outer circles. The celebrations for the marriage of Maria, the daughter of the general Stilicho, to the Emperor Honorius in 398 have been discussed above in the context of the court poet Claudian’s celebratory verses. This was a relationship which was clearly at the forefront of Stilicho’s public presentation, as seen in the inscription on the base of one of the statues dedicated to him in the Forum Romanum. It records that Stilicho was: advanced to royal kinship by marriage as son-in-law of the deified Theodosius, comes of the deified Theodosius Augustus in all wars and victories, and appointed by Theodosius to a second royal relationship by marriage as fatherin-law of our lord Honorius Augustus.¹⁰⁹

Remarkably, in the first half of the fifth century, only three officially sanctioned marriages of women born to the imperial house occurred at all: the union of Thermantia, the grand-niece of Emperor Theodosius I, with her cousin Emperor Honorius in early 408, after the death of his first wife, her sister Maria; the marriage of Galla Placidia, the sister of Emperor Honorius, with the magister militum Constantius in 417; and finally the wedding of Licinia Eudoxia, the daughter of Theodosius II, in 437 with her cousin, Valentinian III.¹¹⁰

¹⁰⁹ CIL VI 1730 = LSA 1436 (ed. Machado): | adque victoriis et ab eo in adfinitatem | regiam cooptato itemque socero d(omini) n(ostri) | Honori Augusti. Stilicho’s first advancement to royal kinship, noted in the inscription, was his marriage to Serena, the niece and adopted daughter of Theodosius I, in 384 (see Oost 1968: 41; Cameron 1970: 56). ¹¹⁰ Thermantia and Honorius: Zos. 5.28.1; Olymp. fr. 3; McEvoy 2013: 180. Galla Placidia and Constantius: Olymp. fr. 33.1.1–7: Oost 1968: 139–44; Matthews 1975: 354–5, 377; McEvoy 2013: 213–15; Leonard 2019. Galla Placidia’s marriage to the Gothic leader Athaulf in early 414 was, of course, not imperially sanctioned: see Olymp. fr. 24.1–6; Oost 1968: 127–8; Matthews 1975: 316; and McEvoy 2013: 200–1. Licinia Eudoxia and Valentinian III: Matthews 2000a: 1–9. For further discussion of imperial marriages, see Croke 2015b; McEvoy 2019a; and Chapter 9 by Anja Busch.

194

   

In the second half of the fifth century, we know of eleven marriages of imperial women, and with a few senatorial exceptions, the majority of these saw the union of imperial women with military men, such as Ariadne and Zeno in c. 466 and Alypia and Ricimer in 467.¹¹¹ On this last occasion, as noted above, Sidonius Apollinaris describes how he arrived in Rome in late 467, amidst the celebrations for the wedding, which was seen as ‘a hopeful guarantee of the safety of the state’. Yet the alliance between Ricimer and his bride’s father, Emperor Anthemius, would soon turn sour. According to Ennodius, this later led Anthemius to lament that: we have reason for resentment against Ricimer. We have given him the greatest favours, but all for nothing. We have even (not without dishonour, it must be said, to our throne and to our own bloodline) allowed him to join our family through marriage, which we conceded for the love of our country though it would inflict shame on us. For who, among past emperors has ever done this before: that, among other gifts that needed to be made to a skin-clad Goth, an emperor included his daughter to gain peace for his people?¹¹²

Despite Anthemius’ protests, the intermarriage of imperial dynasties of the fifth century with the families of their key military leaders—Roman or otherwise—is a recurrent theme of the era. Stilicho had married each of his daughters in succession to Emperor Honorius in the 390s and early 400s; Aetius had betrothed his son Gaudentius to a daughter of Valentinian III in the 450s; and Aspar had seen his son Patricius married—although not for long—to a daughter of Emperor Leo I in c. 470.¹¹³ Indeed, there were precedents for such imperial marriages to military men (or members of their families) already in the late fourth century, as the weddings of Stilicho to Serena in the 380s and of Emperor Arcadius in 395 to Eudoxia, the daughter of the general Bauto, attest.¹¹⁴ Such occasions entailed ¹¹¹ The imperial marriages were those of Justa Grata Honoria and Fl. Bassus Herculanus; Pulcheria and Marcian; Euphemia Marciana and Anthemius; Licinia Eudoxia and Petronius Maximus (which took place against the Augusta’s will); Eudocia and Huneric; Placidia and Olybrius; Ariadne and Zeno; Alypia and Ricimer; Leontia and Patricius; Leontia and Fl. Marcianus; and Ariadne and Anastasius. ¹¹² Ennod. V. Epiph. 67–70: aduersus Ricemerem causa doloris sit et nihil profuerit maximis eum a nobis donatum fuisse beneficiis, quem etiam (quod non sine pudore et regni et sanguinis nostri dicendum est) in familiae stemma copulauimus, dum indulsimus amori reipublicae quod uideretur ad nostrorum odium pertinere: quis hoc namque ueterum retro principum fecit umquam, ut inter munera, quae pellito Getae dari necesse erat, pro quiete communi filia poneretur? (trans. Cook 1942). There was a precedent for the marriage in the betrothal of Eudocia, the daughter of Valentinian III to Huneric, the son of the Vandal king Geiseric, in the 440s. The imperial princess was married to Huneric after her kidnap by the Vandals during their sack of the city of Rome in 455. See Conant 2012: 20–36; Merrills and Miles 2014: 117. ¹¹³ Stilicho’s daughters: McEvoy 2013: 159–62, 180. Gaudentius: Clover 1971: 29–30; McEvoy 2013: 291–2. Patricius and Leontia: Croke 2005a: 191–3; McEvoy 2016a: 490–1. ¹¹⁴ On the marriage of Arcadius and Eudoxia, see Holum 1982: 51–3. Bauto was, however, already dead by the time of his daughter’s marriage, so he cannot be accused of benefiting politically from the arrangement.

   

195

major imperial celebrations in which, as we have already seen above, these generals could play a highly visible role that left no doubt of their political influence. Moreover, intermarriage with the imperial family surely sent an unmistakeable message about the centrality of magistri militum to the court community, and contributed to the public awareness of their role as partners of the emperors.

5. The Politics of Consensus? With such a wealth of evidence which highlights the consistent and public prominence of military men in the fifth century—celebrations in imperial panegyrics, honorific titles in legislation, acclamations, consulships, statues, and marriages into the imperial family—it is very difficult to view men such as Stilicho, Aetius, Aspar, or Ricimer as somehow hiding behind the throne to exercise power. In fact, their wielding of power was carried out in full view of the Roman court and the public, and was surely impossible to miss—and, indeed, was not intended to be missed. It should be characterized as a form of ‘power-sharing’ with the emperor rather than exercising ‘power behind the throne’. These magistri militum were at the very centre of the court society, positioned essentially as the equals of the emperors, and recognizing the true nature of the relationship between emperors and generals fundamentally alters our understanding of court politics. For how are we to interpret the fact that the influence of these dominant generals was publicly acknowledged and accepted by the emperors and their peers? Do we understand it as representing the politics of consensus—that is, that powerful courtiers from the civilian and bureaucratic elite accepted and were content with the partnership rule of generals and emperors? Or was it simply that the military elite had grown far too powerful to oppose? The evidence available suggests that the politics of consensus were in operation. Andrew Gillett has persuasively argued that the fifth-century trend of magistri militum retaining their own panegyrists represented the development of a new form of political communication directed at the senatorial aristocracy at court, reflecting their importance as a constituency whose views and opinions mattered to the generals.¹¹⁵ The significance of the aristocracy’s support can be seen in the interactions between the magister militum Stilicho and the senate of Rome in the late 390s and early 400s. Stilicho was particularly careful in his respect for senatorial dignity, frequently deferring to the opinion of the amplissimus ordo even when it was not strictly necessary. The events of early 408 show the mortal danger that Stilicho faced if he lost senatorial consent for his authority.¹¹⁶ In this year, Stilicho attended a meeting of the senate at Rome, where he insisted upon the

¹¹⁵ Gillett 2012: 280–90.

¹¹⁶ Matthews 1975: 278–83; McEvoy 2013: 177–80.

196

   

payment of reparations to the Visigoth Alaric, but was met with opposition from the senators who were more inclined to make war upon Alaric. According to Zosimus, after Stilicho explained his views: Everyone at this thought that Stilicho spoke reasonably and the senate voted that four thousand pounds of gold should be paid to Alaric, although most voted that way not from preference but out of fear of Stilicho. So Lampadius, a man of high birth and reputation, who murmured in Latin ‘non est ista pax sed pactio servitutis’, which means ‘this is slavery rather than peace’, fled to a nearby Christian church as soon as the senate was dismissed, for fear of punishment for his frankness.¹¹⁷

It is no coincidence that it was only a few weeks later that a court bureaucrat succeeded in inciting an army mutiny which brought about the execution of Stilicho and a major purge of his supporters. The steps taken by a general to maintain support for his position and politics, and the dangers which the loss of such backing posed, are both amply illustrated by Stilicho’s ultimate fate. Yet the remarkably long tenures of the magistri militum of this era reflect an awareness on the part of the court community that the partnership of a dynastically secure noncampaigning emperor (child or adult) and a dominant general was the most viable leadership option to ensure the continued stability of the state in the face of ongoing military threats. Consensus did not, of course, mean that every individual was content with the dominance of these generals, or remained so. In the case of Valentinian III and Aetius, the emperor himself was ultimately deeply unhappy about the powersharing relationship with which he had grown up. And as the senators’ fear of Stilicho at the senate meeting of 408 indicates, a general’s support from other vested interests might also be to some degree prompted by intimidation or awareness of the military resources at his disposal. Yet it is an important nuancing of our understanding of fifth-century politics to note that as a rule, these powerful generals did not simply impose themselves on an unwilling and resistant court. Furthermore, the remarkably long terms of office of magistri militum must have meant that they were frequently so entrenched that many at court (including the emperors themselves) might never have known a different situation. The fact that these generals benefited from broad consensus support also explains why, when some emperors resorted to murder to remove their magistri militum, they risked ¹¹⁷ Zos. 5.29.8–9: Πᾶσι τοίνυν δόξαντος δίκαια λέγειν Στελίχωνος, ἐδόκει τῇ γερουσίᾳ χρυσίου τετρακισχιλίας ὑπὲρ τῆς εἰρήνης Ἀλαρίχῳ δίδοσθαι λίτρας, τῶν πλειόνων οὐ κατὰ προαίρεσιν ἀλλὰ τῷ Στελίχωνος φόβῳ τοῦτο ψηφισαμένων, ὥστε ἀμέλει Λαμπάδιος γένους καὶ ἀξιώματος εὖ ἔχων, τῇ πατρίῳ φωνῇ τοῦτο ὑποφθεγξάμενος ‘non est ista pax sed pactio seruitutis’, ὃ δηλοῖ δουλείαν μᾶλλον ἤπερ εἰρήνην εἶναι τὸ πραττόμενον, ἅμα τῷ διαλυθῆναι τὸν σύλλογον, δέει τοῦ μὴ παθεῖν τι διὰ τὴν παρρησίαν, εἴς τινα πλησιάζουσαν τῶν Χριστιανῶν ἀπέφυγεν ἐκκλησίαν (trans. Ridley 1982).

   

197

violent responses from other interest groups. After Valentinian III slaughtered Aetius in 454, he immediately displayed his dead body publicly and called a senate meeting to denounce the deceased general out of fear that Aetius’ senatorial supporters might revolt.¹¹⁸ Moreover, when Leo I engineered Aspar’s murder in 471, the general’s military supporters launched a full-scale attack on the palace, while his relative Theoderic Strabo started sacking Roman cities in retaliation.¹¹⁹ The deaths of these powerful men—especially when orchestrated by the emperors themselves—sent shockwaves through the Roman court and wider civilian and military networks. We can identify some of those individuals who were invested in the partnership rule between emperors and generals, since the fall of the magister militum often brought about their demise. At the time of the coup which dislodged Stilicho in 408, for example, we learn that at least nine other named high officials in the court and the wider imperial administration were murdered at the same time— including the praetorian prefects of Gaul and Italy, the magister officiorum, the comes domesticorum, and the quaestor sacri palatii.¹²⁰ This represented not merely the removal of a powerful general, but the full-scale destruction of his regime and the liquidation of large numbers of high office-holders who had presumably accepted and supported his power-sharing arrangement with Emperor Honorius. The issue of access is again pivotal to understanding how this network developed: the officials who fell with Stilicho were men who had the general’s ear, and Stilicho himself had regular private access to the emperor, which had allowed him to exert influence on behalf of himself and his clients. In this way, we see the ‘politics of intimacy’ at work: although access to the emperor’s person, in terms of being in his presence in audiences or meetings, did not always guarantee influence, close and protracted intimacy usually did. It was this close, even dominant, intimacy that Stilicho had shared with Honorius and that had allowed him to exercise control through the emperor.¹²¹ Stilicho’s death, and the deaths of his supporters, brought about a period of great turmoil at the court of Honorius, which was only rectified when another magister militum, Fl. Constantius, established a new power-sharing arrangement with the emperor, which was also ultimately based on sustained proximity to, and influence with, the ruler.¹²² The evidence presented here, which highlights the very public prominence of senior magistri militum such as Stilicho, Aetius, Aspar, and Ricimer at the imperial court in the fifth century, and the open acknowledgement and celebration of their influence, suggests that there was a degree of consensus among the court and within the government and aristocracy at large, which represented ¹¹⁸ ¹¹⁹ ¹²⁰ ¹²¹ ¹²²

Prisc. fr. 30.1.39–51; Joh. Ant. fr. 224.3 (ed. Mariev 2008); McEvoy 2013: 296–7. Chron. Pasch. s.a. 467; Malch. fr. 2; Theoph. AM 5964. Zos. 5.32.3–7; Olymp. fr. 5.2.12–21; Matthews 1975: 278–81; McEvoy 2013: 182–4. On the ‘politics of intimacy’, see Starkey 1987b and the Introduction. O’Flynn 1983: 64–73; McEvoy 2013: 187–204.

198

   

support for, or acquiescence to, the power-sharing arrangement that such generals achieved. Though some scholars have chosen to term these generals—Stilicho in particular—as ‘regents’, such a term is misleading and misrepresents the reality of their positions. There was no such position in the Roman imperial government and no constitutional provision was ever made by law for an emperor who was unfit—whether by age or incapacity—to rule.¹²³ The position of the magister militum as a quasi-imperial figure was therefore potentially open to challenge, and this made broad support for their power-sharing arrangements still more important. This raises a further question: if generals were so powerful, and if other vested interests around the throne accepted that, then why did these magistri militum not seek to become emperors themselves? I would suggest that the answer is not that these generals could not become emperor, but that so long as the Theodosian dynasty remained in power—as it did in the West until 455 and the East until 457 (including Marcian)—this was never their aim. Nor was it the goal of Aspar in the East, even though he outlived the Theodosians, since he wished to maintain the same power dynamic under subsequent emperors whom he placed on the throne.¹²⁴ The case of Constantius III is the exception which proves the rule. A magister militum who joined the imperial family in 416 with his wedding to Galla Placidia, sister of Emperor Honorius, Constantius became co-Augustus with Honorius in 421, only to die six months later. However, according to Olympiodorus, he regretted his accession due to the limitations it placed on his freedom.¹²⁵ It was actually much more common for magistri militum to try to marry their children into the imperial house, as we have outlined above. This move was akin to an insurance policy for aging generals who might thereby reinvent their relationship with the emperor and maintain their influence. But they did not aim for the throne themselves, even if they hoped to see their grandchildren occupy it.

6. The End of the Age of the Magister Militum Since 363, the Roman Empire had been ruled by only two imperial houses, those of the Valentinians and Theodosians, which had been intertwined by both collegiate rule and marriage. The stability offered by dynastic rule came to an abrupt

¹²³ As highlighted originally by Mommsen 1903: 101–2, and see also Straub 1952: 108; Cameron 1969: 276; Cameron 1970: 39; and McEvoy 2013: 9–12. ¹²⁴ I am not convinced by arguments that men like Stilicho or Aspar were prevented from assuming the throne due to their religious or ethnic identities (McEvoy 2016a: 498–502). ¹²⁵ Olymp. fr. 33.1. On Constantius’ short reign, see Lütkenhaus 1998; Halsall 2007: 224–34. It is also highly relevant to Constantius’ elevation that he was by 421 the father of Honorius’ heir, the future Valentinian III.

   

199

and dramatic end in both the East and West the 450s, creating new opportunities for ambitious individuals to seek the purple.¹²⁶ It also paved the way for the end of the power-sharing arrangements between emperor and magister militum set out in this chapter. In 450, Theodosius II died after being thrown from his horse without having made any provision for a successor in the East, resulting in the stop-gap solution of the accession of the officer Marcian, followed by his marriage to Theodosius’ virginal sister Pulcheria. This peaceful transition of power was engineered by a court circle which included the magister militum Aspar. After Marcian’s death in 457, Aspar would then manage the accession of the dynastically unconnected soldier Leo.¹²⁷ Yet Aspar’s murder in 471 created a power vacuum which allowed other generals to step forward, with the ambition not of serving as a partner in imperial rule, but of wearing the purple themselves. This is demonstrated by the accession of the magister militum Zeno in 474, the usurpations (or attempted usurpations) of the magistri militum Basiliscus in 475 and Marcianus in 479, and the proclamation of the comes excubitorum Justin I as emperor in 518. With the exception of the short-lived Leo II, the eastern emperors of the second half of the fifth century were all adults,¹²⁸ not infrequently with military backgrounds themselves, even if they had not reached the highest levels of army leadership. This would have made any general’s efforts to establish himself as the loyal long-term military partner of an emperor very challenging.¹²⁹ Moreover, the activities of eastern generals after Aspar demonstrate a disinclination to serve emperors with the same degree of loyalty as their predecessors.¹³⁰ Basiliscus’ rebellion was supported by another magister militum, Armatus, while the general Illus backed the challenge of Leontius against Zeno in the 480s.¹³¹ This represents a clear contrast to the generals who had served under Theodosius II across the first half of the fifth century; he did not face a single usurpation attempt across the course of his long reign.¹³² Perhaps the fates of Aetius in the West and Aspar in the East—both slain as a result of imperial

¹²⁶ McEvoy 2022. ¹²⁷ For the late fifth century marking a new phase in imperial rule from Constantinople, see Meier 2017b: 524–9. On the accession of Marcian and the political machinations surrounding it, see Burgess 1993/1994: 59–68 and Zuckerman 1994: 172. For the accession of Leo I, see Croke 2005a. ¹²⁸ On Leo II, see McEvoy 2019b. ¹²⁹ One should note, however, that military men who became emperor in the East in the late fifth century nevertheless still followed the practice of delegating active campaigning duties to their generals once they assumed the purple, a pattern that would not be broken until the end of the sixth century. ¹³⁰ Meier 2012: 216–17. ¹³¹ Basiliscus: PLRE II Basiliscus 2; Kosiński 2010: 79–97. Armatus: PLRE II Armatus. Illus: PLRE II Illus 1; Brooks 1893; Elton 2000. ¹³² Holum 1981: 82 discusses one potential threat to the emperor during Theodosius’ youth. In the later years of his reign, some sources do refer to a threat of usurpation from the general Zeno, which did not eventuate: Prisc. fr. 16; Joh. Ant. fr. 223.1 (ed. Mariev 2008).

200

   

plots—suggested that such staunch allegiance might not be worth the risks, or at least that it might not ultimately be rewarded. Successive eastern emperors reacted to the potential threats posed by their generals by appointing their own relatives to the military high command. Zeno, whose reign was marked by ongoing instability, selected his brother Longinus as magister militum in 485, while both Anastasius and Justin I elevated nephews to this rank.¹³³ As Michael McCormick has argued, placing imperial relatives in charge of the armies strengthened the bond between the ruling house and the military, thereby potentially diluting any threats from other magistri militum.¹³⁴ The influence of magistri militum in the East was further diminished in the middle decades of the sixth century when Justinian increased their number from five to seven, thus making it more difficult still for any one man to establish a powersharing relationship with the emperor.¹³⁵ After Theodosius II’s unfortunate riding accident in 450, his western colleague and son-in-law Valentinian III had no doubt hoped to rule the East as well.¹³⁶ Yet this never came to pass, and Valentinian’s ill-fated attempt at independence—the murder of the magister militum Aetius at his own hands— was to bring down his regime in the West. The assassination of Valentinian III in 455 ushered in a twenty-one-year period of crisis in imperial leadership, which ended with the deposition of Romulus Augustulus in 476.¹³⁷ The ‘age of the magister militum’, as I have termed it, might be said to have extended until the death of magister militum Ricimer in 472, since he eschewed the emperorship himself, preferring to share power with the western rulers.¹³⁸ Yet these men were not child, nor child-turned-adult, emperors, but ambitious military men, such as Avitus, Majorian, and Anthemius. Their backgrounds and qualifications often clashed with Ricimer’s attempts to maintain the system of partnership rule which had been fostered under Honorius and Valentinian III (hence Anthemius’ chafing at the need to ally himself with a ‘skin-clad Goth’). Moreover, Ricimer’s own loyalties were not those of western generals before him, men like Stilicho, Constantius, and Aetius, who had not wavered in their long-term commitment to the sole emperors they served.¹³⁹ In contrast, Ricimer and the generals who continued to operate in former Roman territory as the

¹³³ Longinus: PLRE II Longinus 6. For details of alliances between imperial families and the military, see Cameron 1978, and Chapter 14. ¹³⁴ McCormick 2000: 150. This policy could still backfire, as we see with Basiliscus, Armatus, and Marcianus, generals and relatives of Emperor Zeno who rebelled against him in 475 and 479. ¹³⁵ Boak 1915: 122. ¹³⁶ Burgess 1993/1994; McEvoy 2013: 296–7. ¹³⁷ McEvoy 2017. ¹³⁸ See further O’Flynn 1983: 104–28 on the activities of Ricimer. ¹³⁹ Aetius had originally supported the usurper John as western emperor in 424–5, but after his defeat transferred his allegiance to Valentinian III and served loyally over the following three decades. See further O’Flynn 1983: 74–103; Stickler 2002; McEvoy 2013: 223–97; Wijnendaele 2017b.

   

201

West fragmented were devoted to their own cause, rather than to the Roman state and its leader.¹⁴⁰

7. Conclusion Thus, it was from the late fourth century and across much of the fifth that the magistri militum were a dominant force at the Roman imperial courts in both East and West. These generals were far from acting as powers behind the throne— indeed there was nothing at all covert about their power and influence, for it was an acknowledged factor in imperial presentation and policy, as the literary and epigraphic evidence makes clear. This is not to say that there were not others whom we might wish to label as ‘powers behind the throne’—imperial family members or palace eunuchs, for example—but such a term implies an illegitimate and unofficial influence, a lurking and hidden power, and this was emphatically not the way that dominant generals operated in this period. One major reason for the generals’ dominance of government in this particular period of Roman history, I have proposed, was a collision of systemic elements of the Roman imperial system: the creation of the supreme command of magister militum itself in the fourth century, followed by the unexpected but repeated succession of underage, non-campaigning emperors in the late fourth and fifth centuries, and the consequent changes in the nature of the emperor’s role and function. The effects of this partnership continued to be felt, even after the end of this particular series of child emperors, in both the East and West, until the downfalls of Aspar and Ricimer in 471 and 472, respectively. The partnership system developed in conjunction with magistri militum offered the Roman Empire stability, with emperors providing a dynastic centre to the regime while generals continued to lead the imperial armies on the emperor’s behalf. To acknowledge this reality offers a different picture of the Roman imperial court in this era. It is not one of the overwhelming dominance of a military with which an inadequate ruler could not cope and which other constituencies were unable to oppose. Instead, we have a picture of greater consensus, in which members of the court—including civilian office-holders, aristocrats, and other military leaders— often accepted and worked with this arrangement of power-sharing between a noncampaigning emperor and a campaigning magister militum. This consensus was not always maintained, as the downfalls of Stilicho, Aetius, and Aspar confirm, but the long tenures of these men suggest that such arrangements could and did function successfully for decades, sustaining the power of general and emperor alike. Indeed, ¹⁴⁰ See MacGeorge 2002 and Wijnendaele 2017a for studies of these ‘warlords’. This term has been used to describe western Roman military leaders who were still operating in former Roman territories in the second half of the fifth century.

202

   

there is a story that when Valentinian III, flushed with his success in killing Aetius, turned to a man present and asked, ‘Was the death of Aetius not well accomplished?’, he received the following reply: ‘Whether well or not I do not know, but I do know that you have cut off your right hand with your left.’¹⁴¹ There is no doubt that the crucial role played by magistri militum as military commanders, leading figures at court, and, indeed, the emperor’s right-hand men, was well recognized by their contemporaries.

¹⁴¹ Joh. Ant. fr. 200.1 (FHG IV, ed. Müller): ὁ δὲ, > (trans. Gordon 1960).

9 Representatives and Co-Rulers Imperial Women and the Court in Late Antiquity Anja Busch

1. Introduction In Late Antiquity, imperial women, and in particular the Augustae, played important roles within the imperial household and the court community at large.¹ In this chapter, I will focus on three aspects in particular: (i) the official representation of imperial women; (ii) the different roles played by the emperors’ wives and sisters at court and in public; and (iii) the interplay between official representation, imperial tradition, and public expectations, which imperial women exploited to create new opportunities for themselves in Late Antiquity. The latter factor becomes most visible in the case of Pulcheria Augusta, sister of Theodosius II, and subsequently wife of his successor, Marcian. Pulcheria is the first imperial woman for whom there is substantial evidence that she was recognized as her husband’s imperial colleague. Pulcheria would subsequently become a model for other imperial women whose position enabled them to create the next emperor.² The reign of the Theodosians introduced remarkable novelties in the role and function of imperial women, in terms of their official representation, public appearances, and political activity inside and outside the imperial court.³ The establishment of the dynasty in the city of Constantinople under Theodosius I, and the consolidation of its role as the key eastern imperial residence under Arcadius and Theodosius II, enabled the women of the family to have increasing contact with the people of the capital, as well as with representatives of relevant political groups who came to court. Women were publicly portrayed on coins, artworks, and in ceremonial performances as key representatives of the imperial family. This marked a real change from the role of imperial women under ¹ I use the term ‘imperial women’ to refer to any woman belonging to the imperial family, whereas the term ‘empress’ relates to an Augusta or an imperial wife, according to common scholarly usage, even if she was not granted the title of Augusta. ² See Herrin 2000; Herrin 2016 on the important place of fifth-century imperial women in the evolution of the role of Byzantine empresses. ³ Holum 1982. Anja Busch, Representatives and Co-Rulers: Imperial Women and the Court in Late Antiquity In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0010

204

   

previous fourth-century dynasties, allowing the Theodosian empresses to achieve public prominence that had not been seen since Constantine elevated his wife Fausta and his mother Helena to the rank of Augusta.

2. Imperial Women from the Tetrarchs to the Valentinians Women did not figure significantly in official imperial representation, such as coins and public monuments, during the Tetrarchic period.⁴ In his polemical work, On the Deaths of the Persecutors, the Christian author Lactantius reveals the weakness of the positions of Diocletian’s wife Aurelia Prisca and their daughter Galeria Valeria at the imperial court in Nicomedia.⁵ Lactantius, who held the chair of Latin in the city, describes both women as Christians, who were forced to sacrifice to the pagan gods by Diocletian.⁶ Valeria was married in 293 to Galerius, who held the rank of Caesar within the Tetrarchic system. After Galerius himself was raised to the rank of Augustus in 305, he decided within a few years— probably in 308—to make Valeria an Augusta herself, and issued coins minted in her name.⁷ She was the first woman to hold this rank since Magnia Urbica, the wife of Carinus, in 283–5. Galerius’ elevation of Valeria to Augusta emphasized his relationship by marriage to Diocletian and his primacy within the Tetrarchic imperial college, perhaps in answer to the recent proclamation of Constantine by his father’s troops. Galeria Valeria appears to have been merely a pawn in Tetrarchic marriage politics, like many imperial women before and after her. To stabilize their mutual loyalty and bind themselves to each other, the Tetrarchs married their daughters or sisters to other members of the imperial college. The title of Augusta did nothing to strengthen the position of Valeria or her mother Prisca. After the death of Galerius in 311, both women were supposed to be handed over to Licinius for protection, but they fled to Maximinus Daza instead.⁸ Daza banished both women from his court when Valeria refused to marry him, and they were forced to wander through the provinces, eventually making their way from Syria to Thessalonica in Greece, where they were recognized and killed on the orders of Licinius.⁹ The murders of Valeria and Prisca were motivated by the struggle for

⁴ Hekster 2015: 313–14 notes the ‘systematic exclusion of imperial women’ from the Tetrarchic imperial representation, which for the contemporary observer ‘implied a departure from the notion of a ruling family’. This break with Roman imperial tradition did not prove successful, according to Hekster. Cf. Jeličić-Radonić 2009: 322. See further Chen 2018. ⁵ Jeličić-Radonić 2009: 312–14 discusses a recent inscription from Salona giving Diocletian’s wife Aurelia Prisca her full name and describing her as nobilissma femina. As she points out (322), the inscription is found on the pedestal of a lost statue, which means that Prisca was represented in art. ⁶ Lact. DMP 15. ⁷ RIC VI, see plate 9 no. 196; plate 10 no. 43; plate 4 no. 58. ⁸ Lact. DMP 35, 39. ⁹ Lact. DMP 39, 41, 50–1.

  -

205

primacy among members of the imperial college following the breakdown of the Second Tetrarchy, and the removal of individuals who could be perceived as threats to Licinius. Women were also used as pawns by Tetrarchic emperors based in the West. Constantine, the son of Constantius I, married an imperial woman in an attempt to strengthen his position, as he was considered to be a usurper by the other Tetrarchs. In 307, he wed Fausta, the daughter of Maximian, who had officially retired and become a senior Augustus alongside Diocletian.¹⁰ Maximian himself wished to return as a full-time emperor, and needed Constantine’s support against Galerius.¹¹ So, just like Valeria, though for different reasons, Fausta started her life as an empress being a pawn in male power politics. It was not until after his victory over Licinius in 324, when he became sole ruler of the empire, that Constantine began to promote Fausta and his mother Helena, his women, as an integral part of the imperial dynasty. They were both raised to the rank of Augusta, either on 8 November 324 or soon after, and were featured on imperial coins showing their profiles, names, and new imperial title.¹² By elevating his mother Helena to the status of Augusta, Constantine emphasized his dynastic intentions to promote his own sons, Helena’s grandchildren, rather than his half-brothers and sisters sprung from his step-mother Theodora. Helena was of humble origin and had lived with Constantine’s father, Constantius I, in the 270s and 280s. In 289, Constantius had left Helena to marry Theodora, the daughter or stepdaughter of Maximian. He was thereafter adopted, elevated as Caesar, and designated Maximian’s successor as Augustus. We do not know where Helena lived before Constantine seized the purple after the death of Constantius I in 306, and her life between then and 324 is shrouded in darkness. It is probable that she became a Christian under the influence of Constantine.¹³ While Constantine’s successors did not promote their women to the status of Augusta or feature them prominently in official imperial imagery, such as coins, some of them did act as public representatives of the imperial household, especially Constantine’s daughter Constantina;¹⁴ Constantius II’s wife Eusebia

¹⁰ Drijvers 1992b and Harries 2012 discuss the lack of sources for Fausta’s life. ¹¹ Drijvers 1992b: 502–3. ¹² There are small quantities of earlier coins for Helena and Fausta, which bear the legends HELENA or FAUSTA NF (nobilissima femina). The year of their minting is unknown, but the use of the abbreviation NF indicates that the women were not yet honoured with the title of Augusta, and thus they must date prior to 324. See Drijvers 1992a: 39–41. ¹³ Cf. Eus. V. Const. 3.47. See further Drijvers 1992a: 35–8. ¹⁴ It is debated whether Constantina had received the title of Augusta and was crowned with a diadem c. 335 by her father Constantine, as stated in Philost. HE 3.22, 26. Bleckmann 1994: 36–40 (contra Holum 1982: 31–4) shows that neither the elevation nor coronation of Constantina would have been anachronistic, and that Philostorgius’ remarks on this matter are plausible with regards to Constantine’s dynastic policy. Harries 2014: 197–8 assumes (against Bleckmann) that Philostorgius’ report was influenced by the Theodosian empresses, in particular Pulcheria.

206

   

and daughter Constantia;¹⁵ and Justina, successively wife of Magnentius and Valentinian I.¹⁶ This paved the way for the increasing prominence of imperial women in the fifth century.

3. Theodosian Women and the Imperial Insignia At the end of the fourth century, Theodosius I decided to re-emphasize the role of the empress, styling his wife Flaccilla as Augusta in 383 and issuing coins showing her portrait, as well as her full name and title.¹⁷ These coins assimilated the empress’ iconography to that of the emperor himself, depicting Flaccilla with the same insignia as an Augustus, namely the imperial diadem and the emperor’s cloak. These were unmistakably symbols of the imperial power and authority.¹⁸ Eudoxia, the wife of Theodosius’ son and successor Arcadius, is the first imperial woman shown with yet another imperial attribute, the hand of God (Figure 9.1). The hand reaches down from the sky and holds a crown over the empress’s head, indicating the divine legitimation of her position.¹⁹ From the middle of the fifth century onwards, numismatic depictions of the Augustae show them together with new imperial insignia, such as the orb surmounted with a cross (globus cruciger) and the long sceptre.²⁰ The fifth-century historian Priscus, who wrote a contemporary secular history of the events between c. 430 and 474, connects the insignia of the Augustae with the ability to confer imperial power.²¹ He describes how Honoria Augusta, sister of Valentinian III, was deprived of her imperial status after she had been convicted of having a secret love affair.²² She therefore sent her ring to Attila, leader of the Huns, and asked him for help, an act which Attila interpreted as a marriage proposal. Since Honoria, as an emperor’s daughter, possessed the imperial sceptre, Attila wrote to both emperors—Valentinian III in the West and Theodosius II in the East— making claims to Roman territory as her future husband.²³ It is important to note that, even if the aforementioned story was exaggerated by eastern Roman

¹⁵ Eusebia: Tougher 1998; Wieber 2010. Constantia: Harries 2014: 209–12; McEvoy 2016b. ¹⁶ For Constantina’s and Justina’s church-building activities as acts of imperial euergetism and patronage, see Dirschlmayer 2015: 52–81. ¹⁷ See e.g. RIC IX, Constantinople, no. 72. ¹⁸ Holum 1982: 33–4; Angelova 2015: 185–94. ¹⁹ MacCormack 1981: 245; Angelova 2015: 198–201. ²⁰ See e.g. RIC X, nos. 2016 and 2023 (solidi of Valentinian III’s wife, Licinia Eudoxia, the reverse showing her nimbate, enthroned, and holding globus cruciger and sceptre); cf. the so-called ‘Diptych of Ariadne’ showing the empress enthroned, holding the globus cruciger (Kunsthistorisches Museum, Vienna). Similar to this is the second diptych of Ariadne, standing, holding sceptre and globe, now in the National Museum of Bargello, Florence. See further Angelova 2004: 1–14. ²¹ Prisc. fr. 17, 20.1, 20.3 (ed. Blockley). For discussion, see Bury 1919; Busch 2015: 166–76; Croke 2015b: 112–16. ²² Prisc. fr. 17 (ed. Blockley); Joh. Ant. fr. 223 (ed. Mariev 2008). Cf. Marcell. com. s. a. 434. ²³ Prisc. fr. 17, 20.1 (ed. Blockley).

  -

207

Figure 9.1 Bronze coin of Eudoxia (RIC X Arcadius 102, Nicomedia), Dumbarton Oaks BZC.1948.17.1122. © Dumbarton Oaks, Byzantine Coins and Seals Collection, Washington DC.

authors,²⁴ Priscus thought it was conceivable that Honoria possessed a claim to, and the ability to delegate, imperial authority.²⁵ The imperial insignia were worn by the Theodosian empresses on public ceremonial occasions in Constantinople. This is illustrated by a speech of John Chrysostom in which he discusses the appearance of Arcadius’ wife Eudoxia. At the end of a procession, in which the empress Eudoxia had not only taken part but pointedly displayed her humility in the presence of the relics of some unknown Christian martyrs, John praises Eudoxia as follows: she, who wears the diadem and has put on the purple, . . . followed these saints like a servant, . . . and despising every earthly pomp, in the midst of such a location she appeared before the people . . . and she threw aside the imperial diadem, and dressed herself in the cloak of humility instead of the purple one.²⁶

Eudoxia performed an impressive act of self-humiliation in front of her subjects by casting aside her imperial insignia. In the presence of the martyrs’ relics, she took part in the procession as an ordinary yet exemplary Christian woman rather ²⁴ Meier 2017a. ²⁵ Busch 2015: 189; see further Maslev 1966; Bensammar 1976. ²⁶ Joh. Chrys. Homilies 2.2 (PG 63: 470): καὶ αυτὴ ἡ τὸ διάδημα περικειμένη καὶ τὴν πορφυρίδα περιβεβλημένη, . . . ὥσπερ θεραπαινὶς παρηκολούθει τοῖς ἁγίοις, . . . βασιλίαν μὲν . . . διαδὴματα . . . ῥίψασα . . . , ἐνδυσαμένη δὲ τὲν τῆς ταπεινοφροσύνης στολὴν ἀντὶ τῆς πορφυρίδος. See further Angelova 2015: 181–2, 187.

208

   

than as an empress. As Mischa Meier has argued, Eudoxia’s performance must have made a strong impact on contemporary observers. Her ostentatiously humble appearance was effective because it was performed by a representative of the highest secular power,²⁷ and because the insignia she laid aside were unmistakably recognized by the spectators as the distinct symbols of imperial authority.²⁸ Emperor Arcadius did not take part in these events, since he was not expected to arrive until the next day. Therefore, Eudoxia’s public performance occurred in her capacity as his imperial representative.

4. Dynastic Potential: Fertile Wives and Virgin Sisters The title of Augusta did not confer any particular constitutional and legal status, nor did it define certain duties or functions at court or within the imperial regime at large. But it was, nevertheless, a significant honour, only to be awarded at the emperor’s behest, and not every imperial woman received the title (as shown by the lack of Augustae under the First Tetrarchy, and in the fourth century between the reigns of Constantine and Theodosius I). It has been proposed that an emperor’s wife needed to have given birth to children in order to be honoured with the title of Augusta.²⁹ However, this does not actually seem to have been the case. All imperial wives were expected to bear children (ideally including at least one male heir). But not all Augustae actually gave birth to children, nor did every imperial mother bear the title of Augusta.³⁰ Nor was it the case that an imperial woman who was ‘born to the purple’ (porphyrogenita), as the daughter of a reigning emperor, automatically became an Augusta.³¹ The lack of consistent criteria for the award of the title can be seen in the following examples. Galla, the daughter of Valentinian I and second wife of Theodosius I, who was born to the purple, bore children to her husband, but was never elevated to the rank of Augusta, as Theodosius I’s first wife Flaccilla had been. Pulcheria, daughter of Emperor Arcadius and sister of Theodosius II, became Augusta in 414, even though, having taken the vow of chastity, she never bore children. However, her younger sisters, Arcadia and Marina, who both had also devoted themselves to virginity, never received the title. Finally, Leontia, daughter of Emperor Leo I, was not honoured as Augusta, even though she was born in the purple, whereas her

²⁷ Meier 2007: 149–50; Kelly 2013b: 223–4. ²⁸ Cf. Diefenbach 1996: 36–9; Meier 2007: 148–51; Kelly 2013b: 228–9. ²⁹ Holum 1982: 30–1; Missiou 1982: 494. ³⁰ James 2001: 119–24. See also Chen 2018, who discusses how the role and significance of imperial women extended beyond child-bearing. ³¹ Cf. Missiou 1982: 494, according to whom either motherhood or imperial birth qualified an imperial woman to become Augusta.

  -

209

older sister Ariadne, who did become Augusta, was born before their father had become emperor. Motherhood was nonetheless an important characteristic for an emperor’s wife, as bearing children helped to secure a legitimate dynastic succession to the throne. The fertility of imperial women played a significant part in their reception by contemporaries and later commentators. The two wives of Emperor Honorius, Maria and Thermantia, neither of whom had children, are almost completely ignored by most of our narrative sources, though at least Maria is prominent in court poetry, as are expectations that she would bear a son.³² In contrast, Theodosius’ I wife Flaccilla, according to Gregory of Nyssa’s funeral oration, was the ‘model of love towards her husband’ (τῆς φιλανδρίας ὁ τύπος) as she bore three children, a daughter, Pulcheria, and the future heirs to the throne, Arcadius and Honorius.³³ Gregory praised Flaccilla’s ‘love of her husband’ (φιλανδρία) because she had left her sons to stay with their father, ‘so that they might be the pillars of his empire’ (ὥστε εἶναι αὐτοὺς τῆς βασιλείας ἐρείσματα).³⁴ Eudoxia, wife of Emperor Arcadius, was the mother of no fewer than four surviving children: Pulcheria, Theodosius II, Arcadia, and Marina (in order of age). According to the literary sources, Eudoxia emphasized her fecundity and even used the children themselves to assert her own or her protégés’ interests at court.³⁵ On one occasion, when the influential eunuch Eutropius was trying to have her expelled from the court, she brought her two infant daughters to Arcadius to persuade him to remove Eutropius instead.³⁶ Then, during the public baptism of Theodosius II, she manipulated the procedure so that it seemed to the spectators as if the new-born heir gave his consent to a request made by Porphyry of Gaza and his fellow monks. Shortly before this event, Eudoxia had agreed to these monks’ petition to destroy a pagan sanctuary without the approval of her husband. But after the baby Theodosius II had apparently agreed in front of the people of Constantinople, the emperor could not refuse that request any longer.³⁷ Finally, during her long-running conflict with John Chrysostom, bishop of Constantinople, on one occasion she even put her son Theodosius II on the bishop’s knees in a public act of reconciliation.³⁸ In these ways, Eudoxia exploited her status as a mother to strengthen her position at court and in the public sphere.

³² Busch 2015: 53–8. For an analysis of the depiction of the first generation of Theodosian women, see now McEvoy 2021: esp. 133–4 on expectations of Maria’s child-bearing, and for Stilicho’s wife Serena, see also Washington 2020. ³³ Greg. Nyss. Or. fun. Flacc. 480.17. ³⁴ Greg. Nyss. Or. fun. Flacc. 488.19. See Leppin 2000 on imperial women in Gregory of Nyssa’s works. ³⁵ Brottier 1996. ³⁶ Philost. HE 11.6. ³⁷ Marc. Diac. V. Porph. 48–9. For a discussion of this incident in the context of visits to the imperial court, see Chapter 13. ³⁸ Soc. 6.11.20; Soz. 8.10.6. On Eudoxia’s relationship with John Chrysostom, see van Nuffelen 2012; Busch 2021.

210

   

While motherhood remained an important virtue for an imperial woman in Late Antiquity, the fifth century in particular saw the introduction of the quality of religiously motivated virginity, which was espoused by Theodosius II’s sisters, Pulcheria, Arcadia, and Marina. As the church historian Sozomen explains in his discussion of the sisters’ vow of 414, this step was taken so that a male competitor, who could potentially challenge the emperor’s authority and become a rival for his throne, could not enter the palace.³⁹ In the Theodosian era, no imperial sisters were married so long as there was a chance that the emperor’s own wife would bear him an heir. When the childless western Emperor Honorius gave his sister Galla Placidia in marriage to his trusted general Constantius in 417, he had already been a widower for years and was not expecting any children.⁴⁰ He later, in 421, designated Constantius his successor to the throne by nominating him as Augustus, and he granted Placidia the title of Augusta, following the birth of a son and heir to the couple.⁴¹ In this way, Honorius ensured that the western throne would remain with a member of the Theodosian house. This was possible because Galla Placidia possessed ‘dynastic potential’,⁴² a woman’s ability to confer imperial authority and dynastic legitimacy on a man, usually a husband or son.⁴³ All imperial women possessed dynastic potential. This could be a risk to the emperor either if a female relative became disloyal to him, as in the case of Honoria, or if there were reasons to doubt his wife’s marital fidelity. Constantine is said to have murdered his significantly younger wife Fausta in 326, because there were rumours about her having a love affair with his son Crispus, who had been born to him by a previous consort, Minervina.⁴⁴ The exile of Eudocia, wife of Theodosius II, to Jerusalem from the early 440s onwards also seems to have been caused by suspicions that she had betrayed her husband.⁴⁵ In both cases, it is impossible ever to know what had really happened, but both occurrences show how important it was for an empress to be above any such suspicion and to appear, like Flaccilla, as a ‘model of married love’. The threat posed to an emperor by his sisters’ dynastic potential could be curbed if, as in the case of the family Theodosius II, they all remained virgins. As Meaghan McEvoy has argued, the celibacy of Theodosius II’s sisters was chiefly designed to protect the young emperor’s rule from competitors to his throne.⁴⁶ ³⁹ Soz. 9.1.3; Holum 1982: 93–6; McEvoy 2019a. ⁴⁰ On marriages between imperial women and generals, see Chapter 8. ⁴¹ Olymp. fr. 33; Philost. HE 12.12. ⁴² Busch 2015: 214–17. Cf. the concept as defined by Priwitzer 2010: 238–9. ⁴³ Sometimes it could be a brother, as in the case of Basiliscus, who, with the help of his sister Verina Augusta, usurped Emperor Zeno’s throne in 475/6. ⁴⁴ Zos. 2.29.1–2. For details, see Drijvers 1992b. Barnes 2011:144–50 discusses the possibility that Fausta became pregnant by Crispus and died accidentally as the result of an abortion. Alternatively, he suggests that she could have committed suicide because of her involvement in the death of Crispus, whom she might have falsely accused of rape. However, Harries 2012: 259–6; 2014: 202–6 points out that we do not have reliable knowledge about the circumstances of Fausta’s death. ⁴⁵ Holum 1982: 176–9; Busch 2015: 157–62. ⁴⁶ McEvoy 2019a.

  -

211

This had the effect of limiting their connections with members of the court community, in contrast with previous emperors, such as Marcus Aurelius and Constantine, whose female relatives had married leading senators.⁴⁷ Pulcheria’s vow increased her standing to a great extent; unfortunately, we are less informed about her sisters Arcadia and Marina, who were never given much attention by late antique and Byzantine historians. Pulcheria, however, knew how to draw prestige from her status as an imperial virgin and to present herself as an authority figure in matters of religion. But in 414, when Pulcheria, Arcadia, and Marina took their vow, nobody could have foreseen that their brother Theodosius II would never sire a son. When the emperor died suddenly in 450 during a hunting accident, Pulcheria—the only sister still living—functioned as a bearer of imperial authority, and her dynastic potential was used to legitimate the claim of Marcian to the purple.

5. From Representation to Participation In this section, we will consider in detail how the prominent public representation of imperial women was mirrored by power and influence at court. They did not have the constitutional ability to rule the empire independently of their husbands, fathers, or brothers. But they were often regarded as exercising a share in imperial power, being described in our literary sources as coregnantes (‘co-rulers’)⁴⁸ or consortes imperii (‘partners in empire’).⁴⁹ Their marriage to the emperor is sometimes portrayed as resulting in a κοινωνία (‘joint rule’).⁵⁰ These sources indicate that the prominence of imperial women was broadly accepted, and moreover that influence at court was possible even without a constitutionally defined role. As the first Christian Augusta, Helena represented her son Constantine in her travels to Jerusalem and through her church-building activity in the Holy Land and in Rome.⁵¹ Eusebius of Caesarea’s Life of Constantine, written about ten years after Helena’s death in the late 320s, reports that Constantine had given Helena access to ‘imperial resources’ (βασιλική ἐξουσία). She used this to perform charitable deeds, caring for the poor, freeing captives, and letting the banished return

⁴⁷ Marcus Aurelius: Pflaum 1961. Constantine: Harries 2014: 206–9. ⁴⁸ Paul. Nol. Ep. 31.4; Sulp. Sev. Chron. 2.33.3, both retrospectively referring to Helena, might have been influenced by contemporary imperial women. ⁴⁹ Claud. Epithal. 277, referring to Maria, the first wife of Emperor Honorius. ⁵⁰ Jul. Or. 3.114B–C on Eusebia and Constantius II; Greg. Nyss. Or. fun. Flacc. 478.20–479.1, 488.7–8 on Flaccilla and Theodosius I; Joh. Chrys. Homilies 2.3 (PG 63: 472) on Eudoxia and Arcadius. Cf. Angelova 2015: 183–4. ⁵¹ Dirschlmayer 2015: 32–52, however, has shown however that none of the churches Helena is said to have built can definitively be attributed to her. Cf. Angelova 2015: 138–46.

212

   

from exile.⁵² Gregory of Nyssa’s funeral oration for Theodosius I’s wife Flaccilla shares some similarities with Eusebius’ presentation of Helena.⁵³ Humbly devoting herself to Christianity, Flaccilla is said to have cared for the poor and those in need.⁵⁴ Gregory repeatedly claimed not only that Flaccilla was ‘the very image of philanthropy’ (ἡ τῆς φιλανθρωπίας εἰκών),⁵⁵ but also that she assisted her subjects by constantly reminding her husband to be a just and mild ruler.⁵⁶ Eusebius’ and Gregory’s presentation of these women outlines some of the expectations of Christian imperial women.⁵⁷ In the fifth century, our sources give more detailed information about the activities of imperial women at court. There are several reports about individuals or interest groups directly addressing empresses and seeking their patronage. Imperial women frequently acted as intermediaries between petitioners and the emperor, since they not only had access to the emperor, but also shared intimacy with him (to use the terms outlined in this volume’s Introduction).⁵⁸ As we discussed above, Eudoxia supported the destruction of a pagan temple in Gaza at the request of Porphyry and his colleagues, despite her husband’s initial reluctance.⁵⁹ When she was later addressed by a group of monks, who had been expelled from Alexandria by the local patriarch Theophilus, she instantly agreed to authorize a synod.⁶⁰ Of course, imperial women could also refuse to advocate for petitioners, a stance which did not earn them plaudits from contemporaries. Olympiodorus expressed his disgust with Galla Placidia, as she allegedly dismissed petitioners who entreated her instead of pressing their case with the emperor.⁶¹ Similarly, Nestorius complained about the influence of Pulcheria, sister of Theodosius II, since she had been involved in the process that led to his deposition as bishop of Constantinople in 434 and his condemnation as a heretic.⁶² In the context of the so-called ‘robber synod’ of 449, Nestorius argued that Pulcheria had not exercised her authority to prevent the synod taking place or at least to influence the synod’s highly controversial decisions, including the deposition, maltreatment, and subsequent death of the Constantinopolitan bishop Flavianus.⁶³ We can deduce from these examples that it was not only common knowledge that imperial women influenced the emperor, but that it was publicly accepted and, moreover, even expected.⁶⁴ However, contemporaries did not approve of this influence when it did not assist their own cause.⁶⁵ ⁵² Eus. V. Const. 3.43.4–3.44. ⁵³ Cf. Holum 1982: 24 n. 68; Drijvers 1993: 86. ⁵⁴ Greg. Nyss. Or. fun. Flacc. 487.13–17 and passim; cf. Theod. HE 5.19.2f.; Suda, π 1685. ⁵⁵ Greg. Nyss. Or. fun. Flacc. 480.11. ⁵⁶ Greg. Nyss. Or. fun. Flacc. 478.20–479.1. ⁵⁷ On this, see further McEvoy 2021: esp. 134–6. ⁵⁸ For stories of imperial women acting as intermediaries at court, see further Chapter 13. ⁵⁹ Marc. Diac. V. Porph. 37–48, 75, 78, 84, 92. ⁶⁰ Soz. 8.11. Cf. Pall. Dial. 8 (PG 45: 26). ⁶¹ Olymp. fr. 37. ⁶² See Chapter 11. ⁶³ Nest. Herac. pp. 342–3 (trans. Driver and Hodgson). ⁶⁴ There were many other individuals and interest groups at court who could be in competition with the imperial women for influence over the emperor. See further Millar 2006 and Elton 2009. ⁶⁵ Busch 2021.

  -

213

Pulcheria’s position in the Theodosian dynasty was quite extraordinary, and it is in her activities and presentation that we see the dynastic potential of an imperial woman reach full fruition. She was granted the title of Augusta after devoting herself to virginity at the age of fourteen.⁶⁶ A contemporary admirer, the church historian Sozomen, writes that Pulcheria took a leadership role in her brother’s regime: After having assumed the responsibility for the empire (ἡγεμονία) she governed the Roman world in the best possible way and with great prudence, as she reasoned well, and accomplished everything, that had to be done, quickly and recorded it.⁶⁷

It is doubtful that Pulcheria, herself only a teenager, held guardianship over her only slightly younger brother, in whose name the ecclesiastical historian said she governed the empire. What is more important is that Sozomen, a contemporary writer, thought that Pulcheria was able to exercise influence and authority at court, though, as Jill Harries has argued, this had to be done outside the consistorium.⁶⁸ This authority was supported by the way in which Pulcheria was depicted in various forms of official representation. Ever since Pulcheria’s designation as Augusta in 414, coins were issued in her name, which showed her, like her predecessors, wearing a diadem and imperial purple cloak and accompanied by the hand of God, which places a crown upon her head (Figure 9.2).⁶⁹ The connection between the award of the title of Augusta and Pulcheria’s public vow of chastity demonstrates that her prestige was largely dependent on religious selfexpression.⁷⁰ What is striking in Sozomen’s report is that the church historian connects Pulcheria’s vow with the longevity of her brother’s empire. Elsewhere, he explains that it was God himself who put Pulcheria in charge of the empire, and moreover that her vow served as a guarantee for the reign of Theodosius II.⁷¹ In Late Antiquity, Christological disputes had an important impact on political discourse. This is because the emperor’s rule, and that of his dynasty at large, was regarded as being legitimized by God. By presenting herself as a model Christian, Pulcheria was able to exercise authority in religious political debates. She identified herself especially with the Virgin Mary,⁷² and, as a result, claimed to be an advocate of Christian orthodoxy. In the early 430s, Pulcheria engaged in the theological debate about the nature of the Theotokos. The bishop of Constantinople, Nestorius, had opposed the idea that the Virgin Mary was the ⁶⁶ Soz. 9.1.4. ⁶⁷ Soz. 9.1.5: ὑπεισελθοῦσα δὲ τῆς ἡγεμονίας τὴν φροντίδα, ἄριστα καὶ ἐν κόσμῳ πολλῷ τὴν Ῥωμαίων οἰκουμένην διῴκησεν, εὖ βουλευομένη ὡς ἐν τάχει τὰ πρακτέα ἐπιτελοῦσα καὶ γράφουσα. ⁶⁸ Harries 2013. ⁶⁹ RIC X Constantinople nos 10–15. ⁷⁰ On the nature and limitations of Pulcheria’s authority, see Harries 2013: 67–73, 88–9. ⁷¹ Soz. 9.1.2–3. ⁷² Holum 1982: 153.

214

   

Figure 9.2 Solidus of Pulcheria (RIC X Theodosius II 205, Constantinople), Dumbarton Oaks BZC.1948.17.1182. © Dumbarton Oaks, Byzantine Coins and Seals Collection, Washington DC.

mother of God, and therefore had come into conflict with bishops from different parts of the empire. The most prominent among them was Cyril of Alexandria, who wrote several letters on this matter to the women around Emperor Theodosius II, namely his wife, Eudocia, and his sisters, Arcadia, Marina, and Pulcheria. Cyril hoped to win them all as allies to his cause.⁷³ We do not know exactly how the relationship between Pulcheria and Nestorius evolved. But in 431, the first Council of Ephesus condemned Nestorius for heresy and deposed him from his position as bishop of Constantinople. In his exile, he wrote an apologetic book, in which he states clearly that in his opinion it had been Pulcheria who had been the driving force that led to his deposition.⁷⁴ Religious conflicts did not come to an end after Nestorius’ removal from the see of Constantinople. Pulcheria, due to internal conflicts at court—possibly, as later sources report, involving Theodosius’ praepositus sacri cubiculi Chrysaphius⁷⁵— was not living in the imperial palace when, in the late 440s, religious controversies escalated again and culminated in the so-called ‘robber synod’ of 449. As a result, the letters from the bishop of Rome, Leo I, to Pulcheria did not reach the Augusta. Apparently Pulcheria was not in a position to influence the emperor’s decisions, because she was estranged from court circles: even Nestorius later accused her of ⁷³ ACOec 1.1.5, 26–61 no. 149; 62–118 no. 150; Wessel 2004: 98–9. These events are discussed in Chapter 11. ⁷⁴ Nest. Herac. pp. 96–7 (trans. Driver and Hodgson). ⁷⁵ Joh. Mal. 14.19; Theoph. AM 5940.

  -

215

inactivity in this matter. But Leo’s letters show that the bishop of Rome still thought that Pulcheria could have some influence, and moreover that he believed it was appropriate to ask her to act as an intermediary with the emperor. Even after the council, Leo continued to plead with Pulcheria, imploring her to exert her influence over the emperor to allow a new council, which would reverse the decisions of the synod.

6. Making the Emperor: The ‘New Helena’ Theodosius II, who had resisted calling a new ecclesiastical council, died in the summer of 450. Since he did not have a male heir, and his wife Eudocia was living in Palestine, it was up to Pulcheria, as the last member of the Theodosian dynasty still resident in Constantinople, to play a role in the selection of a new emperor. As an Augusta and sister of Theodosius II, Pulcheria possessed ‘dynastic potential’, even though she was a consecrated virgin. Marcian was most probably the choice of the eastern magister militum, Aspar, but Pulcheria was certainly involved in the process that led to Marcian’s accession.⁷⁶ It is notable that there was apparently no consultation about the choice with Valentinian III, emperor of the West, who became senior Augustus on the death of Theodosius II. Given the circumstances, Valentinian was not willing to accept Marcian as his co-emperor.⁷⁷ In the East, however, Valentinian’s discontent seems not to have had any effect on Marcian’s legitimacy, which the latter gained through his marriage to Pulcheria Augusta. As Marcian’s wife, Pulcheria was able to become politically active again and organized the meeting of a new synod, as the bishop of Rome had asked her so urgently to do. This corresponded with her own religious interests and beliefs. The synod, which aimed to reverse the decisions of 449 and to restore Christian orthodoxy, took place in the autumn of 451 at Chalcedon. Letters that can be found in the Acts of the Council of Chalcedon point to Pulcheria’s significant influence in the synod’s organization.⁷⁸ During the course of the council, she was kept informed in writing about individual regulations and decisions made by the assembled bishops.⁷⁹ Finally, on the sixth day of the council’s meeting, Pulcheria appeared at Chalcedon together with Marcian to attend the reading of the new Nicene formula of the Christian faith.⁸⁰ A woman’s attendance at a church council was unusual in and of itself. But it does demonstrate Pulcheria’s particular ⁷⁶ On the various reports and traditions concerning Pulcheria’s involvement, see Burgess 1993/1994, who interprets Marcian’s elevation to the throne as the work of Aspar. Cf. Watts 2013: 275. In my opinion, Burgess underestimates Pulcheria’s bargaining power. Had Pulcheria not consented to Aspar’s plans, she could easily have refused to marry Marcian by referring to her vow. She could not have been forced as she had strong support from the people of Constantinople and the anti-Miaphysite/antiNestorian clergy. ⁷⁷ Evagr. HE 2.1. ⁷⁸ ACOec 2.3.2, 21 no. 33. ⁷⁹ ACOec 2.3.2, 86 no. 103. ⁸⁰ ACOec 2.1.2, 53 no. 15; Coll. Avell. 99.11.

216

   

interest in religious matters and that her authority in Christological disputes was acknowledged by the clergy. The latter becomes obvious by the acclamations addressed to her: Many years to the Augusta . . . God bless her, who descends from an orthodox family, God bless the protector of faith . . . all heretics you have expelled. You have prosecuted Nestorius and Eutyches . . . Marcian, the new Constantine, Pulcheria, the new Helena!⁸¹

Pulcheria’s presence at the council’s meeting was also of particular importance for Marcian. The imperial pair were celebrated by the assembled clergy as New Constantine and New Helena, the ‘ideal Christian couple’.⁸² The acclamation cited here shows that the victory of orthodoxy was attributed to Pulcheria herself. Through her particular form of religious self-representation, Pulcheria had gained a reputation for being a campaigner for orthodoxy both at court and beyond the city of Constantinople. As a member of the Theodosian dynasty, born to the purple during her father’s reign as Augustus, she possessed dynastic potential, which gave her the opportunity to secure the continued existence of the dynasty, even without giving birth to a child herself. As an Augusta crowned and thus legitimized by the Christian God, she was also a guarantor of orthodoxy. In the context of finding a new orthodox creed, Pulcheria was identified with the ideal Christian empress, Helena. There is no direct reference to the mother of Constantine as an imperial role model before the Council of Chalcedon in 451. Why, therefore, does she appear on this occasion? Helena was by this time acknowledged as the first Christian empress, the finder of the cross, builder of churches in both the Holy Land and Rome, and exceedingly philanthropic in all her public actions. She was now the point of reference for imperial women just as her son Constantine was for imperial men. We must remember that Marcian had only recently become emperor and had not yet developed his religious authority. To legitimize the decisions of the Council held at Chalcedon (instead of at Nicaea, where it originally was meant to be placed) and, most importantly, to legitimize the Nicene formula renewed on this occasion, it was important to identify the new emperor with Constantine and Pulcheria with Helena. Moreover, Pulcheria’s presence on the sixth day of the assembly was crucial to supporting Marcian’s authority as the legitimate successor of the Theodosian emperors, as well as his religious authority as a Christian emperor. Pulcheria thus emerges as a new type of imperial colleague for her husband Marcian. ⁸¹ ACOec 2.3.2, 175–6 no. 6.11: Augustae multos annos . . . quae ex genere orthodoxa est, deus eam custodiat. custodem dei deus custodiat . . . omnes hereticos tu expulisti. Nestorium et Eutychen tu persecuta es . . . Marcianus novus Constantinus. Pulcheria nova Helena; cf. 151–2 no. 6.4–5; 177–9 no. 6.13; 6.15; 6.20. ⁸² Drijvers 1993: 88.

  -

217

As a maker of emperors, Pulcheria was followed by the Augusta Ariadne, the daughter of Emperor Leo I and widow of Emperor Zeno.⁸³ In 491, only one day after Zeno had died, Ariadne appeared before the people of Constantinople, who cried out that she should choose a new emperor.⁸⁴ Ariadne’s mother Verina (the widow of Leo I) had twice attempted to replace her son-in-law Zeno, first in 474 with either her brother Basiliscus or her lover Patricius, and again in 483 with a general by the name of Leontius.⁸⁵ Even though Zeno faced many challenges to his power throughout his reign, Verina’s efforts were without success. But when Zeno died on 9 April 491, his subjects craved a stable government, which, they thought, could only be provided by an undeniably orthodox Roman emperor. They therefore asked the empress Ariadne to choose such a man. Ariadne, in establishing her own profile as wife of two Roman emperors, as a Christian empress, and especially as a guarantor of orthodoxy, followed in the footsteps of Pulcheria. As a devout Christian, opponent of the Nestorian and the Miaphysite heresies, defender of orthodoxy and the Nicaean/Chalcedonian Creed, and even a ‘new Helena’, Pulcheria had carved out a new role for the ideal Christian empress that resonated both within the imperial court and beyond.⁸⁶

⁸³ Croke 2015a: 293, 307. ⁸⁴ Const. Porph. de cer. 1.92 (1.101); Meier 2010; Croke 2015a. The crucial role of Ariadne in the transfer of imperial power is discussed further in Chapter 4. ⁸⁵ Leszka 1998: 128–36; Meier 2010: 280–4; Croke 2015a: 300–5. ⁸⁶ For Christian imperial women in the sixth century, see Chapter 14.

PART III

T H E P O L I T I C S OF A C C E S S

10 Beyond the Veil Athanasius at the Court of Constans Fabian Schulz

1. Introduction In Late Antiquity, bishops had spiritual, ascetic, and pragmatic authority.¹ The office of bishop, which had its origins in the early church, received a significant boost in status when Emperor Constantine chose to adopt Christianity.² As a consequence, the office started to appeal to members of the urban elite to whom it promised social promotion.³ While most bishops never encountered the emperor, some did manage to meet him in person. In religious disputes, leading bishops often tried to entice the emperor to support their theological view, which would give them an edge over their opponents. The emperor could choose to take sides motivated by personal religious conviction, the hope of restoring peace and order, or both. While all emperors ratified and enforced the decisions of church councils, some even tried to influence and shape the outcomes. Thus, a complex pattern of relationships developed between bishops and emperors over the course of Late Antiquity. During the fourth century, certain bishops appear to have had a strong and lasting influence on emperors, for example, Ossius of Cordoba on Constantine, Ursacius of Singidunum and Valens of Mursa on Constantius II, and Ambrose of Milan on Gratian and Theodosius I. It is misleading to label these individuals as ‘court bishops’, as was the case in early twentieth-century scholarship because bishops held no official positions at the imperial court.⁴ However, we should not go too far in the other direction, and argue that the relationship between emperors and bishops was largely impersonal.⁵ Some bishops did indeed earn the emperors’ trust.

¹ Rapp 2005. I would like to thank the Editors, as well as my colleagues Kamil Choda and Maurits de Leeuw, for their helpful comments and the German Research Foundation for funding our project ‘Power and Influence: Influencing Emperors between Antiquity and the Middle Ages’. ² Leppin 2018a: 135–254. ³ Haensch 2007. ⁴ The classic article is Hunt 1989. See further McLynn 1994: 330; Smith 2007: 187; Schulz 2014: 220, 241; Bernard 2017: 14–15. ⁵ As does Just 2003: 224–9. Fabian Schulz, Beyond the Veil: Athanasius at the Court of Constans In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0011

222

   

The writings of Athanasius of Alexandria are an important source for understanding the relationship between emperors and bishops in the fourth century.⁶ David Gwynn and Richard Flower have illustrated how Athanasius accused his rivals of using the court to boost their position, characterizing Ursacius and Valens as the mighty bishops who controlled Constantius II.⁷ Athanasius even claimed that their efforts to win over the emperor to their side represented a new development, because emperors and bishops had previously kept to themselves.⁸ This was obviously not true, as we know from Emperor Constantine’s concerted involvement in church affairs, including convening the Council of Nicaea in 325.⁹ Furthermore, Athanasius himself often sought the help of emperors both in writing and in person.¹⁰ Athanasius did not usually like to admit this, but he had to make an exception when his opponents alleged that he had instigated Constans against his brother, Constantius II. In order to acquit himself of this charge, Athanasius described his meetings with Constans in detail in a work known as the Apologia ad Constantium imperatorem (‘The defence addressed to Emperor Constantius’). In this chapter, I will first describe the fourth-century political and ecclesiastical developments and analyse the context and character of the work, building on the scholarship of Timothy Barnes.¹¹ I will then turn to examine what Athanasius has to say about his experiences at the court of Constans. The bishop’s account sheds light on important aspects of the meetings between emperor and bishop, such as the initiative, time and place, participants, protocol, topic, and outcomes. These descriptions, when analysed together with other sources such as ancient Church historians and the canons of the western Council of Serdica, reveal how Athanasius portrays his visits to Constans’ court to fit his ecclesiastical and political agenda.

2. Constans, Constantius II, and Athanasius In the fourth century, the Christian church was shaken by a theological standoff regarding question of Christology—that is, the relation between God the Father and his son Jesus Christ. The question of whether they were coeternal and what substance they shared had no easy answers but significant implications. Arius, a presbyter in Alexandria, thought that God created Jesus first, even before time; his enemies alleged that he made Jesus into an ordinary man, which would put the ⁶ Martin 2017: 63–4 argues that while there certainly were bishops who had a lasting influence on emperors, Athanasius was not one of them. ⁷ Gwynn 2007: 147–58; Flower 2013: 78–126. ⁸ Athan. Hist. Ar. 52. ⁹ Drake 2000. ¹⁰ See Barnes 1993: 132; Gwynn 2007: 148, with examples; and Martin 2017: 64. ¹¹ Barnes 1993: chapter VII ‘The Intervention of Constans’ 63–70; Martin 2017: 52 touches on these encounters only briefly.

  

223

whole doctrine about his death into question. The so-called Arian controversy, which started as a struggle within the clergy of Alexandria, eventually came to split the whole empire. In 325, Arius was officially condemned at the ecumenical Council of Nicaea, but this did not end the issue. The Council of Serdica (modern Sofia), instead of bringing the two sides together, ended with a schism in the year 343. The religious dispute was enhanced by political division. The two sons of Constantine, who had ruled over the empire since the death of their brother Constantine II in 340, leaned towards different theological positions: Constantius II, who ruled the East, was in favour of the Homoeans, who believed that God and his son Jesus Christ were of similar substance; Constans, who ruled the West, was inclined to the Nicaean cause, whose adherents believed that they were the same substance. The actions and writings of Athanasius of Alexandria were at the heart of the Arian controversy. In his polemical works, he coined the label ‘Arians’ as an umbrella term for all people he considered to be heretics, most of whom would not have actually subscribed to the teachings of Arius. As a staunch supporter of the decrees of the Council of Nicaea, he became a hero to his followers, who regarded him as a bishop who fought heretics and defied the emperor. However, Athanasius was not well regarded in most other eastern provinces, in which the Homoeans dominated. He was thus exiled a number of times. In 339, the Synod of Antioch deposed him from the see of Alexandria and expelled him from the church. When imperial troops pursued him, he fled to the West for the second time and settled in Rome. There Athanasius found supporters in the Church who received him into their community and declared him reinstated in his home bishopric of Alexandria, which was not accepted by his colleagues in the East.¹² At this point, the theological question became further complicated by the issue of who was actually entitled to make such decisions. Were the western bishops entitled to overrule the decision of the Synod of Antioch? And if Constans, who ruled the western half of the empire, decided to take up Athanasius’ cause, how would that secure his return to the East, which was under the sway of his brother Constantius II? At first, Constans was reluctant to take up Athanasius’ cause, but after he decided to do so, he grew more and more determined. Nicaean lobbyism helped in that. In the end, it was not the declaration of the western Council of Serdica but Constans’ threat of force that secured Athanasius’ return.¹³ Athanasius’ supposed influence on Constans’ temper later became a topic of hot dispute.

¹² Barnes 1993: 47–55; Martin 2017: 51. ¹³ There is an ongoing discussion as to whether the friction between the courts of Constans and Constantius II was confined to the religious sphere or extended beyond it. Callu 1992: 56 and 63 and Davenport 2020b think the former, Barnes 1993: 91, 98 and Portmann 1999: 305–19 the latter.

224

   

3. Apologia ad Constantium Imperatorem Constantius II became sole ruler of the Roman Empire in 353, following the murder of his brother Constans by the forces of Magnentius, and the subsequent suppression of Magnentius’ rebellion. The emperor’s conquest of the West and his theological position meant that the fortunes of the Homoeans were now in the ascendant, which put Nicaeans, and Athanasius in particular, in a difficult position, because they had lost their safe haven. This vulnerability was exploited by Athanasius’ opponents, who revived old allegations and presented new ones. Constantius initially summoned Athanasius to his court at Milan, but when he failed to appear, the Synods of Arles (353) and Milan (355) condemned him in his absence. The bishop preferred to stay in Alexandria, where he enjoyed strong support,¹⁴ and instead chose to defend himself to the emperor in writing. The Apologia ad Constantium imperatorem deals with allegations relating to Athanasius’ actions between 339 and 357, which he addresses in chronological order. Athanasius attributes these charges to the ‘Arians’ in general and to the bishops Ursacius and Valens in particular, singling them out for particular comment in the first chapter. Scholars agree that the Apologia is actually a fusion of two texts, but disagree where to place the division between the parts and how to date them.¹⁵ In any case, Athanasius’ account of his meetings with Constans belongs to the first and earlier part, which is usually dated between 353 and 356, the year in which Athanasius was exiled for the third time. The second part, which dates to the third exile (356–62), justifies his flight. The Apologia addresses the emperor in second person, just like a speech or letter. It is an open question whether this means that the text was actually meant to be read by the emperor himself or whether it is purely a rhetorical device.¹⁶ While it is clear that the second part of the work is aimed squarely at Athanasius’ Egyptian followers, I think that the audience for the first half is to be found in the West, particularly Constantius and his court and the Nicaean bishops who had condemned him at Arles and Milan.

4. Audiences with Constans In the Apologia, Athanasius describes at length his meetings with Constans, specifying when, where, why, and with whom they met. This level of detail is remarkable, as none of these audiences is mentioned in Athanasius’ other writings and few are mentioned by other authors. The omission of these court audiences from Athanasius’ other works is due to the fact that proximity to Constans was ¹⁴ Watts 2010: 180–1. ¹⁵ Brennecke 2006; Gwynn 2007: 37–9. ¹⁶ Barnes 1993: 63, 113; Brennecke 2006: 74, 77, 79; Gwynn 2007: 39.

  

225

regarded by his opponents as suspicious, but the details are vital to the argument of the Apologia. The first allegation was that Athanasius had instigated Constans against his brother. This claim was a rebuke to the charges that Athanasius had himself made about his opponents and their machinations, and posed a serious threat to his credibility.¹⁷ The bishop therefore denied the allegations, arguing that in fact they had not met often and that when they did, he never spoke negatively about Constantius.¹⁸ To prove his point, Athanasius refers to many witnesses, and provides a detailed account of his travels, to show where his audiences with Constans fitted into his itinerary.¹⁹ I will now proceed to analyse key aspects of Athanasius’ visits to court, as the bishop himself presents them.

(i) Time and Place According to the account of his travels, Athanasius only met Constans for the first time after he had been staying in the West for more than three years; the audience took place in Milan. Contrary to what is implied in the Apologia itself, the reason for this delay was probably not on account of the exiled bishop’s self-restraint, but rather a lack of interest on the part of the emperor. Constans had only taken up Athanasius’ cause when Paul of Constantinople was deposed and fled to the West in 342.²⁰ The second meeting happened at Trier in Gaul, before the Council of Serdica, the third in Aquileia in North Italy after the council, and the last one again in Trier. After departing from Alexandria, I did not go to the court of your brother, nor to any others, only to Rome. Entrusting my cause to the church (for I was concerned for this alone), I spent my time in public worship. I did not write to your brother except [on the occasions] when the Eusebians wrote to him against me and I was compelled to defend myself while I was still in Alexandria, and when, at his command that I prepare copies of the holy scriptures, I produced and sent them. [I say this because] in my defence I must tell the truth to your piety. So, after three years had passed, in the fourth year he wrote ordering me to present me before him. (He was then in Milan.) When I enquired into the reason (for I did not know the Lord is my witness), I learned that certain bishops had gone there and requested him to write to your piety so that a council might be held. Believe me, emperor, it happened like this; I am not lying. So I went down to Milan and experienced great generosity; for he graciously saw me and said that he had written and sent to you asking for a council to be held. I was still residing

¹⁷ For example, in the Apologia contra Arianos, on which see Gwynn 2007: 148–9. ¹⁸ Athan. Ap. Const. 2.1, 3.3–4. ¹⁹ Athan. Ap. Const. 3–4. ²⁰ Just 2003: 62; Parvis 2006: 202–6.

226

   

in the aforementioned city when he sent for me [to come] to Gaul, since Father Ossius was going there too, so that the two of us could travel [together] from there to Serdica. After the council, he wrote to me while I was residing in Naissus, and after going up to Aquileia I then remained there [until] the letters of your piety reached me there. And after being summoned again from there by your departed brother, I went to his court in Gaul and so came to your piety.²¹

Barnes dates the four audiences to the autumn of 342, the summer of 343, the spring of 345, and the autumn of 345.²² The cities of Milan, Trier, and Aquileia were all key imperial residences frequented by Constans and his court as they moved through the territory under his control.²³ Here and elsewhere, Athanasius calls the court the στρατόπεδον, a term which literally means ‘military camp’, and reflects the mobile nature of the fourth-century imperial court.²⁴

(ii) Initiative By denying that he had written to Constans after his arrival in the West, and by stressing that he came to court on each occasion because he was summoned by the emperor, Athanasius makes it clear that he did not take the initiative in these discussions. This was a very important point, because the canons of the western Council of Serdica (343) prohibited bishops from visiting the emperor on their own initiative.²⁵ In the eastern Church, there was a similar

²¹ Athan. Ap. Const. 4: Ἐξελθὼν ἀπὸ τῆς Ἀλεξανδρείας οὐκ εἰς τὸ στρατόπεδον τοῦ ἀδελφοῦ σου οὐδὲ πρὸς ἄλλους τινὰς ἢ μόνον εἰς τὴν Ῥώμην ἀνῆλθον καὶ τῇ ἐκκλησίᾳ τὰ κατ’ ἐμαυτὸν παραθέμενος—τούτου γὰρ μόνου μοι φροντὶς ἦν—ἐσχόλαζον ταῖς συνάξεσι. τῷ ἀδελφῷ σου οὐκ ἔγραψα ἢ μόνον, ὅτε οἱ περὶ Εὐσέβιον ἔγραψαν αὐτῷ κατ’ ἐμοῦ καὶ ἀνάγκην ἔσχον ἔτι ὢν ἐν τῇ Ἀλεξανδρείᾳ ἀπολογήσασθαι, καὶ ὅτε πυκτία τῶν θείων γραφῶν κελεύσαντος αὐτοῦ μοι κατασκευάσαι, ταῦτα ποιήσας ἀπέστειλα· χρὴ γὰρ ἀπολογούμενόν με ἀληθεύειν τῇ σῇ θεοσεβείᾳ. τριῶν τοίνυν ἐτῶν παρελθόντων τῳ ἐνιαυτῷ γράφει κελεύσας ἀπαντῆσαί με πρὸς αὐτόν· ἦν δὲ ἐν τῇ Μεδιολάνῳ. ἐγὼ δὲ διερωτῶν τὴν αἰτίαν—οὐ γὰρ ἐγίνωσκον, μάρτυς ὁ κύριος—ἔμαθον, ὅτι ἐπίσκοποί τινες ἀνελθόντες ἠξίωσαν αὐτὸν γράψαι τῇ σῇ εὐσεβείᾳ ὥστε σύνοδον γενέσθαι. πίστευε, βασιλεῦ, οὕτω γέγονε καὶ οὐ ψεύδομαι. κατελθὼν τοίνυν εἰς τὴν Μεδιόλανον εἶδον πολλὴν φιλανθρωπίαν· κατηξίωσε γὰρ ἰδεῖν με καὶ εἰπεῖν ὅτι ἔγραψε καὶ ἀπέστειλε πρὸς σὲ ἀξιῶν σύνοδον γενέσθαι. διάγοντα δέ με ἐν τῇ προειρημένῃ πόλει μετεπέμψατο πάλιν εἰς τὰς Γαλλίας—ἐκεῖ γὰρ καὶ ὁ πατὴρ Ὅσιος ἤρχετο—, ἵνα ἐκεῖθεν εἰς τὴν Σαρδικὴν ὁδεύσωμεν. μετὰ δὲ τὴν σύνοδον ἐν τῇ Ναισσῷ μοι διάγοντι γράφει, καὶ ἀνελθὼν ἐν Ἀκυληίᾳ λοιπὸν διέτριβον, ἔνθα με τὰ γράμματα τῆς σῆς θεοσεβείας κατέλαβον, κἀκεῖθεν κληθεὶς πάλιν παρὰ τοῦ μακαρίτου καὶ ἀνελθὼν εἰς τὰς Γαλλίας οὕτως ἦλθον παρὰ τὴν σὴν εὐσέβειαν. The translation follows Barnes 1993: 67; I have only made it more literal in 3, where I translate ἀνελθόντες with ‘had gone there’ instead of ‘had gone to court’ (Barnes). ²² Barnes 1993: 70, 224–5. ²³ For the establishment of these cities as principal residences under the Tetrarchy, see Chapter 3. On the movements of the sons of Constantine, see Barnes 1993: 218–28 (Appendix 9). ²⁴ Athan. Ap. Const. 4.1; for other cases in the works of Athanasius, see Müller 1952: 1356; and more generally Smith 2007: 188. ²⁵ Serdica Canon 8 (Latin) and Canon VII (Greek); Hess 2002: 203–4, 216–17, 230–1. Ossius of Cordoba, who proposed this measure, gives more details about the reasoning behind it: the great

  

227

prohibition.²⁶ These canons show that a mere thirty years after Constantine’s victory at the Milvian Bridge, there was a growing desire to regulate the court appearances of clerics, who formed a new ‘lobbying group’ that did not exist in the early Roman Empire. The problem was that anybody, even a ‘heretic’, could attempt to gain the emperor’s favour. Instead of curbing the emperor’s power, the aim of the Council of Serdica was to establish a procedure of selection by mandating that petitioners should be invited by the emperor and have the consent of important bishops. If Athanasius wanted to convince the audience of his innocence, his apology had to deal with this sensitive topic. The implicit message is clear: even before this code of conduct was established in the West, he was himself in full compliance.

(iii) Witnesses Athanasius emphasizes that there are many witnesses who can attest to the veracity of his testimony about his meetings with Constans. He states: I never went alone to see your brother, nor did he ever converse with me alone. I always entered his presence with the bishop of the city where I was and other bishops who happened to be there: we saw him together and we departed again together. On this matter Fortunatianus, the bishop of Aquileia, can bear witness, and Father Ossius is capable of speaking, as are Crispinus, the bishop of Patavium; Lucillus of Verona; Dionysius of Elis; and Vincentius, the Campanian bishop. And since Maximinus of Trier and Protasius of Milan have died, Eugenius too who was magister can bear witness. For he stood before the veil and heard the requests we made of Constans and what he graciously said to us.²⁷

This passage serves as evidence that Athanasius never met Constans alone. This was not necessarily self-evident, as there were private meetings between emperors frequency and ‘bad foundation’ of petitions, particularly from African bishops (probably concerning the Donatists), had caused the court to overlook and disregard ‘legitimate’ petitions. Bishops like Ossius, who already knew the emperor or lived close to him, counted on being invited more often. ²⁶ Antioch Canon 11; Stephens 2015: 66 and 243. The relationship between the Canons of Antioch and those of Serdica is complex and the topic of Stephens 2015. ²⁷ Athan. Ap. Const. 3: οὐ γὰρ μόνος ἑώρακά ποτε τὸν ἀδελφόν σου οὐδὲ μόνῳ μοί ποτε ἐκεῖνος ὡμίλησεν, ἀλλ᾿ ἀεὶ μετὰ τοῦ ἐπισκόπου τῆς πόλεως ἔνθα ἦν καὶ ἄλλων τῶν ἐκεῖ παρατυγχανόντων εἰσηρχόμην κοινῇ τε αὐτὸν ἐβλέπομεν καὶ κοινῇ πάλιν ἀνεχωροῦμεν. δύναται Φουρτουνατιανὸς ὁ τῆς ᾿Ακυληίας ἐπίσκοπος μαρτυρῆσαι περὶ τούτου, ἱκανός ἐστιν ὁ πατὴρ ῞Οσιος εἰπεῖν καὶ Κρισπῖνος ὁ τῆς Πατάβων καὶ Λούκιλλος ὁ ἐν Βερωνὶ καὶ Διονύσιος ὁ ἐν ῎Ηλιδι καὶ Βικέντιος ὁ ἐν Καμπανίᾳ ἐπίσκοπος. καὶ ἐπειδὴ τετελευτήκασι Μαξιμῖνος ὁ Τριβέρεως καὶ Προτάσιος ὁ τῆς Μεδιολάνου, δύναται καὶ Εὐγένιος ὁ γενόμενος μάγιστρος μαρτυρῆσαι· αὐτὸς γὰρ εἱστήκει πρὸ τοῦ βήλου καὶ ἤκουεν ἅπερ ἠξιοῦμεν αὐτὸν καὶ ἅπερ αὐτὸς κατηξίου λέγειν ἡμῖν. The translation follows Barnes 1993: 65, except for the last sentence, where he chooses the pluralis majestatis and translates ‘I . . . me’. I would prefer the collective plural, because Athanasius wants to play down his own part.

228

   

and bishops.²⁸ We learn from this passage that Athanasius was always introduced at court by the local bishop. The bishops of the principal imperial residences, such as Trier and Milan, were always likely to know the emperor better, since he attended church during major Christian festivals and invited the bishop to celebrations.²⁹ Athanasius could benefit from their familiarity with the emperor in gaining access to court and, in his subsequent defence, use their hospitality as proof that he was no foreign trouble-maker. The bishops in question were Fortunatianus of Aquileia (mentioned first in the list of witnesses), Maximinus of Trier, and Protasius of Milan (mentioned last). Maximinus was the most prominent of the three. He had received Athanasius in Trier during his first exile (335–7) and had supported his cause ever since. What about the other bishops mentioned in Athanasius’ catalogue? Ossius of Cordoba was a prominent Nicaean veteran who had presided over the Council of Nicaea, on account of his positive relationship with Constantine. Like Maximinus, he was a long-time supporter of Athanasius. The other bishops, who, with the exception of Dionysius of Elis,³⁰ all came from Italy, are not so well-known. Barnes has argued that they all participated in the third audience in Aquileia during the spring of 345.³¹ This means that Athanasius’ account in the quoted passage moves back in time, skipping the fourth and last meeting in Trier, because the local bishop, Maximinus, was already dead and therefore could not be called upon as a witness, and ending with the first meeting in Milan, whose bishop was also long dead. It would make sense for Athanasius to concentrate on the third meeting, because it occurred during the period in which relations between the imperial brothers were at their worst (as we shall see shortly). The next chapter states, however, that Ossius and Athanasius met Constans in Trier, which would be the second audience in the summer of 343.³² So it may have been the case that other bishops were present at more than one meeting as well, although we cannot say this with certainty. Information from other sources can help us to reconstruct Athanasius’ journeys to court and in whose company he might have seen the emperor. In addition to Athanasius and Ossius, the bishops Vincentius, Lucillus, and Dionysius participated in the Council of Serdica (autumn of 343), while Crispinus signed the letters later.³³ After Serdica, Vincentius, a colleague, and a magister militum led a

²⁸ Not only between the ‘Arians’ and Constantius, as Athanasius claims. In the 380s, Ambrose tried to meet the usurper Magnus Maximus in private, but only managed to see him in the consistory. See Smith 2007: 203, on Amb. Ep. 24. ²⁹ McLynn 2004: 246. ³⁰ According to Barnes 1993: 258 n. 9, this Dionysius, whose affiliation is garbled in the manuscripts, should be identified with Dionysius of Elis in Achaia, who was deposed, but then took part in the western Council of Serdica (Greece belonged to the part of the empire under Constans’ rule). ³¹ Barnes 1993: 66; Martin 1996: 440 n. 229; Just 2003: 213. ³² Athan. Ap. Const. 4.4. According to Clercq 1954: 316, this was the only meeting. ³³ Athan. Apol. c. Ar. 48–50, nos. 9, 15, 63, 256.

  

229

delegation charged with informing Constantius about the outcome and delivering a supporting letter by Constans. According to Athanasius, it was the council which despatched the delegation, while Theodoret said it was Constans himself.³⁴ The delegation arrived in Antioch around Easter 344. Athanasius says in the next chapter, that after Serdica he stayed in Naissus (which is 90 miles or 150 km northwest of Serdica), where he is known to have celebrated Easter, and then travelled to Aquileia (on the north-eastern edge of Italy), and that he stayed there.³⁵ According to Barnes, the actual audience with Constans only took place there a couple of months later, in early 345.³⁶ The long interval between Serdica and the meeting between Athanasius and the emperor has some significant implications. With the exception of Athanasius, all of the bishops in question who attended the council had bishoprics to look after, so they would not have been able to remain nearby afterwards. At Serdica, a canon was enacted against bishops who spent too much time away from their bishoprics.³⁷ So Lucillus, Dionysius, and Crispinus had probably travelled home and stayed there until they were summoned to Aquileia. As Maximinus of Trier and Protasius of Milan, who introduced Athanasius to audiences in their cities, were already dead, Athanasius had to resort to different witnesses for the meetings there (i.e. autumn of 342 and summer of 343 in Milan, and autumn of 345 in Trier). He therefore names the magister Eugenius as an eyewitness.³⁸ As Athanasius describes Eugenius as standing in front of the veil which concealed the emperor, he was either magister officiorum, the highest court official, who was a permanent member of the imperial council and responsible for the reception of embassies,³⁹ or the magister admissionum, who was charged with admissions.⁴⁰ The fact that Athanasius only names Eugenius when he cannot produce a living bishop makes one wonder if Eugenius was also present in the meeting at Aquileia that happened in between (spring of 345). Since Eugenius mediated several audiences or was at least present several times, he was probably responsible for ecclesiastical matters in the circle of the emperor’s confidants and was possibly favourably disposed towards Athanasius’ cause. As his links to Corinth suggest, Eugenius likely came from Greece.⁴¹ So for Athanasius, who did not know Latin well, a native Greek speaker was a useful contact.

³⁴ Athan. Hist. Ar. 20.2. Theod. HE 2.6 states that Constans himself chose the two bishops from the participants of Serdica. ³⁵ Athan. Ap. Const. 4.5. ³⁶ Barnes 1993: 70, 89, 225; Athan. Ap. Const. 15.4 and Festal Index 17. ³⁷ Serdica Canon 14 (Latin); Canon XI (Greek); Hess 2002: 222–3, 234–5. ³⁸ Olszaniec 2013: 142–3, however, thinks that Athan. Ap. Const. 3.6 only attests to a single audience that could not have taken place in 342. The problem is that Olszaniec does not read the passage in context. ³⁹ Clauss 1980: 60–72; Delmaire 1995: 75–96; Tantillo 2015: 555. ⁴⁰ PLRE I Eugenius 5 suggests he was already magister officiorum, while Olszaniec 2013: 142 proposes that he was still magister admissionum. Barnes 1993: 66 leaves the question open. On the magister admissionum, see Clauss 1980: 19, 137–8, 152; Delmaire 1995: 43, 74; Tantillo 2015: 552–3. ⁴¹ Olszaniec 2013: 142–4.

230

   

Supporting the emperor in church politics was one of the core tasks of a magister officiorum.⁴² Athanasius knew that. A predecessor of Eugenius in office had been Philumenus, Constantine’s magister officiorum from 325 to 331, who was something like the emperor’s special representative for Church affairs.⁴³ This is what his appearances in the Arian controversy suggest. At the end of the Council of Nicaea in 325, Constantine decreed that whoever did not sign the creed should be banished. Philumenus was to carry out the order. The magister presented the creed to Arius and his sympathizers for signature. When they refused, they had to go into exile.⁴⁴ A couple of years later, Athanasius was being accused of having bribed Philumenus to join a conspiracy against Constantine and stood trial before the emperor, who dismissed the charges (331); allegedly the trial was instigated by ‘Arians’, who wanted to get rid of the new bishop of Alexandria, because he opposed the imperial pardon of Arius.⁴⁵ Given that Athanasius had been accused of bribing the magister officiorum of Constantine (and in turn accused his rivals of using the court to boost their position), we understand why Athanasius only reluctantly names Constans’ magister officiorum as a witness, although Eugenius evidently was an important facilitator for the petitioning bishops.

(iv) Ceremony Athanasius’ mention of the veil and the magister are important indications that the meetings with Constans were formal audiences which took place in a reception hall. This is designed to contrast with the emperor’s private meetings with other bishops, which Athanasius insists he did not have. Those encounters took place in the emperor’s chambers, the realm of the despised eunuchs.⁴⁶ The bishop emphasizes the public nature of his reception with the emperor, because plotting cannot take place when one is being watched.⁴⁷ We are well-informed about how the official admission (admissio) and veneration (adoratio) worked.⁴⁸ Ammianus ⁴² Clauss 1980: 82–98. ⁴³ PLRE I Philumenus; Olszaniec 2013: 321–3; Clauss 1980: 82–4, 87–8, 183. ⁴⁴ This is only attested by Philost. HE 1.9a. ⁴⁵ Athan. Apol. c. Ar. 60.4; Soz. 2.22.8; Soc. 1.27.15: all authors call these allegations baseless. See Barnes 1993: 21; Olszaniec 2013: 322–3. ⁴⁶ Barnes 1993: 127; according to Athan. Hist. Ar. 6, Eusebius of Nicomedia, Athanasius’ archenemy, was introduced to Constantine by imperial women. Hist. Ar. 38 deals with eunuchs and Constantius. The infamous dance of the ‘Arian’ bishops refers to private meetings (Hist. Ar. 52, quoted in the Introduction of this book). In pagan sources there are similar allegations against Constantius and his eunuchs; Hunt 1989: 86 on Amm. 15.3.5; 18.3.6. On court eunuchs, see Scholten 1995; Tougher 2008: 163–83. ⁴⁷ See the discussion of the politics of access in the Introduction of this book. ⁴⁸ Alföldi 1970: 28, 36–7, 62–4; Kolb 2001: 39–40, 118–21; Smith 2007: 209–25. Christian Rollinger’s upcoming Habilitationsschrift Gesten der Macht. Das Zeremoniell des Kaiserhofs in Konstantinopel will also deal with this topic.

  

231

describes the ceremony in some detail: when the general Ursicinus was received by Constantius in Milan (355), the magister admissionum called him into the consistory-chamber, where the emperor and the high officials were assembled.⁴⁹ The consistory (consistorium) was the imperial council.⁵⁰ An example of the audience chamber can still be seen today in the form of the Basilica Constantiniana at Trier, which was discussed by Verena Jaeschke and Caillan Davenport in Chapter 3. The emperor would have sat at the far end of the hall in the apse. He was concealed by a veil, which shielded his sanctity from the world. Petitioners had to traverse these curtains. That is what is implied here when Athanasius refers to ‘entering’ and ‘seeing’ (εἰσηρχόμην—ἐβλέπομεν) Emperor Constans.⁵¹ Roman court ceremonial was very much a watching game, as members of the audience waited to see who was admitted, and in what order, until it was finally their turn, and they were invited to gaze on the emperor.⁵² It is interesting that in order to describe the veil, Athanasius uses the word βῆλον, the Greek rendering of Latin velum, and not a synonym like κάλυμμα or καταπέτασμα.⁵³ Does this give the scene, which took place in the West, a local touch? I do not think this is the case: the palace was a Latin place even in the East.⁵⁴ While in Greek Christian literature the nouns κατα-, παρα-, or περιπέτασμα refer to veils or curtains in various places, the loanword βῆλον is almost a technical term relating to the imperial court or a court of justice.⁵⁵ In a court of justice, the veil protected the secretarium and the judge, in the imperial court the consistorium and the emperor.⁵⁶ Most prominently, the βῆλον features in the Acts of Christian martyrs, which are often set at court.⁵⁷ The Passion of Euplus, which takes place in Catania during the Diocletianic Persecution, begins with the martyr professing his faith in the council-chamber ‘before the veil’ (πρὸ βήλου); upon hearing this, Calvisianus, the governor of Sicily, invites him in for questioning.⁵⁸ As in Athanasius’ Apologia, the veil may be opaque, but it does allow sound to pass through. Both the Acts of Trypho (set at Nicaea during the Decian persecution) and the Acts of Barbarus (set during the first year of the reign of Julian, when he was campaigning against the Franks) come to an end when the governor or ruler (in the first, Aquilinus, governor of Bithynia; in the second, Emperor Julian) ⁴⁹ Amm. 15.5.18. ⁵⁰ Schlinkert 1998; Smith 2007: 168–71, 198–203, 214–20. ⁵¹ Athan. Ap. Const. 3.5; cf. Athan. Ap. Const. 4.4: εἶδον πολλὴν φιλανθρωπίαν. ⁵² See Wallace-Hadrill 2011, and for similar ceremonies in other monarchical cultures, see Duindam 2016: 255–73. ⁵³ In his works, Athanasius uses βῆλον only a second time (see Müller 1952: 197); in Athan. Hist. Ar. 56, it refers to the curtains that separated the altar and the presbyterium from the nave. At church, therefore, the veils protected the sanctity of God from the gazes of laymen. However, κάλυμμα and καταπέτασμα are attested a number of times (see Müller 1952: 712 and 744). ⁵⁴ This was still the case under Theodosius II (408–50), as highlighted by Millar 2006. ⁵⁵ Lampe 1969 s.v. βῆλον; καταπέτασμα; παραπέτασμα; περιπέτασμα. ⁵⁶ Färber 2014: 297–312. ⁵⁷ Caillan Davenport discusses court scenes in other martyr Acts in Chapter 13. ⁵⁸ Acta Eupli 1.1; Färber 2014: 297.

232

   

orders the veil to be closed (συρθῆναι τὸ βῆλον), after the martyr’s confession of faith has angered him. He then deliberates and delivers his verdict.⁵⁹ It is interesting that both Acts use the same vocabulary—this detail of court protocol was probably intended to authenticate the setting of the legendary stories. In the Apologia, Athanasius seems to be using βῆλον in a similar way, namely in order to authenticate the account of his visits. The use of curtains at the imperial court pre-dates Late Antiquity. Officials known as velarii are attested in the palace on the Palatine in Rome, though their exact ceremonial function is unknown.⁶⁰ The introduction of the veil to the ritual of meeting the emperor is usually associated with Diocletian and his reforms of court ceremonial, including the introduction of the ‘adoration of the purple’ (adoratio purpurae).⁶¹ By the Byzantine period, the curtains were formally opened with the shouting of the command ‘Lift!’ (leva), according to the tenth-century Book of Ceremonies.⁶² Athanasius’ testimony in the Apologia is often overlooked in this discussion of the development of court ceremonial, despite the fact that it is one of our earliest pieces of evidence for the post-Diocletianic court.⁶³ The bishop’s audiences with Constans occurred only shortly before the publication of the Codex-Calendar of 354, which depicts the emperor’s brother, Constantius II, sitting on the throne under a curtained baldachin (Figure 10.1).⁶⁴ In his account, Athanasius stresses that on every occasion when entering, seeing, talking, and leaving he did not act alone, but was always together with his colleagues (οὐ . . . μόνος . . . οὐδὲ μόνῳ μοί . . . , ἀλλ᾿ ἀεὶ μετὰ . . . κοινῇ . . . κοινῇ). This emphasis is meant to prove that there was no private time with the emperor in which he could have incited Constans against his brother and imperial colleague. However, it is probable that the actual entrance by the bishops into the emperor’s presence would not have been quite as unified as Athanasius implies. Imperial audiences were all about hierarchy. For imperial officials, there were even laws regulating precedence according to rank.⁶⁵ How was that supposed to work for bishops who were equal, at least in theory? The canons of the western Council of Serdica show that in practice, there were marked hierarchical differences ⁵⁹ Passio sancti Tryphonis 10; Passio sancti Barbari 11. The meaning of συρθῆναι τὸ βῆλον is not easy to grasp. The Latin version of the Passio sancti Barbari implies that the curtain was closed (ante faciem suam adtrahi velum). The Latin version of the Acta Eupli, which Färber 2014: 303–4 discusses, seems to imply a similar procedure. ⁶⁰ Winterling 1999: 98 n. 73; Färber 2014: 307. In later periods, the curtains were operated by eunuchs (Ringrose 2003: 174). ⁶¹ Alföldi 1970: 37–8; Tantillo 2015: 574. The bottom of the curtains is visible in the Tetrarchic fresco on the south wall of the Luxor imperial cult chamber (McFadden 2015a: 121). ⁶² Const. Porph. de cer. 1.47 (1.56); Coripp. Iust. 3.255–63. For discussion, see Kelly 2004: 22–6; Pazdernik 2010: 63–5. ⁶³ For other Constantinian evidence, see Färber 2014: 305. ⁶⁴ The Codex-Calendar of 354 is only extant in Renaissance copies. Salzman 1990: 34 and Appendix 1 discusses the complicated transmission and the image of Constantius II. See further Eberlein 1982 on the representation of the veil in the visual arts. ⁶⁵ Kolb 2001: 117–19; Smith 2007: 215.

  

233

Figure 10.1 Constantius II as depicted in the Codex-Calendar of 354. MS Barb. lat.2154.pt.B (f. 13r). © Bibliotheca Apostolica Vaticana.

between provincial and metropolitan bishops, and that the bishop of Rome had a special position.⁶⁶ It is likely, therefore, that when Constans held court in cities such as Trier, Milan, and Aquileia, it was the bishop of the city who was admitted ⁶⁶ Serdica Canon 9b and 10a (Latin); Canon IXa and IXb (Greek); Hess 2002: 216–19, 232–3.

234

   

first. Had the bishop of Rome participated, he would likely have taken precedence. Hence, Athanasius probably elides some of the hierarchical differential between the bishops to emphasize his presence as part of an ecclesiastical collective. He also prefers not to discuss the next part of the ritual, the adoratio purpurae, of which Ammianus has left us a detailed account.⁶⁷ Athanasius was probably happy to pass over this stage in silence, as venerating a worldly ruler in this way was difficult for a man of God.⁶⁸

(v) Discussions with the Emperor Athanasius reveals little about the content and effect of the audiences with Constans, except that he did not slander Constantius II and therefore played no part in inciting enmity between them. Furthermore, Athanasius claims that Constans spoke of his brother’s generosity and that he himself prayed for the ruler of the East.⁶⁹ However, we also learn that the assembled bishops made requests (ἠξιοῦμεν) and that the emperor replied (κατηξίου λέγειν ἡμῖν—note the imperfect of duration and repetition).⁷⁰ This suggests that the audiences were not as straightforward as Athanasius simply responding to the summons of Constans. He wanted something in return. This was only natural: emperors were supposed to be generous to petitioners.⁷¹ The Codex-Calendar of 354 depicts Constantius enthroned with coins pouring from his right hand and ceremonial plates stacked behind his throne (Figure 10.1). Imperial generosity, however, was not limited to material gifts, but also encompassed requests for favours and privileges. We do not know what Athanasius and the other bishops asked of Constans, though they may have discussed the restitution of exiled bishops. Athanasius is more willing to discuss other audiences when it suits the purpose of his argument. According to his own account of his travels, when he arrived in Milan in 342, Constans had already decided to write to Constantius about holding an ecumenical council, which certain bishops had requested at an earlier audience.⁷² At this point in the Apologia, Athanasius implores Constantius to believe the veracity of his account. That the bishop actually succeeded in dispelling the emperor’s doubts is questionable, however, because his narrative is highly tendentious. The eastern faction at Serdica counted Athanasius among the initiators of the council.⁷³ And this was not only an ‘Arian’ allegation;

⁶⁷ Amm. 15.2.27. ⁶⁸ Alföldi 1970: 64; Kolb 2001: 118–19. ⁶⁹ Athan. Ap. Const. 3.3. ⁷⁰ Athan. Ap. Const. 3.6. ⁷¹ Athan. Ap. Const. 4.4; Schulz 2019: 217. On imperial gifts, see Beyeler 2011. ⁷² Athan. Ap. Const. 4.3. However, Athan. Hist. Ar. 15.2 has Constantius take the initiative. ⁷³ The other members included Julius, Maximinus, and Ossius; cf. letter of the eastern synod (Hilarius, Collectanea antiariana A IV 1 = 58, ed. Feder 1916). Clerq 1954: 310 thinks that Ossius and Maximinus were the leading figures and that they presumably met Constans in Trier.

  

235

the major Nicaean church historians thought that Athanasius played a prominent role. Sozomen and Socrates hold Athanasius (and Paul of Constantinople) responsible.⁷⁴ Theodoret even summarizes the arguments that Athanasius allegedly used to convince Emperor Constans.⁷⁵ So at this point, we may doubt the veracity of the Apologia, since the weight of testimony indicates that Athanasius had some part in the summoning of Serdica, at least indirectly. Furthermore, Athanasius alludes to an embassy from Constantius II, led by a certain Thalass(i)us, which met Constans at Poetovio in Pannonia during the winter of 344/5, when Athanasius was not present. According to Athanasius, it was this delegation that had the greater opportunity to instigate Constans against his brother.⁷⁶ This eastern delegation is otherwise unattested, but Thalassius is well-known to have been a confidant of Constantius.⁷⁷ Athanasius’ implication is clear: at the decisive (or divisive) moments at court, he himself was not there.

5. Alternative Narratives Athanasius’ Apologia offers a very different account of the bishop’s encounters with Constans from that found in other ecclesiastical historians. Most church historians leave out the meeting between Athanasius and Constans altogether, and attribute the hostile letter to Constantius to Constans’ own initiative. According to Socrates, Sozomen, and Theodoret, Constans was so annoyed about the outcome of the Council of Serdica that he was driven to write to his brother.⁷⁸ This was meant to remove Athanasius as a figure of influence from their works. In contrast, Rufinus and Philostorgius write that Constans and Athanasius had a meeting in an undisclosed location, which then resulted in Constans writing a harsh letter to his brother in which he threatened to restore Athanasius by force.⁷⁹ Barnes has proposed that this letter was in fact the result of the embassy from Constantius II led by Thalassius, and that Athanasius met Constans in Aquileia a little later.⁸⁰ Athanasius then used the Apologia to assure Constantius that Constans had less reason to be angry with his brother after his meeting than after receiving the embassy of Thalassius.

⁷⁴ Soz. 3.11; Soc. 2.20. ⁷⁵ Theod. HE 2.4. Parvis 2006: 201 takes this account at face value, although Theodoret is known to embroider his accounts of meetings between emperors and bishops, for instance, in the encounter between Theodosius and Ambrose (Theod. HE 5.17–18). ⁷⁶ Athan. Ap. Const. 3.3. ⁷⁷ PLRE I Thalassius 1. Athan. Hist. Ar. 22 mentions him among Constantius’ comites in 345. ⁷⁸ Soc. 2.22; Soz. 3.30; Theod. HE 2.6. ⁷⁹ Ruf. HE 10.20; Philost. HE 3.12. Soc. 2.22 gives the longest summary of the letter, which Barnes 1993: 89 accepts as historical. According to Soz. 3.20.1 (followed by Barnes 1993: 66, 89), this letter came after a milder one. ⁸⁰ Barnes 1993: 66, 89.

236

   

Barnes’ reconstruction fits the known movements of Athanasius and Constans well. Constans is attested in the autumn of 344 in Pannonia, in April of 345 in Aquileia, and in May the same year in Trier, which means that the reception of Thalassius at Poetovio should have taken place in the winter of 344/5, before the emperor headed west. It is remarkable that while the emperor was busy at Poetovio, Athanasius was already at Aquileia, where the emperor had invited him to an audience. The distances involved are revealing. Poetovio was some 125 miles (200 km) from Aquileia, while Naissus, where Athanasius was staying before he was invited to court, was 550 miles (900 km) from Aquileia.⁸¹ This means either that Constans invited Athanasius and the other bishops months ahead of schedule or that he was delayed. In any case, the bishop(s) had to wait for a long time until the emperor received them. It is probably safe to assume that bishops were kept waiting more often than their own accounts and those of the Church historians would want to admit. Athanasius’ defence that he did not compel Constans to write the threatening letter was much needed. Firstly, if it was the first time that the emperor had met Vincentius since the bishop had returned from Antioch, the envoy would have given him a first-hand account of the gruesome reception of his delegation.⁸² The apparent mistreatment of the imperial delegation must have angered Constans. Secondly, according to Philostorgius, Athanasius obtained the threatening letter from the emperor after bribing figures who were influential at court, foremost Eustathius, the comes rerum privatarum.⁸³ The comes rerum privatarum was in charge of the imperial estates and an ex officio member of the imperial consistory.⁸⁴ It was not the first time that Athanasius had been accused of bribery. During the reign of Constantine, he had been charged with bribing Philumenus, the then magister officiorum (see above). Of course, allegations of bribery were commonplace in such conflicts.⁸⁵ The veracity of such claims is hard to assess. In these two cases, we can only say that according to Nicaean Church historians, Constantine rejected the charges against Athanasius concerning Philumenus, whose fate remains unclear, and that Eustathius’ career does not seem to have suffered from the accusation. While bribery was officially frowned upon, the exploitation of patronage and connections was a natural part of court life. The canons of Serdica even encouraged bishops with ‘legitimate’ requests to use their friends at court to ensure that their petitions were heard.⁸⁶ Bishops who took advantage of court connections and/or made gifts are well-attested in the

⁸¹ I calculated the distances using an interactive road map of the Roman Empire designed by the Stanford Orbis Project (https://orbis.stanford.edu). ⁸² Athan. Hist. Ar. 20; Theod. HE 2.7. ⁸³ Philost. HE 3.12; PLRE I Eustathius 2. Martin 1996: 422 n. 155 and Martin 2017: 52 think that Philostorgius refers to the audience in Milan (342). ⁸⁴ Delmaire 1995: 119–47; Smith 2007: 199f. ⁸⁵ See especially Kelly 2004: 138–85. ⁸⁶ Serdica Canon 9b (Latin); Canon IXa (Greek); Hess 2002: 204–7, 218–19, 232–3.

  

237

later fourth and early fifth century, as Daniëlle Slootjes will discuss in the next chapter.⁸⁷ So it is likely that Athanasius and his colleagues used court contacts in some way. At the same time, it becomes understandable why Athanasius mentions just one of the court officers, of whom several were present, only when he has to, since these witnesses could be used against him. In the end, Constans’ threats caught Constantius off guard, as he was campaigning against Persia. The eastern emperor decided to back down. In a personal letter, he asked Athanasius to return. Before leaving the West and heading to Alexandria, Athanasius paid a visit to Constans in Trier, their fourth and final meeting, where he probably thanked him for his support.

6. Conclusion During the sole reign of Constantius II, the Homoeans extended their influence to the western provinces, where the Nicaeans had previously been dominant under his brother Constans. The Synods of Arles (353) and Milan (355) condemned Athanasius, the Nicaean champion who had been exiled to the west by Constantius (339) and restored to his see thanks to the support of Constans (345). As the decisions of these synods deprived Athanasius of his western support and threatened his already feeble position in the East, the bishop had to rebut the allegations which charged him with both heresy and treason. In the first half of the Apologia ad Constantium imperatorem, Athanasius defends himself against the accusation that during his exile to the West, he had incited Constans to threaten his brother and imperial colleague. It was a difficult task to convince Constantius, his court, and the western clerics, the explicit and implicit addressees of the Apologia, that he did not do this, because Constans, whose anger would secure Athanasius’ return, had actually met the bishop a number of times. Furthermore, rumours about Athanasius’ involvement were widespread. Athanasius’ line of defence was to provide counter-evidence, including details of his visits to court and the names of witnesses who could verify his account. This description is invaluable, because it gives insights into episcopal audiences. At the same time, the defence contains distortions and omissions, which are worth uncovering. The Apologia ad Constantium details four audiences in Milan, Trier, and Aquileia between 342 and 345, in which Athanasius and other bishops met Constans. On each occasion, the small group was headed by the local bishop. The meetings were formal receptions, which took place in the reception hall and had a fixed protocol (admissio): the bishops had to wait and line up; the emperor sat on the throne behind a veil and was surrounded by officers. In stressing the ⁸⁷ See also Weber 1997: 76, 85 on the followers of Priscillian; Pfeilschifter 2013: 491–2 on Porphyry of Gaza; Millar 2006: 219–20 on Cyril of Alexandria.

238

   

collective audience before Constans, Athanasius was at pains to show that his meetings were not private, as had been the case with the ‘Arian’ bishops Ursacius and Valens, and that they were in line with the canons of the western Council of Serdica, which barred clerics from visiting the emperor on their own initiative. The veneration (adoratio) which Athanasius passes over in silence was ambiguous: on the one hand, it was an honour to be received in that way and to kiss the purple; on the other hand, it delineated difference and hierarchy. Athanasius stresses that these meetings happened at the emperor’s invitation, but also admits that the bishops made requests of Constans. We do not know what exactly occurred beyond the veil. We can presume that Athanasius and his colleagues discussed the restitution of the exiled bishops and the importance of the Nicaean faith. Therefore, Athanasius’ argument that in the decisive and divisive moments he was not present is questionable. Athanasius was part of the collective lobbying effort, which led to the summoning of the Council of Serdica and to his eventual reinstatement to the see of Alexandria. Court officials, particularly members of the imperial consistory like the magister officiorum, appear to have been important intermediaries. But as allegations of bribery were powerful and incendiary, Athanasius only mentions them sparingly. Athanasius’ account of visiting Constans at court reveals the important role played by local bishops as brokers of patronage in the major imperial residences, such as Milan, Aquileia, and Trier in the West. These bishops stood above the metropolitan bishops, who, according to the canons of Serdica, decided whether to forward a provincial petition to the emperor and whether to write a letter of recommendation to the bishop of the city in which the emperor was residing at the time. Moreover, according to the same canons, bishops were not supposed to spend too much time away from their bishoprics, which suggests that the average bishop’s time at court would always be limited. This emphasizes the potential power and influence of bishops in the key imperial cities, which could be exploited by individuals such as Ambrose of Milan in the late fourth century.⁸⁸ Soon after finishing the Apologia, Athanasius must have realized that his attempt to placate Constantius had failed. That is when he changed his tone from apologetic to aggressive: Constantius was not just a possible victim of ‘Arian’ influence, but became their puppet or even their master. In the Historia Arianorum, which dates to 357, the meetings between Constantius and the Homoean bishops are portrayed as the exact opposite of the audiences described in the Apologia. The meetings happen in shadowy chambers and eunuchs conspire with the ‘Arians’. Athanasius’ own meetings with Constans are left out. This distorted image resonated with later ecclesiastical historians, who frequently

⁸⁸ McLynn 1994; Schulz 2014.

  

239

discussed the presence of ‘Arians’ at court and said little about Athanasius’ own actions beyond the veil. In the decades to come, accounts of ecclesiastical court visits were often distorted to present a particular version of events. For example, the so-called Petitiones Arianorum (‘Petitions of the Arians’) which were transmitted among Athanasius’ writings, feature Emperor Jovian, who had just recalled Athanasius from his fourth exile in 363. Jovian is shown rejecting the pleas of ‘Arian’ clerics and laymen from Alexandria, even as they beg him at the imperial gates in Antioch to revoke his decision. He also asks them to convert to the Nicene doctrine. This little-studied document is not an authentic protocol of an actual encounter, as has sometimes been assumed, but a masterpiece of negative campaigning.⁸⁹ As Athanasius was about to reclaim his see, the aim of the document was to demonstrate which camp enjoyed imperial support and which did not. In truth, Jovian was much less determined in promoting the Nicene faith. But the fact of the matter was that very few people really knew what happened at court, and depended solely on the stories of those who had been there. As the writings of Athanasius show, such accounts were often grounded in plausible details of the court experience, but their arguments were highly tendentious.

⁸⁹ See Schulz 2019.

11 Dynamics of Power The Nestorian Controversy, the Council of Ephesus of 431, and the Eastern Imperial Court Daniëlle Slootjes

1. Introduction In recent years, ideas and concepts from the field of court studies have started to influence the way in which historians analyse the functioning of the Roman, late antique, and Byzantine imperial courts. The influence of Norbert Elias’ groundbreaking work The Court Society is especially felt to this day, even though various scholars, in particular Jeroen Duindam, have presented different perspectives on, and refinements of, Elias’ model of the court as a society and institution.¹ For the ancient world, the work of Rowland Smith has attempted to compare Elias’ ideas about the court of Louis XIV of France with the workings of the late antique court. The starting point for this chapter is Smith’s suggestion that an analysis of the interactions between Emperor Theodosius II (408–50), his sister Aelia Pulcheria, and ecclesiastical authorities could offer further insights into the functioning of the imperial court in the first half of the fifth century. For the emergence of Christian leaders as influential figures in Late Antiquity had introduced a new dynamic to the court environment, which had to be negotiated and confronted.² This chapter focuses on the events surrounding Theodosius’s decision to summon the Council of Ephesus of 431 and the outcome of the meeting.³ The principal players in these events were the emperor; his female relatives, in particular his sister Pulcheria; Cyril, bishop of Alexandria (412–44); and Nestorius, bishop of Constantinople (428–31). While the controversy which prompted the Council of Ephesus appears, at first sight, to be a theological dispute, its course and developments had much broader repercussions, which were felt at the imperial court and beyond.⁴ For many generations, scholars considered Theodosius to be a ¹ Elias 1969 (German edition), 2006 (English edition by Dublin University Press); Duindam 1995; 2003; 2016; Smith 2007. ² Smith 2007: 230–1. ³ The Greek and Latin documents of the Council of Ephesus, published in the ACOec, have now received an English translation and commentary in Price and Graumann 2020. ⁴ See for instance Grant 1975; Luibhéid 1982; Ulrich 1997; Pietras 2016. Daniëlle Slootjes, Dynamics of Power: The Nestorian Controversy, the Council of Ephesus of 431, and the Eastern Imperial Court In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0012

  

241

weak emperor whose regime was dependent on figures such as imperial women, generals, and eunuchs. Yet, in recent years, scholars such as Fergus Millar and Hugh Elton have shown that Theodosius was no mere figurehead, but a ruler who played an active role in matters of state. This is particularly relevant to a study of the Council of Ephesus, for, as Thomas Graumann has argued, the imperial documents in which the emperor set out his rationale for calling the meeting demonstrate that Theodosius carefully considered how best to bring peace to the warring Church.⁵ This chapter will contribute to this ongoing reassessment of Theodosius’ reign by exploring the way in which the theological conflict between Nestorius and Cyril, and the emperor’s attempts to arbitrate, played out at the imperial court in Constantinople.

2. Methodological Considerations The chapter will examine three features of court life—alliances, patronage, and accessibility—that emerge in almost all modern investigations of royal courts. Elias’ The Court Society provides a helpful framework for conceptualizing some of these aspects, such as the way in which courtiers sought to gain a ruler’s favour in formal receptions. However, he did not, for example, consider the way in which alliances and groups (both formal and informal) shaped the court community. Nor did Elias discuss two important aspects of court society that are fundamental to this study. The first is religion and the influence of ecclesiastical officials, an omission which is surprising given the importance of royal confessors at the court of Louis XIV and other monarchs in Early Modern Europe.⁶ The second is the role of outsiders, individuals, and groups at or beyond the margins of the court, whose voices and actions could still play a role in decision-making.⁷ Questions of Christian imperial piety and orthodoxy loomed large in Late Antiquity, and emperors were publicly held to account for their religious beliefs and policies both by bishops and by the residents of Constantinople. The role of the non-elites is crucial here. The Constantinopolitans gathered and protested not only in public locations such as the hippodrome, but also in and around the city’s churches. Religious passion ran so high that it even led to outbreaks of public violence, either spontaneously or as the result of goading by ecclesiastical figures. The relationship between Christian leaders and their congregations (or followers) was thus crucial

⁵ Wessel 2001; Millar 2006; Elton 2009; Graumann 2013; Kelly 2013b; Price and Graumann 2020: 24. ⁶ Duindam 2011: 7. For example, Elias 2006: 176–7 briefly discusses France’s religious wars in his evolution of the court society in general. On confessors, see Reinhardt 2007; 2009; and Chapter 14. ⁷ Elias’ work was firmly focused on the court society itself, e.g. 2006: 73–85 on the relationship between the aristocracy and the court.

242

   

in swaying public opinion and provoking protests outside the court arena with a view to having an impact on political judgements.⁸ We shall discuss these various elements in turn. The ‘mechanics of decisionmaking’ at court under Theodosius—to use Hugh Elton’s phrase—depended on the complex interaction of formal power (bestowed by the holding of high office) and informal power (exercised through access, influence, and intimacy).⁹ As Elton has rightly pointed out, any model that we develop for understanding the workings of the imperial court needs to consider how, and by whom, these types of power were exercised. He has drawn attention to two groups of courtiers. Firstly, there were the imperial officials and members of the sacred consistory who advised the emperor on a routine basis. We should not underestimate the role played by these officials, invested in formal power by their office, in ensuring the functioning of Roman imperial government, even though their routine labours are not usually featured in our major literary sources. These courtiers need to be contrasted with the second group, composed of exceptional figures who wielded outsized power for particular periods. Their power was often based on personal charisma or other forms of informal authority, rather than deriving from their office. One such individual was Theodosius’s eunuch Chrysaphius, the praepositus sacri cubiculi. These imperial ‘favourites’, to use a term popularized by Early Modern court studies, dominate our literary narratives.¹⁰ In this chapter, we shall see how both types of officials were strategically courted by ecclesiastical leaders seeking to establish connections and alliances to help their cause. Christian bishops did not hold formal government or court offices but were nevertheless deeply enmeshed in court politics. This was because the power of their ecclesiastical office gave them the moral and spiritual authority which could challenge the emperor himself. The problem may be set out as follows. The imperial office was endowed with a set of prerogatives which, since the first century, had given the emperor absolute power in the political, military, legal, and religious affairs of his empire. However, in the later Roman Empire a fundamental difference developed between the level of control that emperors had over the political, military, and legal domains on the one hand and the religious realm on the other. While emperors continued to be the highest political and military leaders of the empire, the emergence of Christianity as the dominant religion led to their loss of the highest religious position, which now lay in the hands of Christian bishops. Consequently, emperors were regarded no longer as

⁸ Smith 2007: 163. See Gregory 1979: 106–8, on the notion that violence carried out by bishops’ followers was often accepted, and Drake 2011 for the idea that violence was a ‘by-product’ of the struggle between bishops and emperors. ⁹ Elton 2009: 133. ¹⁰ Elton 2009: 137. Cf. Millar 2006: 204–7; Smith 2007: 198; Harries 2013: 7–89. For the concept of the ‘favourite’ and the ‘minister-favourite’, see Chapter 14.

  

243

the primary source of religious power, but as the embodiment of the highest secular power. This crucial shift had consequences for the way in which an emperor could control the various aspects of his imperial position. On the one hand, an emperor delegated his supreme political and military power through the appointment of high civil and military commanders such as the urban prefect, the praetorian prefect, or the magister militum. For the duration of their usually short terms of office, these officials could be held directly accountable by the emperor.¹¹ In the religious realm, on the other hand, an emperor had to recognize a level of autonomy and independence on the part of the Christian Church, in terms of both the election of the bishops and the religious authority that these bishops held.¹² In other words, although Christian bishops were required to accept the emperor’s supreme worldly authority, his imperial power was not absolute as far as matters of the Christian Church and its leadership were concerned. The extent of the bishops’ power depended on their official position and the boundaries of their church community. Bishops of small communities exercised authority over relatively circumscribed areas, but the metropolitan bishops had rights and privileges that extended over a wide geographical area. While the highest secular power was invested in one man, the emperor, the highest religious authority was shared by the bishops of the most important sees of Rome, Alexandria, Antioch, Jerusalem, and Constantinople.¹³ In practice, the boundaries between ecclesiastical and secular authority were not always clearly defined. Bishops, especially in smaller communities, started to take over the duties and even some of the court cases of local or imperial officials, as their permanent local presence generated a certain level of trust and stability. Such involvement in civic matters blurred the official limits of their ecclesiastical powers.¹⁴ Sometimes bishops would ask emperors or other imperial officials to interfere in ecclesiastical affairs. If they thought this was to their own or the Church’s advantage, they regarded such intervention as beneficial. We have already seen in Fabian Schulz’s examination of the fourth-century affair of Athanasius (Chapter 10) how the competition between different ecclesiastical pressure groups led to intense lobbying of emperors at court. Therefore, the view of the Church leadership towards the emperor was not always completely straightforward: he could be a valued arbiter or troublesome challenger. These dynamics were given an added layer of complexity by the actions of urban Christian populations. Emperors and bishops could encourage or goad Christian communities to engage in demonstrations of support or opposition, ¹¹ Though see Chapter 8 on the long terms of office of magistri militum in the fifth century. ¹² See Canons 5 and 15 of the Council of Nicaea; Herrmann-Otto 1980: 298–302; Liebeschuetz 2001: 141; Norton 2007. ¹³ Millar 2006: 133–7 provides a helpful overview. ¹⁴ Millar 1971; Brown 2002; Slootjes 2004; Slootjes 2011.

244

   

but equally such actions could emerge spontaneously from popular will. Crowds could never be completely controlled, which made them not only potent weapons, but also extremely unstable ones. For instance, while the bishop Cyril of Alexandria instigated a campaign of intimidation against the philosopher Hypatia, her murder at the hands of the mob might be interpreted as a sign of the bishop’s loss of control over the masses rather than his actual intent.¹⁵ Nevertheless, the behaviour and protests of urban populations, especially in the city of Constantinople, formed an important barometer of public opinion for the emperor and his court.¹⁶

3. The Key Players We now turn to examine the relationship between the key players, beginning with the emperor and his family. The positions of Theodosius II and his sister Pulcheria were intimately connected, as Anja Busch has explored in Chapter 9. Their father, Emperor Arcadius, had appointed Theodosius as co-Augustus in 402 when he was still a baby. After Arcadius’ death in 408, Theodosius, aged seven, became the sole ruler of the eastern empire. In 414, Pulcheria, only a little older than her brother, took a vow of virginity, and was proclaimed as Augusta.¹⁷ The visible expression of her imperial position was confirmed by the placement of her portrait bust in the senate house of Constantinople alongside the Augusti, Theodosius and his uncle Honorius (who ruled the western empire).¹⁸ The eastern imperial family comprised not only Theodosius and Pulcheria, but also their sisters Arcadia and Marina, and eventually the emperor’s wife Aelia Eudocia and their daughters Licinia Eudoxia and Flaccilla (the latter sadly short-lived). The Theodosian dynasty was particularly successful in marshalling the different forms of power and influence wielded by its male and female members into what Jill Harries has called a ‘collective family enterprise’.¹⁹ The Theodosian family and its operation within the wider court society can be profitably considered from the perspective of the study of ‘group behaviour’ undertaken by scholars in the field of social psychology.²⁰ Their work has demonstrated that all groups consist of smaller sub-groups that function slightly differently, with their own aims and agendas, even though they continue to be part of the larger group.²¹ We should envision the Theodosian dynasty—the ¹⁵ Rist 1965: esp. 222–3; Russell 2000: 8–9; Wessel 2004: 46–57; Irshai 2013. ¹⁶ Price and Graumann 2020: 23–4 emphasize the importance of public sentiment. ¹⁷ Chron. Pasch. a. 414; Soz. 9.1.5–6. ¹⁸ Holum 1982: 97. ¹⁹ Harries 2013: 69. See also Holum 1982: 79–111; McLynn 2018: 16. ²⁰ There is an extensive bibliography on group behaviour from the field of social psychology. Important contributions include Tajfel 1982; Tajfel and Turner 1986; Noel 1995; Brown 2000; Hogg and Tindale 2001; Brown 2020. ²¹ Brown 2000; Hogg and Tindale 2001.

  

245

emperor himself, his sisters, wife, and children—as constituting a core group that courtiers and outsides alike wished to associate with and attempt to influence. Yet the family unit also consisted of a number of smaller groups. The household of Pulcheria, for example, was separate from that of Theodosius II himself, and staffed by the Augusta’s own praepositus sacri cubiculi who decided who would have access to her.²² Pulcheria and her sisters Arcadia and Marina were public figures, particularly involved in matters of religious piety and the patronage of Christian communities.²³ As Anja Busch points out in Chapter 9, Pulcheria’s status and representation as an Augusta meant that petitioners sought her support, sometimes for the favours and perquisites she could grant herself, at other times for the influence she could wield with her brother. Pulcheria’s household represented an important group in its own right, as well as a sub-group of the larger Theodosian family unit. We turn now to the ecclesiastical players. Cyril succeeded his uncle Theophilus as bishop of Alexandria in 412.²⁴ Apparently not all parties involved in Alexandria wanted Cyril to obtain the bishop’s seat, but his power base in Alexandria and beyond was strong enough to secure his nomination.²⁵ In his description of Cyril’s appointment, Socrates presents the bishop as someone who immediately started to overstep the limits of his authority.²⁶ Socrates does not mention the precise ways in which Cyril violated the boundaries of his office, but his statement certainly illustrates the contemporary awareness of the tension between ecclesiastical and imperial power.²⁷ Cyril’s actions in his first years as bishop of Alexandria did not go unnoticed in Constantinople. In 416, an Alexandrian delegation arrived at the imperial court to complain about interference of the bishop and his clergy in worldly affairs, even implying Cyril’s involvement in the murder of Hypatia in the previous year.²⁸ Here we see the emperor exercising his traditional authority to arbitrate conflicts between his subjects.²⁹ Theodosius and his western colleague, his uncle Honorius, may have indirectly referred to Cyril’s behaviour in a law of that same year that was directed to the praefectus praetorio Orientis Monaxius. In the law, which focused on problems in Alexandria, they reaffirmed the boundaries between secular and ecclesiastical affairs as follows: It is the pleasure of Our Clemency that clerics shall have nothing to do with public affairs and with matters pertaining to the municipal council.³⁰

²² Harries 2013: 88. On the role of the praepositus, see Smith 2007: 202–3, 206–7, 230. ²³ Soc. 7.22.4–6; Soz. 9.1.10–11; Holum 1982: 91; Harries 2013: 74, 88. ²⁴ Russell 2000: 3–6; Wessel 2004: 15–22. ²⁵ Russell 2000: 6. Cf. Wessel 2004: 22 in particular on Cyril’s support. ²⁶ Soc. 7.7. ²⁷ Wessel 2004: 16–17 has argued that Socrates’ negative presentation of Cyril needs to be understood as the historian’s personal opinion. Cf. Wallraff 1997: 79, 112. ²⁸ Holum 1982: 98–9. ²⁹ Millar 1977. ³⁰ PLRE II Monaxius; CTh. 16.2.42 (416), placet nostrae clementiae, ut nihil commune clerici cum publicis actibus vel ad curiam pertinentibus habeant (trans. Pharr 1952).

246

   

If, indeed, the emperors had Cyril in mind while pronouncing this law, then Theodosius’ view of the behaviour of the Alexandrian bishop was not particularly positive, though at the same time he did not give in to the delegation’s wishes or publicly admonish Cyril.³¹ The final figure, Nestorius, appeared on the stage in early 428 when Emperor Theodosius approached him to fill the see of Constantinople, which had lain vacant since the death of the bishop Sissinius in December 427.³² Sissinius’ passing had triggered serious rivalry between possible candidates and various religious groups in Constantinople, forcing the emperor to intervene.³³ The election of a bishop was not strictly within the emperor’s purview, but he was certainly empowered to arbitrate disputes between subjects. Moreover, it demonstrated the emperor’s eagerness to continue his traditional highest religious function as mediator between the divine and the general population of their empire.³⁴ The emperor’s choice of Nestorius is described by the ecclesiastical historian Socrates as follows: They [= the emperors] therefore sent for a stranger from Antioch, whose name was Nestorius, a native of Germanicia, distinguished for his excellent voice and fluency of speech; qualifications which they judged important for the instruction of the people.³⁵

Theodosius had personally appointed Nestorius, who was very much an outsider to Constantinople, as Peter Van Nuffelen has noted.³⁶ Whereas Socrates merely states that Nestorius was brought to the city, the Life of Hypatius by Callinicus mentions that the emperor sent Dionysius, probably his magister militum per Orientem, to accompany the incoming bishop of Constantinople to his see.³⁷ In other words, the emperor decided to use an official, and one of his generals at that, to escort Nestorius, in order to demonstrate his support for the new prelate. After he assumed office, Nestorius continued to draw support from prominent office-holders. For instance, by 429, when the theological dispute about the nature of Mary started to cause serious tension in Constantinople, and the monastic party in the city publicly opposed Nestorius,

³¹ Holum 1982: 99. ³² Soc. 7.29. McGuckin 1996: 7–9; Russell 2000: 31; Wessel 2004: 4. ³³ Gregory 1979: 81–2; Holum 1982: 148–9. ³⁴ Theodosius II clearly cultivated this image of mediator, as expressed in his Sacra ad Cyrillum Alex. et ad singulos metropolitas (of 19 November 430), ACOec. 1.1.1 pp. 114–15. Cf. Wessel 2001: 287; 2004: 92–5, 99, 140. ³⁵ Soc. 7.29: Ἐπήλυδα δὲ ἐκ τῆς Ἀντιοχείας καλεῖν ἐβούλοντο· ἦν γὰρ τις ἐκεῖ Νεστόριος τοὔνομα, τὸ μὲν γένος Γερμανικεύς· εὔφωνος δὲ ἄλλως και εὕλαλος· διὸ καὶ ὡς προσεπιτήδειον εἰς διδασκαλίαν ἔγνωσαν μεταπέμπεσθαι (trans. Walford 1853). ³⁶ Van Nuffelen 2010b: 442–7. See also Holum 1982: 147–51; Wessel 2004: 84–5, 92. ³⁷ Callin. Vita Hypat. 32.1. See Gregory 1979: 82; Holum 1982: 148–9.

  

247

the bishop had people who protested against him flogged and treated as if they were criminals by calling in the city prefect to deal with them.³⁸ It was hardly normal practice for religious leaders to take prisoners as if they had secular authority, but the bishop could evidently count on the urban prefect to punish his opponents. Yet Nestorius turned out to be a problematic bishop who did not bring the stability that Theodosius had hoped for. He managed to irritate and anger many constituencies in Constantinople, including a strong urban community of monks, the general populace of the city, not to mention members of the imperial family.³⁹ For example, at the beginning of his appointment, Nestorius promised that he would deal with heresies, targeting religious groups such as the Arians and Novatians. However, Nestorius’ efforts threatened the emperor’s power base, since many of the emperor’s German and Gothic troops, including some generals, were members of the Arian community.⁴⁰ The bishop also quickly earned the wrath of the Augusta Pulcheria.⁴¹ Nestorius took away several of Pulcheria’s privileges, such as the right to dine in the episcopal palace after Sunday service and to enter the sanctuary to receive communion. Furthermore, he removed her portrait hanging above the altar of Hagia Sophia, and he openly denied her close connection to the Virgin Mary.⁴² This represented a public denial of the religious position that Pulcheria and the imperial house had carefully cultivated over the previous twenty years.⁴³ Socrates observed that Theodosius himself was the only one who was able to keep Nestorius in check.⁴⁴ As Theodosius had brought Nestorius to Constantinople, one can imagine that Nestorius would have been keen on maintaining the emperor’s support and therefore more willing to listen to him than to other parties, if only for the sake of keeping his position. But at the same time, the public nature of the opposition to Nestorius within Constantinople and within his own family meant that Theodosius had to consider how long he would continue to support the bishop, and what form this support would take. This was an issue that the emperor was forced to confront during the so-called ‘Nestorian controversy’.

³⁸ ACOec. 1.1.5 pp. 7–10; Russell 2000: 38. Cf. Holum 1982: 157–8. ³⁹ McGuckin 1996; Wessel 2004: 86–90. ⁴⁰ Holum 1982: 151. See McEvoy 2016a: 498–502 on the Arian faith of Theodosius II’s generals, and Lee 2013 on the way in which the emperor successfully managed relationships with his senior commanders. ⁴¹ Gregory 1979: 82–5; Russell 2000: 31–3; Wessel 2004: 84, 101–2; Van Nuffelen 2010b: 446. ⁴² Nest. Herac. pp. 96–7 (trans. Driver and Hodgson); Holum 1982: 152–8, 163–4; McGuckin 1996: 16–19; Russell 2000: 31–3; Wessel 2004: 101–2. ⁴³ Cf. Price and Graumann 2020: 24–5, who regard the relationship between the two as more ‘ambivalent’. ⁴⁴ Soc. 7.29.4–5. Cf. CTh. 16.5.65.

248

   

4. Cyril and Nestorius in Search of Support: Alliances, Patronage, and Accessibility In 428, a theological dispute began to foment which would bring Theodosius II, Pulcheria, Cyril, and Nestorius into conflict with each other, and envelop members of the imperial court more widely.⁴⁵ Even though this chapter is not focused on the contents of the theological dispute itself, it is necessary to outline the key issues. At its heart stood the figure of Mary and her status within the Church, either as theotokos (‘she who gave birth to God’), Christotokos (‘she who gave birth to Christ’), or anthropotokos (‘she who gave birth to man’).⁴⁶ Cyril of Alexandria regarded the title theotokos as the most appropriate, but Nestorius—although he seems to have considered both terms acceptable—opted for Christotokos as the most suitable one because of its closeness to the language of the New Testament.⁴⁷ Even though Nestorius tried to convince his opponents of his theological position, resistance against him grew steadily within the city of Constantinople.⁴⁸ The crisis soon reached the ears of other church leaders, including Cyril. Nestorius and Cyril agreed on the fundamentals of the Nicene Creed, but they profoundly differed in explaining the consequence of the consubstantiality of the Father and Son for the nature of salvation.⁴⁹ Cyril penned public letters, designed for wide circulation, in which he tried to justify the use of the term Mary theotokos.⁵⁰ Nestorius felt provoked by the writings of Cyril, considering the bishop of Alexandria as guilty of openly questioning his wisdom and spiritual authority.⁵¹ An exchange of public letters between the two bishops followed, allowing the wider Christian community to catch a glimpse of the competition between them.⁵² As holder of the Alexandrian see, Cyril would have been regarded as a formidable opponent to Nestorius, since Constantinople was still the newest of the eastern bishoprics, its youth not necessarily outweighed by the presence of the emperor and his court in the city.⁵³ Cyril and Nestorius both realized they needed to gather support, first and foremost from fellow church leaders and their congregations.⁵⁴ Cyril compiled a dossier about Nestorius’ unorthodox beliefs and sent it to Rome in order to offer bishop Celestine insights into the Constantinopolitan prelate’s theological ⁴⁵ Cf. Wessel 2004: 90–103. ⁴⁶ Price and Graumann 2020: 67–84. ⁴⁷ Wessel 2004: 1, 85, 126–37; Beeley 2012: 258–64. ⁴⁸ Holum 1982: 158; Russell 2000: 34–5. ⁴⁹ Russell 2000: 39–44; Wessel 2004: 2–4. ⁵⁰ ACOec. 1.1.1 pp. 23–5 (= McEnerney 1987 no. 2); ACOec. 1.1.1 pp. 25–8 (= McEnerney 1987, no. 4; cf. Tanner 1990: 40–4 for translation of the Greek version); ACOec. 1.1.1 pp. 33–42 (= McEnerney 1987 no. 17; cf. Tanner 1990: 50–61); Wessel 2004: 76–83. ⁵¹ Russell 2000: 34–6. Cf. letter of Cyril to Celestine, the bishop of Rome; ACOec. 1.1.5 pp. 10–12 (= McEnerney 1987 no. 11, here 11.3). ⁵² Two letters by Nestorius to Cyril: ACOec. 1.1.1 p. 25 (= McEnerney 1987 no. 3) and ACOec. 1.1.1 pp. 29–32 (= McEnerney 1987 no. 5; cf. Tanner 1990: 44–50), and another letter by Cyril to Nestorius: ACOec. 1.1.1 pp. 25–8 (= McEnerney 1987 no. 4; cf. Tanner, 1990: 40–4); Russell 2000: 36. ⁵³ Holum 1982: 152; Wessel 2004: 83–4, 98. ⁵⁴ Russell 2000: 37.

  

249

position.⁵⁵ As the proud holder of the oldest Christian see, Celestine then urged Nestorius to correct his views.⁵⁶ Nestorius was formally condemned at a meeting of bishops in Rome on 10 August 430.⁵⁷ To Cyril’s relief, the western ecclesiastical authorities had chosen his side in the dispute. Celestine characterized Cyril’s writings as ‘a most ready cure through which the pestilential disease may be completely driven out by a wholesome remedy’ and encouraged the bishop of Alexandria to carry on with his work, ‘since the authentic teaching of our see is in harmony with you, using our apostolic authority you will carry out this decree with accurate firmness’.⁵⁸ As Susan Wessel rightly points out, the Roman support for Cyril’s position against the see of Constantinople must have worried Theodosius II and his advisors, as well as others who supported Nestorius.⁵⁹ Even those parties in Constantinople that were annoyed by Nestorius might still have been concerned at this joining of the two powerful ecclesiastical forces of Rome and Alexandria, which potentially threatened the religious position of Constantinople. With the Roman support of Celestine in hand, Cyril started to seek backing at the imperial court for his theological position.⁶⁰ This was quite a formidable task. For Nestorius, as the bishop of Constantinople, had been able make use of his physical proximity to the imperial court to exert his influence.⁶¹ Cyril’s correspondence shows that he was aware from a very early stage that his situation was being discussed at court. Some discontented Alexandrians had brought a case against him to Constantinople, sending petitions to both the emperor and to Nestorius asking for restitution.⁶² Cyril decided not to travel to Constantinople to press his case in person. Instead, he used letters to try to persuade Theodosius of the rightness of the cause, since he believed that even the emperor was subject to God.⁶³ This epistolary strategy may well have been prompted by problems of access. Not only was it difficult to obtain personal audiences with the emperor, but also suppliants were uncertain as to the form such an audience would take.

⁵⁵ ACOec. 1.1.5 pp. 1–12 (= McEnerney 1987 no. 11); Holum 1982: 158; Russell 2000: 37; Wessel 2004: 105–7. ⁵⁶ Epistula Caelestini papae ad Nestorium (judgement on Roman synod) (CPG 8639); ACOec. 1.1.1 p. 82; Wessel 2004: 106–11; Price and Graumann 2020: 29–30. ⁵⁷ Holum 1982: 161; Wessel 2004: 111. ⁵⁸ Letter of Celestine, bishop of Rome to Cyril: ACOec. 1.1.1 pp. 75–7 (= McEnerney 1987 no. 12, quotes from 12.1 and 12.6); ἑτοιμοτάτη θεραπεία, δι’ ἧς ἡ λοιμώδης νόσος ὑγιεινῶι φαρμάκωι ἀπελαθείη and συναφθείσης σοι τοίνυν τῆς αὐθεντίας τοῦ ἡμετρου θρόνου, τῆι ἡμετέραι διαδοχῆι χρησάμενος ταύτην ἐκβιβάσεις ἀκριβεῖ τὴν ἀπόφσιν, with Russell 2000: 6, 37. Wessel 2004: 103–5 shows that in the West, the emperors’ interference was only desired in the case of heresies, while in the East, there seems to have developed a greater acceptance of imperial intervention in a range of matters. ⁵⁹ Wessel 2004: 111. ⁶⁰ Wessel 2004: 90–103. ⁶¹ Cf. Russell 2000: 31. On the influential position occupied by the bishop of the city which the emperor resided in, see Chapter 10. ⁶² Nest. Herac. pp. 100–1 (ed. Driver and Hodgson); ACOec. 1.1.1, pp. 110–12 is the crucial letter. ⁶³ ACOec. 1.1.1 p. 42 l. 13 to 43 l. 9, especially 43 l. 7; Wessel 2004: 97.

250

   

As Fergus Millar has shown, religious petitioners did not know whether they would be heard by the emperor alone, by the emperor and his consistorium, or by the senate without the emperor present at all. Sending a letter might not have as much impact as seeking an audience, but it also lessened the risk of being denied access or humiliated publicly in Constantinople.⁶⁴ In 430, the bishop penned a series of letters, one to Theodosius II himself, another ‘to the Augustae’ (ad Augustas), presumably Eudocia and Pulcheria, and a third ‘to the imperial mistresses’ (ad Dominas), which included the emperor’s other sisters Arcadia and Marina (who, like Pulcheria, were consecrated virgins).⁶⁵ In the letter ‘to the imperial mistresses’, he appealed directly to the pious virginity of Pulcheria, Arcadia, and Marina, which had been impugned by Nestorius’ public rebukes to Pulcheria. Cyril’s strategy shows that he recognized that the virginal sisters constituted a separate and important sub-group within the larger Theodosian family unit. Whereas Cyril encouraged the emperor to follow the orthodox position, in his letter to the female members of the dynasty, he was quite clear about his expectation of the imperial women: ‘decorate your holy heads with apostolic laurels by leading whomever you decide to help to the correct faith’.⁶⁶ In other words, Cyril expected the female members to be able to take their own, independent stance in the religious conflict. Indeed, Pulcheria had already used bishops to press her own beliefs against those of Nestorius. She employed Proculus, bishop of Cyzicus, to deliver a sermon in a church near her residence which exhorted the congregation to share Pulcheria’s point of view in the Marian controversy.⁶⁷ Cyril may well have hoped that Pulcheria would engineer another such public demonstration, this time in support of him. However, Cyril’s epistolary lobbying was not the success he had clearly hoped it would be. The emperor’s wrath was aroused when Cyril sent further letters, one addressed to Theodosius and his wife Eudocia, and another to his sister Pulcheria alone.⁶⁸ In his reply to the bishop, Theodosius asked whether he thought the members of the family had different views, or might be prompted to different views. Even though the dynasty consisted of different sub-groups, with their own aims and agendas, the emperor wanted to project an image of dynastic unity. In his letter to the bishop, Theodosius expressed considerable irritation at Cyril’s strategy:

⁶⁴ Millar 2006: 205. ⁶⁵ Cyril, Oratio ad Augustas (ACOec. 1.1.5 pp. 26–61) and Oratio ad Dominas (ACOec. 1.1.5 pp. 62–118). Letter of Theodosius to Cyril: ACOec. 1.1.1 p. 73. For discussion, see Holum 1982: 159–61; Wessel 2001: 287–8; 2004: 98–9, 139. ⁶⁶ ACOec. 1.1.5 p. 70: προσβαλοῦσαι δὲ τοῖς ἐφεξῆς κεφαλαίοις εῦθῆ καὶ ἀπεξεσμένον εἰς ἀλήθειαν τὸν ἐμὸν εὑρησετε λόγον (trans. Holum 1982: 160). ⁶⁷ Holum 1982: 157, who notes that the date is Christmas in the years 428–30. ⁶⁸ ACOec. 1.1.1 pp. 73–4; Millar 2006: 153–4, cf. Holum 1982: 159–60.

  

251

Your goal and purpose are the same in seeking to separate members of the church from its body and members from the one body of the imperial house, as if there were no other way to win a reputation.⁶⁹

In other words, Theodosius felt it was important to give the appearance of one imperial voice and Cyril’s letters publicly implied that there was discord. The emperor’s annoyance was increased by Cyril’s elevation of himself above the ranks of his fellow bishops, as if his cultivation of imperial connections gave greater weight to his theological position, all the while ignoring and threatening the ecclesiastical hierarchies of the church.⁷⁰ Cyril had furthermore sought to blur the boundaries between ecclesiastical and imperial authority by trying to involve the emperor and his family in a theological dispute that was not necessary theirs to solve. In this particular instance, Cyril found himself up against a universal aspect of group behaviour identified by social psychologists. When the interference or influence of an outsider threatens group cohesion, the group engages in defensive behaviour by closing ranks and confirming their united position.⁷¹ The words of the emperor, the leader of the Theodosian family, illustrate this process in action. Nestorius, as the bishop of Constantinople, and the emperor’s own choice for the see, thought that he could count on the support of Theodosius against Cyril. However, as we have already seen, it had become clear early on that the bishop was not diplomatically gifted and he had already alienated members of the imperial house, particularly Pulcheria. Thus, when Nestorius decided to gather a council of his own in Constantinople to resolve the theological quarrel, the assumption that he would automatically have imperial support turned out to be a misjudgement. In November 430, Theodosius made use of the imperial prerogative and called for a church council. The imperial decree is a restrained and diplomatic piece of rhetoric, as Thomas Graumann has acutely observed. The emperor does not assign blame, but remains aloof and impartial, desirous of an end to the dispute for the good of the empire.⁷² Theodosius also declined to hold the council in Constantinople, instead opting for the more neutral city of Ephesus. It was geographically positioned between Constantinople and Alexandria, was easily accessible by land and sea, and had a long tradition of as a prominent seat of Roman provincial government and jurisdiction.⁷³ But it is telling that Ephesus was ⁶⁹ ACOec. 1.1.1 pp. 73–4 (Sacra ad Cyrillum Alex.): ὁρμῆς μέντοι μιᾶς καὶ τῆς αὐτῆς προθέσεως τά τε τῶν ἐκκλησιῶν τά τε τὼν βασιλέων μέλη χωρίζειν βούλεσθαι ὠς οὐκ οὔσης ἀφοφμῆς ἑτέρεας εὐδοκιμήσεως (trans. Holum 1982: 165, slightly modified: τὼν βασιλέων has been translated as ‘of the imperial house’ to show that it encompasses both the male and female members of the imperial family); Wessel 2004: 99–100; Graumann 2013: 116. ⁷⁰ Cf. Wessel 2004: 100. ⁷¹ Noel 1995; Brown 2000: 44–6. ⁷² Graumann 2013: 112–15. ⁷³ ACOec. 1.1.1 pp. 1, 114–16 (Sacra ad Cyrillum Alex. et ad singulos metropolitas of 19 November 430), and ACOec. 1.1.3 p. 31 ll. 19–22 (Sacra directa per Iohannem comitem concilio). For discussion, see Gregory 1979: 101; Holum 1982: 17–18 (on the emperor’s power to convoke a church council), 155–6, 162, 164; Wessel 2001: 290–1; 2004: 99, 141–2; Graumann 2013: 110–15. Cf. Burton 1975.

252

   

also a city with a clear connection to the Virgin Mary with whom the empress Pulcheria had strongly associated herself.⁷⁴ The choice of city may therefore have been a sign of Theodosius’ support for his sister against Nestorius. The emperor had decided not to publicly oppose Nestorius’ revocation of his sister’s privileges in Constantinople, but he could signal his disapproval of the bishop’s actions by moving the council out of the capital to neutral territory. If we view this action through the lens of group behaviour, we see Theodosius II once again exercising his authority to emphasize the unity of the imperial family against outside challenges.

5. The Council of Ephesus (431) Theodosius called together the Council of Ephesus but did not preside over it himself. Instead, he dispatched Flavius Candidianus, the comes domesticorum, as his representative, not only to read aloud the imperial instructions (sacra) at the opening of the council, but also to regulate the meeting.⁷⁵ By ordering Candidianus to keep the peace in Ephesus, Theodosius demonstrated that he was responsible for creating and facilitating the best conditions for the bishops to solve their theological dispute. The emperor’s instructions for Candidianus illustrate the awareness of his own imperial power as well as his respect for ecclesiastical authority. In particular, he forbade Candidianus to be physically present at the bishops’ meetings that discussed ecclesiastical affairs, ‘for it is not lawful that one who is not on the list of holy bishops should meddle in ecclesiastical questions’.⁷⁶ At this stage, there are no signs that Theodosius was worried that Nestorius’s position as bishop of Constantinople would come under serious threat and that his personal intervention would be needed. At the beginning of June 431, bishops from around the empire, but particularly from the East, started to arrive in Ephesus.⁷⁷ While the emperor had ordered the council to start on 7 June, a substantial number of intended participants arrived late or had even been prevented from travelling entirely. For example, the invitation to the African delegation reached them too late on account of the Vandal invasion of North Africa.⁷⁸ Moreover, travel conditions must have been

⁷⁴ Cf. Cooper 1998; Price and Graumann 2020: 44–5. ⁷⁵ PLRE II Candidianus 6; ACOec. 1.1.1 pp. 120–1 (Sacra ad synodum per Candidianum); Gregory 1979: 101; Holum 1982: 163; Wessel 2001: 291; 2004: 143–5. ⁷⁶ Sacra ad synodum per Candidianum (CPG 8668), ACOec. 1.1.1 p. 120 ll. 14–15: ἀθέμιτον γὰρ τὸν μὴ τοῦ καταλόγου τὼν ὁσιωτάτω ἐπισκόπων τυγχάνοντα τοῖς ἐκκλησιαστικοῖς σκέμμασιν ἐπιμίγνυσθαι (trans. Wessel 2004: 143). Cf. Nest. Herac. pp. 109–11; Wessel 2004: 143–6. ⁷⁷ See Driver and Hodgson 1925: xvii–xxix for an overview of the day-to-day events of the council and the period leading up to it. Cf. Crabbe 1981 on the invitation list, and also Russell 2000: 46–52. ⁷⁸ ACOec. 1.1.1 p. 115. ll. 19–26. Cf. Nest. Herac. p. 133; Wessel 2001: 292; Wessel 2004: 138–9, 146–7, 159.

  

253

particularly difficult in June 431, as many bishops and in particular John, the bishop of Antioch, did not arrive in time for the designated first day of the council, and thus the opening had to be adjourned.⁷⁹ The irregularities of the speed of travel clearly had further effects on the proceedings of the church council. Those who arrived early could start to set up alliances, whereas those who arrived late might feel left out or caught unprepared when joining the precouncil discussions. The first of the important metropolitan bishops to arrive in Ephesus was Nestorius, whose company notably comprised both bishops and a military escort led by Irenaeus, the comes Orientis. Irenaeus not only was the bishop’s friend, but also had explicit imperial permission to attend.⁸⁰ The presence of fellow bishops in Nestorius’ party was to be expected, as he would have wanted to bring religious supporters. The military presence might have been seen as a bold infringement upon the boundaries between religious and secular authority. Cyril arrived in Ephesus with a large party of fellow bishops and monks, as well as other followers who were not necessarily invited, but who accompanied Cyril in support and who could be perceived as an intimidating presence.⁸¹ As Susan Wessel has argued, the timely arrival of Nestorius could imply that he wanted to meet with Cyril before the start of the council in order to solve issues beforehand.⁸² As might have been foreseen, the bishops who supported Cyril and their followers received Nestorius with hostility, even denying him access to the churches in Ephesus until the dispute was settled.⁸³ While waiting for those bishops—especially John of Antioch and his followers—who were still on their way to Ephesus, Cyril grew increasingly impatient and managed to convince those present that they could no longer wait. He even persuaded—some would call it ‘tricked’—Candidianus to read the second imperial sacra.⁸⁴ Candidianus felt uncomfortable about proceeding without waiting for the delegates who had been delayed and made this known in a letter to Cyril, warning him not to go against the wishes of the emperor who wanted a general council.⁸⁵ Nevertheless, on 22 June 431, the council started in the Great Church of St Mary, without John of Antioch and all others who had yet to arrive. Even more remarkably, Nestorius, one of the key figures in the dispute, was not

⁷⁹ Cf. ACOec. 1.1.1 p. 119, referring to the difficulty of travel; Russell 2000: 46; Wessel 2004: 4, 147, 170–1. ⁸⁰ PLRE II Irenaeus 2. See ACOec. 1.1.1 pp. 93, 121, 1.1.5 pp. 129–35; Gregory 1979: 101–2; Wessel 2004: 139. ⁸¹ Gregory 1979: 101; Holum 1982: 166. ⁸² Wessel 2004: 139. ⁸³ Gregory 1979: 102, 106–7; Holum 1982: 166; Wessel 2004: 138–41; Russell 2000: 46. ⁸⁴ ACOec. 1.1.5 pp. 119–24; Nest. Herac. pp. 106–18, especially 113–16 (letter of Candidianus), and pp. 117–18 (letter of imperial official Palladius). See further Holum 1982: 166; Russell 2000: 47–8; Wessel 2001: 293; 2004: 149. ⁸⁵ Nest. Herac. pp. 113–16 (letter of Candidianus), and pp. 117–18 (letter of imperial official Palladius) (trans. Driver and Hodgson).

254

   

present at the council. While his adversaries claimed that he declined to come to the council, his own version was that he feared for his life.⁸⁶ In Ephesus, Cyril was triumphant. He assumed a leading role in the council, in clear contravention of the emperor’s wish that all parties, including Nestorius, be present.⁸⁷ Throughout the proceedings, Cyril emphasized that he had the support of Celestine, the bishop of Rome (who could not be present in person but had sent several delegates). By the time John of Antioch arrived in the middle of the proceedings, he found that Cyril had managed to persuade 197 bishops to have Nestorius deposed on the basis of his rival’s sermons and writings.⁸⁸ In his report of the meeting to the population of Alexandria, Cyril emphasized how the people of Ephesus rejoiced at Nestorius’ deposition: And all the faithful of the city remained from early morning to early evening awaiting the decision of the holy council. When they heard that the blasphemer had been deposed, all with one voice began to praise the holy synod and to glorify God, because the enemy had fallen. When we were coming out of the church, they went before us with lamps even unto our lodgings—for it was evening. And there arose much joy and illumination in the city, so that even the women went before us carrying censers.⁸⁹

The popular reaction of the Christians of Ephesus could be used to pressure Theodosius to see the righteousness of the decision of the council.⁹⁰ Nestorius, for his part, wrote to Theodosius II to ask him not to accept the decision, since he himself had not been able to be present, and the council had proceeded while John of Antioch was likewise absent. In this letter, the Constantinopolitan bishop emphasized that Cyril and his followers had ‘trampled on ecclesiastical and imperial laws’.⁹¹ John of Antioch refused to accept the decisions made without him and decided to organize a counter-synod in Ephesus together with forty-three bishops who had also been left out of the council. John managed to have Candidianus, who already publicly had objected to the course of the first council,

⁸⁶ Nest. Herac. p. 135 (trans. Driver and Hodgson). Cf. ACOec. 1.1.2 p. 10; Gregory 1979: 102; Russell 2000: 48–9; Wessel 2004: 147, 150–2. ⁸⁷ McGuckin 2004: 75; Wessel 2004: 148–50, 153. ⁸⁸ ACOec. 1.1.3 pp. 3–5 (Relatio ad imperatores de depositione Nestorii); ACOec. 1.1.1 pp. 33–42; Frend 1972: 22; Gregory 1979: 104; Holum 1982: 166; Russell 2000: 49–50; Wessel 2004: 4, 159–60, 163. ⁸⁹ ACOec. 1.1.1 p. 118 ll. 3–10: ἐπέμεινε δὲ πᾶς ὁ λαὸς τῆς πόλεως ἀπὸ πρωὶ ἕως ἑσπέρας περιμένων τὸ κρίμα τῆς ἁγίας συνόδου· ὡς δὲ ἤκουσαν ὅτι καθηιρέθη ὁ δύσφημος, πάντες μιᾶι φωνῆι ἤρξαντο εὐφημεῖν τὴν ἁγίαν σύνοδον καὶ δοξολογεῖν τὸν θεὸν ὅτι πέπτωκεν ὁ τῆς πίστεως ἐχθρός. ἐξελθόντας δὲ ἡμας ἐκ τῆς ἐκκλησίας προέπεμψαν μετὰ λαμπάδων ἕως τοῦ κααγωγίου (λοιπὸν γὰρ ἤν ἑσπέρα) καὶ γέγονε πολλὴ θυμηδία καὶ λυχναψία ἐν τῆι πόλει, ὥστε καὶ τὰς γυναῖκας θυμιατήρια κατεχούσας προηγεῖσθαι ἡμῶν (trans. Gregory 1979: 104). ⁹⁰ Russell 2000: 50. ⁹¹ ACOec. 1.1.5 p. 14 ll. 7–13, here 9–10 (Epistula Nestorii ad Theodosium imperatorem), τοὺς ἐκκλησιαστικοὺς καὶ τοὺς βασιλικοὺς θεσμοὺς πατήσαντες (trans. Wessel 2004: 167).

  

255

add the imperial sacra to endorse this counter-synod, as if this constituted imperial permission to hold the meeting. The meeting started on 26 June and the bishops present moved to depose Cyril, as well as Memnon, the bishop of Ephesus.⁹² Messengers with letters announcing the outcome of both councils hurried to Constantinople to convince Theodosius II of the righteousness of their respective decisions.⁹³ At this time, public protests broke out in Constantinople in support of Cyril. People assembled in Hagia Sophia and shouted acclamations in support of Pulcheria, who had been maltreated by Nestorius. ‘Many years to Pulcheria! She it is who strengthened the faith!’ they cried.⁹⁴ One of the most prominent antiNestorian groups was the monastic party, whom the bishop had alienated from the beginning of his tenure.⁹⁵ They staged protests throughout the city, being abetted in this by the emperor’s own courtiers, according to Nestorius’ own bitter account of the situation.⁹⁶ The archimandrite Dalmatius, who had been living in seclusion for almost half a century, even emerged from his monastery to lead public protests against Nestorius in an attempt to influence Theodosius.⁹⁷ Cyril was not present in Constantinople himself, but he had reported the events of Ephesus by letter to Dalmatius, which encouraged the archimandrite to take action. The protests against Nestorius functioned as expressions of popular and religious will, which Theodosius could not ignore, especially since they included members of his own court.⁹⁸ It was the emperor’s prerogative to accept the outcome of a church council, and Theodosius decided not to endorse the outcome of either meeting at Ephesus. He ordered the bishops to remain and convene again to solve the issues.⁹⁹ New meetings took place between 10 and 22 July. Notably, John of Antioch and Nestorius were excluded from these discussions, and Cyril managed to have Nestorius’ deposition confirmed with approval of the legates from Celestine of Rome. This meant that Cyril could present the result as a universal agreement.¹⁰⁰ Not all his contemporaries were convinced. Isidore of Pelusiota (died c. 450) wrote that Cyril was ridiculed for using Ephesus to fight his personal battles and thereby failing to solve the theological dispute that should have been the central issue of the council.¹⁰¹ Rather than achieving the unity that Theodosius had hoped for, Ephesus revealed the continuing discord within the church establishment. ⁹² ACOec. 1.1.5 pp. 119–24 (119–20 on Memnon’s reaction); Gregory 1979: 106; Holum 1982: 167; Russell 2000: 50; Wessel 2004: 168–70. ⁹³ Holum 1982: 169–70; Price and Graumann 2020: 312–14. ⁹⁴ ACOec. 1.1.3 p. 14. ⁹⁵ Gregory 1979: 109–10; Wessel 2004: 74–5, 82–3, 164–5. Cf. Frend 1972: 27; Russell 2000: 6–7. ⁹⁶ Nest. Herac. pp. 271–2 (trans. Driver and Hodgson). ⁹⁷ Nest. Herac. pp. 271–8 (trans. Driver and Hodgson). ⁹⁸ See Millar 2006: 233–4 on the gradual development of Theodosius’ views during and after these events. ⁹⁹ Soc. 7.34; Gregory 1979: 112; Holum 1982: 168–74; Wessel 2004: 171. ¹⁰⁰ Russell 2000: 51; Wessel 2004: 173–9. ¹⁰¹ Isid. Pel. Ep. 310 (CPG 5557), PG 78.361c (trans. Wessel 2004: 23 n. 30).

256

   

The bishops had demonstrated that they could not solve matters independently and autonomously.¹⁰² In the end, Theodosius declined to approve the results of all the councils and instead confirmed the deposition of three bishops: Nestorius of Constantinople, Cyril of Alexandria, and Memnon of Ephesus.¹⁰³

6. Courting Favour How had Theodosius come to abandon Nestorius in this fashion? The fact that the public protests in Constantinople included courtiers is indicative of a wider change in sentiment at the imperial court. Indeed, the court was soon propitiated by a third lobby group in the summer of 431. This was the group known as the ‘The Easterners’, led by John of Antioch, who had convened the counter-synod at Ephesus against Cyril.¹⁰⁴ They tried to influence Theodosius by asking Nestorius’ friend, the comes Irenaeus, to carry letters to Constantinople outlining Cyril’s outrages. But they also wanted the comes himself to speak directly with the emperor in order to give him the full account. Irenaeus had not only the necessary access to Theodosius which would enable him to hand over the letters, but also, they hoped, the influence to convince him in person. The Easterners had provided Irenaeus with various ‘treatments’ (θεραπείας) to resolve the issue. ‘We beseech Your Clemency to ascertain these treatments from him without prejudice,’ they wrote.¹⁰⁵ Letters from the Easterners were also despatched to the Augustae Eudocia and Pulcheria and to court officials, including one to a praetorian prefect and a magister (perhaps the magister officiorum) and another to the praepositus sacri cubiculi and a second chamberlain, Scholasticius.¹⁰⁶ These were not random choices, but carefully selected targets reflecting different groups within the imperial court—the emperor’s family, a member of the consistorium, and the household staff who controlled access. In each case, the intentions were different. Eudocia and Pulcheria received an account of the crimes of Cyril and Memnon, without any instructions, whereas the prefect and magister were asked to provide them with safe conduct to Constantinople to press their case. Most remarkable are the instructions to two cubicularii, who were to ‘read aloud’ (ὑπαναγνῶναι) the letter to the emperor himself.¹⁰⁷ In fact, Scholasticius was such an important figure that Nestorius himself wrote a detailed letter to the cubicularius—also delivered by Irenaeus—in which he outlined his own theological position and the wrongs ¹⁰² Cf. Graumann 2013: 117–20, 123. ¹⁰³ Nest. Herac. p. 279 (trans. Driver and Hodgson); Wessel 2001: 295; Wessel 2004: 179–80. ¹⁰⁴ Price and Graumann 2020: 296–7, 321–3, 346–7, 413–14, 420–1. ¹⁰⁵ ACOec. 1.1.5 p. 131: οὓς ἱκετεύομεν ἀνεξικάκως τὴν ὑμετέραν παρ’ αὐτοῦ μαθεῖν ἡμερότητα; Millar 2006: 172. ¹⁰⁶ ACOec. 1.1.5 pp. 131–3. ¹⁰⁷ ACOec. 1.1.5 p. 133.

  

257

committed against him.¹⁰⁸ Nestorius concluded the letter with his firm hope that the accusations against him would ‘be introduced into open discussion before the most pious and Christ-loving emperor’ so that they could be refuted and those who spread them destroyed.¹⁰⁹ Such written missives were always intended as preludes to, and ammunition for, personal conversations with the emperors. Nevertheless, despite these efforts at building alliances across a broad spectrum of the court community, support for Nestorius and the Easterners at court was ebbing away, as Irenaeus himself wrote in a letter to the bishops.¹¹⁰ Scholasticius, in particular, had been won over by adherents of Cyril who were also present in Constantinople; their influence soon resulted in a divided court community, the members of which expressed a range of different opinions about the bishops and whether or not their deposition should be confirmed. The emperor convened a session of the consistorium to consider the matter, in which Irenaeus and the followers of Cyril each had to defend their position. The comes protested that he was only there as a bearer of letters, for which he was met with condemnation. Nevertheless, the decision went in the Easterners’ favour, until Cyril’s supporter John arrived and turned the court against Nestorius again. Irenaeus ended his letter with a play on 2 Timothy 4:17, stating that God had watched over him during these trials at court and had rescued him, not from the mouth of one lion, ‘but from the (mouths of ) many thousands of lion cubs’.¹¹¹ Nestorius, in his own later account of the affair, depicted his removal from the see of Constantinople not as a disgrace, but as a success. For, he wrote, he had often lobbied courtiers ‘who had freedom of speech with the emperor and who were supposed [to be] my friends’ to request that he be allowed to retire from the bishopric to his monastery in Syria.¹¹² Nestorius wished it to appear that he had successfully negotiated the delicate ‘politics of intimacy’ to ensure that those closest to Theodosius could press his case. We must set this claim against the uphill struggle which Irenaeus faced to advance the cause of John of Antioch and the Easterners against Cyril. Nestorius himself was by this point in time unpopular among the court society, with the emperor even forbidding his name to be spoken around him, according to a letter of Theodoret.¹¹³ Cyril had indeed been busy in his attempt to turn around opinion at court. Once again, rather than travelling to Constantinople himself and negotiating the tricky business of access, Cyril decided to keep himself at arm’s length. He had sent representatives to the capital who complicated Irenaeus’ mission. But he also ¹⁰⁸ ACOec. 1.1.4 pp. 51–3. ¹⁰⁹ ACOec. 1.1.4 p. 53: venire ad discussionem coram piissimo et amatore Christi imperatore (this letter only survives in the Latin version). ¹¹⁰ ACOec. 1.1.5 pp. 135–6; McGuckin 1996: 103; Millar 2006: 173. ¹¹¹ ACOec. 1.1.5 p. 136: μᾶλλον δὲ σκύμνων μύρίων. ¹¹² Nest. Herac. p. 281 (trans. Driver and Hodgson). On the retirement, see Price and Graumann 2020: 536. ¹¹³ ACOec. 1.1.7, p. 80; Holum 1982: 172.

258

   

went one step further. There survives today among the Acts of the Council of Ephesus a remarkable catalogue of gifts sent by Cyril in the late summer or early autumn of 431 to prominent members of the court community, both men and women, who had access to members of the imperial family.¹¹⁴ The use of gifts represents a clear change in strategy, since writing directly to Theodosius II and other members of the imperial family had angered the emperor. Paying to influence officials at court was standard practice in the late Roman Empire, but what was remarkable was the size of Cyril’s largess, which must have come from the treasury of the see of Alexandria.¹¹⁵ Cyril employed the priest Claudianus to take charge of the distribution of his gifts, which he referred to as benedictiones (‘blessings’).¹¹⁶ They were sent to Chryseros and Paul, praepositi sacri cubiculi in the households of Theodosius II and Pulcheria, respectively, and to a host of other chamberlains and domestic officials, including the wavering Scholasticius. Paul received countless priceless trinkets and items of furniture ranging from ivory chairs and stools to rugs, tablecloths, and curtains, along with fifty pounds of gold. Chryseros, who was regarded as a tougher target, was to be gifted double the amount of presents along with two hundred pounds of gold, provided that his support for Cyril came through.¹¹⁷ There were three women on the list who received gifts. Two of them, Marcella and Droseria, were part of the household of Pulcheria, and received fifty pounds of gold each for persuading the Augusta. The third was Heleniana, the wife of the praetorian prefect, whose bounty was one hundred pounds of gold as well as numerous items of finery and furniture identical to those bestowed on Chryseros. Heleniana was already known to be an opponent of Nestorius, having abused him publicly in church.¹¹⁸ The list ends with three high-ranking government officials: Artabas, styled magnificentissimus (‘most magnificent’), a flattering epithet rather than a precise title; a magister, perhaps the magister officiorum; and the quaestor, undoubtedly the quaestor sacri palatii. All three were to receive the same gifts as Chryseros and one hundred pounds of gold apiece.¹¹⁹ There was nothing secret about Cyril’s ‘blessings’—they were meant to be displayed by the recipients, as a sign that they had received the favour of Cyril and, by extension, the Church itself.¹²⁰ As the variety of people on the list makes clear, Cyril did not take chances by only trying to persuade one or two people at court. In particular, he not only sought to influence prominent individuals, but also to secure the support of their assistants and subordinates, such as Paul and Chryseros’ deputies, Heleniana’s assessor, and the domestici of the magister and quaestor. The list encompassed

¹¹⁴ ¹¹⁶ ¹¹⁷ ¹¹⁸

ACOec. 1.4 pp. 224–5; Smith 2007: 203. ¹¹⁵ Kelly 2004: 171–5. ACOec. 1.4 p. 222; de Leeuw 2019: 151–2. PLRE II Paulus 10; Chryseros 1; ACOec. 1.4 pp. 224–5 (= McEnerney 1987 no. 96). Holum 1982: 152. ¹¹⁹ ACOec. 1.4 p. 222. ¹²⁰ de Leeuw 2019: 157, 165.

  

259

household officials of the emperor and Pulcheria as well as high-ranking government dignitaries who sat on the consistorium. The recipients were carefully chosen, just as Nestorius and his supporters had selected those to whom they would send letters. Cyril and his agents did not, for instance, deliver gifts and money to all the comites consistoriani; the comes sacrarum largitionum and comes rei privatae go unmentioned in the list (unless one is the magnificent Artabas). Nor were generals considered worth approaching, despite their prominence within the court and official hierarchies. The text that accompanied the list shows that Cyril was very specific about whom he wished to influence. Several individuals were, on receiving these gifts, to act as they had been instructed by letter, whereas Marcella was to speak directly to Pulcheria ‘so that she might persuade the Augusta by entreating’ (ut Augustam rogando persuadat), while Heleniana was to talk to her husband, the prefect. Theodore, the domesticus of the chamberlain Scholasticius, was to receive fifty pounds of gold if he convinced his superior ‘to desist in his friendship with Cyril’s opponents’ (ut ab amicitiis aduersariorum desistat).¹²¹ This remark suggests that Cyril possessed quite detailed knowledge of the relationship between selected courtiers and other lobbyists who might be trying to influence them. The catalogue of gifts and the instructions that accompanied it demonstrates Cyril’s continued strategy of influencing court politics from afar. He did not seek to engage personally in the ‘politics of access’—that is, to gain a personal audience with Theodosius II at which he would have to press his case. Instead, he tried to leverage a range of relationships—the chamberlain and his assistant, the Augusta and her cubicularia, the praetorian prefect and his wife—in order to ensure that arguments in his favour would be made in personal and professional interactions at court. This approach shows that Cyril was acutely aware of the ‘politics of intimacy’, that decisions were not always made in the consistorium or in public audiences, but resulted from whispers, conversations, and approaches made outside formal occasions. Although it is not stated in the catalogue itself, the ultimate aim of Cyril’s strategy must have been that some of these individuals would have conversations with the emperor himself. In this way, the bishop targeted specific individuals and sub-groups within the court (the chamberlains, Pulcheria’s household) in order to gain the broad support of the court community as a whole, which would lead to them persuading the most important member, Theodosius. We do not know how these conversations, either between the courtiers themselves, or between the courtiers and Theodosius, played out. But Theodosius soon called for a delegation of both sides to come to Chalcedon in September of 431. The emperor presided over the meeting himself, which eventually led to the

¹²¹ ACOec. 1.4 p. 224.

260

   

replacement of Nestorius as bishop of Constantinople by Maximian. Cyril and John were asked to resolve their theological issues in private meetings and Cyril was—triumphantly—restored to his see of Alexandria.¹²² Theodosius had faced various types of public and private pressure from Cyril and his supporters in the preceding years. There were letters dispatched to the emperor and his family members, lavish bribes sent to courtiers, and protests in the streets and churches of Constantinople. Not all of these were successful, the letters standing out as being particularly poorly received by Theodosius. But, ultimately, Cyril was able to amass enough support at court to ensure that he remained bishop, while Nestorius was deposed.

7. Conclusion This emergence of a Christian ecclesiastical establishment separate from, and yet intertwined with, the court and government, was one of the primary ways in which the world of Late Antiquity differed from that of the Principate. The Nestorian Controversy and the Council of Ephesus illuminate how theological disputes could play out at the imperial court, as bishops on opposing sides competed to influence the emperor and his courtiers. In this chapter, we have paid attention to the various ways in which bishops tried to cultivate alliances and patrons at court, through the use of embassies, letters, and gift-giving. The strategies show the various approaches available to bishops who wished to have their voices heard and the pay-offs and pitfalls involved with each one. One could seek an audience in person with the emperor, but this might be blocked or end in having one’s case rejected as one dealt with the ‘many thousands of lion cubs’, to use Irenaeus’ arresting phrase. Sending a letter might side-step the possibility of such humiliation, but undiplomatic approaches could still earn the emperor’s wrath. Lavish gifts might be a more promising avenue to win support for the cause, provided that the beneficiaries actually came through with their promises. We have also seen how outside actions and protests might exert additional influence on the emperor and his court; some of these public actions might have been initiated by bishops and their supporters, while others arose spontaneously. Indeed, the interplay between the court and the outside world has been a recurrent theme of this chapter. One did not have to live at court, or even be present in Constantinople, to be part of court politics. We might have expected Nestorius’ position as patriarch of Constantinople, which gave him physical closeness to the emperor and the imperial court, to confer an advantage to him. Indeed, in the previous chapter, Fabian Schulz demonstrated the powerful intermediary role

¹²² Wessel 2001: 295–301.

  

261

which bishops in imperial cities could potentially play at court. But support for Nestorius’ position slowly ebbed away over the years 428–31. In contrast, Cyril never travelled to court personally, yet he ultimately won restoration to his see through his lobbying efforts. One of the major strategies pursued by Nestorius, Cyril, and the Easterners led by John of Antioch was the targeting of key individuals who were thought either to be receptive to their ideas or to be particularly influential at court. Letters or gifts did not tend to be sent to all members of the consistorium or the imperial household. Instead, the bishops used their knowledge of court politics to influence selected players who could be reasonably expected to come into personal contact with the emperor or one of his relatives. Indeed, Cyril discovered the danger of writing letters to all members of the Theodosian dynasty at once, as the emperor regarded this as an attempt to sow discord among the family. We have interpreted these events in the light of psychological studies of group behaviour: groups often contain sub-groups with their own interests and agendas, but when the group as a whole is threatened, it closes ranks. While Theodosius’ support for Nestorius initially led him to overlook the bishop’s treatment of his sister Pulcheria in the interests of ecclesiastical harmony, his later actions show that he was ultimately concerned for the stability of his family and the empire.

12 Splendid Isolation Secluded Emperors and the Spectre of Oriental Despotism Martijn Icks

1. Introduction Emperor Carinus is usually not counted among the most exemplary rulers of Rome. Nevertheless, he features as the hero in an anecdote told by Synesius of Cyrene in his speech De regno (‘On Kingship’). Leading a military campaign against the Parthians, the emperor ordered a stop when the army had reached the borders of enemy territory. While the soldiers were having a bite to eat: an embassy appeared from the enemy lines, thinking on their arrival to have the first conversation with the influential men who surrounded the king, and after these with some dependents and gentleman ushers, but supposing that only on a much later day would the king himself give audience to the embassy.¹

To their astonishment, however, the Parthians were immediately led before Carinus, who was sitting in the grass enjoying some pea soup. Without bothering to get up, the emperor threatened that all Parthian forests and plains would within a month be barer than his head if they did not surrender. To underline the point, he took off his cap to show the ambassadors his bald head. When this message was conveyed to the Parthian king, he was so terrified that he immediately came to discuss terms. Hence ‘he of the tiara and robes’ yielded to ‘one in a simple woollen tunic and cap’.² We can safely assume that this display of imperial machismo never actually took place. For one thing, the Parthian Empire had long been replaced by Sasanian Persia in Carinus’ day, and for another, Carinus never launched a campaign ¹ Syn. de regno 12.4: πρεσβείαν ἐκ τῶν πολεμίων παρεῖναι καὶ οἴεσθαι μὲν ἥκουσαν προεντεύξεσθαι τοῖς βασιλεῖ παραδυναστεύουσι, καὶ τούτων γε αὖ πελάταις τιςὶ καὶ εἰσαγγελεῦσιν, ὡς εἰς ἡμέραν πολλοστὴν ἀπ᾽ἐκείνης τοῦ βασιλέως τῇ πρεσβείᾳ χρηματιοῦντος (trans. A. Fitzgerald 1930). See Lamoureux and Aujoulat 2008 for a detailed commentary on the speech; also Cameron and Long 1993: 103–42. Translations of other texts come from the Loeb Classical Library, sometimes adapted. ² Syn. de regno 12.7: τὸν ἐν τιάρᾳ καὶ κάνδυϊ . . . τῷ μετὰ χιτῶνος φαύλων ἐρίων καὶ πίλου. Martijn Icks, Splendid Isolation: Secluded Emperors and the Spectre of Oriental Despotism In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0013

 

263

against Rome’s eastern neighbour.³ But the truthfulness of the story is less important than the message it meant to convey. Synesius allegedly delivered De regno before Arcadius during his embassy to Constantinople at the end of the fourth century, although its blunt tone makes it unlikely that it was spoken to the emperor’s face.⁴ The speech criticizes the young ruler for his passive, secluded lifestyle, comparing him and other emperors from the recent past to splendorous peacocks and, less flatteringly, lizards who never peep out of their lair to enjoy the sun. Unlike the gruff warhorse Carinus, who spent his days among his men and was content to live on army rations, Arcadius wallows in luxury in the palace, far away from the eyes of hoi polloi. ‘You rejoice only in the pleasures of the body, and the most material of these, even as many as touch and taste offer you,’ Synesius admonished the emperor. Unable to resist yet another animal simile, he added: ‘and so you live the life of a polyp of the sea’.⁵

2. Imperial Visibility and Oriental Seclusion De regno presents the most elaborate treatment of a theme which frequently features in speeches and histories from the late fourth and fifth centuries: the princeps clausus.⁶ This is the emperor who is locked away in his palace and, blissfully unaware of the outside world, fails to interact with his subjects. In these texts, imperial seclusion is often associated with excessive pomp, the dominance of eunuchs at court, and an inclination on the part of the emperors to prefer luxury and indolence to military toil.⁷ While this is undoubtedly a distorted and exaggerated image, it does signal how concerned late antique authors and orators were about imperial visibility and accessibility. Of course, such concerns were hardly new. Complaints about rulers who made themselves unavailable to their subjects can be traced back to the early days of the Principate, although they may have been expressed in different ways.⁸ From the reign of Augustus onwards, visibility and accessibility had served as important markers of good emperorship. Various scholars have explored how the Romans connected ‘seeing’ and ‘being seen’ to notions of power.⁹ On the one ³ It is possible that Synesius was thinking of Carinus’ father, Carus. ⁴ See Cameron and Long 1993: 127–42, who argue that the speech was aimed at a court faction whose favour Synesius hoped to win. (This is contested by Hagl 1997: 76–82, but see Lenski 1998.) ⁵ Syn. de regno 10.3: μόνας ἡδομένους τὰς τοῦ σώματος ἡδονάς, καὶ τούτων γε τὰς ὑλικωτάτας, ὅσας ἁφή τε καὶ γεῦσις πορίζουσι . . . βίον ζῶντας θαλαττίου πνεύμονος. ⁶ The words are first used by Sulpicius Alexander (preserved in Greg. Tur. HF 2.9) with regard to Valentinian II: clauso apud Viennam palatii aedibus principe (‘the emperor was cloistered in the halls of the palace at Vienne’). ⁷ See Icks 2017; also Stroheker 1970; Chastagnol 1985. ⁸ See further the discussion of this topos in historical context in Chapter 14. ⁹ Parker 1999; Fredrick 2002; Hekster 2005; Icks 2020a. See Duindam 2016: 227–85 for a crosscultural perspective on the (in)visibility of monarchic rulers, in particular during ritual performances.

264

   

hand, emperors, consuls, and other prominent figures had to put themselves on display to show off their high status and personal qualities, impressing the soldiers and people on whose support their power ultimately relied. As part of these displays, they had to signal their readiness to listen to and address the needs and grievances of ordinary citizens. On the other hand, the high and mighty made themselves vulnerable through such displays, since they had to subject themselves to the public gaze and hence to public judgement. If they did not behave according to social norms and expectations, they were seen as dishonourable and lost their legitimacy as political leaders.¹⁰ In the endless game of watching and being watched, of judging and being judged, the imperial court played a central role. After all, this was the place where elites gathered, where alliances were forged or broken, and favours granted or withheld. As Andrew Wallace-Hadrill has remarked, the court was a fundamentally theatrical place, a glamourous stage on which actors could show themselves, but also play roles and veil their true intentions, all the while eyeing each other intently. Above all, it was the place where one strove to gain the attention and hopefully the favour of the emperor.¹¹ From the perspective of courtiers, imperial visibility and accessibility at court were thus of paramount importance. If one could not see the emperor, be seen with the emperor, or be seen by the emperor, one had no opportunity to forge close ties, boost one’s prestige, and advance one’s interests, let alone acquire the sort of intimacy with the ruler that enabled real influence over the longer term.¹² Taking these things into consideration, it is not hard to see why Greco-Roman orators and authors presented imperial seclusion as problematical. According to Synesius, Arcadius not only failed to appear in public, but was even difficult to approach within the confines of the court: surrounded as he was by unworthy favourites, only a few privileged senators were allowed to greet him, while generals and captains were hesitant to enter the palace.¹³ The anecdote about Carinus and the Parthian embassy provides an important counterpoint to this image of an unapproachable ruler. Synesius’ point is not just that Carinus showed active leadership and martial valour, but also that his active campaigning and closeness to the soldiers made him highly accessible—even to his enemies. As the speech emphasizes, the Parthians were highly surprised by their speedily admission and the lack of ceremony in their encounter with the leader of the Roman world.

¹⁰ See in particular Parker 1999: 167–8 for the dual nature of visibility as a source of power as well as vulnerability in Roman culture. ¹¹ Wallace-Hadrill 2011: 97–8. ¹² For the concepts of the ‘politics of access’ and ‘politics of intimacy’, see the Introduction of this book. ¹³ Syn. de regno 10.4, 11.3. Synesius is painting a caricature; see Pfeilschifter 2013: 76–122 for the actual circumstances of imperial accessibility in late antique Constantinople, and McEvoy 2020a for an examination of Arcadius himself.

 

265

Presumably their own master, ‘he of the tiara and robes’, would not have been so readily available. In another passage of De regno, Synesius likewise frames imperial seclusion as Oriental, holding up another historical figure for Arcadius to emulate: Agesilaos II, ‘the lame king whom Xenophon praises throughout all his history’. Allegedly, this monarch was ‘most conspicuous in all his actions to everyone who desired to see the leader of Sparta’. Because he was such an excellent general, he almost managed to drive King Artaxerxes II from his kingdom.¹⁴ Once again, an active and accessible Western ruler is (almost) triumphant over an Oriental ruler who is imagined as his opposite. As we will see, Synesius of Cyrene was not the only author who thought there was something distinctly un-Roman about emperors who lived in splendid isolation. Either implicitly or explicitly, Greco-Roman sources frequently associate the figure of the secluded emperor with the world of the East, in particular with the Persian court. In fact, the notion that Oriental kings sealed themselves off from their subjects has deep roots in Greek literature, going all the way back to Herodotus, Ctesias, and Xenophon. In this chapter, I will explore imperial seclusion as a marker of ‘bad’ emperorship in Greco-Roman literary discourse, with a particular focus on the image of the secluded Oriental court as a parallel or counterpoint to the Roman court. First, I will briefly discuss the roots and characteristics of his image. Then, I will analyse when and how it was employed in depictions of invisible, inaccessible emperors from the early Principate to Late Antiquity.

3. The Invisible Kings of the East Scholars of the Achaemenid court have commented on the limited visibility and accessibility of the Persian king, who appears to have been ‘a shadowy figure even to longstanding courtiers’.¹⁵ Although he travelled between several capitals to make himself available in different locations, the Great King took care to maintain an aloof status and placed himself at a great distance from his nobles and common subjects through the unmatched splendour of his appearance and highly formalized court ceremony. Those who approached him without permission risked death. Access to the monarch was controlled by eunuchs, who held sway over the royal bedchamber and served him at dinner.¹⁶ ¹⁴ Syn. de regno 13.8: τοῦ χωλοῦ βασιλέως ὃν ἐπαινεῖ Ξενοφῶν ἐν ὅλῳ συγγράμματι . . . ἐν ᾧ πάντα ποιῶν καταφανέστατος ἧν οἷς ἐπιμελὲς τὸν ἡγεμόνα τῆς Σπάρτης ὁρᾶν. ¹⁵ Llewellyn-Jones 2013a: 44. ¹⁶ Wiesehöfer 1996: 29–55; Briant 2002: 259–61, 274–7; Brosius 2007; 2010; Llewellyn-Jones 2013a: 44–8. Much of our information about the Achaemenid court comes from Greek sources, but palace architecture and reliefs confirm the impression of the king’s aloof status and limited accessibility.

266

   

Like all things Persian, the court of the Great King was fascinating to Greeks of the classical age, prompting admiration as well as scorn.¹⁷ As the royal seat of a continent-spanning empire, it was by definition a very un-Greek place, unlike anything to be found in the world of self-governing poleis. In classical Greek literature, the Persian court is famed for its incredible wealth, which not only symbolized the Persian king’s unsurpassed power, but could also be interpreted as a sign of Oriental tryphē, of indulgence in luxury and bodily pleasure, and hence as a sign of weakness and decadence.¹⁸ The Greeks famously rejected proskynesis, the ritual of obeisance to the monarch, as a gesture of submission to a despotic ruler that was incompatible with Hellenic freedom.¹⁹ Both these things—despotism and tryphē—play a large role in Greek literary images of the secluded Persian (or, more generally, Oriental) court, where the king’s luxurious life is wholly detached from that of ordinary mortals. As Lloyd Llewellyn-Jones has pointed out, Greek discourse was fixated on the invisibility of the Persian monarch, while at the same time imagining him as all-seeing, since his numerous spies and informers allowed him to keep a close eye on the doings of his subjects. Frequent reference is made to criminals and traitors having their eyes gouged out. In short, the Great King controlled not only his own visibility, but also the very sight of his subjects.²⁰ Herodotus traces the origins of the Near Eastern court back to the predecessors of the Persians, the Medes. Right from the start, it seems, seclusion was a vital ingredient of Oriental court life. As the historian has it, Deioces, the first Median king, lived in a palace that was enclosed by seven strong concentric walls and communicated with the outside world through intermediaries. Significantly, ‘he had spies and eavesdroppers everywhere in his domain’, strengthening the impression of a monarch firmly in the grip of paranoia, while drawing attention to the striking imbalance in visibility between ruler and ruled.²¹ Deioces seems to suffer from delusions of grandeur, thinking he is a being of a higher order, or at least working hard to create this impression in others. His isolation is thus not only a matter of physically distancing himself from his subjects, but also of claiming a status that places him above and beyond conventional social hierarchies.

¹⁷ Much has been written about Greek perceptions of Persians and the Persians’ role as the ‘Other’ in Greek culture; e.g. Hall 1989; Isaac 2004: 257–303; Gruen 2011: 9–75; Lenfant 2012. See Strootman and Versluys 2017 for a wide-ranging treatment of the reception of Achaemenid Persia in antiquity and beyond. ¹⁸ Briant 2002: 255–6; Jacobs 2010: 395–7. Persian kings feature prominently in Book 12 of Athenaeus’ Deipnosophistae, which is entirely devoted to tryphē; see Lenfant 2007. ¹⁹ E.g. Hdt. 7.136; Plut. Artax. 22.4; Just. Epit. 6.2.13. ²⁰ Llewellyn-Jones 2013b: 173–7; also Llewellyn-Jones 2013a: 47–8. The concept of an ‘empire of the gaze’ has been borrowed from Grosrichard 1979, which explores early modern discourses on Oriental despotism, including the notion of a secluded court. ²¹ Hdt. 1.99–100: οἱ κατάσκοποί τε καὶ κατήκοοι ἦσαν ἀνὰ πᾶσαν τὴν χώρην τῆς ἦρχε.

 

267

Ctesias also discusses the Near Eastern court in his Persica, but traces its origins to the Assyrians.²² As the author claims, the first Assyrian king to isolate himself from his subjects was Ninyas, the son of the legendary warrior-queen Semiramis.²³ His example was followed by his successors, culminating in the reign of Sardanapalus, the last Assyrian monarch. Much more so than Herodotus, Ctesias associates royal seclusion with great luxury and passive ‘feminine’ behaviour. The prime motive for the Assyrian kings to seek solitude appears to have been not a concern for their safety, but the creation of an environment where they were not bothered by the troubles of the outside world. Sardanapalus in particular is presented as the very embodiment of tryphē, a man deprived of any masculine qualities who wallows in luxury and has no other goal than indulging in bodily pleasures.²⁴ A similar theme is present in Xenophon’s Agesilaos, where the visibility and active leadership of the Spartan king are contrasted with the seclusion of the Persian monarch, with the former sharing the toil and hardships of his soldiers, while the latter spends his days in comfort and limits his availability to his subjects as much as possible.²⁵ Obviously, not all classical Greek authors judged the Persian or Oriental court in the same manner, but the seclusion of the Oriental king frequently serves as a marker of the distance and alienation between the monarch and his subjects. This theme was still current in the literature of the Roman Empire, especially in Greek sources. Pseudo-Aristotle’s De mundo (‘On the Universe’), a philosophical treatise from the Hellenistic or early imperial age, describes Cambyses, Xerxes, and Darius as leading secluded lives, ‘invisible to all, in a marvellous palace with a surrounding wall flashing with gold, electrum and ivory’, while their intermediaries ensure that they ‘might see everything and hear everything’.²⁶ The story of the effeminate, wool-spinning Sardanapalus was retold by Diodorus Siculus and Athenaeus and was still known to Paulus Orosius in the early fifth century. Many more stories about invisible, androgynous kings can be found in Book 12 of the Deipnosophistae (‘Sophists at Dinner’).²⁷

²² Lanfranchi 2010: 56–7. See Llewellyn-Jones and Robson 2013: 1–87 for a good discussion of Ctesias’ use of Persian sources and his perspective on the East. ²³ Preserved in Ath. 12.528e–f (Ctesias, Persica F1n in Lenfant 2004). ²⁴ Ctesias’ account of Sardanapalus has been lost, but several adaptations in later works allow us to reconstruct the gist of the original text: Diod. 2.23.1–4; Nicolaus of Damascus, FGrH 90 F2 (Ctesias, Persica F1pδα); Ath. 12.528f–529a (Ctesias, Persica F1pα). ²⁵ Xen. Ages. 9.1–5. The author gives a more nuanced verdict on the Persian court in his Cyropaedia, where Cyrus the Great is said to have limited royal visibility in order to rule more effectively (7.5.37ff.); see Vlassopoulos 2017: 370–2. For a discussion on the various interpretations of Xenophon’s Cyrus as an ideal ruler, see Tamiolaki 2017: 190–2. ²⁶ Aris. [Mund.] 398A: παντὶ ἀόρατος, θαυμαστὸν ἐπέχων βασίλειον οἶκον καὶ περίβολον χρυσῷ καὶ ἠλέκτρῳ καὶ ἐλέφαντι ἀστράπτοντα . . . πάντα μὲν βλέποι, πάντα δὲ ἀκούοι. Interestingly, the author employs the image of the secluded Persian king in a positive context, as a metaphor for the divine; see Betegh and Gregoric 2014. ²⁷ Oros. 1.19. See Gambato 2000 for secluded ‘female-kings’ in Book 12 of the Deipnosophistae.

268

   

4. Secluded Emperors in the Principate As is well-known, Augustus founded his rule on the notion that he was a princeps rather than a rex, communicating to the senators that he regarded them as his peers. Consequently, most emperors from the time of the Principate cultivated their civilitas. This meant that they rejected excessive luxury and pompous displays, preferring to project the image of a modest, accessible prince.²⁸ However, that was not their only mode of representation. As Christoph Michels has explored in Chapter 1 in this volume, even an exemplary princeps like Antoninus Pius was at the centre of an ostentatious court and could be unabashedly referred to as μέγας βασιλεὺς by senatorial courtiers like Fronto. On the other hand, the Historia Augusta praises Antoninus for his abolishment of imperial pomp and reducing the influence of courtiers who acted as intermediaries.²⁹ Civilitas, then, can be understood as just one among several imperial strategies of communication, to be employed for particular audiences on particular occasions. A civilis princeps displayed himself in ways that emphasized his closeness and accessibility to his subjects, in particular the senatorial aristocracy that would posthumously chronicle and judge his reign. However, this emphasis on the emperor’s modesty, closeness, and accessibility was at odds with the splendour of the aula Caesaris that developed over the course of the Principate. For a keen observer like Marcus Aurelius, there was no fundamental difference between the Roman court and those of monarchic figures like Philip II of Macedon, Alexander the Great, and Croesus of Lydia.³⁰ By its very nature, these comparisons suggest, the Roman court had a whiff of the Oriental about it. But was it also characterized as a place of Oriental-type seclusion by Greco-Roman authors? Among the rulers of the first and second centuries, two stand out for their alleged isolation from their subjects: Tiberius and Domitian.³¹ In the case of the former, the emperor’s withdrawal to Capri was a major point of criticism in the literary sources. Suetonius discusses it in terms of a withdrawal from the public gaze, remarking that Tiberius, ‘being as it were out of sight of the citizens . . . at last gave free rein at once to all the vices which he had for a long time ill-concealed’.³² Tacitus makes a similar comment, speculating that the emperor left for Capri ‘to find an inconspicuous home for the cruelty and lust which his acts proclaimed to the world’.³³ The underlying assumption is that sovereigns who ²⁸ Wallace-Hadrill 1982. ²⁹ Fronto, Ad M. Caes. 2.3 (vdH² p. 21, l. 20, p. 23, l. 17); HA Ant. Pius 6.4, 7.7–8. See also M. Aur. Med. 1.17.5 for the lack of pomp at Antoninus’ court. ³⁰ M. Aur. Med. 10.27. ³¹ Another good example might be Commodus, who became estranged from the senate and ruled through intermediaries like Perennis and Cleander for large parts of his reign. See Hekster 2002: 60–77. ³² Suet. Tib. 42.1: quasi civitatis oculis remotis, cuncta simul vitia male diu dissimulata tandem profudit. ³³ Tac. Ann. 4.57, 6.1: saevitiam ac libidinem cum factis promeret, locis occultantem.

 

269

were not held in check by the public eye might be up to anything. Although there are obvious parallels to the secluded hedonism of Oriental kings like Ninyas and Sardanapalus, Tiberius’ seclusion is never framed as such in the literary sources. One reason for this may be that the emperor made no attempts to elevate his status through pomp and splendour, a recurring feature in stories about secluded Oriental kings.³⁴ Things are different in Domitian’s case. Unlike Tiberius, this ruler did adopt a more explicitly autocratic style, downplaying his civilitas in favour of a grand, superhuman aura.³⁵ In the works of Martial and Statius, emphasis is laid on the emperor’s awe-inspiring presence and majestic appearance.³⁶ Only after Domitian’s assassination and Nerva’s and Trajan’s return to a more modest style of representation did Martial openly associate his former patron with Oriental despotism, bidding all flatterers to ‘go far away to the turbaned Parthians and, with base and servile supplications, kiss the feet of their pageant kings’.³⁷ Such remarks signal the restoration of cordial relations between emperor and senate after a period of alienation under Domitian. In Pliny’s famous Panegyricus to Trajan, the deceased tyrant’s social isolation is described in vivid terms: this is the place where recently that fearful monster built his defences with untold terrors, where lurking in his den he licked up the blood of his murdered relatives or emerged to plot the massacre and destruction of his most distinguished subjects. Menaces and horror were the sentinels at his doors, and the fears alike of admission and rejection; then himself in person, dreadful to see and to meet, with arrogance on his brow and fury in his eye, a womanish pallor spread over his body but a deep flush to match the shameless expression on his face. None dared approach him, none dared speak; always he sought darkness and mystery, and only emerged from the desert of his solitude to create another.³⁸

³⁴ In fact, as Cassius Dio remarks, while Tiberius was still in Rome he was ‘extremely easy to approach and easy to address’, making an effort to ensure that ‘the people might meet him with as little difficulty and trouble as possible’ (57.11.1–7). ³⁵ For Domitian’s reign and representation, see Jones 1992; Southern 1997; Leberl 2004; Gering 2012. ³⁶ E.g. Mart. 5.3, 7.5, 9.24; Stat. Silv. 4.1.1–4, 4.2.10–17. ³⁷ Mart. 10.72: ad Parthos procul ite pilleatos et turpes humilesque supplicesque pictorum sola basiate regum. ³⁸ Plin. Pan. 48.3–5: quam nuper illa immanissima belua plurimo terrore munierat, cum velut quodam specu inclusa nunc propinquorum sanguinem lamberet, nunc se ad clarissimorum civium strages caedesque proferret. Obversabantur foribus horror et minae et par metus admissis et exclusis; ad hoc ipse occursu quoque visuque terribilis: superbia in fronte, ira in oculis, femineus pallor in corpore, in ore impudentia multo rubore suffusa. Non adire quisquam non adloqui audebat, tenebras semper secretumque captantem, nec umquam ex solitudine sua prodeuntem, nisi ut solitudinem faceret.

270

   

Domitian, then, is not compared to a splendorous Oriental monarch, but to a savage, bloodthirsty beast that no one dares approach.³⁹ Obviously, Trajan is his opposite: the optimus princeps is of such excellent character that he is ‘capable of living in the public eye’ (habitare ut in publicis posset) and is on such friendly terms with the senators that his palace feels to them like ‘a home we share’ (communi domo).⁴⁰ Where Domitian sought ‘solitude behind closed doors’ (solitudine et claustris), Trajan freely mingles with ‘the thronging crowds of his subjects’ (civium celebritate). All this leads Pliny to conclude that the only protection on which an emperor can truly rely is the love of his people.⁴¹ In Dio Chrysostom’s Orations on Kingship, the image of the secluded Oriental monarch features prominently. Although Domitian is nowhere mentioned by name, his shadow looms large in these works. All four of the speeches can be dated to the reigns of Nerva and Trajan and some of them, at least, appear to have been delivered before the emperor and his court.⁴² In the course of these orations, Dio paints a picture of the ideal king, often through dialogues between historical characters such as Philip II of Macedon and his son Alexander. Hence, a model emerges of a monarch of superior virtue who enjoys the approval of the gods, controls his impulses and desires, dearly loves his people, and makes every effort to provide them with justice and protection.⁴³ As a counter-example, the orator brings up the indolent Sardanapalus, whose unlimited power had made him so depraved that he ‘would never have been aroused to leave his chamber and the company of his women’.⁴⁴ Elsewhere, Dio mentions the Syrian king, ‘who spent his life in his harem with eunuchs and concubines without ever a sight of army or war or assembly at all’.⁴⁵ Without doubt, such bad exempla were meant to evoke the memory of the secluded Domitian, while Trajan was supposedly much closer to the ideal of the accessible king. In the First Oration on Kingship, Dio conjures up the image of two mountains, Peak Royal and Peak Tyrannous, where Royalty and Tyranny reside. The latter is described as sitting on a splendid throne embellished with gold, ivory, amber, and other precious materials, with sceptres, tiaras, and diadems symbolizing ‘vainglory, ostentation, and luxury’ (δόξαν . . . καὶ ἀλαζονείαν καὶ τρυφήν). He continues:

³⁹ The association of tyrants with savage beasts goes back to Plato (Resp. 9.588b–589a) and Aristotle (Pol. 3.1287a). ⁴⁰ Plin. Pan. 47.4, 48.3. ⁴¹ Plin. Pan. 49.2–3. ⁴² See Gangloff 2009: 6–14 for the dates and circumstances of the speeches. According to Moles 1983: 177–8, the fourth oration was also delivered before Trajan, at a very early point in his reign. ⁴³ Gangloff 2009: 17–29. ⁴⁴ Dio Chrys. Or. 1.3: οὐκ ἄν ποτε ἤγειρεν ἐκ τοῦ θαλάμου καὶ παρὰ τῶν γυναικῶν. ⁴⁵ Dio Chrys. Or. 4.113: μετ᾿ εὐνούχων καὶ παλλακῶν ἔνδον διαβιοῦντι, στρατοπέδου δὲ καὶ πολέμου καὶ ἀγορᾶς ἀθεάτῳ τὸ παράπαν. Presumably the ‘Syrian king’ also refers to Sardanapalus. In another speech, Dio (through the mouth of Socrates) speaks of the possibility that the Persian king is a tyrant living a life of wealth and pleasure while suppressing his subjects (3.40–1).

 

271

in order to assume the appearance of pride, she would not glance at those who came into her presence but looked over their heads disdainfully. And so everybody hated her, and she herself ignored everybody.⁴⁶

Although the East is not mentioned in this passage, Tyranny’s haughty demeanour and luxurious surroundings immediately bring to mind the secluded Persian kings from classical Greek discourse. However, Dio adds a remarkable detail that is not usually present in classical accounts of secluded kingship: And instead of Friendship, Flattery was there, servile and avaricious and no less ready for treachery than any of the others, nay rather, zealous above all things to destroy.⁴⁷

The importance of friendship (philia) as a royal virtue is a recurring theme in Dio’s oeuvre, especially in the Third Oration on Kingship. In the orator’s view, it is friendship which inspires rulers to prefer altruistic toil over selfish pleasure and guarantees a close, stable bond between sovereign and subjects. Surrounding himself with true friends rather than untrustworthy flatterers brings a ruler delight and dispels loneliness. Moreover, Dio argues, loyal hearts provide better protection from harm than arms or fortresses.⁴⁸ As we have seen, this final point was also made by Pliny. The latter likewise emphasized the importance of amicitia in his speech to Trajan, so that some relation between the works of both orators is often assumed.⁴⁹ Evidently, the importance of friends at close quarters was a timely topic in the years immediately following Domitian’s autocratic regime, in which the emperor had become increasingly alienated from his senatorial peers. As Dio points out on more than one occasion, loyal friends are a ruler’s eyes and ears, allowing him to see, hear, and think about everything that goes on around him.⁵⁰ A tyrant, however, is ‘the most friendless man in the world’ (πάντων γὰρ ἀπορώτατός ἐστι φιλίας), incapable of even making friends. Picking up a theme from classical discourse, Dio remarks that the Persian king may have had a special official called the King’s Eye, but reproachfully adds that ‘he did not know that all the friends of a good king are his eyes’ (ἀγνοῶν ὅτι τοῦ ἀγαθοῦ βασιλέως οἱ φίλοι πάντες εἰσὶν ὀφθαλμοί).⁵¹ This remark undermines the traditional image of the invisible but all-powerful and all-seeing Oriental monarch.

⁴⁶ Dio Chrys. Or. 1.78–82: ἵνα δὲ φαίνοιτο μεγαλόφρων, οὐ προσέβλεπε τοὺς προσιόντας, ἀλλ᾿ ὑπερεώρα καὶ ἠτίμαζεν, ἐκ δὲ τούτου πᾶσιν ἀπηχθάνετο, πάντας δὲ ἠγνόει. ⁴⁷ Dio Chrys. Or. 1.82: ἀντὶ δὲ Φιλίας Κολακεία παρῆν, δουλοπρεπὴς καὶ ἀνελεύθερος, οὐδεμιᾶς ἦττον ἐπιβουλεύουσα ἐκείνων, ἀλλὰ μάλιστα δὴ πάντων ἀπολέσαι ζητοῦσα. ⁴⁸ Konstan 1997: 131–8. ⁴⁹ Konstan 1997: 126. Pliny develops the theme of amicitia in Pan. 85–7. ⁵⁰ Dio Chrys. Or. 1.32, 3.104–7. ⁵¹ Dio Chrys. Or. 3.116–18.

272

   

Trajan would be well advised to take a different course. Only through the cultivation of philia could he hope to keep a grip on his empire and govern it well.

5. Secluded Emperors in Late Antiquity On the whole, direct comparisons between the Roman court and the secluded courts of Oriental kings appear to have been sparse in the literature of the Principate. However, the theme became more prominent in Late Antiquity. An important reason for this was the rise of Sasanian Persia in 224, which posed a serious challenge to Roman hegemony and led to several military conflicts between the two superpowers in the third and fourth centuries.⁵² At the same time, as Matthew Canepa has argued, Roman and Sasanian modes of imperial representation influenced each other deeply through diplomatic contacts: both the Roman emperor and the Persian king gradually adopted many of the same rituals and symbols, including the diadem, the nimbus, red or purple shoes, prostration, and ceremonial silence.⁵³ In short, the Sasanian court became an important point of reference for the Roman emperor—much more so than the Parthian court had ever been. For late antique Greco-Roman authors, the splendorous Persian court provided an apt parallel to assess the elaborate ceremony that came to envelop their own sovereign. Critical voices contended that the emperorship had been infected with Oriental tryphē. Eutropius scorned Diocletian’s splendid adornment and above all his introduction of prostration, a custom which he considered ‘suited rather to royal usages than to Roman liberty’ (regiae consuetudinis formam magis quam Romanae libertatis). He recounted how many Persian courtiers and treasures had fallen into Diocletian’s hands as war booty, strongly suggesting a connection between both events.⁵⁴ According to the Historia Augusta, it was the Syrian Elagabalus who first demanded imperial adoration ‘in the manner of the king of the Persians’ (regum more Persarum), although the custom was abolished by his successor, the ‘good’ Emperor Severus Alexander.⁵⁵ Doubtlessly, many Greeks and Romans would also have been reminded of Alexander the Great, whose infamous attempt to introduce proskynesis was recounted in Greco-Roman literature as having deeply offended his freedom-loving Greek and Macedonian followers.⁵⁶

⁵² For Roman–Persian relations, see Wiesehöfer and Huyse 2006; Dignas and Winter 2007; Canepa 2009. ⁵³ Canepa 2009: 188–223. ⁵⁴ Eutr. 9.26; see also Lact. DMP 21.2, where it is Galerius who introduces prostration after a victory over the Persians. ⁵⁵ HA Alex. Sev. 18.3. ⁵⁶ For example, see Curt. 6.6.1–11; Arr. Anab. 4.9–12. See Alföldi 1970: 9–25 for the literary contrast between Romana libertas and servitus Persica.

 

273

Modern scholarship points out that the ceremonialization of the imperial office was not achieved in one sudden shift under Diocletian, but was a gradual development over the course of centuries.⁵⁷ As Canepa suggests, courtly rituals like proskynesis may actually have been adopted by the Sasanians from the Romans rather than the other way around.⁵⁸ But realities mattered less than perceptions. We can assume that many late antique Romans accepted imperial pomp and ceremony without question, taking great pride in kissing the emperor’s robe and other ceremonial privileges.⁵⁹ After all, the courts were bustling with new elites such as high-ranking eunuchs and other social climbers who had attained their privileged position through the fulfilment of a range of civic or military offices. Unlike the traditional senatorial aristocracy, many of these people did not attach much value to the old ideal of imperial civilitas and were quite happy to work in a system that emphasized the great social distance between themselves and their sovereign.⁶⁰ Yet others, including but not limited to members of the senatorial aristocracy, continued to cherish the notion of a modest, accessible princeps who treated distinguished men as equals. In their eyes, the elevation and ceremonialization of Roman emperorship was a lamentable innovation whose origins were to be found in Persian despotism. Emblematic of this ‘Orientalization’ were the eunuchs who came to dominate court life in Late Antiquity.⁶¹ As a result, there was a conspicuous gap between literary discourses on ‘good kingship’ continuing in the vein of earlier Greek and Roman models and the practice of the increasingly ceremonious and (especially after 395) sedentary emperorship of the late third and the fourth and fifth centuries. Needless to say, new narratives were developed which were better suited to the emperor’s changing role. A good example can be found in the Panegyrici Latini, where the splendorous appearance of Diocletian and Maximian is described in terms of an epiphany.⁶² Nevertheless, ideals of imperial civilitas and accessibility, often combined with active military leadership, are expressed too frequently and persistently in late antique literature and oratory to be dismissed as outdated and irrelevant. They remained viable as an alternative discourse for authors to criticize the pomp of an aloof, ritualized court and, at times, for emperors to employ before a receptive audience.

⁵⁷ See Alföldi 1970 for the development of Roman imperial ceremonial; also Kolb 2001: 38–46. See further the discussions in Chapter 3 and Chapter 10. ⁵⁸ Canepa 2009: 64–6, 149–52. ⁵⁹ Kissing the emperor’s robe as a privilege: Matthews 1989: 246–9. A fine example of the rejection of proskynesis is provided in the Historia Augusta, where Maximinus Thrax exclaims that he would never let free men kiss his feet—unlike his son, who acted ‘exceedingly haughty’ (superbissimus) (Max. duo 28.7). ⁶⁰ Noethlichs 1998; Smith 2007; Icks 2020b: 181–3. ⁶¹ Icks 2017: 467–72; also Stevenson 1995. For the intensification of ceremonial in historical context, see Chapter 14. ⁶² Pan. Lat. 11(3).10.4–5.

274

   

The continuing relevance of imperial civilitas is well illustrated by Pacatus’ panegyric on Theodosius, delivered in the curia at Rome in 389. The orator’s emphasis on this particular virtue was not only especially appropriate in the urban setting of Rome, where the sovereign had traditionally assumed the role of ‘first citizen’, but also well suited to an emperor who liked to present himself as a new Trajan.⁶³ The speech, which was strongly influenced by Pliny’s panegyric, recounts how Theodosius freely mingles with the people of Rome and is happy to grant their requests. In contrast, his unnamed predecessors are denounced as principes clausi who ‘considered their royal majesty diminished and cheapened unless they were shut up within some remote part of the palace’, where they lay ‘buried in the shade of their abode’.⁶⁴ This description does not evoke the splendour of Oriental seclusion so much as it does Domitian’s gloomy solitude. Theodosius was a military active emperor who claimed victories against barbarians and usurpers, but his sons Arcadius and Honorius became ‘palace emperors’ who hardly left their capitals and entrusted their battles to right-hand men like Stilicho, Gainas, and Eutropius. In doing so, they cemented a trend towards a more passive, ceremonious emperorship that had already been initiated under Gratian and Valentinian II.⁶⁵ When adult commanders such as Majorian and Anthemius became Western emperors in the late fifth century, they found themselves part of—and struggled against—a system in which many of their military responsibilities were invested in the general Ricimer, as Meaghan McEvoy discussed in Chapter 8. Evidently, this state of affairs exposed emperors to criticism, as we have already seen in Synesius’ comments on Arcadius in De regno at the start of this chapter. In this speech, imperial seclusion was closely associated with military passivity. Synesius’ text contains several references to the splendorous courts of the East. For instance, the orator remarks upon Arcadius’ gold-bedecked, gem-studded dress and claims he could only walk on earth that had been sprinkled with gold dust imported from far-away continents. Both aspects evoke the image of the Persian king, who was famous for his magnificent clothes and the fact that his feet were not allowed to touch the ground.⁶⁶ Like Xenophon in the case of Artaxerxes II, Synesius interprets Arcadius’ alleged tendency to stay hidden as a misguided attempt to boost his prestige and project a superhuman aura.⁶⁷ This was imprudent behaviour because, as the orator explains, the king should display his virtue ⁶³ See Schmidt-Hofner 2012 for Trajan as a role model of civilitas in late antique Rome. ⁶⁴ Pan. Lat. 2(12).21.2–3: qui maeiestatem regiam imminui et vulgari putabant, nisi eos intra repositum Palatinae aedis inclusos . . . intra domesticam umbram iacentes. ⁶⁵ See McEvoy 2010; 2013 for developments in the West; see Pfeilschifter 2013 for sedentary emperorship in Constantinople. ⁶⁶ Syn. de regno 11.2–4; see also Chastagnol 1985: 158–60 for Arcadius’ splendid dress. For the feet of the Persian king not touching the ground: Ath. 12.514c. ⁶⁷ Syn. de regno 10.2 (14.3). Cf. Xen. Ages. 9.1: ‘the Persian thought his dignity required that he should be seldom seen’; also Hdt. 1.99.

 

275

in public to inspire emulation in his subjects. In particular, Arcadius is urged to become acquainted with his soldiers and play an active role in warfare, since a king is ‘a craftsman of war’ (τεχνίτης . . . πολέμων), just like the cobbler is with shoes.⁶⁸ In fact, the ideals of kingship espoused in De regno are heavily inspired by Dio Chrysostom’s kingship orations, with which Synesius was well acquainted. Like the addressees of Dio’s speeches, Arcadius is urged to set a moral example through personal excellence, to reject indolence in favour of toil and to cherish true bonds of friendship, rejecting the flatterers at court who corrupt his soul.⁶⁹ Particularly, Synesius echoes Dio with his remark that friends allow a king to ‘see with the eyes of all’ and to ‘take counsel from those opinions of all which tend to one conclusion’.⁷⁰ Many of the points in De regno about the importance of imperial visibility can also be found in a contemporary speech by the western court poet Claudian.⁷¹ In his panegyric on Honorius’ fourth consulship, which was held in Milan at the dawn of 398, the poet recounts the parental advice that the old Emperor Theodosius had supposedly given his son, explaining to him how to be a good ruler. As Catherine Ware has pointed out, this speech combines elements from Pliny’s panegyric with Dio’s kingship orations, thus associating the young Honorius with the optimus princeps Trajan.⁷² One such element is the notion that imperial rule was not a privilege to be enjoyed at leisure, but something that the emperor had to earn through moral excellence, which should always be on display: Had fortune, my dear son, given thee the throne of Parthia, hadst thou been a descendant of the Arsacid house and did the tiara, adored by Eastern lands afar, tower upon thy forehead, thy long lineage would be enough, and thy birth alone would protect thee, though wantoning in idle luxury. Very different is the state of Rome’s emperor. ’Tis merit, not blood, must be his support. Virtue hidden hath no value, united with power ‘tis both more effective and more useful. Nay, o’erwhelmed in darkness it will no more advantage its obscure possessor than a vessel with no oars, a silent lyre, an unstrung bow.⁷³

⁶⁸ Syn. de regno 9.1–7, 11.5. ⁶⁹ See Konstan 1997: 125–6 for Synesius’ reliance on Dio for the friendship theme. ⁷⁰ Syn. de regno 7.3: τοῖς ἁπάντων μὲν ὀφθαλμοῖς ὄψεται . . . ταῖς ἁπάντων δὲ ἀκοαῖς ἀκούσεται καὶ ταῖς ἁπάντων γνώμαις εἰς ἕν ἰούσαις βουλεύσεται. See note 50 for the relevant passages in Dio Chrysostom. ⁷¹ Although there is no reason to assume that one copied the other: Cameron 1970: 321–3. ⁷² Ware 2013: 315. However, I disagree with Ware’s contention that the inspiration from Pliny and the inspiration from Dio can be neatly assigned to separate halves of Theodosius’ speech. See also McEvoy 2013: 162–6 for a good discussion of the oration and its representation of the Honorius– Stilicho partnership, and see further Chapter 8. ⁷³ Claud. De IV Cons. Hon. 214–24: si tibi Parthorum solium Fortuna dedisset, | care puer, terrisque procul venerandus Eois | barbarus Arsacio consurgeret ore tiaras: | sufficeret sublime genus luxuque fluentem | deside nobilitas posset te sola tueri. | altera Romanae longe rectoribus aulae | condicio. virtute decet,

276

   

Once again, seclusion is associated with indolence. Only when Honorius can keep his passions under control is he fit to be emperor: ‘When thou canst be king over thyself then shalt thou hold rightful rule over the world’ (tunc omnia iure tenebis,| cum poteris rex esse tui).⁷⁴ By being visible, Claudian explains, the young ruler can prove his worthiness and hence win the loyalty of his subjects, which will keep him safer from harm than guards or weapons. In fact, seclusion is typical of tyrants, who ‘live hedged about with swords and fenced with poisons, dwelling in a citadel that is ever exposed to danger, and threaten to conceal their fears’.⁷⁵ A few lines further, Theodosius again brings up the East, pressing upon his son that ‘they are not submissive Sabaeans whom I have handed over to thy rule, nor have I made thee lord of Armenia; I give thee not Assyria, accustomed to a woman’s rule’.⁷⁶ Since he rules over Romans, who resent tyranny, Honorius should not claim an exceptional status and ‘overstep the limits established for mankind’ (praescriptos homini transcendere fines), but respect his subjects and the laws that he himself has set.⁷⁷ In short, he should not be like the Persian king, whom the Greeks and Romans mistakenly believed claimed a divine status. The speech ends with the old emperor urging Honorius to personally lead his troops into battle and share all of their hardships, while avoiding a spacious royal tent or other forms of luxury that would only slow the army down—tropes associated with the Persian king, especially Darius III on his fatal campaign against Alexander.⁷⁸ Of course, as it turned out Honorius would never follow up on this advice, even when he reached adulthood and could feasibly have taken to the field. Yet in 398, Claudian apparently considered it prudent to associate the underage son with his father’s active military leadership and to raise the expectation that he would follow in his footsteps. Evidently, ‘palace emperorship’ was not yet the established new model, although it was already on its way to set the standard in both halves of the empire. It is probably no coincidence that two of the texts that are most vocal in rejecting this type of rule stem from this transitional period, when memories of military active emperors were still fresh.

non sanguine niti. | maior et utilior fato coniuncta potenti, | vile latens virtus. quid enim? submersa tenebris | proderit obscuro veluti sine remige puppis | vel lyra quae reticet vel qui non tenditur arcus (trans. Platnauer, Loeb). ⁷⁴ Claud. De IV Cons. Hon. 261–2 (trans. Platnauer, Loeb). ⁷⁵ Claud. De IV Cons. Hon. 293–4: muniti gladiis vivant saeptique venenis, | ancipites habeant arces trepidique minentur (trans. Platnauer, Loeb). Cf. Plin. Pan. 49.2–3. ⁷⁶ Claud. De IV Cons. Hon. 306–8: non tibi tradidimus dociles servire Sabaeos, | Armeniae dominum non te praefecimus orae, | nec damus Assyriam, tenuit quam femina, gentem (trans. Platnauer, Loeb). The Sabaeans were governed by a secluded king living in ‘effeminate luxury’ (Strabo 16.4.9). The female Assyrian ruler is probably the legendary warrior queen Semiramis. See Lanfranchi 2010: 57–60 for the Persian king’s supposed divine status. ⁷⁷ Claud. De IV Cons. Hon. 304 (trans. Platnauer, Loeb). ⁷⁸ Claud. De IV Cons. Hon. 320–52. See Briant 2015: 282–6 for the excessive luxury of Darius’ campaign.

 

277

6. Conclusion Classical Greek literature developed the image of the invisible and inaccessible Oriental king in his gem-studded court, a symbol of autocracy as well as tryphē. By the turn of the first century, this theme was picked up by Dio Chrysostom to contrast the aloof, secluded Domitian with the accessible Trajan. Yet whereas Oriental kings had often been imagined as all-seeing to underline their godlike power, Dio stressed that seclusion led to ignorance and that good governance was only possible with the help of true friends. In doing so, he advocated the importance of close ties between emperors and elites. The same theme was developed by Pliny, although without reference to the East. In Late Antiquity, the increasing ceremonialization of the imperial court—regarded as Orientalization by hostile sources—and the fact that most emperors stopped waging war in person prompted critical comparisons between Roman emperors and secluded Oriental kings. These were particularly prominent in Synesius’ De regno and Claudian’s panegyric on Honorius’ fourth consulate, both addressed to ‘palace emperors’ who abandoned active military leadership and were content to play a mostly ceremonious role. Hence, a very old literary theme gained new relevance. While many of the new elites who frequented late antique courts were doubtlessly appreciative of the changing modes of imperial display and behaviour, there was still an expectation among traditionally minded audiences, such as members of the senatorial aristocracy, that Roman emperors were active and accessible, the opposite of Oriental despots. If they failed to uphold these standards, a vital distinction became dangerously blurred.

13 Envisioning Audiences at the Roman Imperial Court Caillan Davenport

1. Introduction The Roman Empire was a land of many marvels and wonders, which attracted travellers and pilgrims from across the Mediterranean.¹ They journeyed to Egypt to hear the singing Colossi of Memnon, to Greece to heal at the sanctuary of Asclepius at Epidaurus, and, in Late Antiquity, to the Holy Land to travel in the footsteps of Christ.² Poets and panegyricists competed to laud new imperial constructions, such as the Colosseum, the palace of Domitian, and the Church of the Holy Sepulchre, as monuments which rivalled all previous natural and man-made wonders.³ Yet one of the Roman world’s premier attractions was not a monument, but a man: the emperor himself. Petitioners and embassies from cities and communities made the long journey to Rome, Constantinople, or wherever the emperor and his court happened to be, in order to make appeals for civic privileges, financial assistance, religious freedom, and judicial arbitration.⁴ Other, less fortunate individuals were arrested and brought before the emperor to stand trial, and in these cases his court became a court of law, the highest in the land.⁵ Before a petitioner or embassy was able to meet the emperor, they had to negotiate the vexed question of access, making the appropriate connections or

¹ The research for this chapter was supported by an Australian Research Council Discovery Early Career Researcher Award (DE150101110) and an Alexander von Humboldt Fellowship. For their advice and feedback on this material at different stages, I am grateful to Benjamin Kelly, Hartmut Leppin, Shushma Malik, Chris Mallan, audience members at the 2016 Celtic Conference in Classics, the peer reviewers for Oxford University Press, and, above all, Meaghan McEvoy. ² Colossi of Memnon: Strabo 17.46 [816]; see Rosenmeyer 2018 for the inscriptions left by travellers. Sanctuary of Asclepius: Paus. 2.26.1–28.1. Holy Land: Itinerarium Burdigalense (333–4); Itinerarium Egeriae (381–4). ³ Colosseum: Mart. Spect. 1. Flavian Palace: Mart. 7.56, 8.36; Stat. Silv. 1.1.34, 4.2.18–31. Church of the Holy Sepulchre: Eus. SC 11–12, cf. V. Const. 4.45–6. ⁴ Millar 1977: 38–9, 375–85. For Late Antiquity, see Syn. de regno 17.1–2 and Chapter 12. On the locations in which emperors delivered justice, see Färber 2014: 67–122. ⁵ Millar 1977: 236–40 discusses the imperial cognitio. Caillan Davenport, Envisioning Audiences at the Roman Imperial Court In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0014

      

279

greasing the right palms to secure an audience. Accused criminals had no such problems getting into the court, but their primary concern was the same as that of all other petitioners: that the emperor should act as the model of a good ruler and prove to be a merciful judge.⁶ One did not usually encounter the emperor alone: instead, he was surrounded by relatives, advisors, officials, guards, slaves, and other members of the court community. Courtiers could prove to be impartial and reasonable sources of advice for the emperor or malign influences who led him astray. Once one was admitted to court, the experience of being in the presence of the emperor, gazing upon his face, and speaking with him, was regarded by Romans as something akin to meeting a god.⁷ All who entered hoped that the emperor, like a god, would listen to their prayers and render decisions that were just and fair. In this chapter, I shall examine the ways in which the emperor and his court were portrayed between the first and sixth centuries in accounts written by, or featuring, individuals who had audiences with the emperor, either as petitioners and ambassadors or as accused criminals committed to stand trial.⁸ These individuals were representatives of different religious, political, and cultural communities, which I shall call ‘interest groups’. These interest groups included sophists, philosophers, the Greek community of Alexandria, Jews, and Christians. Some visitors have left us their own account of their experiences, such as Philo’s memorable report of his meeting with Caligula in the Embassy to Gaius, Eusebius’ descriptions of Constantine’s court in the Life of Constantine, and Athanasius’ account of his audience with Constans in the Defence Addressed to the Emperor Constantius, which Fabian Schulz has already discussed in Chapter 10.⁹ However, most of our stories feature in works written by other members of the interest groups. Examples of these include Philostratus’ Life of Apollonius of Tyana and Lives of the Sophists, Eunapius’ Lives of the Sophists and Philosophers, the Acts of the Alexandrians, Jewish rabbinic literature, and Christian apocryphal texts and hagiographies. I shall argue that these disparate narratives, despite their obvious differences in time period, genre, and outlook, exhibit shared concerns and interests which enable us to reconstruct how Romans from across the empire envisioned an audience with the emperor at court.¹⁰

⁶ For the association between kingship and justice, see Noreña 2009: 273–4. ⁷ Two examples from different time periods: Plin. HN pref. 11 (Titus); Aus. Grat. Act. 1.2 (Gratian). The late antique vision of emperors is deftly sketched by Kelly 1998: 139–40. ⁸ The focus is on individuals and communities within the Roman Empire itself, which means that the chapter does not consider embassies from foreign nations. ⁹ Since Schulz devotes considerable attention to Athanasius’ account, I will not deal with it in this chapter, though it shares many similarities with other Christian narratives discussed here. ¹⁰ The selection of texts is based on wide reading across the relevant genres; while some texts discussed are unique examples of their genre (for example, the Acts of the Alexandrians), others, such as hagiographies, have been chosen as representative samples of the genre.

280

   

2. Stars and Storytellers I would first like to consider the different interest groups and their representatives, and the various narrative contexts in which these stories appear. The most prevalent genre is biography, which focuses on the life and deeds of a particular individual or a group of individuals (though not always in a chronological birth-to-death narrative).¹¹ These biographies take different shapes. There are autobiographical narratives included in a larger work, as in Philo’s account of meeting Caligula in the Embassy to Gaius or Eusebius’ encounters with Constantine in his Life of the emperor; biographical novels such as Philostratus’ Life of Apollonius of Tyana; collective biography, like Eunapius’ Lives of the Sophists and Philosophers; Christian hagiography, as in Sulpicius Severus’ Life of St Martin of Tours; and Christian Acts which tell the story of a particular apostle, such as the Acts of Paul. All of these biographical forms focus on a particular representative of the interest group—whom I shall call the ‘star’ of the narrative—and their meeting with the emperor at court.¹² Other key writings that we shall consider, although they do not necessarily take the form of biography, share the biographical impulse in that they focus on a particular star.¹³ We find this impulse, for example, in the Acts of the Alexandrians, which are written in the form of ‘court acta’ (trial transcripts), but which nevertheless concentrate on the confrontation between a particular Alexandrian citizen and the emperor.¹⁴ Some stories of court audiences are best classified as anecdotes, in that they are tales embedded in larger texts to illustrate a specific point.¹⁵ The anecdote and its close companion, the apophthegm, are not always reliable sources for historical events, but they can be very useful for reconstructing ‘attitudes and ideologies’—in our case, those of the different interest groups.¹⁶ Some anecdotes circulated orally for more than a hundred years before being written down, such as the stories of Roman emperors and rabbis found in Jewish exegetical literature.¹⁷ Even though these anecdotes are not complete biographies of individuals, they are still character-focused, because they concentrate on the encounter between the star and the emperor. We should not draw too fine a line between the Jewish tales and, for example, the collective biographies of Philostratus and Eunapius and the lives

¹¹ For considerations on the form of biography, and comparisons with ancient and modern expectations, see Hägg 2012: 1–8. ¹² The use of the term ‘star’ is borrowed from Brown 1971: 82, describing holy men as ‘ascetic stars’, and Swain 1996: 397, who refers to the ‘sophistic stars’ in Philostratus’ Lives of the Sophists. ¹³ On the growth of biographical narratives in the Roman imperial period, see Swain 1997. ¹⁴ For the form of trial acta, see Musurillo 1954: 249–52. ¹⁵ On anecdotes in Roman imperial literature, see Saller 1980. ¹⁶ Saller 1980: 82. See also Laurence and Paterson 1999: 194–7; Langlands 2018: 330–1. ¹⁷ Moeser 2002: 107–49. On Jewish oral tradition, Jaffee 2001 is fundamental.

      

281

of pagan holy men, which are also heavily comprised of anecdotes.¹⁸ We might say, therefore, that all our stories represent imaginings of the court which are focalized through one particular character, the ‘star’. The characterization of the emperor and his court as either positive or negative depends on their treatment of the star and, by extension, the interest group that he or she represents.¹⁹ We shall now briefly review the different interest groups and their aims and agendas, beginning with the Greek world. Philostratus, writing in the third century, and Eunapius, writing in the fourth, highlight the important place of Greek intellectuals (including sophists, philosophers, and pagan holy men) within Roman society.²⁰ Good emperors are portrayed as lending a sympathetic ear to petitions from these men and depending on their sage advice in order to rule wisely.²¹ There is a certain degree of literary invention in the works of both authors. For example, although Philostratus had first-hand knowledge of the Severan court,²² his works created ‘fictional worlds’ inhabited by real historical figures in order to demonstrate the vitality of Greek intellectual culture in the Roman Empire.²³ A different perspective on the imperial court emerges from works featuring another Greek interest group, the members of the gymnasial class in Alexandria. Their accounts of embassies to, and trials before, Roman emperors are told in a group of texts which circulated widely in Roman Egypt; these are collectively known today as Acts of the Alexandrians.²⁴ Although inspired by real events—the conflict between the Greeks and the Jews of Alexandria over their rights and privileges in the first century²⁵—the narratives are fictionalized descriptions of events at the imperial court, given in the form of trial transcripts.²⁶ The Acts of the Alexandrians thus share some themes with the works of Philostratus and Eunapius in that they celebrate Greek champions. But the Alexandrian ambassadors are prepared to die for their social position and privileges, a sacrifice which is usually associated with religious groups such as Jews and Christians. The Jewish depictions of embassies to the Roman imperial court range from written eyewitness accounts to oral testimonies. A first-hand narrative is provided ¹⁸ On Philostratus, see Bowersock 1969: 15; Goldhill 2009: 105–9; cf. Hägg 2012: 344–51. For similar adaptions of the biographical format in the lives of holy men, see Cox Miller 1983: 55–60. ¹⁹ For focalization as a key choice in historical storytelling, see Marincola 1999: 303–4. ²⁰ On Philostratus’ life and career, see Anderson 1986: 3–8; Bowie 2009. For Eunapius, see Penella 1990; Cox Miller 2000; Watts 2005. ²¹ For Philostratus, see especially Flinterman 2004, who traces the intellectual heritage and role of Philostratus’ heroes as royal advisors. Rouiller 2010: 369–73 argues that the court of Septimius Severus was a primary audience for the Life of Apollonius of Tyana. Eunapius’ Lives were designed to compete with, and in many ways emulate, the hagiographies of Christian holy men (Cox Miller 2000: 222, 235–41). ²² Philostr. VA 1.3.1, with Bowersock 1969: 15; Bowie 2009: 19–20. ²³ Kemezis 2014: 156. For Philostratus’ investment in the world he describes, see Schmitz 2009: 60–2. ²⁴ Musurillo 1954. ²⁵ Harker 2008: 10–24. ²⁶ Musurillo 1954: 249–52; Harker 2008: 1–4.

282

   

by Philo of Alexandria, the philosopher who led a Jewish delegation to Emperor Caligula in 39/40.²⁷ His Embassy to Gaius, although based on his own experiences, is not without an agenda, as Philo seeks to demonize the emperor, his courtiers, and the opposition embassy from the Greek community of Alexandria.²⁸ The other significant Jewish depictions of the emperor and his court appear in exegetical writings of the rabbis, who often used contemporary examples set in the Roman world to illustrate their interpretations of the Hebrew Bible.²⁹ The rabbinic commentaries date from the period of Late Antiquity, but they preserve oral traditions about meetings between Jews and earlier emperors, such as Trajan, Antoninus Pius, and Marcus Aurelius.³⁰ The specific examples discussed in this chapter come from the midrash, Jewish commentaries which are first attested in the third century, and the Babylonian Talmud, which dates from the sixth century.³¹ These tales feature similar themes to Philo’s Embassy, in that they critique anti-Jewish emperors and celebrate rulers sympathetic to their religion, but they are also heavily fictionalized, featuring events that could never have taken place in reality. We turn now to Christian narratives. The first category of texts is the Apocryphal Acts, which record the experiences of Christians before pagan emperors in the pre-Constantinian period. The term ‘apocryphal’ is used to refer to these narratives because, although they describe the lives of apostles and early Christian missionaries, they were not included in the New Testament, unlike the late first-century Acts (also known as the Book of Acts or Acts of the Apostles).³² The circulation of the Apocryphal Acts from the second century onwards reveals that Christians needed texts to document and commemorate their struggles against pagan emperors.³³ The second category is composed of encounters with Christian emperors. These stories are found in texts such as Eusebius’ Life of Constantine, an encomiastic biography of the emperor,³⁴ and in the new genre known as hagiography, the writing of lives of holy men and women, which emerged in the fourth century.³⁵ There are certain rhetorical and fictional ²⁷ For Philo’s life and background, see Schwartz 2009. The date of the embassy is discussed by Smallwood 1970: 47–50. ²⁸ On Philo’s rhetoric, see Leon 2016; Gruen 2016: 400–9. ²⁹ Lieberman 1944. ³⁰ On Jewish oral and scribal traditions, see Jaffee 2001; and for the incorporation of anecdotes in Jewish exegetical literature, see Moeser 2002: 120–4 For encounters with Roman emperors, see Wallach 1941; Herr 1971. ³¹ For the midrash, see Fraade 2007: 99–100; and for the Babylonian Talmud, see Kalmin 2006. ³² See Tuckett 2015 and Pervo 2015. Cf. Junod 1992, who discusses the problems in categorizing ‘canonical’ and ‘apocryphal’ texts. ³³ Davenport 2020a. We will examine the Acts of Peter and the Acts of Paul, which were probably written in Asia Minor in the late second century (Elliott 1993: 229–30, 357, 392). Another text, the Acts of John at Rome, dates to the period between Constantine and the early sixth century (Junod and Kaestli 1983: 835–61). ³⁴ Cameron and Hall 1999. ³⁵ On the rise of the hagiographical discourse, see Van Uytfanghe 1993. For the connections between the Life of Constantine and hagiography, see Cameron 2000.

      

283

elements in all these Christian narratives. Despite the fact that he was a sometime visitor to the court,³⁶ Eusebius embellished his account to emphasize the Christianity of the court environment, which fitted with his portrayal of Constantine as the new Moses.³⁷ All hagiographies, regardless of the many claims to autopsy that they contain, are likewise literary imaginings of saints’ lives, often with fantastical and supernatural elements.³⁸ Sometimes the saints receive very positive receptions at court, as in the fifth-century Life of Melania the Younger, which describes Melania’s encounter with the pious Serena.³⁹ On other occasions, however, the holy men and women need to win over emperors who are opposed to them, as in the Life of St Martin of Tours and the supplementary Dialogues, written by Sulpicius Severus in the late fourth century.⁴⁰ The most extreme confrontations feature in texts which dramatize encounters between emperors and holy men who espoused different Christological beliefs. Such religious divisions consumed the imperial and ecclesiastical establishments in Late Antiquity (as we shall see in John of Ephesus’ sixth-century Lives of the Eastern Saints). The common theme of all these hagiographical texts is their meditation on the role of spiritual leadership within the context of a monarchical state.⁴¹ The saints are depicted as paradigms of piety, and the devoutness (or lack thereof ) of the emperor and his family members is reflected in their treatment of them. The purpose of this necessarily short survey has been to situate the different interest groups, and the stories that they told about their representatives encountering the emperor at court, in their historical and literary context. The key point that I wish to emphasize is that all our narratives are, to one degree or another, imaginings of the court, containing some fictional or fictionalized elements. Real historical figures appear even in the fictional stories in order to give them a sense of authenticity.⁴² The tales emphasize the role of the star as the champion of his or her interest group, and their reception and treatment at the imperial court reflects the emperor’s view of the community that they represent. We shall now proceed to examine how these features play out within the context of the narratives themselves, and what they tell us about how Romans imagined the imperial court.

³⁶ Eusebius only visited the court on four occasions and was not a close confidant of the emperor, despite wishing to appear so (Barnes 1981: 253–4, 265–7; Drake 1988: 26–8, 32–4). ³⁷ Important discussions include Rapp 1998; Dagron 2003: 127–35; Johnson 2014: 160–2. ³⁸ See Viggiani 2009 for a detailed exploration of these different elements in the Life of Melania the Younger. ³⁹ The Life, which exists in both Greek and Latin versions, was written by Gerontius, a monk who took charge of Melania’s monastery on the Mount of Olives after her death (Gorce 1962: 54–62). ⁴⁰ Sulpicius was a disciple of Martin of Tours who learned some stories from the saint himself (Sulp. Sev. V. Mart. 25.1–8, see also the Preface 1.7–9). For the dates of these works, see Labarre 2004: 102; Barnes 2016: 215–24. ⁴¹ Fontaine 1976: 116–21. For the major hagiographical motifs in the text, see Labarre 2004: 112–16. ⁴² For this aspect of Eurasian narratives of kingship, see van Leeuwen 2017: 15.

284

   

3. Discussion and Dialogue The principal feature shared by all our stories is their focus on the star who meets and speaks with the emperor or a member of his family. This manifests itself in three different ways, depending on the context in which the star appears at court. The first is the trial scene, in which the accused is questioned by the emperor and sometimes by one or more of his advisors as well. This is often given in transcriptstyle dialogue, as in the Acts of the Alexandrians. These Acts usually feature an introductory prose preamble which sets the scene, before providing what purports to be a word-for-word reproduction of the conversation. The following example comes from the Acts of Appian: : ‘. . . who sending the wheat (?) to the other cities, sell it at four times its price, so as to recover their expenses.’   : ‘And who receives this money?’  : ‘You do.’  : ‘Are you certain of this?’ : ‘No, but that is what we have heard.’  : ‘You ought not to have circulated the story without being certain of it. Executioner!’⁴³ In some of our other texts, the dialogue between the accused and the emperor retains its prominence, but it is embedded within a larger prose narrative. The Christians were particularly interested in describing trials before the emperor, given the importance of martyrdom narratives to the shaping of their community and identity.⁴⁴ In the apocryphal Acts of Paul, Paul is brought before Emperor Nero in chains. The emperor asks him: ‘Man of the great king, now my prisoner, what induced you to come secretly into the Roman Empire and to enlist soldiers in my territory?’⁴⁵ Paul launches into a short speech, in which he replies that he has enlisted men to serve Christ throughout the entire world, and that Nero will only be saved if he too embraces Christianity. The same narrative interest in the trial dialogue can be found in the pagan tradition in the form of Apollonius of Tyana’s arraignment before Domitian, which occurs in Philostratus’ Life of Apollonius. Domitian asks the sage four questions: why he wears strange clothing; why people ⁴³ Acta Appiani [Musurillo, Acta XI], col. ii, ll. 3–12: Ἀππιανὸς . . . πεμπόμενοι εἰς τὰς ἑτέρας πό[λεις] τοῦ τετραπλοῦ πωλοῦσι ἵνα σώσω[σι] ἃ ἐδώκασι. αὐτοκράτωρ εἶπεν: κα[ὶ] τ[ίς] ἐσ[τ]ιν ὁ τὰ χρήματα λαμβάνων; Ἀππιανὸς εἶπεν: σύ. [αὐτ]οκράτωρ: καὶ τοῦτο πέπεισαι; Ἀππιανὸς: οὔ, ἀλλ᾽ ἠκούσαμεν. αὐτοκράτωρ: καὶ πρὸ τοῦ σε πεισθῆναι οὐκ ὤφιλας τοῦτον τὸν λόγον ἀποφήνασθαι. σπεκουλάτωρ. (trans. Musurillo 1954, slightly modified). ⁴⁴ Perkins 1995. ⁴⁵ Acta Pauli 11.3 = Martyrium Pauli 3: Ἄνθρωπε τοῦ μεγάλου βασιλέως, ἐμοὶ δὲ δεθείς, τί σοι ἔδοξεν λάθρα εἰσελθεῖν εἰς τὴν Ῥωμαίων ἡγεμονίαν, καὶ στρατολγεῖν ἐκ τῆς ἐμῆς ἐπαρχίας; (ed. Lipsius 1891: 110, trans. Elliott 1993).

      

285

call him a god; why he predicted a plague at Ephesus; and on whose behalf he had sacrificed a boy. Apollonius’ answers—and the positive response they receive from the audience—compel Domitian to declare the sage innocent of all charges.⁴⁶ In all three of these trial narratives, the encounter between the star and the emperor is either entirely imaginary or heavily fictionalized. But the citation of the verbatim dialogue gives the moment immediacy, vividness, and a sense of authenticity. The back-and-forth between emperor and sage in Philostratus’ Life of Apollonius dramatizes the climactic moment in which Apollonius finally comes face to face with Domitian. In other cases, the pseudo-verbatim record provides inspiration to fellow members of the interest group by allowing them to see how their peers stood up to Roman emperors. By using the trial transcript format, the authors of the Acts of the Alexandrians (who remain unknown) wanted their accounts to be regarded as authentic records of appearances before the emperor, and portrayed the Alexandrian stars as role models for their fellow citizens.⁴⁷ The same impulse lies behind the creation of a dialogue between Nero and Paul in the Acts of Paul. In the canonical Acts and 2 Timothy, the two men never meet. By filling in this lacuna within the Christian tradition, the author of the Acts of Paul created an inspiring moment in which the apostle challenges the authority of the Roman emperor. The second type of dialogue takes the form of a philosophical discussion between emperor and star. In Philostratus’ Life of Apollonius, the newly acclaimed Vespasian summons the sage to his side in Alexandria. The emperor asks Apollonius and two other philosophers to advise him as to how he should avoid ruling as a tyrant in the manner of his predecessor Nero.⁴⁸ There is also a cycle of stories in the Babylonian Talmud consisting of dialogues between a Roman emperor called ‘Antoninus’ and Rabbi Judah the Prince. Antoninus has been identified with Antoninus Pius, Marcus Aurelius, or one of the Severan emperors, but certainty is improbable.⁴⁹ The key point is that the dialogues depict the Jewish rabbi as an advisor and confidant of the Roman emperor: Antoninus even has a cave with a secret passageway leading from his palace to the rabbi’s house. To use the terminology of David Starkey, the rabbi has not only access to the ruler, but also real intimacy with him.⁵⁰ As with Vespasian’s discussion with Apollonius, many of the emperor’s questions have wider philosophical application: he asks what he should do about difficult advisors, his daughter’s adultery, and whether he should make his son ‘Asverus’ his heir.⁵¹ This type of discussion between a king and a wise man had precedents in earlier Hellenistic dialogues, such as the encounter between Alexander and the gymnosophists.⁵² The continued popularity ⁴⁶ ⁴⁸ ⁵⁰ ⁵¹ ⁵²

Philostr. VA 8.5.1–3. ⁴⁷ Davenport 2020a: 271–2, 278–80. Philostr. VA 5.27–38. ⁴⁹ Wallach 1941: 259–60; Jacobs 1995: 125–9, 153. Starkey’s concepts are discussed in the Introduction of this book. Babylonian Talmud, Avodah Zarah 10a. Wallach 1941; Herr 1971: 123–7; Jacobs 1995: 129–31.

286

   

of this genre and its reappearance in new forms under the Roman Empire reveals perennial interest in the philosophy of kingship and the behaviour of the good ruler. The third and final type of dialogue with the emperor occurs during embassies to the court, in which the star makes a request on behalf of his or her interest group. One of the most memorable encounters is that between the Jewish ambassadors and Emperor Caligula in Philo’s Embassy to Gaius. Philo writes that he had hoped he would be received by Caligula seated like a judge with his learned assessors beside him. The emperor would then give both parties the proper time to speak and consult with his advisors before delivering the verdict.⁵³ This was not what happened at all. The discussion is prompted at each stage by a (rather aggressive) question or statement by Caligula: (1) ‘Are you the god-haters who do not believe me to be a god, a god acknowledged among all the other nations but not to be named by you?’ (2)

‘Why do you refuse to eat pork?’

(3)

‘We want to hear what claims you make about your citizenship.’⁵⁴

Philo details the responses of his embassy, noting when they were not allowed to answer questions fully because the Alexandrians interjected, and describing how Caligula attempted to catch them off guard by constantly moving throughout one of his residences.⁵⁵ The record of this conversation underlines Philo’s argument that Caligula was unfit to be emperor, because he turned their audience into a theatrical event which made a mockery of proper procedure.⁵⁶ These narratives all share a coherent vision that the ideal emperor was an impartial and generous listener. The emperor’s engagement with, and positive response to, the star’s appeals function as signs of both his fitness to rule and the star’s ability to represent his or her interest group properly. In Philostratus’ Lives of the Sophists, the Greek sophists often petition the emperor for favours or ask for assistance for their own city or province.⁵⁷ Unlike the examples above, Philostratus rarely presents these encounters with verbatim dialogue, preferring to summarize the contents of the sophists’ pleas and their effect on the emperor. But the interaction between the emperor and star remains key to the narrative. For

⁵³ Philo Leg. 180, 183, 350. For other sources attesting similar expectations, see Dio 77(76).17.1; Plin. Pan. 79.6–7. ⁵⁴ Philo Leg. 353: ‘ὑμεῖς,’ εἶπεν, ‘ἐστὲ οἱ θεομισεῖς, οἱ θεὸν μὴ νομίζοντες εἶναί με, τὸν ἤδη παρὰ πᾶσι τοῖς ἄλλοις ἀνωμολογημένον, ἀλλὰ τὸν ἀκατονόμαστον ὑμῖν;’ 361: ‘διὰ τί χοιρείων κρεῶν ἀπέχεσθε;’ 363: ‘βουλόμεθα μαθεῖν,’ ἔφη, ‘τίσι χρῆσθε περὶ τῆς πολιτείας δικαίοις’ (trans. Colson, Loeb). ⁵⁵ Philo Leg. 363–7. ⁵⁶ Philo Leg. 351, 359, 368. For attempts to weigh up the accuracy of Philo’s account, see Barrett 1993: 182–91; Gruen 2002: 66–8; Schwartz 2009: 30–1. ⁵⁷ Bowersock 1969: 44–7.

      

287

example, Philostratus describes how the eloquent oration of Scopelianus of Clazomenae convinced Emperor Domitian to rescind his famous vine edict in Asia.⁵⁸ Scopelianus emerges as a heroic advocate for his home province. Even in narratives that depict genuine historical events, there is often a considerable degree of fictionalizing in order to represent the star as a persuasive individual. In his Life of Constantine, Eusebius describes Constantine’s reaction to the oration On Christ’s Sepulchre, which the bishop himself delivered at Constantinople in November 335: He listened with rapt attention, and where a large audience was standing round right inside the palace he stood up and listened with the others. When we begged him to rest on the imperial throne which was nearby, he would not do so, but made a shrewdly considered critique of the speech, and affirmed the truth of its doctrinal theology.⁵⁹

Eusebius’ account of his time at court is not entirely reliable, since he omits the primary reason for his visit. He was a member of a delegation of bishops sent to Constantine to convey the decision of the Council of Tyre that Athanasius should be deposed from the see of Alexandria.⁶⁰ The embassy was an embarrassing farrago for Eusebius and his colleagues, and consequently was not included in the Life.⁶¹ Instead, Eusebius used his own oration as the basis of an anecdote to portray Constantine as a receptive listener. The emperor’s response validates Eusebius’ status as a representative of the Christian church. A similar treatment of this theme appears in a later work of Christian hagiography, Gerontius’ Life of Melania the Younger. Melania visits Serena, the cousin of Emperor Honorius and the wife of the magister militum Stilicho, to ask for her assistance in arranging the sale of her properties, so that she and her husband can embrace the ascetic lifestyle.⁶² The conversation between the two women is given in great detail by Gerontius: Serena duly promises to intercede with Honorius to ensure that Melania’s properties are sold and that the money is distributed to worthy individuals. The historical role of imperial women as influential intercessors with the emperors has been explored in several chapters of this volume.⁶³ In the context of the Life of Melania the Younger, the dialogue between the two ⁵⁸ Philostr. VS 520. On the edict, see Jones 1992: 77–8. ⁵⁹ Eus. V. Const. 4.33.1–2: προθυμότατα μὲν τὰς ἀκοὰς ὑπεῖχεν, πλήθους δ᾽ἀκροατῶν περιεστῶτος ἔνδον ἐν αὐτοῖς βασιλείοις ὄρθιος ἑστὼς ἅμα τοῖς λοιποῖς ἐπηκροᾶτο, ἡμῶν δ᾽ἀντιβολούντων ἐπὶ παρακειμένῳ τῳ βασιλικῷ θρόνῳ διαναπαύεσθαι, ἐπείθετο μὲν οὐδαμῶς, συντεταμένῳ δὲ λογισμῷ τὴν διάκρισιν ἐποιεῖτο τῶν λεγομένων, ταῖς τε δογματικαῖς θεολογίαις ἀλήθειαν ἐπεμαρτυρεῖ (trans. Cameron and Hall 1999). ⁶⁰ Drake 1988: 22–3; Cameron and Hall 1999: 328–9. ⁶¹ On the omission, see Drake 1988: 22–3; Cameron and Hall 1999: 328–9, and on the events of the embassy, see Barnes 1981: 239–40; 1993: 23–5. The Council of Tyre itself is described in rather oblique terms at Eus. V. Const. 41.1–42.5. ⁶² Vit. Mel. Gr. 11–13. ⁶³ See the Introduction of this book, Chapter 9, and Chapter 11.

288

   

women serves as a demonstration of both Melania’s and Serena’s piety and devotion to God. There is a clear message here: when a petitioner and the imperial house share the same Christian qualities, justice will be done. In the next section of the chapter, we will see what happens when their values are not so closely aligned, even though the emperor might still be Christian. The focus on the dialogue with the emperor or a member of his family in our narratives reflects the fact that this was the pivotal moment of any audience at court. The common theme is the notion of justice and fairness: embassies and petitioners hope that they will receive a generous hearing and have their requests granted; defendants expect a fair trial from an objective judge; and philosophers want to guide the emperor to make the correct decisions and to rule in the manner of an ideal king. The narratives not only are concerned with their portrayal of the emperors—they also want to depict their stars as passionate and fearless advocates for their interest group.

4. Persuasion and Punishment It was rare for any emperor or imperial representative to agree automatically to a petition or request: they usually required convincing arguments to be offered, as shown in the stories of Scopelianus and Domitian and Melania and Serena. Nor did they always do so in a generous spirit, with Theodosius II once characterizing appeals from Jewish representatives as ‘pitiable entreaties’ (miserabiles preces).⁶⁴ And so, in many of our narratives, emperors fail to be convinced of the righteousness of a cause by mere words, forcing the stars to try other divine or supernatural methods of persuasion. Sulpicius Severus’ account of the visit of Martin of Tours to the court of Magnus Maximus in Trier in 387 forms an appropriate starting point, as it depicts Martin skilfully manipulating court protocol to gain the upper hand with the emperor. Severus describes how Martin, unlike other less righteous bishops, was initially unwilling to visit the court because Maximus had staged a usurpation against Emperor Gratian. Eventually, Maximus persuades Martin that his soldiers had chosen him through the will of God, and the bishop consents to come to Trier.⁶⁵ There he dines with the emperor and his courtiers, including Evodius, the praetorian prefect; Marcellinus, Maximus’ brother; and the emperor’s uncle (whose name is unknown).⁶⁶ Martin is placed next to the emperor, as Maximus wants to put the bishop in a position to honour him. When the ‘loving cup’ is brought to the table to be shared between the guests, Maximus gestures that it should be given to Martin first, with the intention that he should receive it from ⁶⁴ CTh. 16.8.26 (423). ⁶⁵ Sulp. Sev. V. Mart. 20.1–4. For the date, see Barnes 2016: 231. ⁶⁶ PLRE I Evodius 2, Marcellinus 12, Anonymus 36.

      

289

the bishop in turn. However, after drinking, Martin instead decides to pass the cup to his own priest.⁶⁷ This was a significant break with protocol, but it did not anger the emperor—quite the contrary. As Severus writes: The emperor and all those present were so amazed by what he had done that they actually approved of an act which humiliated them. And the news spread through the palace that Martin, at the emperor’s banquet, had done something which none of the bishops had done at dinners hosted by lesser officials.⁶⁸

Martin’s treatment of Maximus, and the emperor’s subsequent acknowledgement of the bishop’s holiness through this action, is designed to represent the triumph of the spiritual over the secular world.⁶⁹ The bishop, Severus writes, was subsequently invited back to the palace at Trier on many occasions, becoming a regular confidant of the emperor.⁷⁰ In some Jewish and Christian texts, the stars perform miracles to win the emperor’s favour. A story in the Babylonian Talmud tells how Rabbi Simeon ben Yohai and Rabbi Eleazar travelled to Rome to ask Antoninus Pius to rescind Hadrian’s ban on circumcision, which had made all who practised it liable to the death penalty.⁷¹ During the visit to Rome, Simeon ben Yohai successfully exorcises a demon from the emperor’s daughter.⁷² The Jews therefore win Antoninus’ gratitude: He [the emperor] said to them: ‘Request whatever you desire’. They were led into the treasure house to take whatever they chose. They found that bill [Hadrian’s law], took it, and tore it to pieces. It was with reference to this visit that R. Eleazar son of R. Jose related: ‘I saw it [sc. the curtain taken by Titus from the Temple of Jerusalem] in the city of Rome and there were on it several drops of blood’.⁷³

We know that Antoninus Pius did issue a rescript which allowed Jews to practise circumcision again, but continued to forbid it among other peoples.⁷⁴ This suggests that the emperor was probably petitioned by members of the Jewish community to grant them an exception to his predecessor’s decision.⁷⁵ In the story from the Babylonian Talmud, however, the regular process of petition and ⁶⁷ Sulp. Sev. V. Mart. 20.5–6. ⁶⁸ Sulp. Sev. V. Mart. 20.7: quod factum imperator omnesque qui tunc aderant ita admirati sunt ut hoc ipsum eis, in quo contempti fuerant, placeret. celeberrimumque per omne palatium fuit, fecisse Martinum in regis prandio, quod in infimorum iudicum conviviis nemo episcoporum fecisset (trans. White 1998). ⁶⁹ Fontaine 1967: 915; 1976: 118–19; Malmberg 2007: 75–6, 88–9. ⁷⁰ Sulp. Sev. Dial. 2.6.1–3. ⁷¹ Birley 1997: 229, 270–1; Horbury 2014: 311–16. ⁷² Although this story is apocryphal, there is a historical episode in which a Jewish leader does perform an exorcism before Emperor Vespasian, though not on an imperial relative (Joseph. AJ 8.42). ⁷³ Babylonian Talmud, Me’iloh 17b (trans. I. Epstein 1932–52). ⁷⁴ Dig. 48.8.11.pr; Smallwood 1981: 467–70. ⁷⁵ Smallwood 1959: 343; Horbury 2014: 314.

290

   

response becomes a larger than life tale. The Jewish heroes save the emperor’s daughter and win the right to enter the imperial treasury and tear up a law, even viewing the stolen spoils of the Temple of Jerusalem in the process. The tale is, of course, a fantastical one, which uses the common motif of a holy man performing an exorcism on an imperial daughter to vivid effect.⁷⁶ But rather than dismissing the tale altogether, we should ask why it developed in the first place and why this particular motif possessed explanatory power. We know that Hadrian’s harsh and uncompromising treatment of the Jewish people had lived long in their collective memory.⁷⁷ It may have been the case that many Jews did not believe that his successor would simply rescind the law on the basis of a formal petition. The story of Rabbi Simeon ben Yohai and Rabbi Eleazar’s visit to Rome might therefore have developed as a way to explain the motivations behind Antoninus Pius’ decision. This version has the added resonance of envisioning the Roman emperor as a figure who is not almighty, but a human father who needs the assistance of Jewish religious leaders to save his daughter’s life. The message, I would argue, brought reassurance to the Jews that their community and religion could be needed and respected by Roman emperors.⁷⁸ The encounter between John and Domitian in the late antique Acts of John at Rome show that Christians also thought miracles were a very effective way of winning over a pagan emperor.⁷⁹ In the story, Domitian summons John from Ephesus to Rome after he hears his prophecy that the empire will fall. He is expected to be put to death because of Domitian’s decree that Christians should be executed.⁸⁰ When John has his audience with the emperor, he informs him of the coming of a new king, Jesus Christ.⁸¹ However, Domitian is not convinced, and

⁷⁶ On exorcism, see Brown 1971: 88–9. In the late fourth-century Life of Saint Abercius, the protagonist comes to Rome, where he is enlisted by Faustina the Younger to expel a demon from her daughter Lucilla (Vita S. Abercii 58–65, ed. Nissen 1912: 76–9; see Thonemann 2012: 260 on how the hagiographer came to compose the story). Similar exorcisms are performed on a daughter of Gordian III by Saint Tryphon (Passio S. Tryphonis Prior 1–3, Altera 3–7, ed. Delehaye and Peeters 1925: 329–31, 344–5) and a daughter of Zeno, the latter of whom was cured by her own sister, Hilaria, who had become a monk (Vita S. Hilariae, trans. Wensinck 1913: 10–3). This latter story, as Wensinck notes, is a Christian adaptation of the Egyptian legend of Bent-resh, attesting to the antiquity of this storytelling motif. For its appearance in international folklore, see TMI D2176.3 (‘Evil spirit exorcised’) and D2176.3.3 (‘Evil spirit exorcised by saint’). ⁷⁷ Smallwood 1959: 337–8 collects stories of Jewish martyrs that may be related to Hadrian’s circumcision edict. ⁷⁸ The story is reprised in Medieval Jewish literature during an embassy to the Byzantine Emperor Basil I, who compelled Jews to convert to Christianity. The demon is expelled from Basil’s daughter by Rabbi Shephatiah (Bonfil 2009: 78–80, 168). ⁷⁹ For the historical background of Domitian’s relationship with the Christians, see Jones 1992: 114–16, and on the Roman tradition of John’s exile, see Malik 2020: 66. ⁸⁰ Acta Iohannis 5 (ed. Junod and Kaestli 1983: 867–9). ⁸¹ This paragraph summarizes the events of Acta Iohannis 7–13 (ed. Junod and Kaestli 1983: 871–9). Strangely, the discussion between Domitian and John does not take place in the palace, since at one point it is mentioned that the emperor will return there (Acta Iohannis 11 [ed. Junod and Kaestli 1983: 877]). There is no topographical precision in the discussion, and it is never stated precisely where Domitian and John meet, despite the fact that the emperor is attended by soldiers and other members of the court.

      

291

demands proof of the Christian God. John therefore orders a poisoned draught to be prepared, drinks it, and survives unscathed, to the amazement of the emperor and his courtiers.⁸² Domitian thinks his staff have failed to prepare the poison properly, so John asks for a criminal to be brought in to try the drink instead. The criminal promptly dies, but through the power of God, John brings him back to life. John then performs a second miracle, resurrecting one of Domitian’s female cubiculariae.⁸³ Amazed by this series of extraordinary events, and won over by the power of the Christian God, the emperor commutes John’s death sentence to exile. The account reminded late Roman Christians of the travails of their forebears under pagan emperors, and assured them of the rightness of their own beliefs, in the same way that the tale of the rabbis at the court of Antoninus Pius offered comfort to the Jews that their religion could help even a Roman emperor. Not all of our stories have such happy endings. When the star does not achieve his goal because the emperor is implacable or hostile, he sometimes decides to criticize them in a striking display of parrhesia (‘frank speech’). It was a wellestablished tenet of Greek monarchical thought that a good king should be ready to accept criticism, and this idea persisted throughout the Roman imperial period to Late Antiquity.⁸⁴ Writing in the reign of Arcadius, Synesius of Cyrene argued that that emperors should not only hear praise at their courts, but also allow freedom of speech.⁸⁵ Yet the reality of autocratic rule made such expressions of honesty extremely dangerous, and only to be attempted by those who were privileged by their relationship with the emperor or their role in society.⁸⁶ In the Roman world, as in many other monarchical states, subaltern or holy figures such as the fool and the monk were allowed to exercise such licence of speech. And so it was that Christian emperors, whose court ceremonial was designed to render them austere and distant, could still find themselves the recipients of blistering diatribes from holy men, whose asceticism and piety meant that they spoke with the true voice of saintliness.⁸⁷ Many Greeks and Romans believed that it was the calling of philosophers and intellectuals to stand up to despots, even though they were somewhat less of a protected group than jesters and ascetics, being frequently executed or exiled for their views. In his Life of Apollonius of Tyana, Philostratus writes: ‘I know that ⁸² Courtiers are indicated by the Greek οἱ παρεστῶτες τῷ Δομετιανῷ πότε: ‘those then standing around Domitian’ (Acta Iohannis 10 [ed. Junod and Kaestli 1983: 875]). ⁸³ She is one of the slaves of Domitian πρὸς τῷ κοιτῶνι, ‘belonging to the bedchamber’, hence one of the cubiculariae (Acta Iohannis 13 [ed. Junod and Kaestli 1983: 879]). This strikes an anachronistic note, since female cubiculariae are only attested at the imperial court in Late Antiquity: see Chapter 14. ⁸⁴ Momigliano 1973; Brown 1992: 61–70; Rapp 2005: 267–73. ⁸⁵ Syn. de regno 1.4–5. ⁸⁶ A good example of this is the case of the Severan senator Pollienus Auspex. Upon learning of Septimius Severus’ adoption by the long-dead Marcus Aurelius, Auspex congratulated the emperor on finding a suitable father (Dio 77[76].9.4). On frankness of speech at Justinian’s court, see Leppin 2018b: 52–3. ⁸⁷ Leppin 2009: 162–4.

292

   

tyranny is the surest test of true philosophers.’⁸⁸ In the novel, when Apollonius is imprisoned by Emperor Domitian for supporting Nerva, an informer attempts to goad him into abusing the emperor behind his back. However, the sage refuses to do so, replying that he will only attack Domitian in person, demonstrating his courage in the face of tyranny.⁸⁹ In the end, Apollonius does not deliver his oration denouncing Domitian in full (for reasons to which we will return later), so Philostratus adds his own carefully constructed version of the sage’s speech as a substantial coda to these events, allowing his readers to savour the full force of the arguments deployed against the tyrannical emperor.⁹⁰ Indeed, parrhesia plays a double role in our court narratives. On one level, the courage of the star in expressing his thoughts freely marks him out as a role model for other members of his interest group. But there was also a subversive pleasure for the readers of the texts in seeing the emperor criticized, often in language that is downright abusive.⁹¹ We see this impulse in the trial narratives that make up the Acts of the Alexandrians. In the Acts of Isidorus, the hero asks Emperor Claudius to rescind the privileges of Jews in Alexandria and downgrade them to the status of other Egyptians. But Isidorus is not even-tempered in making the request, deriding Claudius as ‘the cast-off son of the Jewess Salome’!⁹² In the Acts of Appian, in which a certain Appian is arraigned before Commodus, the Alexandrian accuses the emperor of being an uneducated tyrant and a brigand-chief.⁹³ Both stories end with Isidorus and Appian being executed by the emperors. Such displays of parrhesia were ineffective in persuading the emperors to acquit the Alexandrians, but this does not appear to be the main focus of the trial narratives. Instead, the abuse allows the readers of the Acts of the Alexandrians to imagine what it would be like to tell the emperor what they really thought of him. This was evidently an attractive and popular notion, given the wide distribution of the Acts throughout Egypt. The idea that a tyrannical ruler could be reprimanded or chastised finds its ultimate expression in narratives featuring divine retribution being meted out to the emperor. This motif functions as the opposite of the persuasive miracle stories discussed earlier.⁹⁴ Two examples occur in apocryphal Acts which feature confrontations between Christians and Emperor Nero. In both the Acts of Peter and the Acts of Paul, the apostles fail to escape the long arm of Roman law and are executed by the emperor. But they both return to haunt Nero. In the Acts of Peter, Nero has a dream in which the deceased apostle flogs him and commands him to cease persecuting Christians, while in the Acts of Paul, the recently executed Paul ⁸⁸ Philostr. VA 7.1–2: Οἶδα καὶ τὰς τυραννίδας, ὡς ἔστιν ἀρίστη βάσανος ἀνδρῶν φιλοσοφούντων (trans. Jones, Loeb). ⁸⁹ Philostr. VA 7.27. ⁹⁰ Philostr. VA 8.6–7. ⁹¹ Davenport 2020a: 279–80. ⁹² Acta Isidori [Musurillo, Acta IV], Recension A, col. iii, ll. 11–12: ἐκ Σαλώμη[ς] [τ]ῆς Ἰουδα[ίας υ]ἱὸς [ἀπό]βλητος (trans. Musurillo 1954). ⁹³ Acta Appiani [Musurillo, Acta XI], col. ii, ll. 5–13, col. iv, l. 8. ⁹⁴ Davenport 2020a: 282–5.

      

293

reappears as a ghost to tell the emperor that divine punishment awaits him.⁹⁵ The depiction of these encounters is heavily indebted to prophetic scenes in early apocalyptic literature.⁹⁶ But it also shows a Christian interest in ‘equitable justice’, in which all men—even kings—are subject to the same punishments.⁹⁷ The idea that the Roman emperor, although he was the ultimate arbiter of justice within the empire itself, still had to answer to higher powers is not restricted to Christian literature. In the Life of Apollonius of Tyana, the sage—despite Domitian’s willingness to acquit him on all charges—decides to perform a miraculous vanishing act, disappearing from the emperor’s court before the eyes of the amazed audiences. Such a performance was designed not only to prove the sage’s immortality but also to undermine Domitian’s earthly authority to bring him to trial in the first place.⁹⁸ The Acts of the Alexandrians, the apocryphal Christian Acts, and the Life of Apollonius of Tyana all take the realistic historical scenario of a trial at the imperial court and adapt it to unexpected ends so that the star achieves some measure of satisfaction against the emperor, even if he is not acquitted. Similar themes pervade Christian accounts in which a holy man comes to the court of an emperor who espouses a different Christological belief. The sixth-century historian and hagiographer John of Ephesus describes the visit of the Monophysite monk Zoora to Constantinople, where the sometime stylite complains to Justinian that he had been forced from his column by supporters of Chalcedonian theology. When Justinian proves implacable and vows to uphold the position of the Council of Chalcedon, Zoora tells the emperor that God will send him a sign on his own body as a demonstration of the incorrect nature of his beliefs. Sure enough, Justinian soon begins to suffer from a swelling in his head that becomes so bad that ‘human shape was not recognized in him’ and his wife Theodora is forced to sequester him in his bedroom to stop the spread of rumours. Humbled by God in this fashion, Justinian assents to Zoora’s demands.⁹⁹ Christian narratives show a detailed knowledge of court ceremonial, which is sometimes manipulated in fantastical and miraculous ways in order to persuade the emperor. We can see this in Sulpicius Severus’ account of the meeting between Martin and Valentinian I at Trier, which took place not long after Martin was made bishop of Tours in 370/1.¹⁰⁰ According to Severus—who himself claimed to rely on eyewitnesses—Martin had repeatedly sought an audience with Valentinian to make an unspecified request, but the emperor refused to see the bishop because ⁹⁵ Martyrium Petri 12 (ed. Lipsius 1891: 102); Acta Pauli 11.6 = Martyrium Pauli 6 (ed. Lipsius 1891: 116). ⁹⁶ Rordorf 1982: 368–71. ⁹⁷ Perkins 2009: 101. ⁹⁸ Philostr. VA 8.5.1–4. See further Meyer 1917: 419; Kemezis 2014: 183–4. ⁹⁹ Joh. Eph. V. SS. Or. (ed. Brooks 1923: 24–5). For discussion, see Leppin 2009: 159–60; HasseUngeheuer 2016: 266–9. ¹⁰⁰ Sulp. Sev. Dial. 2.5.5. For Valentinian I’s presence in Trier in these years, see Seeck 1919: 238, 240, 242.

294

   

of his own pride and the urgings of his hostile wife, who followed the unorthodox Arian theology (we shall return to her shortly). In Chapter 10, Fabian Schulz has explored the historical difficulties faced by bishops who wished to achieve audiences at court, showing that they were not given special dispensation simply because they were Christians. In Severus’ story, Martin decides to seek heavenly assistance to gain access. An angel appears to the bishop, telling him to go to the palace, where he will find no doors blocking his way. When Martin arrives, everything is as the angel had said it would be: the gates are open, and no official stymies Martin’s access, so he manages to reach Valentinian’s audience hall unhindered.¹⁰¹ This was not to the emperor’s liking: When Valentinian saw Martin coming from a distance, he began gnashing his teeth, asking why Martin had been admitted. He refused to honour Martin, who stood nearby, by rising from his throne, until fire engulfed the chair and the flame scorched the king in that part of his body upon which he sat. Thus, the proud man was ejected from his seat and, against his will, rose in the presence of Martin.¹⁰²

This memorable scene reverses the expected protocol of an audience before a late Roman emperor. Admission to the emperor’s presence was carefully regulated by his staff, even for bishops. Visitors approached the emperor as he remained seated behind the veil, and they would then perform adoratio in his presence.¹⁰³ In Severus’ story, however, the tables are turned, and it is Valentinian himself who is forced to pay homage to Martin, God’s representative on earth, by a fiery assault on the imperial posterior (a scene later vividly depicted in a fourteenth-century fresco at Assisi, as shown in Figure 13.1). This negative portrayal of Valentinian was probably inspired by the emperor’s well-known recalcitrance to welcome bishops at his court with open arms.¹⁰⁴ It reflects the tensions between imperial and ecclesiastical authority which dominated the politics of Late Antiquity, and the fact that bishops always occupied an ambiguous place within the court community.¹⁰⁵ On a narrative level, Severus’ story provides his Christian readers with the pleasure of seeing this intransigent emperor suffering at the hands of God. These stories of persuasion and punishment do not take place in a vacuum involving only the star and the emperor. They demonstrate an awareness of major ¹⁰¹ Sulp. Sev. Dial. 2.5.1–8. ¹⁰² Sulp. Sev. Dial. 2.5.9: qui cum venientem eminus videret, frendens cur fuisset admissus, nequaquam adsurgere est dignatus adstanti, donec regiam sellam ignis operiret ipsumque regem ea parte corporis, qua sedebat, adflaret incendium. ita e solio suo superbus excutitur et Martino invitus adsurgit: multumque conplexus quem spernere ante decreverat, virtutem senisse divinam emendatior fatebatur: nec expectatis Martini precibus prius omnia praestitit quam rogaretur (trans. Goodrich 2015). ¹⁰³ For these aspects of court protocol, see Chapter 3 and Chapter 10. ¹⁰⁴ McLynn 1994: 80–2; Hunt 2007: 78–9. ¹⁰⁵ This idea is further developed in Chapter 14.

      

295

Figure 13.1 Martini, Simone (1284–1344): Saint Martin and the Miracle of the Fire. Assisi, Church of San Francesco. © 2022. Photo Scala, Florence.

court events (such as banquets, audiences, and trials) and the protocols that they involved. They are populated with figures whom one would expect to find at court, from the emperor’s relatives and advisors to his personal staff. But the stories also subvert the expected course of events by having the star manipulate the situation to their advantage. This could be achieved through a relatively understated sleight of hand with a loving cup or a miraculous performance. All these stories profile the star, their talents, and their status as role models for the interest group that

296

   

they represent.¹⁰⁶ The punishment narratives provide readers with the further pleasure of seeing an emperor abused, terrified, or cut down to size by the star. These were fantasies which did not usually take place at the court in real life, but they could be acted out in the imagination of the emperor’s subjects.

5. Atmosphere and Access In this section, we shall explore more fully the ways in which our narratives describe the experience of being at court, in terms of how they envisage the atmosphere and surroundings and how they depict courtiers and their relationships with each other and the emperor. As we have emphasized in this chapter so far, the central element of all our stories is the encounter between emperor and star, but even the most dialogue-focused trial accounts do feature some scenesetting aspects at the beginning. In the Acts of Isidorus, the star’s trial before Emperor Claudius takes place in the imperial gardens, where the emperor is surrounded by senators, consulars, and imperial women.¹⁰⁷ In one of his sophistic biographies, Philostratus relates how Heracleides the Lycian became flummoxed when he visited Septimius Severus because ‘he was terrified by the court and the guards’.¹⁰⁸ Unsurprisingly, more detailed descriptions occur in narratives written by those who recounted their own audience with the emperor. Philo describes how his embassy first waited for Caligula in Rome outside one of his estates near the Tiber, only to receive a wave from the emperor as he was leaving the complex.¹⁰⁹ The Jewish ambassadors followed Caligula to Puteoli, then back again to Rome, where they finally secured an audience in the horti Maecenatis et Lamiani, but were forced to traipse around the estate to keep up with the emperor.¹¹⁰ Most of our narratives, even those written by eyewitnesses, are never as detailed as we might expect about the physical aspects of the imperial residence or other locations where audiences take place.¹¹¹ We do not usually get thorough accounts of aspects such as palace architecture, furnishings, and decorations, either described with the awestruck wonder of a first-time visitor or as part of a censorious condemnation of imperial excess (the exception here is Philo, for whom Caligula’s orders about palace decoration form an integral part of the ¹⁰⁶ See Pazdernik 2010: 79 on late Roman stories of philosophers and holy men at court. ¹⁰⁷ Acta Isidori [Musurillo, Acta IV], col. ii, ll. 4–8. ¹⁰⁸ Philostr. VS 614: αὐλὴν καὶ δορυφόρους δείσαντα. For the place of guards in the court community, see Chapter 2. ¹⁰⁹ The estate, the gardens of Agrippina, was on the right bank of the Tiber near the Vatican. See Smallwood 1970: 253–4. ¹¹⁰ Philo Leg. 180–2, 185–6, 351, 358, 364–7. ¹¹¹ As Färber 2014: 119–20 shows, imperial judicial hearings were not confined to the palace, at least in the first two centuries.

      

297

emperor’s tactics during the audience).¹¹² This general reticence stands in contrast with the world of court poetry, where we do find instances of palatial ecphrasis, as seen in Martial’s encomium of Domitian’s palace and Statius’ thanksgiving for an invitation to a banquet.¹¹³ Martial and Statius are figures who resist easy classification. They were not imperial mouthpieces, penning carefully vetted verses at Domitian’s command,¹¹⁴ but hovered at the edges of the court community, seeking succour and validation from patrons more favourably positioned, such as the emperor’s chamberlain Parthenius.¹¹⁵ We can speak of them as ‘court poets’, not in the sense of propagandists drawing a salary from the treasury, but as authors whose works unmistakably emerged from, and were fostered by, the court community, its sensibilities, and its aesthetics.¹¹⁶ When Statius pays tribute to Domitian’s palace—‘Awesome and vast is the edifice, distinguished not by a hundred columns but by as many as could shoulder the gods and the sky if Atlas were let off ’¹¹⁷—his flattering lines were crafted to delight the imperial ear, perhaps even at the banquet itself.¹¹⁸ His expressions of wonder reflect both the potential experience of a visitor to the palace and the artistic milieu of Domitian and his court.¹¹⁹ Our narratives dispense with poetic extravagances and instead usually concentrate on the specific details necessary for understanding the relationship between the star and the emperor. For example, Sulpicius Severus’ Life of St Martin of Tours has a key scene set at a banquet: the hagiographer describes who was there and their seating arrangements, but not the décor of the dining hall or the food itself.¹²⁰ Eusebius’ picture of the vicennalia celebrations that followed ¹¹² Compare the use of the topos of luxury in other literary contexts, as discussed in Chapter 12. ¹¹³ Mart. 8.36; Stat. Silv. 4.2. I am grateful to Benjamin Kelly for his observations on Domitian’s court. ¹¹⁴ Compare, for example, the way in which the German Emperor Maximilian I (1459–1519) personally supervised and commented on the productions of his artists and authors, discussed by Silver 2008. ¹¹⁵ Mart. 5.6, 8.28, 9.49. We might also place the poems concerning Domitian’s eunuch Earinus within this courtly context (Mart. 9.11–13, 16–17, 36; Stat. Silv. 3.4; with Vout 2007: 167–212). On Martial’s cultivation of imperial patronage, see Watson and Watson 2003: 10–11. ¹¹⁶ See the rich discussion of Coleman 1988: xxv–xxvii. Newlands 2012: 23–36 argues against the term ‘court poet’, but seems to do so from the belief that it only represents salaried poets. We might compare such court poets, imbibing and replicating court aesthetics, to the artists who designed imperial cameos, on which see the enlightening analysis of Smith 2021. Note, in this context, Zanker 2002: 117: ‘the panegyric formulae of Domitian’s court poets accurately reflect the image that the emperor himself cultivated’. ¹¹⁷ Stat. Silv. 4.2.18–20: Tectum augustum, ingens, non centum insigne columnis, sed quantae superos caelumque Atlante remisso sustentare queant (trans. Coleman 1988). ¹¹⁸ Coleman 1988: 83–4. The poem is styled a eucharisticon, or ‘thank-offering’ to Domitian, in the manuscript, which may be authentic (see Coleman 1988: xxx). Although Newlands 2002: 32–3 plays down the court context of Statius’ patrons, it is worth noting that the dedicatee of Book Four, which begins with not one but three poems in praise of Domitian, is someone who was certainly a member of the court community: the praetor Vitorius Marcellus (see Coleman 1988: xix). ¹¹⁹ For a later example of palatial ecphrasis in court poetry in the sixth century, see Corippus’ poem in praise of Justin II, esp. Iust. 1.97–114. ¹²⁰ Sulp. Sev. V. Mart. 20.2–7.

298

   

the Council of Nicaea in 325 likewise focuses on the human elements, commenting on how the emperor’s use of space showed his favour to the bishops:¹²¹ The event was beyond all description. Guards and soldiers ringed the entrance to the palace, guarding it with drawn swords, and between these, the men of God passed fearlessly, and entered the innermost royal courts. Some then reclined with him, others relaxed nearby on couches on either side. It might have been supposed that it was an imaginary representation of the kingdom of Christ, and that what was happening was ‘dream, not fact’.¹²²

Eusebius’ aim here is to demonstrate how he and his colleagues had privileged access to the emperor, dining with him on intimate terms, either on his own couch or on adjoining ones.¹²³ But the Life does not feature a stage-by-stage description of Eusebius’ evening at court, from admittance to farewells, nor an account of what the dining hall itself looked like. If we are looking for this type of evocative detail, then we need to turn to a text such as the late fourth-century Vision of Dorotheus. Although very fragmentary, this text provides a vivid first-person narrative of Dorotheus’ service as a Christian soldier in Heaven. Although Heaven is not the imperial court, it is clearly very closely modelled on it in both form and detail.¹²⁴ As Christopher Kelly has memorably remarked, in Late Antiquity, ‘Heaven remained a very Roman place’.¹²⁵ It is particularly telling that the one prose narrative from our period that does feature a detailed assessment of a visit to a real royal court is not about the Roman court at all. This is Priscus of Panium’s vivid eyewitness report of his participation in an embassy from Emperor Theodosius II to the court of Attila the Hun in 448/9.¹²⁶ Priscus’ account was originally part of a much longer narrative history, which only survives today in substantial fragments. He describes the exterior and interior of Attila’s residence, the decoration and finery, eating and drinking, and protocol and ceremonial.¹²⁷ Priscus’ work was clearly intended for elite Roman readers, who would have been interested in the experience of a

¹²¹ For the date and place, see Barnes 1982: 219. ¹²² Eus. V. Const. 3.15.2: κρεῖττον δ’ ἦν παντὸς λόγου τὸ γιγνόμενον· δορυφόροι μὲν γὰρ καὶ ὁπλῖται γυμναῖς ταῖς τῶν ξιφῶν ἀκμαῖς ἐν κύκλῳ τὰ πρόθυρα τῶν βασιλείων ἐφρούρουν, μέσοι δὲ τούτων ἀδεεῖς οἱ τοῦ θεοῦ διέβαινον ἄνθρωποι ἐνδοτάτω τ’ ἀνακτόρων ἐχώρουν. εἶθ’ οἱ μὲν αὐτῷ συνανεκλίνοντο, οἱ δ’ ἀμφὶ τὰς ἑκατέρων προσανεπαύοντο κλινάδας. Χριστοῦ βασιλείας ἔδοξεν ἄν τις φαντασιοῦσθαι εἰκόνα, ὄναρ τ’ εἶναι ἀλλ’ οὐχ ὕπαρ τὸ γιγνόμενον (trans. Cameron and Hall 1999). ¹²³ For this ideal of imperial conviviality, and the seating hierarchy, see Malmberg 2007: 79–81, 84–5. ¹²⁴ A scholarly edition and English translation can be found in Kessels and van der Horst 1987. For the date and the description of the court, see Bremmer 1988; 2002: 128–33. ¹²⁵ Kelly 1998: 141–2 (quotation on 141). ¹²⁶ Prisc. fr. 11.2 (ed. Blockley 1983: 246–79), fr. 13.1–14 (ed. Blockley 1983: 282–95). ¹²⁷ Compare also Sidonius Apollinaris’ detailed description of the court and daily routine of another foreign ruler, Theoderic II of the Goths (Ep. 1.2.4–10).

      

299

Hunnic court and the differences between it and their own world.¹²⁸ I suggest, therefore, that Romans did not regard it as necessary to describe their own imperial court and its physical environment in great detail, even if they had never actually visited the court, because they already had fixed ideas, drawn from earlier and contemporary literature, art, and storytelling, or even their own imagination, of what a royal court looked like. Even Statius’ ecphrasis of Domitian’s palace was influenced by the royal dinners and domus conjured up by his poetic predecessors.¹²⁹ It is clear that as far as the narrators and readers of our stories were concerned, the physical space was only important as far as it elucidated the relationships between stars, courtiers, and the emperor. This primary interest in social interactions is shown by the attention paid to the characters such as attendants, advisors, and imperial women. These individuals play an important role because their influence either could assist the star in gaining access to and winning over the emperor, or, as was more often the case, ensured that obstacles were put in their path. Slave and freedman officials emerge as particularly prominent hindrances. Philo famously complained about the power of Caligula’s a cubiculo Helicon, whom he claimed had been bribed by the rival Alexandrian ambassadors.¹³⁰ Once admission to court had been granted, the stars had to reckon with imperial courtiers, who are often mentioned standing around the emperor in both trial and audience scenes. These could function as a spiteful presence, as in Philostratus’ story of Heracleides the Lycian, who feared the maliciousness of Aelius Antipater, the emperor’s secretary for Greek letters, when he came before Septimius Severus.¹³¹ In some narratives, unnamed groups of courtiers are said to conduct whispering campaigns against the star, whose influence with the emperor they regard as inappropriate. For example, in Porphyry’s Life of Plotinus, the philosopher’s plan to build the city of Platonopolis in Campania initially receives the support of Gallienus and his wife Salonina. However, they are soon turned against it by ‘some of those who attended on the emperor’, an obvious reference to hostile courtiers.¹³² Similar explanations are given in Eunapius’ Lives of the Philosophers for the downfall of Sopater at Constantine’s court. Although Sopater was a pagan philosopher, rather than a Christian, he formed a close relationship with the emperor. According to Eunapius, Constantine ‘was captivated by him and publicly made him his assessor, giving him a seat at his right hand, a thing incredible to hear and see’.¹³³ Sopater’s public position and perceived influence ¹²⁸ We might compare the accounts of European ambassadors of their visits to Oriental courts, which emphasized the physical layout of palaces and foreign rituals for the edification of their fellow countrymen back home. See Duindam 2016: 176–88. ¹²⁹ Coleman 1988: 84, 89; Malamud 2001: 33–6; Moormann and Stocks 2021: 83. ¹³⁰ Philo Leg. 167–78. ¹³¹ Philostr. VS 614. ¹³² Porph. V. Plot. 12.10–3: τινες τῶν συνόντων τῷ βασιλεῖ. ¹³³ Eun. V. Soph. 462: ἑαλώκει τε ὑπ᾿ αὐτῷ, καὶ δημοσίᾳ σύνεδρον εἶχεν, εἰς τὸν δεξιὸν καθίζων τόπον, ὃ καὶ ἀκοῦσαι καὶ ἰδεῖν ἄπιστον (trans. Wright, Loeb).

300

   

inflamed tensions at court. Eunapius writes that wicked Christian courtiers ‘were bursting with jealous malice against a court so lately converted to the study of philosophy’.¹³⁴ They therefore contrived to blame Sopater for a grain shortage in Constantinople.¹³⁵ These and other similar tales of conflict are filled with stock characters and play out in a manner that simplifies the complexities of court politics, but they nevertheless reveal a common preoccupation with the court as an arena of competition, as well as a broad understanding of how rivalries functioned. In Eunapius’ narrative, Sopater makes the transition from a visitor granted access to the emperor to an intimate friend who is able to encourage Constantine to share his philosophical preoccupations. Imperial women form a distinct sub-category of courtier, who (with some exceptions) are usually envisioned as problematic individuals working against the star, as they are throughout most Eurasian kingship literature.¹³⁶ In Philostratus’ account of Herodes Atticus in the Lives of the Sophists, he describes how the illustrious rhetor was accused of tyranny by the Athenians and brought to trial before Emperor Marcus Aurelius in Sirmium in 174. Marcus is said to be well-disposed towards the people of Athens, rather than Herodes, because— among other reasons—they are supported by his wife, Faustina the Younger, and his three-year-old daughter Vibia Aurelia Sabina.¹³⁷ Philostratus writes that the little girl ‘above all used to fall at her father’s knees with many blandishments and implore him to save the Athenians for her’.¹³⁸ When Herodes subsequently appears before the emperor to be tried, he is wracked with grief because of the sudden deaths of his own foster daughters. He therefore lashes out at Marcus, crying: ‘You sacrifice me to the whim of a woman and a three-year-old child!’¹³⁹ The idea that a toddler could really influence the course of events is implausible, and reflects the widespread view that females were inappropriate influences on emperors, even when mere babes. In the end, Marcus weighs both sides of the case equally, and passes a merciful judgement on Herodes that, in Philostratus’ view, shows his ability to behave like a true and equitable philosopher.¹⁴⁰ The stereotype of the interfering imperial woman also appears in some Christian narratives. Sulpicius Severus claims that Martin of Tours’ request for an audience with Valentinian I was blocked by his wife, whom the author incorrectly calls ‘Arriana’, rather than Justina.¹⁴¹ The name ‘Arriana’ and her objections to the pious and devout Martin were no doubt occasioned by her adherence to the Arian ¹³⁴ Eun. V. Soph. 462: ῥηγνύμενοι τῷ φθόνῳ πρὸς βασιλείαν ἄρτι φιλοσοφεῖν μεταμανθάνουσαν (trans. Wright, Loeb). ¹³⁵ Eun. V. Soph. 462–3. ¹³⁶ Van Leeuwen 2017: 163, 166–7. ¹³⁷ Philostr. VS 560. For the date of these events, and the identity of Marcus’ daughter, see Birley 1993: 180–1, 248. ¹³⁸ Philostr. VS 560: τοῦτο γὰρ μάλιστα ξὺν πολλοῖς θωπεύμασι περιπῖπτον τοῖς γόνασι τοῦ πατρὸς ἐδεῖτο σῶσαί οἱ τοὺς Ἀθηναίους (trans. Wright, Loeb). ¹³⁹ Philostr. VS 561: γυναικί με καὶ τριετεῖ παιδίῳ καταχαριζόμενος (trans. Wright, Loeb). ¹⁴⁰ Philostr. VS 561. ¹⁴¹ Sulp. Sev. Dial. 2.5.5; PLRE I Iustina.

      

301

faith, rather than the orthodox Christian beliefs enshrined at the Council of Nicaea.¹⁴² But we do also find positive portrayals of Christian imperial women, who, through their own sanctity and piety, help to smooth the way at court for suitably devout suppliants. In the Life of Melania the Younger, Serena is so impressed by Melania’s humility that she often asks bishops to invite Melania to court.¹⁴³ When Melania and Pinianus request Serena’s help in arranging the sale of their properties so they can distribute their wealth to the poor, she swiftly organizes Honorius’ approval.¹⁴⁴ Mark the Deacon’s Life of Porphyry of Gaza portrays the Augusta Eudoxia, wife of the eastern Emperor Arcadius, in a similar vein. It describes how the bishop Porphyry came to the eastern court in Constantinople in 401/2 to protest against the pagan population of Gaza and to ask for the emperor’s assistance in closing the temple of Zeus Marnas in the city. Porphyry prophesies that the pregnant Eudoxia will give birth to a boy (the future Theodosius II), and when this comes to pass, the empress agrees to assist Porphyry in convincing Arcadius to support his claim.¹⁴⁵ The malleability of the female characters to fit the needs of individual stories is shown by two contrasting representations of Plotina, wife of Trajan. The first can be found in the Acts of Hermaiscus, in which embassies from the Greek and Jewish communities of Alexandria arrive in Rome to petition Trajan on an unspecified matter.¹⁴⁶ Plotina is portrayed as pro-Jewish, working behind the scenes to lobby leading senators on Trajan’s consilium to support the Jews and oppose the Alexandrians. Hence, when Alexandrian embassy is led in to see the emperor, Trajan’s mind is already made up as a result of his wife’s malign influence. In the Jews’ own tradition, however, there is a completely different portrayal of Plotina as anti-Jewish. In the Midrash Rabbah, Plotina is said to have given birth to a child, who dies on Hanakkah. Given the empress’ loss, the Jewish population is uncertain as to whether they should light candles as part of their celebrations, but decide to proceed and follow their prescribed rituals. This is reported to Plotina, who writes to Trajan to say: ‘Instead of subduing the barbarians, come and subdue these Jews, who have revolted against you.’¹⁴⁷ In this story, Plotina’s hostility serves as an explanation for Trajan’s harsh subjugation of the Jewish revolts that occurred late in his reign. These different views about Plotina shed little light on the character of the empress, but reveal the widespread perception that imperial women functioned as harmful influences at court. Our court narratives possess a keen sense of the benefits available to the star and the interest group they represent after they have jumped through all the appropriate hoops. There were tangible and immediate rewards, such as having a ¹⁴² See also Hunt 2007: 78–9. For Justina, see McEvoy 2016b, and her subsequent clashes with Ambrose of Milan, see McLynn 1994: 170–80. ¹⁴³ Vit. Mel. Gr. 11. ¹⁴⁴ Vit. Mel. Gr. 12. ¹⁴⁵ Marc. Diac. V. Porph., esp. 39–43, 50. ¹⁴⁶ Acta Hermaisci [Musurillo, Acta VIII], col. ii, ll. 30–2. ¹⁴⁷ Midrash Rabbah, Lamentations 1.16 (trans. M. P. Ben Zeev 2005).

302

   

petition or a special favour granted, as in the case of the Jewish rabbis who were allowed to enter the imperial treasure house and tear up Hadrian’s law. But there was also the potential for the emperor, or a member of his family, not only to embrace the star as a member of their innermost circle but also to express their favour in physical terms. Here we find the stars going beyond achieving mere access to the court to display intimacy with imperial figures. In Gerontius’ account of Melania the Younger, the fact that Serena invites Melania to sit next to her, then embraces her and kisses her face and eyelids, is a sign of how much she esteems the ascetic woman.¹⁴⁸ The high standing of Martin of Tours at the court of Magnus Maximus is likewise demonstrated by the ministrations of the emperor’s wife (whose name is sadly lost to us).¹⁴⁹ In Sulpicius Severus’ story, the empress takes it upon herself to wash Martin’s feet and personally waits upon him at meals.¹⁵⁰ These acts are signs of shared Christian piety and devotion, as we have already noted, but their personal nature signals that the relationship between the star and these imperial women goes far beyond that achieved by other Christians at court. Intimacy between star and the emperor himself is taken to a new level in Philostratus’ Life of Apollonius of Tyana, in which the philosopher-sage counsels Titus and strokes his neck, ‘which was as strong as that of a bodybuilder’.¹⁵¹ This fetishization of Titus’ body represents the influence that Apollonius is able to exert over Vespasian’s son and heir. These representations of physical intimacy reflect the fact that it was a great honour to touch a Roman emperor or a member of his family and to be touched by them in return. In the Principate, the exchange of kisses on the lips between the emperor and senator was a sign of equality among peers. In Late Antiquity, changes in court ceremonial and protocol meant that visitors and courtiers were usually only permitted to kiss the emperor’s purple robe or feet.¹⁵² But in both the Principate and Late Antiquity, the granting or refusal of such contact with the emperor was carefully watched by courtiers to ascertain if an individual was in favour with the emperor or not.¹⁵³ In our narratives, if an emperor kisses or touches the star, it not only functions as a sign of personal favour, but also demonstrates his acceptance of the interest group they represent. The anecdote about Apollonius touching the neck of Titus, although it might seem fantastical, reflects the idea that the emperor’s body was a locus of competition at court. When a consular sucked Caligula’s toes at a banquet or Fronto dreamed about ¹⁴⁸ Vit. Mel. Gr. 12. ¹⁴⁹ Maximus’ wife is called Helena/Elen in the later Welsh tradition, but there is no contemporary Roman source attesting her name (Harbus 2002: 58–61). ¹⁵⁰ Sulp. Sev. Dial. 2.6.4–7. ¹⁵¹ Philostr. VA 6.30.2: γὰρ δὴ ἔρρωτο αὐτὸν ἴσα τοῖς ἀσκοῦσι τὸ σῶμα (trans. Jones, Loeb). I am grateful to Alastair Blanshard for his observations on this passage. ¹⁵² For the purple robe, see Avery 1940, and for the feet, see Const. Porph. de cer. 1.84 (1.93). ¹⁵³ On changes in court ceremonial, see Chapter 14. See Wallace-Hadrill 2011 on court politics as a watching game.

      

303

kissing Marcus Aurelius, their physical actions and desires reflect the need to be close to the emperor, and in particular, the longing for other people to see them expressing such intimacy.¹⁵⁴ What is particularly interesting is that the fascination with imperial bodies continued in Late Antiquity, when the sacred personages of the ruler and their distance from his subjects was emphasized through dress, decoration, and ritual.¹⁵⁵ Hence the court scene in Cyril of Scythopolis’ sixthcentury Life of Sabas is especially remarkable for the way in which Emperor Justinian, inspired by God himself, breaks protocol by rushing forward to greet the monk Sabas and kissing his head.¹⁵⁶ While most scholarly discussions focus on the veneration due to male emperors, there is evidence that imperial women could receive similar honours.¹⁵⁷ When, in the Life of Melania the Younger, Serena sits on her throne, invites Melania to sit beside her, and then kisses the ascetic woman, their physical intimacy transgresses the expectations of ritual and protocol in a way that would not happen publicly at court in real life, but could take place within the fictionalized world of hagiography.¹⁵⁸ When we examine our texts as a group, we can see that they are primarily interested in the people at court and their interactions with each other, rather than in the physical layout or decoration of the residences in which audiences took place. This shows that the imperial court was envisioned as a locus of competing individuals and groups, all jockeying for imperial favour. These court rivalries are portrayed in very broad strokes, as courtiers and imperial women try either to hinder the star’s access to the emperor (in case of embassies) or to secure his condemnation (in the trial narratives). There is, however, differentiation made between access (the ability to gain admission to the emperor’s presence) and real intimacy (the ability to acquire familiarity with a member of the imperial family), which reflects the hierarchy of relationships at court. It is particularly interesting that folkloric stereotypes, such as that of the scheming imperial woman, are found equally among works written by those who had experience of the court (like Philostratus) and in those narratives, including oral traditions, told by individuals who had never been to the court.¹⁵⁹ This suggests that first-hand knowledge of the ¹⁵⁴ Caligula: Dio 59.29.5. Marcus Aurelius: Fronto, Ad M. Caes. 12.6 (vdH² 1988: 66–7). On the emphasis on love and affection at the Antonine court, see further Chapter 5. ¹⁵⁵ See Kelly 1998: 143–4, 151–2. ¹⁵⁶ Cyr. Scyth. V. Sabae 71 (trans. Price 1991); Leppin 2009: 156–7; Hasse-Ungeheuer 2016: 246–7. ¹⁵⁷ Cameron and Long 1993: 401–2 on adoratio before images of an Augusta. On the equation of imperial women with emperors in dress and ritual in Late Antiquity, see Holum 1982: 31–44, 65–7, 185–6 and Chapter 9. See also Proc. Anecd. 15.13–8 for Theodora demanding obeisance, though one should note that this account has many tendentious aspects, since it portrays the empress forcing subjects into servility in opposition to the accessible Justinian. ¹⁵⁸ The fictionalizing element can also be found in stories embedded in chronicles and histories, such as the romantic tale of how the Athenian woman Athenaïs sought an audience with Pulcheria to present a petition regarding her inheritance. Pulcheria thought the beautiful woman would make an excellent bride for her brother, Theodosius II (Joh. Mal. 14.4). In reality, Pulcheria was not simply plucked from obscurity at an imperial audience: see Holum 1982: 112–21. ¹⁵⁹ For the formulaic nature of Philostratus’ court narratives, see Anderson 1986: 52.

304

   

imperial court did not necessarily determine the ways in which narratives unfolded: instead, they were shaped by shared preconceptions about access to power.

6. Conclusion In Book Seven of Philostratus’ Life of Apollonius of Tyana, the sage and his companion Damis find themselves waiting for an audience with Domitian in the forecourt outside the palace in Rome: Standing outside the palace, he [Apollonius] saw some people receiving flattery and others paying it, and the hubbub of people going in and out, at which he said: ‘This place, Damis, seems to me like a bathhouse. I see those outside hurrying to get in and those inside hurrying to get out, and some look washed and others unwashed’.¹⁶⁰

This passage perfectly captures the Roman imperial court as a place of social interaction, distinguished by intense rivalry and shifting relationships between emperor, courtiers, and visitors. It was somewhere—to adapt Apollonius’ bathhouse metaphor—that one needed to get one’s hands dirty to get ahead. This concept of the court as a community of individuals competing for the emperor’s attention is something that Philostratus would have understood as a veteran of the Severan court, but it is also a vision found widely throughout all the different court narratives written in the Roman imperial period through to Late Antiquity. In the course of this chapter, we have examined a diverse range of stories about audiences at the imperial court involving different interest groups, such as sophists, philosophers, holy men, Alexandrian Greeks, Jews, and Christians, spanning a broad chronological range across the Principate and Late Antiquity. The primary point of commonality shared by these different tales is not, as one might expect, an interest in describing the experience of the physical appearance and layout of the emperor’s residence, its architecture, finery, and furnishings. Instead, they focus on what happens when their star encounters the emperor.¹⁶¹ The encounter between monarch and subject is a timeless topos of storytelling without geographical or chronological limits.¹⁶² One very common narrative trope

¹⁶⁰ Philostr. VA 7.31.2: Προσεστὼς δὲ τοῖς βασιλείοις καὶ τοὺς μὲν θεραπευομένους ὁρῶν, τοὺς δὲ θεραπεύοντας, ἐσιόντων τε καὶ ἐξιόντων κτύπον ‘δοκεῖ μοι,’ ἔφη ‘ὦ Δάμι, βαλανείῳ ταῦτα εἰκάσθαι, τοὺς μὲν γὰρ ἔξω ἔσω ὁρῶ σπεύδοντας, τοὺς δὲ ἔσω ἔξω, παραπλήσιοι δέ εἰσιν οἱ μὲν ἐκλελουμένοις, οἱ δ᾿ ἀλούτοις’ (trans. Jones, Loeb). ¹⁶¹ It is worth emphasizing that even Statius’ description of Domitian’s banquet gives pride of place to the poet’s ecphrasis of the emperor’s visage (Silv. 4.2.38–56; Coleman 1988: 88–9). ¹⁶² Burke 2009: 203–8; Duindam 2016: 278–9.

      

305

is the meeting between king and commoner, when the king has gone incognito among his people as a way of finding out what his subjects think of him.¹⁶³ Another, which can be found in monarchical societies where the court functioned as a locus of justice, is the idea that one could always appeal to the king against abuses, especially those carried out by his own officials.¹⁶⁴ It is this concept of the Roman emperor as the ultimate judge on Earth that underlies our court narratives. But judicial systems did not operate exactly the same way across all monarchies, which means we need to ask: what—if anything—is specifically Roman about our narratives? I would suggest that the conception of the court as an arena of justice reflects the fact that, as Fergus Millar has shown, petition and response was the main mechanism through which the Roman people were able to interact with the emperor and his administration.¹⁶⁵ In our stories, this interaction is lifted out of the relatively dry and mundane world of bureaucratic procedure and transformed into a dramatic encounter between emperor and subject.¹⁶⁶ The focus on the actual dialogue spoken by the star and the emperor imbues these encounters with vibrancy, immediacy, and authenticity. The reaction of the emperor to the star’s requests or pleas is taken as representative of their attitude to the interest group at large. The narratives thus conform to pre-existing opinions and biases: in the Acts of the Alexandrians, emperors execute the members of the Greek gymnasial class because they are favourable to the Jews, while Trajan and Hadrian are usually referred to in a hostile manner in the Jewish sources because of their reputation for harshly suppressing revolts. There is a certain element of fictionality in all these accounts, even those based on eyewitness testimony, as the star, the emperor, and the courtiers are all depicted in stereotypical terms. This fictionality is taken to the extreme degree in stories which represent the acting out of fantasies. We see the stars speaking truth to power, abusing and criticizing emperors to their face, and performing miraculous acts to terrify and punish their rulers. The dream of cutting an arrogant or unjust leader down to size is a feature of subversive literature in many monarchical societies.¹⁶⁷ These tales often describe how kings are brought down by supernatural phenomena or killed for

¹⁶³ Burke 2009: 205; with Duindam 2016: 278–9 on stories in European, Middle Eastern, and Asian cultures. ¹⁶⁴ See Burke 2009: 204–5 on European rulers portrayed as ‘Solomon figures’, and Kennedy 2005: 141 on justice at the ‘Abbasid court. ¹⁶⁵ Millar 1977, and for Late Antiquity, see Dillon 2012. On justice as a key feature of kingship, broadly understood, in pre-industrial societies, see Crone 2015: 56–7. Interestingly, while imperial justice plays a significant role in stories (both textual and oral) about the Roman emperor, it does not figure prominently in art (Hekster 2020). The gulf between the two is a topic that demands further research. ¹⁶⁶ The desire to petition the emperor in person in Late Antiquity continued despite the sophisticated system of appellate jurisdiction in the late empire, and the fact that judgements could be made without the emperor present (Dillon 2012: 214–50; Färber 2014: 110–11, cf. his discussion of the reign of Justinian, 111–18). ¹⁶⁷ Scott 1977: 12, 17–18; 1990: 80–2.

306

   

their tyrannical behaviour.¹⁶⁸ The Roman interest in making the imperial court a venue for such acts of revenge reveals that they conceived of it as the empire’s preeminent arena of arbitration and justice. The view of the court as a competitive environment manifests itself in the portrayal of courtiers, such as advisors, officials, and imperial women. These characters either help the star in gaining access to the emperor and winning his favour or, more often, act as obstacles in his path. It is always a great triumph if the star can outwit these individuals and establish a real intimate relationship with the emperor, which is sometimes dramatized by acts of physical affection. I regard it as significant that even authors such as Philo, Philostratus, and Eusebius, who had first-hand knowledge of the imperial court, cleave closely to these traditional narratives of encounters between emperor and subject. Moreover, the attention paid to audiences, trials, and banquets, and the people who attended them, reveals a shared understanding in all our stories of the importance of access, ritual, and competition that transcends author, genre, and time period. When we combine this with the lack of interest shown by our narratives in extended descriptions of the physical environment of the court, we can say that in the Roman world, the court was primarily envisioned not in terms of buildings and structures, but in terms of people and personalities, and the interaction and competition between them. This prevailing perception of the imperial court as a community did not change between the Principate and Late Antiquity. Behind all these fantastical stories of vanishing holy men, abusive prisoners, and fiery thrones, there lies a real sense of what made a court a court.

¹⁶⁸ Johnson 1983: 66 (on the fates of Egyptian pharaohs in the Demotic Chronicle); Thelander 1982: 473–4 (evil kings and queens in French fairy tales); Perrie 1987: 55 (stories about Ivan the Terrible); Andaya 1993: 420–1 (belief in Southeast Asia that portents signified a bad ruler).

PART IV

C O M P A R A T I V E PE R S P E C T I V E S

14 The Evolution of the Roman Imperial Court in Historical Context Caillan Davenport and Meaghan McEvoy

1. Introduction At his open audiences he admitted even the common people, accepting the request of those who approached him with such affability that he jokingly reproved one man for hesitating as he held out his petition to him ‘like a coin to an elephant’.¹ In front of the emperor’s throne there stood a certain tree of gilt bronze, whose branches, similarly gilt bronze, were filled with birds of different sizes, which emitted the songs of the different birds corresponding to their species. The throne of the emperor was built with skill in such a way that one instant it was low, then higher, then quickly it appeared most lofty; and lions of immense size (though it was unclear if they were wood or brass, they certainly were coated with gold) seemed to guard him, and, striking the ground with their tails, they emitted a roar with mouths open and tongue flickering. Leaning on the shoulders of two eunuchs, I was led into this space, before the emperor’s presence. And when, upon my entry, the lions emitted their roar and the birds called out, each according to its species, I was not filled with special fear or admiration, since I had been told about all these things by one of those who knew them well. Thus, prostrated for a third time in adoration before the emperor, I lifted my head, and the person who earlier I had seen sitting elevated to a modest degree above the ground, I suddenly spied wearing different clothes and sitting almost level with the ceiling of the mansion. I could not understand how he did this, unless perchance he was lifted up there by a pulley of the king by which tree trunks are lifted. Then, however, he did not speak at all for himself, since, even if he wished to, the great space between us would render it unseemly, so he asked about the life of Berengar and his safety through a minister. When I had

¹ Suet. Aug. 53.2: promiscuis salutationibus admittebat et plebem, tanta comitate adeuntium desideria excipiens, ut quendam ioco corripuerit, quod sic sibi libellum porrigere dubitaret, quasi elephanto stipem (trans. Wardle 2014). Caillan Davenport and Meaghan McEvoy, The Evolution of the Roman Imperial Court in Historical Context In: The Roman Imperial Court in the Principate and Late Antiquity. Edited by: Caillan Davenport and Meaghan McEvoy, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192865236.003.0015

310

   

answered him reasonably, and when his interpreter gave a sign, I left and was soon received in the hostel assigned to me.²

These stories describe what it was like to encounter a Roman emperor. Separated as they are by a thousand years, they depict two very different worlds. The first is a famous anecdote about Augustus, told here by Suetonius, but which circulated widely throughout the Roman period.³ It characterizes Augustus as an open and accessible ruler, who admitted the people of Rome to public audiences, and encouraged them to have no fear in approaching him. In contrast, the second passage depicts a court with an intimidating and foreboding atmosphere created by all manner of wizardry. This is the account of the Italian emissary Liudprand, later the bishop of Cremona, of his audience before Constantine VII Porphyrogenitus, held in the Hall of the Magnaura. In 949, Liudprand had been sent to Constantinople with a message from his lord, Berengar, Margrave of Ivrea, who was soon to be king of Italy.⁴ In Liudprand’s narrative, the emperor is not open and accessible, but appears guarded by roaring lions and sitting on a levitating throne, and he only speaks to ambassadors through an intermediary.⁵ Constantine VII was known as the ‘Emperor of the Romans’, but the court of tenth-century Byzantium bore little resemblance to that of Augustus and his early imperial forebears. The transformation of court ceremonial illustrated by the passages from Suetonius and Liudprand returns us to the theme with which we opened this book. For Roman authorities and modern scholars alike, the evolution of court ritual is a vivid and effective way of pointing to the changes that occurred between the Principate and Late Antiquity. By the time of the Middle Byzantine world inhabited by Liudprand and Constantine VII, the ‘Empire of the Romans’ was ruled by a recognizably Medieval monarch.⁶ ² Liudprand of Cremona, Retribution 6.5: Aerea, sed deaurata quaedam arbor ante imperatoris sedile stabat, cuius ramos itidem aereae diversi generis deaurataeque aves replebant, quae secundum species suas diversarum avium voces emittebant. Imperatoris vero solium huiusmodi erat arte compositum, ut in momento humile, exelsius modo, quam mox videretur sublime, quod inmensae magnitudinis, incertum utrum aerei an lignei, verum auro tecti leones quasi custodiebant, qui cauda terram percutientes aperto ore linguisque mobilibus rugitum emittebant. In hac igitur duorum eunuchorum humeris incumbens ante imperatoris praesentiam sum deductus. Cumque in adventu meo rugitum leones emitterent, aves secundum speties suas perstreperent, nullo sum terrore, nulla admiratione commotus, quoniam quidem ex his omnibus eos qui bene noverant fueram percontatus. Tertio itaque pronus imperatorem adorans caput sustuli et, quem prius moderata mensura a terra elevatum sedere vidi, mox aliis indutum vestibus poenes domus laquear sedere prospexi; quod qualiter fieret, cogitare non potui, nisi forte eo sit subvectus ergalio, quo torcularium arbores subvehuntur. Per se autem tunc nihil locutus, quoniam, etsi vellet, intercapedo maxima indecorum faceret, de vita Berengarii et sospitate per logothetam est percontatus. Cui cum consequenter respondissem, interprete sum innuente egressus et in datum mihi hospitium mox receptus (trans. Squatriti 2007). ³ Other versions of the story can be found in Quint. Inst. 6.3.59; Macrob. Sat. 2.4.3. ⁴ On Liudprand’s life and career, see Squatriti 2007: 3–37. ⁵ The machinery of the animals is also described in Const. Porph. de cer. 2.15; Dagron and Flusin 2020: IV.2, 687–8. Brett 1954 discusses how these contraptions worked. ⁶ On the similarities between the Byzantine court and those of other Medieval kingdoms, see Macrides 2011: 234–5; Magdalino 2011: 138–44. See McCormick 1985 on the constant change and renewal of ceremonial forms throughout the Byzantine period.

      

311

The studies in this volume lead us to two main conclusions about the Roman imperial court in the Principate and Late Antiquity. The first is that long-term changes in the court as an institution certainly occurred between the first century  and the sixth century , but these developments did not occur as part of a straightforward and linear course (as might be supposed from the contrasting anecdotes about ceremonial included above). Instead, there was always negotiation, experimentation, and adaptation within this larger trajectory; we can discern changes in power dynamics, ritual experimentation, and appointment policies at court, not all of which were necessarily enduring. The second conclusion is that there were many essential continuities of court life that endured despite institutional change, such as the politics of access and intimacy. Indeed, similar continuities can be found in other monarchical societies.⁷ This does not make the study of these factors any less rewarding and important, because the dynamics of the Roman imperial court changed depending on factors such as the personality of the emperor, the location of the court, and the presence and influence of imperial favourites. These dynamics affected how perennial themes of court life, such as patronage and access, played out at different points in time. It is equally important, therefore, to pay attention to court society and culture at the micro-level and to analyze wider developments at the macro.⁸ This final chapter has two main goals. Firstly, it draws together key themes from the chapters into a discussion of the transformation of the Roman imperial court, which pays attention to both the continuities and changes outlined above. Secondly, it endeavours to place the Roman court in a broader historical context, drawing from scholarly studies of other monarchical courts in Eurasia both in Antiquity and in the Medieval and Early Modern periods. Some developments in the Roman world followed a pattern common to all or most royal courts, while others show the Roman court charting a different course.⁹ This comparative material will help us to answer a fundamental question: what was Roman about the Roman imperial court?

2. There and Back Again We can identify four major phases in the location, movement, and number of Roman imperial courts. The first, from the time of Augustus to the late third century, was usually distinguished by one court. The primary residence of this court was Rome, but it was not the only residence, given that the emperor and his court often spent time at imperial villas (where judicial and other government ⁷ Duindam 1995: 192; 2016: 200; Winterling 1997c: 151–3; Vale 2001: 17–18. ⁸ Note the comments of Asch 1991: 4–5 on the problems with creating a ‘universal model’ for all courts, and the necessity of conducting close analysis of particular circumstances. ⁹ See the important methodological remarks of Scheidel 2009a: 5 (general); Kelly 2022a: 11 (on courts). For a comparative approach to royal courts, see Duindam et al. 2011.

312

   

business could be conducted),¹⁰ or accompanied the emperor on provincial tours and military campaigns. This model of courtly residence may be termed ‘court-incapital’, not in the sense that the court never left Rome, but because Rome was the ‘capital’ city to which the court always returned.¹¹ There were some variations to this pattern, such as the joint emperorship of Marcus Aurelius and Lucius Verus (161–6), when Verus spent much of his time on the eastern frontier. In this case, we can speak of two courts. But very often collegiate rule was dictated by the desire to designate a clear succession, as under the Severans, rather than to establish genuinely separate administrations. This configuration was challenged by the demands of warfare in the mid-third century, resulting in the expansion of the imperial college and increasing time spent by the emperors on the frontiers, with particular experimentation occurring during the joint reign of Valerian and Gallienus (253–60).¹² The second phase lasted from Diocletian to the reign of Theodosius I.¹³ The Tetrarchic innovation was to formalize collegiate rulership, in which emperors spent most of their reigns in different regions, a system which gave rise to multiple courts and administrative establishments for each emperor. A series of new principal residences was established at cities such as Trier, Milan, Antioch, and Nicomedia, to enable these cities to house the peripatetic courts, as examined by Verena Jaeschke and Caillan Davenport in Chapter 3. These residences served as real rivals to Rome, a change which meant that the imperial courts would rarely visit the sacra urbs itself in the fourth century.¹⁴ The foundation of Constantinople by Constantine did little at the time to alter the peripatetic nature of imperial rule or the multiplication of imperial courts. Constantine’s sons and successors, and later the imperial brothers Valentinian I and Valens, were all primarily campaigning emperors, even though some of them (notably Constantius II and Valens) did visit and reside in Constantinople for periods. Phase three begins in the late fourth century with the death of Theodosius I in 395 and the accession of his child-emperor sons, Arcadius and Honorius. This

¹⁰ George 2022. ¹¹ See Duindam 2016: 166 who points out that even the Early Modern French court was never permanently resident at Versailles, and that royal courts based in Madrid, Vienna, or Milan often sojourned at different locations, especially in the summer. This pattern can be found in Rome, where emperors often decamped to villas in Campania. ¹² The patterns of imperial journeys and movements are studied in detail by Millar 1977: 15–53. For the argument that Rome remained the city to which the court returned until the Tetrarchic changes, see Davenport 2017. ¹³ See also Pfeilschifter 2013: 14–18, identifying the years 284–395 as a distinct phase in the Roman monarchy. ¹⁴ This does not mean, of course, that Rome was irrelevant to the emperors or the imperial administration at large during the fourth century. See Humphries 2003; McEvoy and Moser 2017: 16–20; Machado 2019: 102–16, 126–32.

      

313

phase is marked by the presence of the court at Rome and Ravenna in the West,¹⁵ and the consolidation of Constantinople as the primary residence in the East.¹⁶ Such changes can largely be attributed to the phenomenon of child-emperor rule: in the western empire, the city of Rome became an important venue for Honorius and Valentinian III to display their piety and ceremonial-style leadership in a way that was acceptable to their military managers.¹⁷ In the eastern empire, the city of Constantinople became a true imperial residence under Arcadius and Theodosius II, which marked the transition back to the court-in-capital mode of rulership. Arcadius and Theodosius II never commanded their armies in battle, but depended on their magistri militum to undertake military responsibilities, as explored by Meaghan McEvoy in Chapter 9.¹⁸ But these emperors were not palace-bound, as envisaged by the critical stereotype of the princeps clausus discussed in Chapter 12 by Martijn Icks. Instead, they had a vital and important relationship with the city of Constantinople and its people.¹⁹ The western imperial court disappears in the late fifth century when there were no longer Roman emperors, following the deposition of the usurper Romulus Augustulus in 476 and the death of the last legitimate emperor, Iulius Nepos, in 480. Phase four therefore only concerns the East. Now there was again only one imperial court, based in one city, Constantinople.²⁰ Emperors of the late fifth and sixth centuries, up to the reign of Maurice (583–602),²¹ did not, as a general rule, campaign in person, preferring to continue to delegate those responsibilities to their generals, even if the emperors had themselves been military men prior to their elevation. The impact which this made on contemporaries is shown by the erroneous remark of John Lydus that Theodosius I had passed a law which prevented emperors from taking to the field.²²

¹⁵ There has been a tendency to assign Ravenna an outsized role in the fifth century as the emperor’s primary residence. As McCormick 2000: 136 has remarked, it ‘resembled rather a glorified military base’. See fully Gillett 2001, a seminal article establishing the importance of Rome for the last western emperors, developed further by McEvoy 2010 (on child emperors) and McEvoy 2017 (on the shadow emperors), as well as Humphries 2012, who provides a specific case study of the reign of Valentinian III in Rome (also considered by McEvoy 2010). ¹⁶ For 395 as a turning point, see Pfeilschifter 2013: 21–2; Destephen 2016: 353–4; Meier 2017b: 513–15. See further Mayer 2002: 105–74; Croke 2010; Destephen 2016: 62–107 on Constantinople and the Theodosians. ¹⁷ McEvoy 2010. ¹⁸ McCormick 1986: 47–8; McCormick 2000: 135–43. ¹⁹ See the magisterial analysis of Pfeilschifter 2013. For further examinations of the role of Arcadius and Theodosius II in Constantinople, see van Nuffelen 2012; Kelly 2013b; McEvoy 2020a. For the importance of the entrenchment of the emperor and court in Constantinople to the role of the people in politics, see Kaldellis 2015: 102–17. ²⁰ Magdalino 2011. ²¹ On the period from Arcadius to Maurice as a distinct phase in Roman imperial history, see Diefenbach 1996: 41; Meier 2017b (the latter with further sub-divisions). ²² Joh. Lyd. de mag. 2.11.4, 3.41.1. Theodosius II had planned to lead the expedition to install Valentinian III in the West, only to change plans at Thessalonica on grounds of illness. As Lee 2013a: 96 remarks, such leadership was probably similar to Claudius on his British campaign—ceremonial

314

   

The connection forged between the court and Constantinople across phases three and four can be seen in the impact of courtiers on the topography of the city. The multiple residences of the Theodosian imperial women—Placidia and Marina in the First Region; Pulcheria in the Third; Placidia, Eudocia, and Arcadia in the Tenth; and Flacilla and Pulcheria in the Eleventh—reflect the investment of the dynasty in Constantinople, rather than in any other eastern cities.²³ Houses and palaces owned by leading courtiers of the late fourth and fifth centuries, such as the magistri militum Promotus and Areobindus, the western Emperor Olybrius, and the praetorian prefects Aurelianus and Rufinus, became so well-known that their names were retained in the titles of districts or churches long after their deaths. In the case of Rufinus, his magnificent palace at Chalcedon passed into imperial hands, but retained his name.²⁴ Even the residences of court chamberlains were so well-known that their owners gave their names to entire quarters of the city, as shown by the cases of Antiochus, who served the emperors Arcadius and Theodosius II, and Narses, the cubicularius of Justin II.²⁵ Constantinople became the primary arena for political and religious ceremonial, which took place in the public spaces such as the Hebdomon parade ground and the hippodrome in the centre of the city, as explored by Christian Rollinger in Chapter 2 and Audrey Becker in Chapter 4.²⁶ Performances of imperial piety and relic translations were an increasingly important part of civic life from the fifth century onwards, a process that has Averil Cameron has termed the ‘liturgification’ of ritual.²⁷ Emperors such as Theodosius II, Marcian, and Justinian famously took it upon themselves to walk barefoot or crown-less through Constantinople in acts of extreme penitence or piety.²⁸ Under the Theodosians, relic translations became major public events which involved members of the court. In 406, the bones of St Samuel were formally received by Arcadius, the praetorian prefect, the urban prefect, and the senate, an event which functioned as a display of the piety of the emperor and his court rather than of the bishop of

rather than practical. The same caveat should apply to Marcian accompanying the expedition against the Huns in 451. He is only actually attested at Heraclea in Thrace, not leading the army in battle (Price and Gaddis 2005 nos 12, 14, 15). ²³ Matthews 2012: 86–98. ²⁴ Magdalino 2001: 57–60; Bowes 2008: 105, 107. Rufinus: PLRE I Rufinus 18; ACOec. 1.1.7, pp. 79–80 (an audience with Theodosius II at the Rufinianae). See further Janin 1964: 313–14, 317, 398–9. ²⁵ Janin 1964: 310–11, 395–6. ²⁶ For the performance of imperial ceremonial in the hippodrome, see McCormick 1986: 64–78; McCormick 2000: 158–9. ²⁷ A. M. Cameron 1987: 126; see also Janin 1966. On the rituals of the fifth century, see further Diefenbach 1996; van Nuffelen 2012; Kelly 2013b; Meier 2017b: 516–24. For liturgification in the sixth century, see A. M. Cameron 1979, which Meier 2012: 222–36 argues occurred in response to the political instability of the late fifth century. ²⁸ Theodosius II: John. Mal. 14.22; Chron. Pasch. s.a. 447. Marcian: Theod. Lect. Epit. 30 [365]. Justinian: Theoph. AM 6050.

      

315

Figure 14.1 Ivory depicting a relic translation in Constantinople, Trier Cathedral Treasury. © akg-images.

Constantinople.²⁹ We have a visual depiction of such a relic translation on the Trier Ivory, which brings to life the arrival of the bones of St Stephen (Figure 14.1).³⁰ The scene shows Theodosius II, his sister Pulcheria, and officials—all arrayed in ranks according to court protocol—receiving the relics and overseeing their placement in the new chapel that bore the saint’s name in the imperial palace.³¹ The most important ritual occasions were the transitional moments of dynastic ceremonial: imperial funerals and accessions. Here we see how both individual courtiers and the court as an institution came to assume a pre-eminent role in the public acclamation of emperors, regardless of whether these events took place in the palace, the hippodrome, the Hebdomon, or other important venues. At the proclamation of Leo I at the Hebdomon in 457, it was the court leaders (archontes) who attended alongside the soldiery and Anatolius the Patriarch, and it was the magister militum Aspar who rode with the new emperor in his carriage through the streets of the city.³² Ceremonies are an indivisible part of every court society, but their significance changes depending on individual historical circumstances. The fact that imperial and dynastic rituals came to be concentrated in

²⁹ Chron. Pasch. s. a. 406; McLynn 2004: 265–6; 2018: 12. ³⁰ Holum and Vikan 1979; Diefenbach 1996: 43–4. ³¹ Holum 1982: 103–9. He also convincingly defends the date of the ivory against attempts to place it much later in the Byzantine period. ³² Const. Porph. de cer. 1.91 (1.101); Feissel in Dagron and Flusin 2020: IV.1, 552–3.

316

   

Constantinople demonstrated, in the words of Paul Magdalino, that ‘the court was symbolically attached to the capital, and could not be detached without violating the proper, constitutional order of things’.³³ As with Rome in the early empire, the location of the court in Constantinople did not preclude journeys to other cities or residences.³⁴ But even when emperors began to take to the field again in the late sixth century, beginning with Maurice, it is telling that this did not lead to the re-establishment of a peripatetic court on the fourth-century Roman model. There were still imperial colleges composed of multiple emperors in Byzantium, but this was more about securing the succession through the appointment of co-Augusti and Caesars rather than establishing other imperial courts in different regions of the empire, much as in the Principate. One might also speculate that the cost of moving the court from Constantinople to the regions and back again or permanently setting up courts elsewhere for other members of the imperial college would have proved prohibitively expensive for the government establishment, especially as the loss of territory and corresponding tax revenue took its toll in the seventh century.³⁵ Indeed, the primacy of Constantinople would only be challenged when the emperor and his court were forced out during the Latin Empire in the thirteenth century, but even then they returned to the Constantinople after this aberrant period.³⁶ When we examine the changing relationship between courts and cities over the long sweep of Roman history in this fashion, phase two—the ‘long fourth century’ from Diocletian to Theodosius I—looks exceptional, rather than the normal, even paradigmatic, pattern of late antique rulership that it is often taken to be.³⁷ The transition between two different modes of courtly residence—from a largely itinerant court to a court-in-capital—is a feature of many monarchical societies, as scholars such as R. J. W. Evans and Jeroen Duindam have explored.³⁸ Sometimes the embedding of a court in a city occurred, as in the Roman world, when monarchs ceased to campaign in person. For example, Mehmed II ³³ Magdalino 2011: 138. ³⁴ McCormick 2000: 137–9. For Justinian’s construction of palaces on the Asian side of the Bosphorus outside the city, see Magdalino 2022: 142–3, and for the use of residences in the vicinity of Constantinople into the Middle Byzantine period, Hellenkemper 2015. ³⁵ See Haldon 2016: 17–18, 61–2, who writes that the loss of territory concentrated pathways to power in Constantinople itself, transforming the state into a ‘metropolitanocentric empire’ (p. 17). For a comparative perspective, see Elliott 1977: 179, who points out that cost was the reason why the seventeenth-century Spanish court remained largely sedentary. ³⁶ On fourteenth-century Constantinople and the court, see Macrides 2011. Geary et al. 2015: 179 describe Constantinople as ‘the most stable court in the world’ between 500 and 1500. ³⁷ For comparable remarks focused on the functioning of the Roman political process, see Pfeilschifter 2013: 21–2; Kaldellis 2015: 109–10. ³⁸ Evans 1991: 488–9; Duindam 2016: 159–68, 296. The entrenchment of the French court at Versailles is the classic example, studied by Elias 2006: 174–5. There are other examples, such as the residence of the Carolingian court at Aachen under Charlemagne (Nelson 2019: 336–45, 356–9), the settlement of the ‘Abbasid court in Sāmmarā and Baghdad in the ninth century (Kennedy 2005: 145–59), and the embedding of the Spanish Habsburg Court in Madrid in the sixteenth century (Elliott 1977: 169–70).

      

317

(1444–6, 1451–81), who captured Constantinople and made Istanbul the capital of the Ottoman Empire, was originally a vital military leader who then gradually retreated behind the new trappings of his monarchical splendour in the Topkapı palace. After the reign of Süleyman I (1520–66), Ottoman sultans stopped leading campaigns altogether, with the court near-permanently ensconced in Istanbul.³⁹ The overall trajectory of Roman imperial history is not, however, of one move from a mobile court to a permanent residential court, but of the transition from court-in-capital (Rome) to an itinerant court—or, more accurately, courts—then to another court-in-capital (Constantinople). This series of changes between different types of courtly modes finds a parallel in imperial China, where there was a mobile court and a range of capitals (with designations such as ‘Supreme’, ‘Central’, and ‘Southern’) between the tenth and fourteenth centuries .⁴⁰ This was a period of intense warfare, and the strategy ‘provided the means to establish centralized agencies in more than one locality’, which mirrors the thinking behind the proliferation of imperial residences under the Tetrarchs.⁴¹ Thereafter, China possessed a court-in-capital under the Ming dynasty from the fourteenth to the mid-seventeenth century, but when the Ming regime gave way to the Qing, the new emperors initially adopted an itinerant lifestyle which deliberately echoed the earlier movements of the court and promoted the idea of martial preparedness.⁴² The transition between different type of courtly residential styles under the Ming and Qing was thus motivated both by regime change and by evolving ideologies of rulership. This is what we see in the Roman world—rather than a gradual and inexorable move from itinerant to (largely) sedentary court—deliberate choices were made about the number of emperors (and hence courts), their location(s), and their mobility, based on both practical and ideological reasons. The different phases outlined here had an impact on the role played by the court in the selection of an emperor, especially when an emperor died without an heir or designated successor. After the murder of Caligula in 41, the Praetorians solved this problem by selecting Claudius, the nearest imperial relative. But thereafter at key moments—the suicide of Nero and the Year of the Four Emperors in 68–9; the crisis that unfolded after Domitian’s murder in 96 and the weakness of Nerva; and a similar turn of events after the murders of Commodus and Pertinax in 192–3—we see senators in Rome and senatorial generals in command of armies competing for the purple, a phase that generated the ideology that the best man (i.e. senator) was eligible to claim the throne. Yet the third century witnessed an increasing need for emperors to campaign on the frontiers, and their tendency to die in battle or to be murdered while doing so, ³⁹ For Mehmed II and Topkapı, see Necipoğlu 1991. On the move to non-campaigning sultans, see Imber 2012: 219–20. ⁴⁰ Twitchett and Tietze 1994: 79–80, 97–8; Franke 1994. There is a list of Chinese capital cities in Franke and Twitchett 1994: xxix. ⁴¹ Franke 1994: 270. ⁴² Rawski 1998: 17–23; Duindam 2016: 165.

318

   

meant that the mobile court—and particularly its military members—assumed a prominent role in selecting successors.⁴³ The emperors Macrinus (217–18), Philip (244–9), Florianus (276), and Carus (282–3) all served as praetorian prefects prior to their elevations, while Claudius II Gothicus (268–70), Aurelian (270–5), Probus (275–82), and Diocletian (284–305) ascended to the purple while they were senior military officers in the comitatus. Although there were many attempted usurpations in the provinces during the third century which did not involve the court, it was the case that imperial acclamations from within the comitatus tended to be more successful. This practice continued into the fourth century: when an emperor died without a clear heir, the members of the court could make a decision, either choosing one of its own members or summoning their preferred candidate, as with the successive accessions of Jovian and Valentinian I in 363 and 364.⁴⁴ Even dynastic succession was not assumed to be a fait accompli: when Valentinian fell ill in 367, the court seriously discussed designating a successor from outside his family. After the emperor recovered, he took the extraordinary step of making his eight-year-old son Gratian Augustus in order to signal his intention that he was not just an emperor, but the father of a new dynasty.⁴⁵ From the fifth century onwards, the court in Constantinople obtained significant influence in manner that outstripped even Rome during its own court-incapital phase. In the first and second centuries, it was acceptable for a Galba or Septimius Severus to be acclaimed in the provinces and march to Rome, or for the city to be stricken with fear that the governor of Syria might declare his hand against Nerva.⁴⁶ But the court in the fifth- and sixth-century East was not so patient: a successor was usually someone already on the scene, not a commander in the field or an imperial relative elsewhere in the empire.⁴⁷ When Theodosius II died in 450, it was not a foregone conclusion that his son-in-law Valentinian III in the West would henceforth rule both East and West, as far as the court in Constantinople and its dominant figure, the magister militum Aspar, were concerned. This resulted in the elevation of Marcian and a convenient, if religiously awkward, marriage to the deceased emperor’s virginal sister Pulcheria to ensure that there was still an incumbent on the throne in the Daphne palace.⁴⁸ Indeed, when there was a succession crisis at the death of the emperor, the court was determined to select its own acceptable candidates, as Audrey Becker has elegantly shown in Chapter 4.

⁴³ On the astonishing rate at which Roman emperors were murdered compared with other monarchs, see Kelly 2022b: 378–83. ⁴⁴ Amm. 25.5.1–5, 26.1.3–6. ⁴⁵ Amm. 27.6.1–16; Zos. 4.12; McEvoy 2013: 49–53. ⁴⁶ Plin. Ep. 9.13.11. ⁴⁷ This did not preclude military revolts, such as those under Zeno, but they tended to be unsuccessful (and still frequently launched in Constantinople). ⁴⁸ McEvoy 2019a: 121 discusses Theodosius II’s failure to provide for the eastern succession.

      

319

This sometimes resulted in relatives of the deceased emperor being passed over. When Marcian died in 457, his son-in-law Anthemius was overlooked in place of Leo I, whose elevation was by assured by Aspar. The nephews of Emperor Anastasius were also passed over on his death in 518:⁴⁹ as Becker shows, no candidates considered by the different constituencies in Constantinople were Anastasius’ relatives, and eventually the court selected Justin, the commander of the excubitores.⁵⁰ This represented the acknowledgement of the centrality of this military commander and his role at court, similar to the proclamations of praetorian prefects as emperors in the third century. Later in the sixth century, both Tiberius II Constantine and Maurice would follow Justin in rising to the purple from the post of comes excubitorum. In the seventh, Phocas ostentatiously married his only child, Domentzia, to his comes excubitorum, Priscus, to bind the commander and his forces to his regime.⁵¹ Of course, this does not mean that dynastic hopes disappeared—indeed, while Anastasius was still alive his brother Paulus and his nephew, the general Hypatius, had been publicly praised by the panegyricist Priscian.⁵² Hypatius, however, was unpopular, and at Anastasius’ death was based at Antioch, far away from Constantinople and the court.⁵³ The descendants of former dynasties were regularly talked about, and feared, as viable challengers for the throne at moments of crisis—indeed, Hypatius became a possible challenger to Justinian during the Nika riots of 532 when he was resident in the city.⁵⁴ But anyone with a dynastic connection would still need to have the consensus of the court community to become emperor. The role of the court in securing an emperor’s succession is brilliantly demonstrated by Corippus’ panegyrical poem on the accession of Justin II. He has the praepositus sacri cubiculi Callinicus proclaim to the new emperor that ‘the laws summon you, the court supports you’ (te iura vocant, te sustinet aula).⁵⁵ There were other relatives of Justinian still living, many of whom had held important military commands—including a great nephew, the general Justin⁵⁶—but it was Justin II, the curopalates, who won out.⁵⁷ Corippus observed that he had been ‘brought up in the midst of the court among the ranks of educators and faithful cohorts’ (media nutritus in aula inter alumnorum numeros coetusque fideles).⁵⁸ A similar view is found in the fragmentary panegyric of Dioscorus of Aphrodito, who

⁴⁹ See A. D. E. Cameron 1978 on the family; McEvoy 2018: 99–107 on explanations that developed for the non-dynastic succession of Justin. ⁵⁰ See Leppin 2011: 32–3, 43–8 on Justin I’s rise to power. ⁵¹ Haldon 1984: 136, 192–3, 408–9, also extending the analysis into the seventh century; Whitby 2000a: 291. On Priscus, see Parnell 2017: 137. ⁵² Priscian, Pan. 290–300. ⁵³ Croke 2007: 16. ⁵⁴ See the explorations of this theme in McEvoy 2018: 99–111; 2019a: 123–5; 2022. ⁵⁵ Coripp. Iust. 1.148; A. M. Cameron 1976: 4–5. ⁵⁶ A. M. Cameron 1976: 130; and on Justinian’s employment of his relatives as generals, Parnell 2017: 81–2, 133–5, and the principle more broadly, Whitby 2000b: 473–4. ⁵⁷ See Lin 2021: 123–9, 134–41 for Justin II’s court connections. ⁵⁸ Coripp. Iust. 4.192–3.

320

   

praised Justin II on the occasion of the reception of his image in Antinoë as ‘the youthful son of the many-sceptred palace’ (νέον υἷα πολυσκήπτρου παλλατίου).⁵⁹ Dynastic ideology was never the sole determining factor at any point in Roman history through to the High Byzantine period.⁶⁰ What these cases show is that throughout the fifth and sixth centuries, the city of Constantinople was the place where imperial, military, and court dynasties, from the Anthemii and the Ardaburii to the houses of Anastasius and Justin I, held office, intermarried, and competed for power. These connections between aristocracy, court, and imperial blood can be detected across several generations in the marriage of Placidia—the great-great-great-great-great-granddaughter of Theodosius II—to the general John Mystacon, who served Tiberius II and Maurice, as well as the marriage of Theodosius, son of Emperor Maurice, to a daughter of Germanus, of the family of Justin I.⁶¹ This meant that the usurpation of Phocas, a rank outsider, in 602 represented a real shock to Constantinopolitan elite, since he was the first emperor since the establishment of Constantinople as a capital under the Theodosians who was not a member of the imperial family or a courtier. This leads us to consider the wider relationships between emperors, courts, and the communities of the empire throughout the different phases. In phase two, from Diocletian to Theodosius, the proliferation of emperors and the selection of provincial cities as principal residences meant that there were more opportunities for elites and communities to be closer to one of several imperial courts. This could manifest itself in different ways, such as the ability of a larger number of provincials to witness ceremonial processions through civic spaces, and greater influence being invested in local authorities such as the bishop, which Fabian Schulz highlighted in Chapter 10. The benefits of proximity to the court were thus made apparent to multiple communities across the empire, in Gaul (Trier), North Italy (Milan), the Balkans (Thessalonica/Sirmium), Asia Minor (Nicomedia), and Syria (Antioch).⁶² Regional courts enabled different groups of officials, chosen by different emperors, to assume authority in different parts of the empire. We might note especially here Valentinian I’s Pannonians and Theodosius I’s Spaniards, whose influence was concentrated at the court of their particular imperial patron, as masterfully shown by John Matthews.⁶³

⁵⁹ P. Cairo Masp. 2.67183, l. 7: The use of ‘many sceptred’ as an adjective first appears in Anicia Iuliana’s dedicatory epigram for the St Polyeuctus church in Constantinople (Anth. Gr. 1.10) (Whitby 2006: 179). ⁶⁰ Dagron 2003: 14–15. Elevation to imperial status from humble background via the court did still require defence in various ways, such was the ingrained fascination with dynasty: see, for example, Agapetus, Advice to the Emperor Justinian 4 (ed. Bell 2009: 101). ⁶¹ Placidia: A. D. E. Cameron 1978: 273. Theodosius: Feissel 2010: 258–67; Lin 2021: 132. ⁶² We should also note that the presence of an imperial court in a provincial city could also have negative effects, as shown for example by the terror inflicted by the Caesar Gallus during his tenure in Antioch (Amm. 14.1.1-10, 7.1-21, 9.1-9). ⁶³ Matthews 1975.

      

321

When the Roman Empire transitioned back to the court-in-capital model, these opportunities were now concentrated in one city, Constantinople. In the fifth century and beyond, the emperor and his court evidently thought that there was less of a need to travel on a regular basis to cities such as Sirmium and Antioch, as Constantinople was positioned as a New Rome above all other cities.⁶⁴ Arcadius and Theodosius II both rarely travelled, and when they did, it was only to a handful of cities in Asia Minor.⁶⁵ Likewise, Peter the Patrician’s sixth-century discussion of the ceremonies staged to mark an emperor’s return to Constantinople suggests that journeys to Thrace, Asia, and Bithynia were the most common.⁶⁶ These restricted itineraries had the effect of altering the how provincials could access the court. Fewer progresses to fewer provinces meant fewer opportunities for elites to petition the emperor in person on his progress; instead, they had to travel to Constantinople and wait there for an audience, competing with all the other parties who descended on the city.⁶⁷ Petitioners would no longer approach the emperor in a familiar context, surrounded by their peers from the same city or region, who would also be hoping for imperial assent to their request.⁶⁸ Instead, they would find themselves at the court of Constantinople competing for an audience with strangers and rivals, much like Philo at the court of Caligula centuries before.⁶⁹ It is worth noting that this development did reintroduce some equality to the petitioning process, because there were some regions—Spain, North Africa, and Egypt, for example—which had rarely seen emperors even in the fourth and fifth centuries.⁷⁰ Visits to the court in Constantinople were never quick and straightforward. The ambassador Synesius, whose case was examined by Martijn Icks in Chapter 12, was forced to spend three years in Constantinople, from winter 397/8 to autumn 400, while lobbying for tax breaks for Cyrene.⁷¹ The situation was even worse for those who tried to rectify abusive treatment by officials. In the sixth century, Dioscorus travelled twice from his village of Aphrodito in Egypt to Constantinople ⁶⁴ On the competition between Constantinople and other major cities of the empire in Late Antiquity, see Grig 2012. ⁶⁵ Destephen 2016: 382–97. ⁶⁶ Peter the Patrician in Constantine VII, Three Treatises on Military Expeditions 690, 699 (ed. Haldon 1990b: 138–9). ⁶⁷ See Morris 2004: 128 for the general principle, and Rapp 2005: 266–7 on the large number of bishops travelling to Constantinople to present petitions. For the excitement that could be generated by a rare imperial visit to the provinces, see Nov. Theod. 23 on petitions handed to Theodosius II when he came to Heraclea (Harries 1988: 164–5). ⁶⁸ See CJ 10.48.2, in which two decurions from Antioch present themselves to Diocletian, with the other members of the local curia in attendance. ⁶⁹ See the important discussion of Andriollo 2020: 256–64 on the constraints on petitioners in the fifth century. This overall line of argument was suggested to us by reading Murphy 2016, who examines the difficulties faced by French town councils when the opportunities to petition the king on regular visits to their own cities diminished. ⁷⁰ Note, for example, Shaw 2011: 517–18 on many ecclesiastical embassies sent from North Africa to Ravenna. ⁷¹ Barnes 1986: 104–6, 111–12; Cameron and Long 1993: 91–102.

322

   

to appeal against the abuses of the local pagarch, even writing a poem in praise of an individual whom he thought could help his case at court.⁷² One interesting part of Dioscorus’ story is that while he was in Constantinople he assisted other Egyptians to write their petitions.⁷³ This example is symptomatic of a wider trend: a law of Justinian expressed concerns about the number of farmers travelling to Constantinople to complain about their overlords. The emperor ordered that most of them were to be sent back home, with only a few allowed to remain in the city to press their case.⁷⁴ Such journeys could be expensive, even ruinous. Justinian decreed that he introduced the legislation ‘so that [the farmers] do not leave their homelands and suffer hardship here and die’.⁷⁵ The relationship between Christian figures and the imperial court also changed with the court-in-capital phase of governance.⁷⁶ From the fifth century, the bishop of Constantinople assumed an importance that far outstripped all other bishops in the eastern empire,⁷⁷ which had the potential to lead to a somewhat antagonistic relationship with the court.⁷⁸ The fact that the patriarch was based in the same city as the emperor and court did not immediately lead to influence—indeed, very often the contrary was true. As with lay petitioners, bishops frequently travelled to Constantinople to petition the emperor. The string of bishops who regularly made their way to the city earned the ire of Justinian, who ordered them not to abandon their churches and spend too long time in the city at their see’s expense.⁷⁹ Often these ecclesiastical figures were pursuing agendas in opposition to the patriarch, as discussed by Daniëlle Slootjes in Chapter 11. Bishops and other ecclesiastical officials focused their energies on winning favour with different members of the court at Constantinople, ranging from imperial women to chamberlains. This was a quite different situation from that faced by Athanasius, the deposed bishop of Alexandria, in the fourth century, when there were two courts centred around the brothers Constans and Constantius II. Athanasius was accused of exploiting the multiplicity of emperors and courts by playing them against each other, as examined by Fabian Schulz in Chapter 10. Here we see how the timeless and universal element of court politics, the competition for access, could be transformed by developments rooted in the specific circumstances of one society. In the case of Rome, the changing number of imperial courts, and the return to the one court-in-capital mode of rulership in Constantinople, had important ramifications that were felt throughout the empire.

⁷² P. Cairo Masp. 1.67024 (the imperial decision Dioscorus receives from Justinian); P. Aphrod. Lit. 4.4 (poem); Bell 1944: 31–5; Ruffini 2018: 175–80. ⁷³ P. Cairo Masp. 3.67352. ⁷⁴ Just. Nov. 80.2; Fournet 2010: 248–9. ⁷⁵ Just. Nov. 80.10: ὥστε μὴ τὰς οἰκείας ἀπολιμπάνοντας πατρίδας ἐνταῦθα ταλαιπωρεῖσθαι καὶ τελευτᾶν (ed. Schoell and Kroll 1895: 396). ⁷⁶ See especially McCormick 2000: 159–60. ⁷⁷ Leppin 2018b: 57. ⁷⁸ For competition between emperor and bishops, see van Nuffelen 2012. ⁷⁹ Just. Nov. 67.3.

      

323

3. The Sun in Splendour The emperor stood at the heart of the Roman court.⁸⁰ Regardless of whether he was a small child or a venerable senator, a shy and retiring intellectual or a formidable general, there was no court without him. The character of the emperor played a major role in his attitude towards accessibility, which in turn shaped the tenor of the court society at large.⁸¹ Would he be a ruler who would listen to advice from a wide variety of sources, as Antoninus Pius is reported to have done?⁸² Or would he be controlled by a select group of intimates, as is alleged of Emperor Claudius?⁸³ The openness of the emperor had ramifications for the permeability of the court. Would he admit the general population to entreat him and hear their complaints with a just and merciful ear? Or would he be cut off from the world at large and ignore his senatorial peers and the people? Access to the emperor was always regulated by various forms of ritual; the less accessible rulers were characterized negatively by their court and by the people at large. But it is difficult to assess changes in imperial accessibility and the ceremonies that regulated it. Our written sources have their own particular aims and agendas in portraying specific emperors as hostile and isolated or welcoming and open, as explored in various chapters throughout the volume, especially those by Fabian Schulz (Chapter 10), Martijn Icks (Chapter 12), and Caillan Davenport (Chapter 13). This problem was not unique to Rome: the dichotomy between the literary depictions of the accessible and welcoming monarch and the isolated despot are a common feature of most monarchical societies.⁸⁴ These stereotypes do not take into account the realities of power. For example, Domitian was criticized for spending hours alone stabbing flies with his stylus, an act which symbolized his withdrawn and suspicious nature.⁸⁵ But perhaps he was conscientiously working through imperial paperwork, like Henry VII, who cloistered himself in his Privy Chamber in order to give his full attention to administration?⁸⁶ Moreover, relaxed imperial behaviour—such as that of Antoninus Pius, discussed by Christoph Michels in Chapter 1—could function as a strategy of domination and control just as much as strict ceremonial regulations. We would do well to remember in this context the words of Richard Wortman about Russian Tsars—regardless of their own personal style of ruling, ‘the instruments of force . . . always remained in evidence’.⁸⁷ Exceptions to the welcoming king/isolated despot discourse occur only in those regimes in which the isolation of the monarch was an accepted facet

⁸⁰ Coripp. Pan. Anast. 19 metaphorically described Emperor Justin II as a nourishing spring, ‘the mighty fountain-head of the court’ (fons maximus aulae). ⁸¹ See Starkey 1987a: 6–10 on the impact of changing English monarchs in the Tudor and Stuart periods. ⁸² M. Aur. Med. 1.16. ⁸³ Tac. Ann. 12.1, 12.3; Suet. Claud. 25.5; Dio 60(60).2.4. ⁸⁴ For example, Weiser 2003: 13, 15–18 (English monarchy); van Berkel 2019 (‘Abbasid court). ⁸⁵ Suet. Dom. 3.1. ⁸⁶ Starkey 1987b: 73–5. ⁸⁷ Wortman 2006: 3–4.

324

   

of the imperial system, as was the case of Japan during the centuries of the Shogunate (bakufu). The Japanese emperor’s isolation was designed to maintain his purity: he had no proper role in government of the empire, with such functions instead devolved to the shogun.⁸⁸ But this was certainly not the case in Rome. We therefore need to ask: which aspects of imperial accessibility and court ceremonial underwent significant changes between the Principate and Late Antiquity, and in what ways? First of all, it is important to emphasize that ritual was always a part of Roman court life. Ritual existed to regulate and give order to court society and the relationship between emperor and subject. It was the nature of ritual that changed and intensified over time.⁸⁹ Augustus’ audiences with the senate and with the people at large (the so-called promiscuae salutationes) followed certain hierarchical and organizational procedures, in the same way that there were specific rules for encountering Constantius II. But in the early empire, there were multiple ways of interacting with the emperor. There are several anecdotes about emperors speaking with petitioners at the public salutatio, one of which concerns a visitor having a private conversation with Claudius by taking him off to the side.⁹⁰ The design and layout of the palace on the Palatine after the Flavian reconstruction meant that there were multiple options for audiences and interactions. The emperors could move from room to room around the peristyle of the domus Augustana during the daily salutatio, greeting callers in pre-arranged smaller groups. Alternatively, an emperor could receive visitors in more majestic fashion, while sitting in the apse of the magnificent Aula Regia.⁹¹ In Late Antiquity, majesty was not merely an option, but the rule. The ideal of the elevated emperor, seated in the apse of a great hall, was enshrined in the architecture of Tetrarchic palaces, as explored by Verena Jaeschke and Caillan Davenport in Chapter 3. One had to be invited to pass beyond the veil, to view and greet the emperor and kiss his purple cloak. Such a ritual emphasized the asymmetry of the emperor’s relationship with his subjects.⁹² Indeed, by the High Byzantine period, the emperor kissed only one man on the lips as a sign of equality: the patriarch of Constantinople.⁹³ When Constantine built his new Daphne palace at Constantinople, it came equipped with a number of ceremonial venues for receptions and meetings, including the throne room, the Consistorium, the Hall of the Nineteen Couches, and the Augoustaios.⁹⁴ Several emperors, such as Constantine and Arcadius, commissioned their own thrones, which were still

⁸⁸ Duindam 2016: 55, 259–60. For the Tokugawa Shogunate, see Hall 1991. ⁸⁹ Matthews 1989: 248. ⁹⁰ For example, Suet. Aug. 53.2, Claud. 37.1. ⁹¹ For the different scenarios and routes, see Pflug 2014; Pflug and Wulf-Rheidt 2022: 221–5; Davenport 2022: 302–3. ⁹² Smith 2007: 215–20. ⁹³ Const. Porph. de cer. 1.14; Macrides et al. 2013: 386. ⁹⁴ Const. Porph. de cer. 1.84, 1.91 (1.93, 100); see also Coripp. Iust. 3.191–203; Dagron and Flusin 2020: V, 29, 40, 48.

      

325

used even in the tenth century.⁹⁵ The court had always been a centre of artistic magnificence—the many cameos of Julio-Claudian emperors are a testament to this⁹⁶—but we see the greater centralization of splendour over time. Emperor Leo I decreed that pearls, emeralds, and hyacinths were to be the exclusive preserve of the imperial family. Moreover, he ordered that ‘imperial ornaments ought only to be made within my court by palatine artists, and not produced indiscriminately in private houses or workshops’.⁹⁷ We might also contrast the simplicity of dress for which Antoninus Pius and other early Roman emperors were praised with the remarks of Peter Chrysologus, a fifth-century bishop of Ravenna: ‘Royal power does not allow everyday costume: imperial dignity is bestowed only with the diadem and the purple.’⁹⁸ The parameters of access to the court and the emperor changed as part of these developments. The daily ritual of courtiers paying their respects to the emperor certainly continued through Late Antiquity and into the Byzantine period.⁹⁹ However, privileges were more clearly linked to rank and hierarchy at court. In the early empire, one could be invited to the emperor’s bedroom even when an equestrian—but by the second half of the fifth century, only patricians were permitted to accompany a newly acclaimed emperor to his apartments.¹⁰⁰ Moreover, the promiscuae salutationes, in which the people could come to the court on holidays and festival days and petition the emperor, ceased in Late Antiquity. One had to be invited to meet with the emperor, as the descriptions of the receptions of Diocletian and Maximian in the Panegyrici Latini make clear.¹⁰¹ The open days of the early empire were replaced by invitations to leaders of the demes (the circus factions) to witness and acclaim set-piece ceremonial events in the palace at Constantinople, a procedure most thoroughly documented from the seventh century onwards.¹⁰² The Roman imperial court regularly functioned as a judicial arena. The emperor was supposed to be prepared to consider written and verbal pleas and petitions from subjects graciously and impartially. This use of the court as a court in the legal sense can be found in many (but not all) monarchical societies.¹⁰³ For

⁹⁵ Const. Porph. de cer. 2.15. ⁹⁶ Smith 2021. Note also from an early stage, goldsmiths (CIL VI 3951, 8741) and silversmiths (CIL VI 8727) belonging to members of the imperial family; later in the third century, a freedman praepositus opificibus domus Augustianae (CIL VI 8648). ⁹⁷ CJ 11.12.1.2: ornamenta enim regia intra aulam meam fieri a palatinis artificibus debent, non passim in privatis domibus vel officiis parari. For goldsmiths and silversmiths as court officials within the department of the sacrae largitiones, see CJ 12.23.7.8, 12.23.7.18. The production of imperial art as gifts in Late Antiquity, a parallel to this phenomenon, is nicely encapsulated by MacMullen 1962. ⁹⁸ Peter Chrysologus, Sermons 23 (PL 52: 265): plebeium cultum regia non admittit potestas, augustus honor non nisi diademate confertur et purpura. ⁹⁹ Featherstone 2006: 54–5. ¹⁰⁰ Contrast Plin. Ep. 3.5 with Const. Porph. de cer. 1.91 (1.100). ¹⁰¹ Pan. Lat. 11(3).11.1–3. ¹⁰² Magdalino 2015: 170–5. ¹⁰³ For example, Geary et al. 2015: 186 note that early Indian royal court did not usually have a judicial function.

326

   

example, the court of the Early and Middle ‘Abbasid Caliphate shares many similarities with Rome of the Principate, since the caliph was supposed to hear cases in person at the maẓālim, which was technically open to all comers to press their case.¹⁰⁴ Medieval French kings received petitions and pardoned criminals as part of their formal entrance into towns, in emulation of the just and pious Louis IX.¹⁰⁵ There were changes, however, in accessibility over time both in Rome and in other societies. In Late Antiquity, officials known as referendarii are attested as intermediaries for the presentation of petitions at court—there was no chance to take the emperor to the side for a private conversation, as under Claudius.¹⁰⁶ In the reign of Leo I, people left petitions at the column topped by the emperor’s statue, where they would be collected by palace officials and given to the emperor.¹⁰⁷ Praetorian prefects sorted through petitions and composed drafts for the emperor’s consideration, limiting genuine discussion in the consistorium.¹⁰⁸ An emperor who did not make himself available to his subjects could even be praised for his inaccessibility, as Evagrius did of Emperor Maurice.¹⁰⁹ Islamic and Christian societies that had once prided themselves on delivering justice in an accessible manner likewise began to institute greater restrictions over time. In the reign of Mehmed II, the Ottoman court sought to control carefully the act of petitioning the sultan. A special ‘Chamber of Petitions’ was built deep inside the palace in Istanbul, and it was there that high-ranking viziers, army judges, and finance offices were entitled to present petitions to the sultan on four days each week.¹¹⁰ Likewise, French monarchs increasingly tried to distance themselves during ceremonial entrances to cities, making themselves visible to the townspeople and councillors, but limiting opportunities for genuine interaction.¹¹¹ The changes in petitioning procedures, particularly the French example noted above, illustrate an important point: there was a significant difference between a monarch’s visibility and his accessibility. Meeting the emperor in a formal situation at court was not enough to exercise real influence. The case of Synesius shows the difficulties involved: he had come to Constantinople to present the gift of aurum coronarium (‘crown gold’) from the cities of the Pentapolis to Emperor Arcadius in the winter of 397/8. Synesius would have delivered his speech, a

¹⁰⁴ Kennedy 2005: 141 The same situation prevailed at the court of the Ummayad rulers in Spain in the eighth and ninth centuries (Müller 2011) and with the Mamluk sultans of Egypt, who heard complaints in the ‘house of justice’ (dār al-’adl) twice per week (Fuess 2011: 156–60). ¹⁰⁵ Murphy 2016: 45–6. ¹⁰⁶ The referendarii first appear in the fifth century (Bury 1910: 25–7). By the sixth century, the emperor’s wife also had a referendarius (Const. Porph. de cer. 1.86 [1.95]). See Macrides 2004: 357–8; Morris 2004 on the office of ‘master of petitions’, attested first in the seventh century, and his increasing status and powers over time. However, as Macrides 2004: 356 points out, no prescribed ceremonial format for the presentation of petitions was ever developed, even in the later Byzantine period. For the limited access of people to present petitions to the emperor in High Byzantium, see Magdalino 2015: 169–70. ¹⁰⁷ Patr. Const. 2.31; Bauer 1996: 216–17. ¹⁰⁸ Jones 1964: 336–7. ¹⁰⁹ Evagr. HE 5.19. ¹¹⁰ Necipoğlu 1991: 20–1, 96–110. ¹¹¹ Murphy 2016: 47–51.

      

327

stephanotikos logos, at court when offering the gold.¹¹² But for whatever reason, his request for tax relief for his patria was not granted by Arcadius on that occasion, and he needed to lobby prominent courtiers to achieve his ends. The same could be true of religious petitioners. Indeed, although Christian emperors were supposed to be welcoming to all holy men—such as the monks who approached Justinian, as Caillan Davenport discussed in Chapter 13—the reality was quite different. Justinian himself issued a law that monks and bishops alike should not travel to Constantinople without permission from their own patriarchs.¹¹³ Frustrations at the accessibility of emperors manifested themselves in the hagiographical conceit that divine help was needed for a righteous bishop to gain access to the emperor’s ear. Martin of Tours had God throw open the doors to Valentinian I’s palace in Trier, while Zoora and God ensured that Justinian would become very disfigured. We do not wish to suggest that the early Roman emperors could always be reached and spoken to easily—this was plainly not the case, since audiences were still regulated in various ways. Nor were public audiences the way to gain any sort of intimacy with the emperor, to return to David Starkey’s useful concept. But the potential for interaction between emperor and subject within the framework of court ceremonial was certainly greater in the Principate than in Late Antiquity. Roman emperors appeared in public for numerous events ranging from coronations and relic processions to feast days and games. The south-eastern face of the base of the Theodosian Obelisk in Constantinople shows the imperial family and their court watching the spectacles in the hippodrome (Figure 14.2). This was a strategy to make the emperors appear accessible to the people, even though accessibility in public spaces was carefully regulated and limited.¹¹⁴ The use of guard units increased the spectacle of parades, as noted by Christian Rollinger in Chapter 2, but at the same time their presence limited genuine interaction with the emperor and ensured that dissent could be suppressed. The strategy of being publicly visible but not very accessible was subsequently borrowed by the pre-Petrine Russian Tsars, who were heavily influenced by the Byzantine conception of rulership. They held magnificent public processions on religious holidays, but largely remained cloistered and inaccessible in their palaces or cathedrals.¹¹⁵ Similar techniques were adopted by other monarchs from the Habsburg Spanish kings to the Islamic Fatimids.¹¹⁶ In Constantinople, distance between the emperor and the people was maintained even at events that featured

¹¹² Barnes 1986: 104–6; Cameron and Long 1993: 91–3. ¹¹³ Just. Nov. 86.8. ¹¹⁴ For a later depiction of the imperial box on a monument of Porphyrius the charioteer, featuring Emperor Anastasius, his wife Ariadne, and their courtiers, see A. D. E. Cameron 1973: 49–55, with Plate 7. ¹¹⁵ Wortman 2006: 15–18. ¹¹⁶ Habsburgs: Elliott 1977: 169. Fatimids: Sanders 1994.

328

   

Figure 14.2 The south-eastern face of the base of the Theodosian Obelisk, Constantinople. Photo by Meaghan McEvoy.

the performance of imperial humilitas and thus abnegated the pomp of most imperial ceremonial.¹¹⁷ When an emperor did appear in public, people would attempt to use his visibility to try to gain access, despite the guards and officials surrounding him. If a petitioner tried the direct approach, and stopped the emperor or a member of his family during public processions, they played a risky game, as Peter van Nuffelen has shown: their request could be granted or they could be publicly humiliated.¹¹⁸ The bishop Athanasius, who confronted Constantine I while he was riding into Constantinople, was at first not even recognized by the emperor and almost dismissed before he had an opportunity to press his case.¹¹⁹ The seemingly impromptu approach made by Porphyry of Gaza to Emperor Arcadius at his son’s baptism was actually stage-managed by the Augusta Eudoxia.¹²⁰ It is unsurprising that some emperors developed a distaste for processions, for fear of how they ¹¹⁷ Van Nuffelen 2012: 186–8; Kelly 2013b: esp. 228. Cf. Diefenbach 1996: 49–52. Kelly 2013b: 238–42 emphasizes that displays of imperial piety which might appear improvised were actually planned out. ¹¹⁸ Van Nuffelen 2010a: 235–8. ¹¹⁹ Soc. 1.34; Soz. 2.38. ¹²⁰ On control, see van Nuffelen 2012: 191–2, and for the baptism incident, see Pfeilschifter 2013: 92–5; McLynn 2018: 14–15.

      

329

might be ambushed or manipulated.¹²¹ The same delicacy is true of appeals made at the hippodrome.¹²² During one set of circus games, members of the Greens asked Emperor Anastasius to release some recently arrested citizens from prison: he responded by setting his guards on them, which resulted in a riot and fire that spread through the hippodrome and the city centre.¹²³ These developments in ceremonial and accessibility were accompanied by religious changes which emphasized the emperor’s Christian faith. When Leo I was made emperor at the Hebdomon in 457, the assembled audience acclaimed him as chosen by God. He then proceeded to the Papilion Church and then the Church of Saint John the Baptist at the Hebdomon, where he placed his crown on the altar. He then came to the Palace of the Helenianai inside the Theodosian walls, where he was joined by a court official (one of the vestitores) whose task it was to carry a cross before his carriage in the procession through Constantinople. After entering the city and meeting the senate in the Forum of Constantine, he proceeded into the Church of Hagia Sophia, where Leo again placed his crown on the altar and took communion.¹²⁴ Prayers became a commonplace part of accession ceremonial. During the imperial proclamation of Leo II in the hippodrome, the patriarch stood next to the boy and said a prayer, which was followed by the assembled crowd all saying ‘Amen’.¹²⁵ After Anastasius’ elevation on a shield, representing his military authority, the patriarch prayed before putting the chlamys and crown on the emperor.¹²⁶ Prayers made personally by the emperors and their wives emphasized their own piety and suitability to rule. After the death of Justinian, his successor Justin II stopped to pray in the chapel of Archangel Michael in the palace, while his wife Sophia prayed to the Virgin at the Church of the Chalkoprateia.¹²⁷ Displays of imperial religious devotion extended across the entire city of Constantinople. Christianity also made a significant impact on the rituals of court life, a process that began as early as Constantine, who led his Christian courtiers in prayers.¹²⁸ Ausonius, poet and tutor to Valentinian I’s son Gratian at the court in Trier, composed Easter Verses to be declaimed in the emperor’s presence.¹²⁹ The courtly context of the poem is demonstrated by its climax, in which Ausonius compared the three Augusti—Valentinian I, his brother Valens, and his son Gratian—to the Holy Trinity.¹³⁰ Our evidence for daily ritual increases in the fifth century with the

¹²¹ Macrides et al. 2013: 399. ¹²² A. D. E. Cameron 1976: 162–73. ¹²³ Joh. Mal. 16.5; Chron. Pasch. s. a. 498. ¹²⁴ Const. Porph. de cer. 1.91 (1.100). ¹²⁵ Const. Porph. de cer. 1.94 (1.103). ¹²⁶ Const. Porph. de cer. 1.92 (1.101). ¹²⁷ Coripp. Iust. 2.4–71; A. M. Cameron 1976: 149, 152. ¹²⁸ Eus. V. Const. 4.17; McLynn 2004: 236–8. There were certainly no churches in the palace under Constantine (Leppin 2015b: 134–5), though this does not mean that religious services could not have been performed in other venues (Bowes 2008: 110, citing Eus. V. Const. 4.22). ¹²⁹ McLynn 2004: 251. ¹³⁰ Aus. Versus Paschales 24–31; Lenski 2002: 32; see Hunt 2007: 76–7 on the Christian atmosphere of his court.

330

   

embedding of the court in Constantinople. Theodosius II engaged in singing responsive psalms with his sisters each morning, perhaps in one of the chapels which were built in the palace at this time.¹³¹ The Syriac Life of Peter the Iberian, who spent time as a hostage at Theodosius II’s court, records daily prayers, fasting, and night services; the commitment to piety was shared by the emperor and his wife Eudocia alike.¹³² Members of the court community were expected to take part in major religious festivals. When Theodosius II’s magister officiorum Paulinus was suffering from a malady of his foot, he had to ask that he be excused from the court’s celebration of Epiphany.¹³³ In September 428, Theodosius II even hosted an event of the memory of John Chrysostom ‘at court’ (apud comitatum). This was to become an annual commemoration, repeated year after year in the palace at Constantinople.¹³⁴ We are much less well informed about the built environment in which these Christian ceremonies took place, due to the paucity of archaeological evidence. However, literary references show that over time emperors did take steps to ensure that the art and décor of their residences in Constantinople reflected their Christian faith. Emperor Anastasius recruited a Manichaean artist to paint religious images in the Palace of the Helenianai and in the Church of Saint Stephen at Aurelianae, while Justin II’s Chrysotriklinos, or ‘Golden Triclinium’, had an icon of Christ enthroned placed above the emperor’s own seat.¹³⁵ All these developments—state ceremonial, daily court rituals, and the Christian palace art—laid the groundwork for the world of tenth-century ritual brought to life in the Book of Ceremonies, where ‘[i]n the palace and the city, imperial ceremony is also religious ceremony’.¹³⁶ Handbooks, manuals, and compilations of court ceremonial, precedence, and routine like the Book of Ceremonies are found in many monarchical societies.¹³⁷ However, they are not a phenomenon attested in the Roman Principate. This can be explained by the fact that court rituals evolved from Republican elite behaviour, and were deliberately portrayed as such. It would have gone against the ‘traditionalizing’ mode of imperial ceremonial for Augustus or any of his immediate successors to record or prescribe standards of ceremonial protocol.¹³⁸ Thanks to the Book of Ceremonies, which quotes excerpts verbatim, we know that the first effort to record Roman state and court ceremonial took place in Constantinople in ¹³¹ Soz. 7.22. ¹³² John Rufus, Life of Peter the Iberian 25 (ed. Horn and Phenix Jr 2008: 33); see also Zach HE. 3.158 (ed. Greatrex et al. 2011: 117). ¹³³ Chron. Pasch. s.a. 444. For Epiphany celebrations, see also John Rufus, Life of Peter the Iberian 27 (ed. Horn and Phenix Jr 2008: 35). ¹³⁴ Marcell. com. s.a. 428; Croke 1995: 78. ¹³⁵ Theoph. AM 5999. Lavin 1962: 22–3; A. M. Cameron 1979: 17. ¹³⁶ A. M. Cameron 1987: 113. ¹³⁷ Vale 2001: 42–56, 201–6 (Europe); Barjamovic 2011: 40 (Assyria); Kropp 2011: 118–25 (Ethiopia). ¹³⁸ Davenport 2022: 293.

      

331

the sixth century.¹³⁹ This was the Treatise on Court Ceremonial in at least nine books compiled by the magister officiorum Peter the Patrician.¹⁴⁰ Peter’s work included sections on the procedures for the appointment of court officials, the reception of portraits of western emperors at court, the admission of Persian ambassadors, descriptions of coronations from Leo I to Justinian, and even an account of the procedures for imperial inspections of the granaries. Constantine VII’s Treatises on Military Expeditions includes further excerpts of Peter’s work, including the rituals performed when an emperor returned to Constantinople from a military expedition.¹⁴¹ Peter the Patrician’s slightly younger contemporary, John Lydus, followed in his footsteps to a certain degree by recording in his On the Magistrates details of procedures, costume, insignia of state, and court officials.¹⁴² This is not to say that they were no rules or protocols before the sixth century— there clearly were—but the entrenchment of emperor and court in Constantinople, and the centrality of its ceremonies to political, religious, and civic life, provided the impetus for their recording. The transformation and intensification of ceremonial over time is a development Rome shares with other monarchical societies. If we look at the continental European world, two particular trends emerge in the transition from the Medieval to the Early Modern period. The first is the emergence of the royal court as the pre-eminent court surpassing all other princely and aristocratic courts.¹⁴³ The second is the development of court ceremonial as a ‘platform for the cult of majesty’.¹⁴⁴ This was accomplished by instituting rules, strategies, and pomp to distance the ruler from his courtiers and subjects, such as limiting who could access specific rooms and regulating who could speak to the monarch and when. In France, for example, although the king was always supposed to be relatively accessible, there is a clear transition from the thirteenth-century Louis IX, famed for interacting with the common people, to the fifteenth- and sixteenth-century monarchs who experimented with greater restrictiveness and using ceremonial as a distancing strategy. Then we move to the opulent world of the Versailles of Louis XIV, ‘the Sun King’.¹⁴⁵ For Louis XIV, ceremonial was not merely a way to regulate the court community, but formed part of a wider ‘image-making

¹³⁹ A. D. E. Cameron 1976: 249–50. ¹⁴⁰ Const. Porph. de cer. 1.84–95 (1.93–104), 2.51. Laniado 1997: 410 suggests that Peter compiled both pre-existing material (the coronation of Leo I occurred well before his own birth, for example) and wrote new sections for this work. The earlier material may well have come from the now-lost history of Candidus, who began with Leo’s elevation (frag. 1, ed. Blockley 1983: 464–5). ¹⁴¹ Haldon 1990b: 136–41 (text and translation), 259–68 (commentary); Sode 2011: 172–3; Feissel in Dagron and Flussin 2020: I, 66*–67*. ¹⁴² For example, Joh. Lyd. de mag. 1.16.3, 2.3.8, 2.11.1, 2.13.1–14.4, 2.26.1, 3.41.1, 3.68.1–5. ¹⁴³ Muir 1997: 257. ¹⁴⁴ Asch 1991: 5–11; see also Stradling 1988: 333–4; Weiser 2003: 5, 8–12. ¹⁴⁵ For Louis IX, see Gaposchkin 2008: 42–3, 46–7, 106–7, 121–2. For the traditional openness of the French court, see Duindam 2003: 161; Spangler 2016: 158. Changes in ceremonial emphasizing distance: Asch 2016: 186–7; Murphy 2016; Spangler 2016. Overall trends: Asch 1991: 28; Muir 1997: 257. French court as a centre of sophistication: Schalk 1991: 248–51.

332

   

of a king’.¹⁴⁶ In the sixteenth and seventeenth centuries, monarchies in Early Modern Denmark, Sweden, and Russia deliberately looked to the French court for inspiration on how to use similar ceremonies in their own countries.¹⁴⁷ The sixteenth-century Swedish Court Master was required to ‘have learned how a court should function at foreign princely courts’ as part of the qualifications for his position.¹⁴⁸ Similar developments can be witnessed in the Islamic world, in which pomp and splendour were initially resisted as unbecoming for a Muslim ruler, since the Prophet Muhammad had reportedly been a model of accessibility.¹⁴⁹ But over the : centuries, elaborate ceremonial procedures and new forms of ritual crystalized around Muhammad’s successors. The caliphs and sultans of the ‘Abbasid, : Fatimid, and Ottoman courts, located in Baghdad, Cairo, and Istanbul, respectively, were inspired both by Sassanian ritual and by contact with the Byzantine Empire and its ceremonial practices.¹⁵⁰ The Topkapı palace, built in Istanbul by Mehmed II, was designed to articulate his claim to domination across Europe and Asia in the manner of Byzantine emperors, and to give him privacy and seclusion as befitted his sanctity.¹⁵¹ Michael Talbot has shown how the rituals for the reception of foreign ambassadors in Topkapı took them through a series of stages designed to embody Ottoman dominance, culminating in the sight of the monarch himself.¹⁵² The increasing emphasis on the sacredness of the royal person meant that the sultan ceased to have a participatory role in audiences. Indeed, in the sixteenth century, Murad III was described as ‘a lifeless paper cutout sitting like an idol’ by the German envoy Salomon Schweigger.¹⁵³ The similarities between the evolution of these monarchical courts and the Roman court are obvious, from the images of the statue-like ruler (shades of Ammianus’ description of Constantius II, examined by Christian Rollinger in Chapter 2) and the use of court spaces to articulate ideology (as explored in Verena Jaeschke and Caillan Davenport’s discussion of Tetrarchic palaces in Chapter 3) to the distancing between king and subject (addressed by both Fabian Schulz in Chapter 10 and Caillan Davenport in Chapter 13). This suggests that long-term ritual change at monarchical courts was driven by an impulse to glorify and sanctify rulers through space and ritual. But this impulse did not always come from the monarchs themselves, as shown by Christoph Michels’ discussion in Chapter 1 of the flattery directed by Fronto at Antoninus Pius. The involvement of courtiers in shaping the glorification of the king finds parallels in ¹⁴⁶ Burke 1992. ¹⁴⁷ Olden-Jørgensen 2002 (Denmark); Wortman 2006: 27–9 (Russia); Persson 2015 (Sweden). ¹⁴⁸ Persson 2015: 123–6. ¹⁴⁹ Cook 2011. ¹⁵⁰ ‘Abbasids: El Cheikh 2004: 154–5; El Cheikh 2014; Marsham 2009: 185, 189, 237, 257, 313; Fuess and Hartung 2011: 8. Fatimids: Canard 1951; Sanders 1994: 6–7. Ottomans: Necipoğlu 1991; Talbot 2016. ¹⁵¹ Necipoğlu 1991: 4–30, 58–9, 64–9, 251–2. ¹⁵² Talbot 2016. ¹⁵³ Necipoğlu 1991: 101–3.

      

333

other societies. The fifteenth-century Castilians Diego de Valera and Enrique de Villena, for example, provided their monarchs with suggestions about coronation ceremonies and dining rituals.¹⁵⁴ Moreover, pomp and ceremonial were not regarded as negative in Late Antiquity—order and ritual were celebrated as an intrinsic part of the imperial office, as shown by Pulcheria’s education of her brother Theodosius II in behaviour and deportment.¹⁵⁵ Emperors who did not uphold proper standards were criticized in private and public.¹⁵⁶ The army officer Phocas, who seized power from Maurice in 602, was ridiculed by chants in the hippodrome for his unfamiliarity with court ritual.¹⁵⁷ As noted earlier in this chapter, the intensification of ceremonial and the diminishing accessibility of the monarch have often gone hand-in-hand with the embedding of the court in a capital city and the ruler ceasing to lead his armies in battle. Child-rulership was sometimes a factor in this. The cloistered rule of the ‘Abbasid caliphs from the tenth century onwards can be attributed to the appointment in 908 of a thirteen-year-old caliph, al-Muqtadir, which resulted in the handing of real power to the vizier.¹⁵⁸ In Rome, there were restrictions on imperial accessibility prior to the age of the child emperors, but sustained minority rule certainly increased the difficulties of interacting with and persuading the emperor. We need to bear in mind, however, that the young ruler might not actually be the individual that a courtier, visitor, or petitioner needed to convince.¹⁵⁹ This is an issue to which we will return later in this chapter. It would actually be unexpected to find any monarchical court in the premodern age in which the level of ceremonial was systematically reduced over the centuries as part of efforts to close (but not remove) the distance between the monarch and his subjects. Even monarchies that underwent modernization in the nineteenth and twentieth centuries decided to adapt ceremonies rather than remove them or reduce them significantly. Indeed, the coming of modernity actually led to an increase in public pageantry outside the palace in Britain.¹⁶⁰ But we need to bear in mind that the intensification of court ceremonial was not always a linear and straightforward process. For example, Charles the Bold, Duke of Burgundy, instituted a new formal public audience with compulsory attendance for all courtiers, but this proved to be unpopular and fell into abeyance later in his reign.¹⁶¹ When Henry III of France erected a barrier around himself and his closest confidants at dinner, there was so much opposition that these regulations ¹⁵⁴ Nader 1988: 299–300. ¹⁵⁵ Soz. 9.1.7–8. ¹⁵⁶ Coripp. Iust. 3.156–90. ¹⁵⁷ Haldon 2002: 64–5. For criticism of later Byzantine emperors who did not uphold ceremonial standards, see Macrides et al. 2013: 1–2. ¹⁵⁸ Kennedy 2016: 161–71. ¹⁵⁹ Note a law of 408, when Theodosius II was seven years old, ordering the praetorian prefect Anthemius to consider requests from provincial embassies before referring any to the emperor (CTh. 12.12.14). ¹⁶⁰ Cannadine 1993. See also Fujitani 1996: 42–55 on a similar transformation in Japan. ¹⁶¹ Paravicini 1991: 72–3, 80.

334

   

soon had to be relaxed.¹⁶² It should be pointed out that these incidents did not change the overall importance of ceremonial in the court of either Burgundy or France, but they show that the process was marked by considerable negotiation between monarch and courtiers. There were some Roman emperors who attempted to reverse the trend of increasing ritual and diminished accessibility, even in Late Antiquity. Emperor Julian was determined to reduce the level of splendour at the imperial court and he dismissed officials whom he regarded as responsible for manipulating his predecessor, Constantius II.¹⁶³ Julian himself wished to behave in the manner of a civilis princeps, a move that proved to be very controversial.¹⁶⁴ It was popular with senators who regarded civilitas as a key imperial virtue, as Martijn Icks explored in Chapter 12. But such views clashed with the opinions of courtiers who were accustomed to different, and more reserved, standards of imperial dignity.¹⁶⁵ We should not forget that court officials were invested in the maintenance of ritual because it acted as a mechanism which allowed them to exert control over outsiders.¹⁶⁶ Parallels to this can be found in other societies: in Western Han China, for example, court ceremonial was refined and reworked by officials not only with the aim of emphasizing the emperor’s majesty, but also of enhancing their own status and prestige.¹⁶⁷ Julian’s efforts did not alter the intensification of ceremonial over la longue durée because they were not supported by subsequent emperors or the majority of their courtiers.¹⁶⁸ The adaptability of ceremonies to serve the needs of emperors is demonstrated by Peter the Patrician’s accounts of imperial acclamations in the fifth and sixth centuries. Peter did not write his Treatise on Court Ceremonial to codify rules: he wanted to present emperors with a series of choices, rather than prescriptions.¹⁶⁹ This is why he decided to record the procedures for acclamations both at both the Hebdomon and in the hippodrome.¹⁷⁰ The movement to the hippodrome for the crowning of Leo II was, at the time, not intended to signal a paradigm shift in imperial ceremonial but reflected the fact that Leo I was old and ill and was reluctant to travel far.¹⁷¹ Infirmity probably also explains why Justin I elevated ¹⁶² Duindam 1995: 56, 111–12, 121; Knecht 2014: 195; Asch 2016: 186–7. ¹⁶³ Lib. Or. 18.130–45; Soc. 3.1.53; Amm. 22.4.1–5, 9–10. ¹⁶⁴ Lib. Or. 1.129; Matthews 1989: 236; Pazdernik 2010: 79. ¹⁶⁵ Amm. 22.7.1–4. ¹⁶⁶ On court officials valuing precedence and hierarchy, see McCormick 2000: 156–7, and Pazdernik 2010: 65 on the veneration due to them. ¹⁶⁷ Habberstad 2017: 80–4. ¹⁶⁸ On political culture under the Valentinians, see Drijvers 2012. ¹⁶⁹ Const. Porph. de cer. 1.91 (1.100); Dagron 2003: 59–60; Pfeilschifter 2013: 90. ¹⁷⁰ This movement from the Hebdomon (where the elevations of Valens, Arcadius, Theodosius II, and Marcian took place) to the hippodrome is explicitly stated in Const. Porph. de cer. 1.91 (1.100). This was the location for the coronation of Leo II in 473 (1.94 [1.103]), Zeno in 474 (Vict. Ton. s. a. 474; Theod. Lect. Epit. 64 [400]; Theophanes AM 5966) and Anastasius in 491 (1.92 [1.101]). Note, however, that the usurper Basiliscus, who is omitted by Peter, was elevated at the Hebdomon (Theod. Lect. Epit. 66 [402]). ¹⁷¹ On coronations as a ‘matter . . . of circumstances’, see Dagron 2003: 65; for Leo II, see McEvoy 2019b: 200.

      

335

Justinian in 527 at the raised tribunal, the Delphax, before the Hall of the Nineteen Couches in the palace.¹⁷² The palace was subsequently used for acclamations from Justinian to Tiberius II Constantine, but then Maurice was crowned at the Hebdomon in 582 precisely because since the sickly Tiberius II Constantine was living in the palace there at the time.¹⁷³ When we examine our period in hindsight, it is clear that Roman state and court ceremonial followed a trajectory, but its route was shaped by manipulation, negotiation, and happenstance. One phenomenon that has often gone hand-in-hand with the intensification of ceremonial and reduced monarchical accessibility is the transfer of some or all of the ruler’s authority to a court official. European monarchical courts of the sixteenth and seventeenth centuries were often dominated by the all-powerful ‘minister-favourite’, to use the term coined by A. Lloyd Moote.¹⁷⁴ Court favourites served as a barrier between the monarch and the court at large, ‘a mechanism of mediation’, which ensured that the ruler was protected from opprobrium or resentment, which was directed at the favourite instead.¹⁷⁵ Minister-favourites were special cases, who operated like ‘surrogate sovereigns’, as L. W. B. Brockliss has put it. These were men like Cardinal Wolsey, the alter rex (‘second king’) during the first half of Henry VIII’s reign, Robert Cecil under Elizabeth I, and Cardinal Richelieu in the French regime of Louis XIII.¹⁷⁶ Brockliss has argued that minister-favourites shared the following characteristics: they were promoted by the ruler because of their organizational abilities; they came to be the centre of patronage and royal administration at court; and they exercised complete control independent of their monarch.¹⁷⁷ There are cases of praetorian prefects who functioned like minister-favourites in the Roman Principate, such as Sejanus under Tiberius and Plautianus in the reign of Septimius Severus.¹⁷⁸ The appearance of such all-powerful prefects was sporadic, however, and they did not form part of a distinct phenomenon as in sixteenth- and seventeenth-century Europe. There are some grounds for detecting similarities with the magistri militum discussed by Meaghan McEvoy in Chapter 8. The generals fit many of Brockliss’ qualifications: they acted as the centre of patronage and government at court and controlled affairs largely independent of the emperor. But there is a key difference, in that generals such as Stilicho, Aetius, Ricimer, and Aspar were often appointed by earlier emperors rather than the one(s) they ended up controlling, or they sometimes came to power against an emperor’s will, as in the case of Aetius. There is, therefore, a stark ¹⁷² Const. Porph. de cer. 1.95 (1.104); Feissel in Dagron and Flusin 2020: IV.1, 610–11. ¹⁷³ Joh. Eph. HE 5.13; Chron. Pasch. s. a. 582; Theoph. AM 6074. ¹⁷⁴ Moote 1989: 89, 104; Moote 1992. For this phenomenon, see Bérenger 1974 and the papers collected in Elliott and Brockliss 1999, especially Thompson 1999. ¹⁷⁵ Evans 1991: 487. ¹⁷⁶ For Wolsey, see Starkey 1987b: 109. ¹⁷⁷ Brockliss 1999: 280–5 (quotation on 283); see also Thompson 1999: 14–15. ¹⁷⁸ Sejanus was a popular paradigm for criticism of the ‘minister-favourite’ in the seventeenth century (Elliott 1999: 2).

336

   

contrast with the conscientious minister who worked his way up the ranks because he had gained the trust of a particular monarch, as in Early Modern Europe. Moreover, the generals’ power flowed from their control not only of the government, but also of the army, which allowed them to dominate the emperor. The magistri militum of the late fourth to late fifth century were actually closer in nature to ‘diarchs’, or officials who divide monarchical responsibilities with the ruler. This is what Alan Strathern has termed ‘the ritualisation trap and the diarchical escape’: the king becomes such a remote, divinized, or ceremonialized figure that his actual powers are instead exercised by a colleague in a form of ‘diarchy’.¹⁷⁹ The most famous example of this occurred in Japan during the period of bakufu or ‘military government’ of the shoguns (the Kamakura, Muromachi, and Tokugawa Shogunates), when the emperor had an important ceremonial role, but true authority lay with the shogun.¹⁸⁰ This was a mutually beneficial relationship, as the emperor’s support was necessary for the shogun’s legitimacy and it was the emperor who allowed a shogun to become divine after death.¹⁸¹ The emperor remained enormously significant as the centre of religious authority throughout Japanese history, which gave him a place and purpose within the framework of the Tokugawa Shogunate.¹⁸² A similar shift in power took place in the Islamic ‘Abbasid Caliphate. In 936, military and economic pressures forced the caliph al-Rādī : to appoint one of his military governors, Ibn Rā’iq, as amīr alumarāʾ (‘commander of commanders’) with full responsibility for all civilian and military affairs.¹⁸³ This position subsequently became institutionalized under the Buyid dynasty, whose members controlled the emirate.¹⁸⁴ Similar to the Japanese emperor, the ‘Abbasid caliphs retained considerable religious authority.¹⁸⁵ One might also note in this context the rise of the ‘mayor of the palace’ (maior domus) in the Merovingian kingdom of Austrasia; these officials came to exercise effective power in place of the kings.¹⁸⁶ The mayoralty became the hereditary office of the Pippinid family, who eventually (unlike the Japanese shoguns) claimed the title of king for themselves.¹⁸⁷ These societies all witnessed the institutionalization and acceptance of the separation of ceremonial rulership and the administration of government. There are certainly parallels here to Meaghan McEvoy’s formulation of the age of the child emperors and their military managers.¹⁸⁸ However, as she argues in this volume, the age of the magister militum lasted only about a century, coming to an end in the 470s. The reason for the change in the western empire is obvious: the disintegration of the West as a coherent political unit. For the eastern empire, several causes can be proposed. Firstly, the demise of the Theodosian ruling house ¹⁷⁹ ¹⁸² ¹⁸³ ¹⁸⁶ ¹⁸⁸

Strathern 2019: 186–9. ¹⁸⁰ Mass 1990; Hall 1990; 1991. ¹⁸¹ Hall 1991: 149, 165. Wakabayashi 1991: 30–1; Kōjirō and Bock 1993: 226–9; Hurst 1999: 580. Kennedy 2016: 168–9. ¹⁸⁴ Kennedy 2016: 187. ¹⁸⁵ Hanne 2007. James 1972: 145–56; Hen 2020: 226–8. ¹⁸⁷ Nelson 2019: 43–50, 54–5, 59–60, 69–72. McEvoy 2013.

      

337

in the 450s opened up new possibilities for the succession, as did the death in 471 of the long-serving magister militum Aspar, who had in 450 and 457 played a pivotal role in the selection of emperors. Secondly, there was a difference in the aims of men who were appointed magistri militum. There was a movement away from loyalist generals, who controlled government in a dominant ‘partnership’ with the emperor, to magistri who were prepared to launch major rebellions against emperors, raise usurpers, or attempt to seize the purple for themselves. Thirdly, a restructure of the army took place in the sixth century, which created more magistri militum, thereby reducing the potential for any one of these generals to assume a controlling position. Fourthly, there was a change in the identity of the men becoming emperors, in that rulers such as Leo I and Zeno came from a military background themselves, and they already had strong links with, and support from, the army. However, these military men did not usually take to the field and campaign once they had become emperor. Instead, they settled down in Constantinople and played the part of an urban, ceremonial emperor on the model established by Arcadius. It is only in the late sixth century that emperors themselves began to lead their armies into battle again. Yet, as we noted above, this did not lead to the resurgence of the mobile imperial court or the establishment of new imperial cities in the provinces. The court was now indivisible from Constantinople, where the emperor would always be visible, if not necessarily always accessible.

4. All the King’s Men The changing spatial and ceremonial dimensions of the Roman imperial court were paralleled by developments in the composition and structure of the court community. ‘Courtier’ remains a valid modern term for describing the members of this community, from high administrators and imperial women to chamberlains and doctors, in both the Principate and Late Antiquity. But from the third century, and particularly the fourth, onwards, there were attempts to demarcate membership of certain segments of the court community in the form of titles, honours, and perquisites. The first development concerns the imperial council. The transition from the ad hoc imperial consilium of the Principate to the consistorium of the late Roman world meant that henceforth senior officials, including military men, now had a fixed place in the hierarchy of the court and the government at large. This change was accompanied by a new system of honours awarded to high-ranking senators and equites to indicate membership of the court: the official grades of comites of the first, second, and third ranks. A comes ordinis tertii such as Symmachus was only rarely at court, but his title gave him a permanent connection to this

338

   

community.¹⁸⁹ The emperors, however, ensured that comites consistoriani who had actually served in the presence of the emperor always outranked any honorific appointments.¹⁹⁰ Of course, the size and purpose of the consistorium was not static and unchanging, showing that the structure of the court continued to evolve in Late Antiquity. By the reign of Justinian, the consistorium appears to have lost much of its deliberative power, largely becoming a body that gave ceremonial approval to the emperor’s decisions. Instead, the emperor preferred to consult with selected high-level officials from within its ranks.¹⁹¹ The second change concerns the bureaucracy at large. In the early empire, imperial administration was undertaken by slaves, freedmen, and high-ranking equites. There was no real opportunity for freeborn men below equestrian rank to serve in official posts either at court or in the government more broadly. This changed across the third century, so that educated freeborn men were henceforth able to climb a new cursus of bureaucratic success, acquiring equestrian and senatorial rank in the process.¹⁹² As Aloys Winterling has noted, this meant that the late Roman court and administration functioned as a locus for social mobility in a way that had not been the case in the Principate.¹⁹³ From the reign of Constantine onwards, officials who formed part of the administrative and financial bureaux attached to the emperor’s comitatus were described as palatini, literally those ‘who served in the sacred palace’ (in sacro palatio militarunt).¹⁹⁴ The palatini were superior to other imperial bureaucrats, as they were ‘endowed with palatine status’ (palatina . . . praediti dignitati), and therefore received numerous financial privileges, such as exemption from curial duties, and were honoured with high standing in their home communities.¹⁹⁵ In the fourth century, the anonymous author of the Account of the Whole World and its Peoples noted that intelligent men from many regions of Asia Minor had found gainful employment at ‘both the eastern and western courts’ (duo comitata orientis quoque et occidentis).¹⁹⁶ It is important to emphasize, however, that thousands of palatini—despite being attached to major departments, such as the scrinia or the sacrae largitiones—did not actually serve at court, but served in the provinces instead.¹⁹⁷ Their designation as palatinus, like the title of comes, was a mark of status that connected an individual to the court. The process of establishing the court as a coherent part of the Roman imperial government is one that we can, following Winterling, appropriately describe as ¹⁸⁹ Sogno 2006: 2–22. ¹⁹⁰ CTh. 6.22.8.1–3 (425). ¹⁹¹ Jones 1964: 338–41. ¹⁹² On this transformation, see Kelly 2004: 110–11; Davenport 2019: 575–91, esp. 584–5, and Haensch 2006 on the replacement of freedmen. ¹⁹³ Winterling 1997c: 164–5. There was, however, no state examination for such posts as in China (Elman 2000: 5–8; Elman 2013; Van Ess 2007: 242–5). ¹⁹⁴ CTh. 6.35, heading. The palatini are first attested under Constantine: CTh. 6.35.1 (314). ¹⁹⁵ CTh. 15.12.2 (357); Jones 1964: 104, 586; Kelly 2004: 187–9. ¹⁹⁶ Expositio totius mundi (ed. Rougé 1966: 180). ¹⁹⁷ MacMullen 1964: esp. 307; McCormick 2000: 136.

      

339

‘institutionalization’.¹⁹⁸ This began, as he demonstrated, with the investment of power and responsibility in members of the familia Caesaris in the Principate. By the middle of the fifth century, Emperor Leo I could be acclaimed as the choice not only of the senate, army, and people, but also of the court.¹⁹⁹ This volume has explored some of the ramifications of these developments, including the institutionalization of aspects of Roman intellectual culture, notably jurisprudence. In Chapter 6, Jill Harries demonstrated how jurists were transformed from imperial friends and ad hoc advisors in the Principate into permanent court officials whose decisions could be taken to reflect the views of the emperor and his regime in Late Antiquity. Indeed, as Kevin Feeney argued in Chapter 7, the institution of the consistorium created the potential for imperial appointments not to be merely clientelist in nature, but to embody the regime’s philosophy of governance. These appointments were not kept beyond closed doors, but publicized far and wide, featuring in imperial panegyrics and in acclamations.²⁰⁰ The growing complexity of the court was not a specifically Roman phenomenon. Parallels can be found in other monarchical societies in Europe, the Near East, and Asia.²⁰¹ Indeed, Norbert Elias identified such expansion and institutionalization as fundamental to the development of the French court between the Medieval and Early Modern periods.²⁰² Such institutionalization had its limits at Rome, however. Despite the closeness that could develop between members of the imperial family and members of the ecclesiastical establishment,²⁰³ there was no official position of ‘court bishop’.²⁰⁴ Nor is there any trace of the phenomenon of the ‘royal confessor’, a counsellor who had near constant access to the king, as in the later Catholic courts of Early Modern Europe.²⁰⁵ We know that the imperial palace at Constantinople was enhanced over the centuries with some thirty religious establishments, beginning with the church of St Stephen in Daphne under Theodosius II,²⁰⁶ a chapel of the Archangel Michael in or adjacent to the palace, attested by the reign of Anastasius,²⁰⁷ and the churches of Saints Peter and Paul and Saints Sergius and Bacchus in the palace of Hormisdas, constructed by Justinian in Justin I’s reign.²⁰⁸ ¹⁹⁸ Winterling 1997b; 1999. ¹⁹⁹ Const. Porph. de cer. 1.91 (1.100). ²⁰⁰ Panegyrics: Priscian, Pan. 239–41; Coripp. Iust. 1.14–27. Acclamations: A. D. E. Cameron 1976: 239–40. ²⁰¹ Duindam 2016: 243–4. See further Strootman 2014: 40–1 on the Hellenistic courts, and Duindam 2003: 12 on Early Modern Europe. ²⁰² Elias 2012: 250–1. ²⁰³ For the Theodosian emperors making appointments to the see of Constantinople from within their circle, see van Nuffelen 2012: 194. Timothy, bishop of Alexandria, was invited to stay in the imperial place by Basiliscus and his wife Zenonis, see Zach. HE 5.210 (ed. Greatrex et al. 2011: 175). ²⁰⁴ Leppin 2018b: 53. ²⁰⁵ Duindam 2003: 236; Reinhardt 2009: 577–8, 581–2; Asch 2016: 183, 190–1. ²⁰⁶ Theoph. AM 5920. That St Stephen was the first is indicated by the Notitia of 425, which lists no churches in the First Region containing the Great Palace (Matthews 2012: 86). This suggests that the palace Church of Our Lord, often given an early date, was not yet built (cf. McCormick 2000: 141). ²⁰⁷ Theoph. AM 6003; Kostenec 2005: 26–35. ²⁰⁸ Proc. Aed. 1.4.1–8; Croke 2006.

340

   

Presumably, these foundations had clergy associated with them, possibly seconded from Hagia Sophia, but there is no evidence of them exercising prominent influence at court in the period covered by this volume.²⁰⁹ Indeed, it is only in the High Byzantine age that we get a real sense of there being specific palace clergy in imperial ceremonies, a situation which exhibits similarities with other Medieval monarchies.²¹⁰ Thus, despite the important role played by Christianity in shaping the image of the emperor, his court, and the government in Late Antiquity, this did not automatically translate into a privileged institutionalized position for ecclesiastical officials. This distance between the Roman imperial court and Christian institutions in Late Antiquity may well have been shaped by the difficult theological politics of the era. The many doctrinal disputes that divided the empire could have made it impolitic or impractical to invite bishops or clerics into the court in a more formal manner. At the same time, it is likely that ecclesiastical office-holders cherished the freedom they acquired from functioning outside the court, as it allowed them to hold the emperor to account more easily. Emperors themselves often sought spiritual advice from holy men and women well outside the palace walls. Someone like Daniel the Stylite, whom multiple fifth-century emperors visited on his column on the outskirts of Constantinople, operated outside of the court society.²¹¹ It is somewhat ironic, therefore, that the most famous depiction of the late Roman court, the mosaics at San Vitale at Ravenna depicting Justinian, Theodora, and their courtiers, feature a named image of Maximianus, the bishop of Ravenna (Figure 14.3). This was an audacious act by the Italian bishop, who had likely never even met the imperial couple or their entourage in Constantinople. However, as the patron of the mosaics, Maximianus evidently decided to have himself portrayed as the court’s representative to the inhabitants of the newly reconquered exarchate of Ravenna. There is a variety of ways in which membership of a monarchical court can be defined—hierarchy and titulature is one, another is culture, language, and art. Elias regarded the evolution of the French nobility from a warrior class to a courtly elite with its own set of refined customs and values as part of the ‘civilising process’ that created the modern world.²¹² There are valid aspects to this argument, chiefly the changes he identified in the relationship between knighthood, aristocratic ²⁰⁹ Just. Nov. 3, which stipulates the number of clergy for Hagia Sophia and its associated churches, does not mention any churches inside the palace, which could indicate that they served there as well. The singing deacons and deaconesses who participated in Justinian’s funeral (Coripp. Iust. 3.42–3) probably came from Hagia Sophia (A. M. Cameron 1976: 181). ²¹⁰ Const. Porph. de cer. 2.8, 2.11 shows that Saint Stephen had its own protopapas and clergy in the tenth century. The ‘palace chapel’ is regarded as a later Byzantine phenomenon by Mango 1976: 89, 96, while Klein 2006: 93 identifies the ninth and tenth centuries as a key phase in the transformation of the ecclesiastical aspects of the palace. The church of Theotokos of the Pharos, first attested in the eighth century, was known as the ‘emperor’s chapel’ (capella imperatoris) by pilgrims (Jenkins and Mango 1956: 134). ²¹¹ McEvoy 2022. ²¹² Elias 2012: 171, 201, 231, 235–7, 253.

      

341

Figure 14.3 Mosaic depicting the Emperor Justinian and his court, Basilica of San Vitale, Ravenna. © Christine Webb, Alamy Stock Photo.

power, and cultural values in France and Early Modern Europe at large.²¹³ But Elias’ model, as Jeroen Duindam has expertly analysed, was founded on troubling methodological assumptions, not only in the determinations he made about what counted as ‘civilized’, but also because he promoted the idea that civilization and its values spread from West to East.²¹⁴ In this volume, we have discussed the Roman imperial court in terms of community values and behaviour, but we have resisted terms such as ‘courtliness’.²¹⁵ As Winterling has pointed out, the Romans did not have such a concept.²¹⁶ He argued that the values and expectations of the Roman imperial court were not unique to the court per se, but were representative of the ideals of the aristocracy writ large.²¹⁷ The court could, of course, promote or foster aristocratic art and literature in a range of ²¹³ Schalk 1991; Muir 1997: 120–2; Duindam 2003: 12; Crone 2015: 210. ²¹⁴ Duindam 1995: 159–71; Duindam 2003: 7–8, 295; Duindam 2011: 23. For a range of criticisms of Elias’ model from other quarters, see Asch 1991: 4; Evans 1991: 484; Bang 2011: 105–7, 126–7. ²¹⁵ OED s.v. ‘courtliness’: ‘The quality of being courtly; courtly civility or “grace of mien”; courtly elegance of manners’. For this phenomenon in western Europe, see Vale 2001: 26–7; Geary et al. 2015: 192, 196. ²¹⁶ Winterling 1997b: 165–6. Note also the important remarks of Noreña 2007a: 312–13 on the focus on ‘virtue and moral value’ at the Roman court as opposed to ‘etiquette and codes of conduct’ at the Early Modern courts. ²¹⁷ See Olson 2022: 476–7 on the difficulties of identifying specific ‘courtly’ dress and adornment during the Principate. For comparative discussion, see Vale 2001 (Medieval courts); Duindam 2003: 287 (Early Modern courts); Geary et al. 2015: 192 (Byzantium). Cf. Strootman 2014: 267, who identifies the spread of an ‘aulic “high culture” ’ in the Hellenistic world.

342

   

different ways, from the coterie of Augustan poets clustered around Maecenas to the Constantinopolitan courtiers who commissioned mosaics of their emperors throughout the city.²¹⁸ Some of the most reviled figures of the late Roman court— the eunuch cubicularii—played a major role in Christian literary patronage. Palladius’ Lausiac History, a collective biography of holy men and women, was commissioned by Theodosius II’s eunuch Lausus, while Zachariah Rhetor dedicated his own Ecclesiastical History to the eunuch chamberlain Eupraxius, who held office under Anastasius.²¹⁹ Imperial tastes could also promote certain types of recreational activities. Theodosius II so enjoyed polo (tzykanion) that he installed a field, the Tzykanisterion, in the palace at Constantinople.²²⁰ Hunting also became a more popular and genuinely courtly pastime in the late antique East than it had perhaps been during the Principate.²²¹ From the late sixth century onwards, emperors even began to construct hunting places and game parks in the vicinity of Constantinople.²²² Yet the Roman imperial court was never a community that was sharply demarcated by the use of a dialect or form of language, as was the case with many other royal establishments.²²³ At the Early Modern French court, specific forms of vocabulary and pronunciation were regarded as markers of ‘courtly’ manner of speech, which functioned as a way of demarcating insiders from the outsiders.²²⁴ Several courts used different languages to honour and respect the sanctity of the ruler. For example, in the Javanese monarchy, there was a specific requirement that the palace language be used ‘in the royal presence’.²²⁵ In the Ottoman court, sign language was introduced to court in the reign of Süleyman I, as it was thought unbecoming to speak before the sultan.²²⁶ Perhaps the most famous example of a court language is the Kyoto dialect, the most refined and cultivated form of Japanese.²²⁷ When Emperor Hirohito delivered a radio address to the Japanese people in August 1945 to announce their surrender to the American forces, he spoke in this dialect, which meant that most of his subjects ²¹⁸ Principate: Bernstein 2022. Mosaics: Joh. Lyd. de mag. 2.20 (Leo I); Zach. HE 3.158 (ed. Greatrex et al. 2011: 280) (Justin I); Theoph. AM 6079 (Maurice). ²¹⁹ Pall. Hist. Laus. Preface 2–3; Zach. HE 3.144, 146 (ed. Greatrex et al. 2011: 95, 97). ²²⁰ Janin 1964: 118–19; Canepa 2009: 180; Kelly 2013a: 4. ²²¹ For Theodosius II’s hunting habits, see Theod. Lect. Epit. 18 [353]; Joh. Ant. fr. 220 (ed. Mariev 2008). The eastern general (and later western emperor) Anthemius hunted with his comrades in his youth (Sid. Ap. Carm. 2.145–55); this was common practice among eastern military officers in the fifth and sixth centuries (Theoph. AM 5943, 6090). Principate to Late Antiquity: Roller 2022: 336–46, esp. 346. ²²² Littlewood 1997: 36–8. ²²³ Note the remarks of Evans 1991: 484–5 on the Early Modern courts of Europe. Cf. John Lydus’ (de mag. 3.20.1) complaints about those who mispronounced official terms, which marked them as outsiders (Kelly 2004: 27–8). ²²⁴ Motley 1990: 70–2; Fumaroli 1994: 521–4, 691–8; Duindam 1995: 100. ²²⁵ Errington 1982. ²²⁶ Necipoğlu 1991: 26–9. ²²⁷ Sakurai 1984; Lillehoj 2011: 66. One of the primary functions of the Japanese imperial court, which had no real authority to govern, was to set standards of taste and refinement (Butler 2002: 248–86).

      

343

could not actually understand him.²²⁸ In Rome, there was a highly rhetorical style to the formal written pronouncements of the Roman emperor and his administration, as witnessed by the florid imperial edicts and letters preserved in the Codex Theodosianus. But this never evolved into a specific court language or dialect that distinguished members of the court community from outsiders. Royal courts have also traditionally expected certain behavioural and emotional standards of kings and courtiers. In the words of Patrick Geary and his colleagues, courts are dominated by a type of ‘moral economy’ based on the monarch’s feelings and emotions, and thus, ‘the transfer and acquisition of power were seen as the results of personal interaction’.²²⁹ In Chapter 5, Benjamin Kelly explored how the behaviour of Roman emperors and their courtiers were shaped by prevailing aristocratic values. The world of the Antonine court, which was populated by a relatively united senatorial and equestrian aristocracy with shared values and expectations, had stable norms of emotional expression focused on love and affection. However, as Kelly has observed, the courts of some fourthcentury soldier emperors, such as Valentinian I, were less stable emotional communities, since the emperor also exhibited negative emotions such as anger and rage.²³⁰ This disruption of the accepted ‘moral economy’ of courtly relationships had the potential to leave members of the court society uncertain as to how they should relate to the emperor.²³¹ Some courtiers adopted and emulated Valentinian’s conduct, while others tried to rein in his excesses and thus reassert norms of imperial behaviour.²³² These developments can be at least partially ascribed to the rival value system espoused by the military-men-turned-emperors in Late Antiquity.²³³ These emperors brought their former army comrades into the court community, as explored in Chapter 3. The clash between established aristocrats and new men from different cultural and social backgrounds, elevated to high status because of a close relationship with the ruler, is a common feature of monarchical societies, especially in periods of regime change.²³⁴ As Kelly emphasized in his chapter, we should resist teleological conclusions, and conclude that the breakdown of ‘normative consensus’ at the courts of several fourth-century emperors was a hallmark of all imperial reigns in Late Antiquity. The situation was indeed more complicated. Many military men who achieved high office sought to identify themselves as aristocrats and adopt the value system

²²⁸ Low 2006: 93. ²²⁹ Geary et al. 2015: 190. ²³⁰ But he was not entirely uncultured: Matthews 1989: 237–8. ²³¹ See Brown 1992: 48–61 for an incisive discussion on the aristocratic virtue of self-control in Late Antiquity. ²³² See Drijvers 2012 on officials who emulated the emperor and Pazdernik 2012: 104–6 on the efforts of the quaestor Flavius Eupraxius to keep him in check. ²³³ Matthews 1989: 249. ²³⁴ See Graff 2002: 92–3 on China in the sixth century and Kumar 2011 on the Delhi sultanate in the thirteenth and fourteenth centuries.

344

   

of senators in an attempt to find acceptance within elite society.²³⁵ The place and power of generals such as Stilicho, Aetius, and Aspar at court was legitimated by the consensus of the court community, as Meaghan McEvoy explored in Chapter 8. Moreover, further study is necessary to establish the emotional tenor of court communities under different late antique emperors. Jovian, a former primicerius domesticorum, did not cultivate a court environment full of Illyrian military comrades as we might have expected. Instead, as Kevin Feeney has shown in Chapter 7, he followed a policy of conciliation, which led him to embrace multiple constituencies in his court appointments. There are no tales of Jovian’s incandescent rages or of a court marked by anger and tension. The court of Theodosius II, while perhaps not the monastic retreat which it was characterized as by ecclesiastical historians, certainly had a different tone from the courts of Constantius II and Valentinian I.²³⁶ The annoyance of Theodosius II was signalled by a sudden shaking of his cloak rather than the apoplectic rage of some of his predecessors.²³⁷ This returns us to the point with which we began this chapter: it is important to examine the character of specific Roman courts within the long-term pattern of systemic changes that occurred between the Principate and Late Antiquity. The history of military men as emperors and courtiers is a complex one, marked by conflict, negotiation, and adaptation. The Roman imperial court community was relatively permeable, in both the Principate and Late Antiquity. In physical terms, this manifested itself in the relationship between the emperor, the court, and urban environments. The court and its members interacted with spaces and people in cities such as Rome and Constantinople, in contrast with the stern isolation of some other monarchical courts, such as that of imperial China, which were largely cloistered behind the walls of their palaces.²³⁸ Aristocrats did not usually reside at court in the manner of Louis XIV’s Versailles; instead, they had their own houses and palaces in Rome and Constantinople.²³⁹ But the permeability is also true on a social level. There were no initiatives that attempted to articulate membership of the court until the fourth century, when we find administrative officials designated as palatini and selected senators and equites made court comites. Even then, there was no all-encompassing measure to define precisely who was or who was not a ‘courtier’. Not all palatini and comites necessarily served at court all year round, or even on a regular basis, nor were all courtiers palatini and comites. Even the privilege of kissing the purple robe, which the emperor could award to his courtiers, was distributed more widely to individuals who held administrative or

²³⁵ Davenport 2019: 591–9. ²³⁶ Soc. 7.22.4. ²³⁷ Nest. Herac. pp. 284–5 (trans. Driver and Hodgson 1925). ²³⁸ Geary et al. 2015: 181. ²³⁹ For these residences, see Eck 1997; Machado 2012; 2019: 231–52. One possible exception, noted in Chapter 3, is the situation of high officials in the Tetrarchic period.

      

345

military positions, but who could not be considered members of the court community, such as the protector Flavius Abinnaeus.²⁴⁰ Comparison with other monarchical societies reveals paths that the Roman imperial state could have taken to formalize recruitment into, or membership of, the court. For example, at the Japanese Heian court, individuals had to be officially granted the title of ‘courtier’ (tenjōbito) and graded in a hierarchy from Senior First down to Junior Eighth. Under Emperor Uda (887–97), promotion through these ranks depended on how often one attended court, reflecting a very regimented approach not seen at Rome, even in Late Antiquity.²⁴¹ In some societies, there were firm measures to delineate and differentiate a court aristocracy from the nobility of the land at large. In Habsburg Spain, there were two groups of courtiers: a select coterie of top-ranked individuals, known as the grandees, who were permitted to wear hats in the king’s presence, which differentiated them from the hatless aristocrats, the títulos (the titled nobles).²⁴² In contrast, the orders of precedence that were instituted by late Roman emperors from Valentinian I and Valens onwards were designed to establish a coherent hierarchy encompassing all administrative, military, and senatorial posts. The decisive change, when it came in the Roman world, happened in the seventh century, with a new system of ranks and titles all based at court, so that, in the words of John Haldon, ‘all forms of social and political advancement now depended on association with the imperial court and its system of precedence’.²⁴³ One recruitment mechanism adopted by many court societies was the institution of male pages.²⁴⁴ In the Ottoman Empire, enslaved Christian boys were educated at court, where they served as pages, converted to Islam, and were trained in the arts of Muslim government. Many pages subsequently joined the army or remained at court in official positions, but the most favoured were sent out to govern the provinces, so that the court effectively fertilized the administration of the empire.²⁴⁵ While the Ottomans preferred outsiders as pages, in other monarchical societies, pages were selected from among the offspring of aristocrats, as at the Early Modern French court, where noble children served the monarch as enfants d’honneur.²⁴⁶ At Hellenistic courts, many aristocratic pages were eventually appointed philoi, official friends of the king, a title that was more exclusive and formal than amicus ever was at Rome.²⁴⁷ The Roman world of the Principate followed neither of these models: the pages known as paedagogiani who were

²⁴⁰ P. Abinn. 1. For these events, see Kelly 2004: 199–200. ²⁴¹ Geary et al. 2015: 184; Steininger 2017: 22–5; Wakabayashi 1991. ²⁴² Elliott 1977: 173–4. ²⁴³ Haldon 1990a: 387–99 (quotation from 398). ²⁴⁴ Duindam 2016: 126, 234–5. See, for specific examples, Strootman 2014: 136–44 on Hellenistic royal pages, Paravicini 1991: 85 on pages at the court of the Duke of Burgundy, and Duindam 2003: 57 on pages at Versailles. ²⁴⁵ Necipoğlu 1991: 111–22; Imber 2012: 215–18. ²⁴⁶ Duindam 2003: 286. ²⁴⁷ Engels 2017: 70–1.

346

   

trained for service at court were slaves.²⁴⁸ Unlike the Christian slaves in the Ottoman world, the Roman pages were never promoted to high office, and they were often mistreated as sexual playthings.²⁴⁹ The situation changed in Late Antiquity, by which time the paedagogiani were free men in the service of the emperor.²⁵⁰ Aristocrats have often been honoured with the privilege of being raised with a future king at court. For example, Hellenistic princes were surrounded by cadres of suntrophoi (literally ‘those raised together’), who were the scions of court families. As Ivana Savalli-Lestrade has aptly put it, ‘they were aulikoi [courtiers] from birth’.²⁵¹ In Rome, the limited evidence suggests that such friendships were more informal and fluid. There are stories of aristocratic young men being brought up and even educated with members of the imperial family at court. These include the future Emperor Titus, son of the senator Vespasian, who was a comrade of Claudius’ son Britannicus; the Augusta Flaccilla’s nephew Nebridius, childhood friend of her sons Arcadius and Honorius; Paulinus, son of a comes domesticorum, who was a fellow pupil and friend of Theodosius II.²⁵² The teaching that did take place must have been organized on an informal basis by individual tutors, and would have been a far cry from the Merovingian aristocratic ‘court school’ (scola regis), which was supervised by the mayor of the palace.²⁵³ This examination of the Roman imperial court community in historical perspective emphasizes the relatively permeable nature of the court society, even in Late Antiquity, when compared with other monarchical states. There was no one formal title for courtier, nor an exclusive court nobility, no institutionalized aristocratic training system, nor a specifically courtly style and manner of language. The lack of an established ecclesiastical presence in the form of court bishops and confessors suggests that it was seen as preferable for the court and church to remain separate—and competing—entities. In the transition from the Principate to Late Antiquity, the greatest change occurred in the administrative sphere, with the institution of the consistorium and the drawing of legal experts into the court. This suggests that there was a drive (conscious or unconscious) to formalize the Roman imperial court as an engine of government, rather than as a social unit.

²⁴⁸ Balty 1982; Keegan 2013. ²⁴⁹ Keegan 2013: 85–9; Chrol and Blake 2022: 358. For the wider context, see Vout 2007, whose account of sex and power at the imperial court features three case studies of unequal sexual or eroticized male relationships: Hadrian and Antinous; Nero and Sporus; Domitian and Earinus. ²⁵⁰ Amm. 29.3.3; CTh. 8.7.5 (354), with discussion in Jones 1964: 571; den Boeft et al. 2014: 121. ²⁵¹ Savalli-Lestrade 2017: 103–6 (quotation from 104). ²⁵² Britannicus: Suet. Tit. 2. Nebridius: Jer. Ep. 97. 5. Paulinus: Chron. Pasch. s.a. 420, 421; Joh. Mal. 14.3–4, 6, 8. ²⁵³ Hen 2007: 101–6; Hen 2020: 230–1.

      

347

5. A Woman’s Place Women have occupied an ambiguous space at all Eurasian monarchical courts which aspired to dynastic continuity.²⁵⁴ Marriage or concubinage with women is necessary for kings to produce heirs, but the position and potential influence— often, but not always, derived from child-bearing—of these women challenged the accepted masculine monopoly on political authority.²⁵⁵ The Roman republican system of government had no place for female officials among the ranks of consuls, legates, and prefects, but the transformation of the state into a monarchical form offered new possibilities for women. Augustus’ wife Livia, in the eyes of the poet Ovid, was nothing less than his female equivalent, a femina princeps.²⁵⁶ The honorific title of Augusta, first granted to Livia in 14, and which was subsequently conferred on some imperial wives, daughters, and sisters, offered no formal power or responsibilities. But Augustae and other imperial women had vitally important public roles to play both as patrons of individuals and communities and as exemplary paradigms of female virtue.²⁵⁷ Within the court community, they acted as benefactors, advisors, and marriage brokers.²⁵⁸ Imperial women were also acknowledged as exercising influence with their husbands or male relatives, advocating for important causes or calming an emperor’s wrath.²⁵⁹ At the same time, however, attempts to engage in overtly political decision-making were roundly condemned by the literary sources.²⁶⁰ In Chapter 13, Caillan Davenport discussed how this prejudice played out in stories told about the Roman imperial court, with girls and women ranging from Marcus Aurelius’ three-year-old daughter to the Arian wife of Valentinian I being portrayed as negative influences on the emperor. The place of women at the Roman imperial court was also shaped by the fact that the emperors always had to engage in monogamous marriages.²⁶¹ There was, therefore, no hierarchy of wives and consorts and their attendant offspring, nor a harem or any formal system of concubinage, as we find at the Islamic ‘Abbasid and Ottoman courts and in many Asian and African societies.²⁶² Roman emperors

²⁵⁴ The exception to this rule is, of course, monarchies that allowed female rulers. Some African kingdoms had a long tradition of female rulership, on which see Duindam 2016: 93–4. ²⁵⁵ van Leeuwen 2017: 163–7. ²⁵⁶ Ov. Tr. 1.6.25. ²⁵⁷ See Purcell 1986 on Livia’s position; and Chen 2018: 62–9; Boatwright 2021: 9. ²⁵⁸ McCormick 2000: 148; Kunst 2010: 148–50. For example, the senator Nebridius (PLRE I Nebridius 3) benefited from the fact that his aunt Flaccilla was Theodosius I’s first wife. Through this connection, the emperor arranged his marriage to Salvina (Jer. Ep. 79.2). ²⁵⁹ IAph2007 8.32 is an early example, a letter from Octavian to Samos, in which he acknowledged that Livia had pressed its cause to be made a free community. For the idea of a woman calming the emperor, see Kunst 2010: 158; McEvoy 2021: 130–1. ²⁶⁰ Angelova 2004: 3; Leppin 2018b: 49; Boatwright 2021: 16–19; Hug 2022: 72–3. ²⁶¹ On monogamy and Roman society, see Scheidel 2009b: 295–6. ²⁶² For comparative material on these various possibilities, see Walthall 2008: esp. 13–14; Scheidel 2009b; Duindam 2016: 108–127, as well as Peirce 2008 (Ottomans) and El Cheik 2014: 366–7 (discussing ‘Abbasid and Byzantine women). See Kaplan 2008 for a fascinating ethnographic account of the harem in a modern-day royal court, that of the Oba of Benin in Nigeria.

348

   

did, of course, often have sexual relations with women (and men) to whom they were not married, but this did not form part of any institutionalized polygyny.²⁶³ High Medieval and Early Modern European courts might initially seem to provide more promising points of comparison, since their monarchs also practised monogamous marriage. However, in these later European courts, the combination of monogamous royal marriage and adultery ‘created two characters absent in harem-based reproduction: the mistress and the bastard’, as Jeroen Duindam has memorably put it.²⁶⁴ During the Principate and Late Antiquity, there was no near-permanent, dominant figure of ‘the mistress’ comparable to that found in later European societies, in which such women and their coteries coexisted openly with an official wife and provided a rival centre of female power.²⁶⁵ There were some cases of this in Rome, such as Nero’s relationship with the freedwoman Claudia Acte, who posed a challenge to his mother Agrippina.²⁶⁶ But other examples of Roman mistresses—notably Antonia Caenis (Vespasian) and Galeria Lysistrate (Antoninus Pius)—were freedwomen who had sexual relationships with emperors whose wives had already died. They were certainly sought after as figures of influence and brokers of patronage, but the role of the emperor’s mistress never approached the semi-institutionalized position that it did in later European monarchies.²⁶⁷ Despite the fact that Roman emperors and their sons certainly had sexual relations with slaves and freedwomen,²⁶⁸ the existence of their illegitimate children goes largely unreported.²⁶⁹ We do not know what happened normally to children born of emperors and slaves and freedwomen: perhaps they were integrated into the imperial slave household, were sent away to serve in other aristocratic houses, or were killed.²⁷⁰ The most famous example of an illegitimate child, though a manufactured one, is the case of Elagabalus, who claimed to be the offspring of a liaison between his mother Julia Soaemias and Emperor Caracalla.²⁷¹ But this was a purported sexual relationship with an aristocratic woman, not with a member of the slave and freedwomen population of the imperial household. There is also the case of the praetorian prefect Nymphidius Sabinus, who, after Nero’s murder, said he was the illegitimate son of Caligula. He

²⁶³ Chrol and Blake 2022: 350–9 offer an important discussion of this point. ²⁶⁴ Duindam 2016: 122; Bartlett 2020: 159–68. ²⁶⁵ Duindam 2003: 238–40. ²⁶⁶ Tac. Ann. 13.12–13, 14.2; Suet. Ner. 28.1, 50; Dio 61(61).7.2. ²⁶⁷ Caenis: Dio 66(65).14.1–4; Suet. Vesp. 3, 21; Dom. 12.3. Galeria Lysistrate: HA Ant. Pius 8.9; CIL VI 8972. See Boatwright 2021: 113–15; Chrol and Blake 2022: 361–6. ²⁶⁸ Marcus Aurelius was rare in noting that he did not sleep with the slaves Benedicta and Theodotus (Med. 1.17). ²⁶⁹ Scheidel 1999: 279; Scheidel 2009b: 301; Chrol and Blake 2022: 356. ²⁷⁰ We would like to acknowledge our debt to the animated discussion about imperial illegitimacy at Benjamin Kelly and Angela Hug’s conference at York University, which stimulated these thoughts. ²⁷¹ Dio 79(78).31–3; Hdn. 5.3.9–10; HA Heliogab. 1.4–5. This relationship even entered official discourse, as Elagabalus was styled the son of the deified Antoninus Magnus (e.g. CIL XVII.2 636), as was his brother and successor Severus Alexander (e.g. CIL VI 40683). See Hekster 2015: 218–21.

      

349

proclaimed that his mother Nymphidia, the daughter of the powerful imperial freedman Callistus, had had an illicit sexual liaison with the emperor. However, as the offspring of a freedman, Nymphidia would herself have been a citizen.²⁷² There is only one recorded example of a child from a sexual relationship between an emperor and a slave or freedwoman. According to Lactantius, the boy Candidianus was born to Galerius ‘from a concubine’ (ex concubina), probably around 296, when Galerius was a Caesar in the First Tetrarchy.²⁷³ It is highly unlikely that Lactantius would have used the word concubina to describe a relationship between Galerius and a high-status or free woman.²⁷⁴ In the Roman world, there was no formal legislation stating that the illegitimate son of an emperor could not become emperor himself. The examples of Nymphidius Sabinus and Elagabalus show that there was no inherent shame to being a bastard claimant to the purple: in fact, it could be an advantage if it helped establish a connection to a previous emperor.²⁷⁵ Candidianus was adopted by Galerius’ wife, Galeria Valeria, and then betrothed to the daughter of Maximinus Daza. Lactantius claims that Galerius had grand plans for Candidianus, which included elevating him into a new imperial college that would exclude both Constantius I and his son, Constantine.²⁷⁶ Even in the European Middle Ages, the premise that illegitimate sons were excluded from the succession depended on the laws of individual kingdoms, and there was little consistency in this regard before the twelfth century.²⁷⁷ But the anxiety that illegitimate sons of emperors could potentially engender in Rome is shown by the case of Caesarion, the offspring of Julius Caesar and Cleopatra. He was executed by Octavian, lest he prove to be a rival as Caesar’s true heir.²⁷⁸ The wider imperial family could also be enveloped in concerns about the succession, hence the panic sparked by a relationship between the Augusta Honoria, sister of Emperor Valentinian III, and her procurator, Eugenius, and any potential offspring from the union.²⁷⁹ Mistresses and imperial bastards emerge as a more prominent phenomenon as we move into the Byzantine period.²⁸⁰ In the seventh century, Heraclius’ illegitimate son John Atalarichus, the product of a liaison with a Gothic woman, rebelled against his father.²⁸¹ In the tenth, Zoe, the mistress and later the fourth wife of Leo

²⁷² Plut. Galb. 8; Tacitus, Ann. 15.72.2. ²⁷³ Lact. DMP 50.2; PLRE I Candidianus 1. ²⁷⁴ OLD s.v. concubina. Treggiari 1981 shows that freeborn concubinae were rare: most were slaves or freewomen. In these relationships, the man was usually of higher status than the woman. The situation was different in Medieval Europe, in which the word concubina could be used to describe an aristocratic mistress (Bartlett 2020: 155–6). ²⁷⁵ Some scholars believe that Constantius I and Helena were unmarried, and thus Constantine was illegitimate (e.g. Drijvers 1992a: 15–19), while others argue that they were in a legitimate union (see Barnes 2011: 30–8). Even if he were illegitimate, it was not an obstacle. ²⁷⁶ Lact. DMP 35.4, 50.2, 50.6. ²⁷⁷ McDougall 2017: esp. 9–21; Bartlett 2020: 165–86. ²⁷⁸ Plut. Ant. 81.2–82.1; Dio 51.6.1–2, 15.5. ²⁷⁹ Marcell. com. s.a. 434 (which even mentions a pregnancy). ²⁸⁰ Garland 1995: 106–7. ²⁸¹ Sebeos, Armenian History 41 (ed. Thomson, Howard-Johnson, and Greenwood 1999: 92–3); Nic. Brev. 13 (ἐκ παλλακῆς, ‘from a concubine’), 24.

350

   

VI (886–912), gave birth to a son prior to their marriage. This marriage itself needed special dispensation from the Church, since it had been established that Christian emperors could only legitimately marry three times. The birth took place in the porphyra, the purple chamber in the palace at Constantinople where imperial sons were traditionally brought into the world.²⁸² The boy was henceforth known as Constantine VII Porphyrogenitus (‘purple-born’) (913–19), a moniker designed to emphasize his legitimacy as Leo’s son and heir.²⁸³ Constantine VII was fortunate in that he was his father’s only offspring, and this status added an urgency to his being recognized as the purple-born heir, despite the irregularity of his parents’ marriage. Bastard sons of Byzantine emperors were normally regarded as potential complications to the imperial succession, as the case of Caesarion had foreshadowed many centuries before. Emperor Romanus I (920–44) had his illegitimate son Basil castrated and employed as a eunuch official to prevent any fraternal rivalry with his legitimate sons.²⁸⁴ Illegitimate daughters were more useful than sons in this regard, since they could be used to form marriage alliances with foreign powers, as occurred during the Palaiologan era.²⁸⁵ These political and familial strategies of the Byzantine imperial court mirrored those of other contemporary Medieval monarchies.²⁸⁶ The fifth century witnessed important developments in the role and prominence of imperial women, as Anja Busch explored in Chapter 9. In the fifth century, Augustae start to be portrayed on coinage sitting on a throne, a representational privilege previously only accorded to the emperors themselves. The sixth-century empress Ariadne is also shown enthroned on a spectacular ivory diptych now in Vienna (Figure 14.4).²⁸⁷ Later that same century, Justin II and his wife Sophia were depicted on coins sitting together on a double throne, emphasizing their equality as co-rulers of the empire.²⁸⁸ These changes in representation took place at the same time that women began to play new roles in the imperial succession. When monarchical dynasties which transfer power through the male line fail to produce a son, as happened in both the western and eastern empire during the Theodosian period, they have traditionally faced a series of choices.²⁸⁹ These included the adoption of a new heir; the marriage of an aristocrat to a royal woman (if one was available), thus transferring power through the female line;²⁹⁰ or the creation of a queen or ²⁸² See Herrin 2013: 223–5; Bartlett 2020: 70–1. ²⁸³ Ringrose 2008: 66–7. ²⁸⁴ Kazhdan and McCormick 1997: 178; Herrin 2013: 227–8. ²⁸⁵ Schrijver 2018: 225–6. ²⁸⁶ Bartlett 2020: 15–26. ²⁸⁷ Angelova 2004: esp. 5–6. ²⁸⁸ A. M. Cameron 1975: 8–16; McClanan 2002: 158–61; their equality is likewise emphasized in Coripp. Iust. 2.169–73. For imperial women seated on the throne in later Byzantine ceremonies, see Ringrose 2008: 69–71. ²⁸⁹ For the reproductive capacities of Roman emperors and their wives, see Scheidel 1999: 266–79. ²⁹⁰ This occurred in Egypt in the Eighteenth Dynasty due to a shortage of male heirs, a crisis that paralleled that of the Theodosian age in Rome. See Spence 2007: 299–30.

      

351

Figure 14.4 Ivory diptych of the Augusta Ariadne, Kunsthistorisches Museum, Vienna. © akg-images.

empress regnant (providing a suitable female candidate existed).²⁹¹ The latter option was not on the table in the fifth century and would not be considered for several hundred years, until Irene ruled the empire as Augusta between 797 and 802. Indeed, in the field of female rulership, Byzantium set the standard for Medieval Europe at large; queens regnant did not appear in western Europe until the twelfth century.²⁹² Roman emperors who lacked sons had often availed themselves of the method of adoption to continue their dynasty, though they usually selected their new heirs from within their extended family.²⁹³ But adoption was not used in the fifth century, probably on account of the primacy of the Theodosian house and the preference that that bloodline continue in some form.²⁹⁴ This meant that option two, marriage of a man to an imperial woman, became a new way of transmitting imperial power. As Busch has argued, the ‘dynastic potential’ of Theodosian women such as Galla Placidia and Pulcheria, and later of Ariadne, the daughter

²⁹¹ For example, in Japan of the Asuka and Nara periods, succession crises resulted in the appointment of empresses regnant. See Hurst 1999: 577–9. ²⁹² Bartlett 2020: 124–46. ²⁹³ Hekster 2001. ²⁹⁴ For discussion, see McEvoy 2019a. Adoption was still employed later in the Byzantine Empire, in contrast with western kingdoms of Medieval Europe (Bartlett 2020: 78–84).

352

   

of Leo I, gave them the ability to bestow legitimate authority on their husbands.²⁹⁵ Such transfers of power had been accepted and commonplace in the Macedonian and later Hellenistic monarchies,²⁹⁶ but had largely been resisted in the Roman Empire up to this point.²⁹⁷ Indeed, Theodosius II’s sister Pulcheria famously ensured, through vows of virginity, that not only she but her two younger sisters would remain celibate so that no other man would be admitted into the palace and pose a threat to their brother’s rule.²⁹⁸ The consecration of imperial women as holy virgins was itself a major break with the past, since this was unthinkable during the Principate.²⁹⁹ The dynastic potential of an Augusta meant that she became a figure to whom the imperial court could turn for leadership during the uncertain period which followed the deaths of emperors. Her selection of a husband could determine the next emperor, as discussed by Audrey Becker in Chapter 4. This method of bestowing imperial power did meet some initial resistance (during the reign of Zeno, for example) and called for negotiation among different groups at court.³⁰⁰ Indeed, the selection process itself consolidated the importance of the court as the primary source for new emperors, which we have already noted. When Ariadne was asked to choose her husband in 491, she selected Anastasius, not an illustrious general with numerous campaigns to his credit, but one of the decurions of the silentarii, and thus a courtier.³⁰¹ Likewise, when Justin II’s health declined, the Augusta Sophia and the senate decided on the promotion of the comes excubitorum Tiberius to the post of Caesar in 574.³⁰² Such developments represented a real change from the world of the Principate, and set the stage for the accepted transfer of imperial power through the female line in later Byzantium.³⁰³ The progressive increase in the status of royal women over time is paralleled in other monarchical societies. In polygynous realms, this change was often driven by a desire to introduce a hierarchy among the king’s women.³⁰⁴ For example, the Hellenistic monarchies bestowed on the king’s first-ranked wife the title of basilissa, the female equivalent of basileus. The first woman so honoured was Phila, the wife of Demetrius Poliorcetes, in 306/305 .³⁰⁵ At other courts, the ²⁹⁵ See the praise of Ariadne and her role in handing over the empire to Anastasius in Priscian, Pan. 301–8. ²⁹⁶ Miron 2000: 44–7; Strootman 2014: 107–8. ²⁹⁷ On potential early imperial precedents that were not followed through, see McEvoy 2019b: 198; Hug 2022: 80. ²⁹⁸ Soz. 9.1; McEvoy 2019a: 118–21. ²⁹⁹ See Hug 2022: 81 on unmarried women in the Principate. ³⁰⁰ McEvoy 2019b: 204–5. ³⁰¹ Zach. HE 7.18 (Greatrex et al. 2011: 228) on the prior (non-sexual) relationship between Anastasius and Ariadne. Silentarii of this period were far from humble officials (Bell 2009: 14; Porena 2018: 179). ³⁰² Joh. Eph. HE 3.5 (Sophia and the senate choose); Evagr. HE 5.13 (Sophia alone persuades Justin II). ³⁰³ Garland 1999: 1–2. ³⁰⁴ Scheidel 2009b: 269–70. ³⁰⁵ Carney 1991: 161. On hierarchy among Hellenistic wives, see Strootman 2011: 77; Strootman 2014: esp. 107, also 109–10, 198–9.

      

353

prominence of women grew in order to fill a void left by male members of the family. Leslie Peirce’s examination of the women of the Ottoman court shows that they obtained greater prestige in the sixteenth century as a result of the embedding of the court and its harem permanently in Istanbul, and the infantilization of male princes, who were cloistered in the palace and forbidden to marry. In contrast, the female princesses had their marriages celebrated publicly in Istanbul.³⁰⁶ The sixteenth century saw the formal title of queen mother (valide Sultan) introduced at the Ottoman court. In the seventeenth century, these women frequently acted as regents for young sultans and were even awarded the title ‘mother of Muslim believers’.³⁰⁷ We do not wish to suggest that the circumstances which brought about the increasing prominence of women at Roman, Hellenistic, and Ottoman courts were exactly the same.³⁰⁸ But what these cases do show is that the awarding of new titles to royal women, and the gradual increase in their influence and authority over time, was both a reaction to and a strategy to cope with other developments at court and in the monarchy at large. The creation of the title of valide Sultan (queen mother) at the Ottoman court draws our attention to the role of royal mothers, who often outranked the ruler’s wives.³⁰⁹ The highest-status woman at the Chinese court was the grandmother empress dowager, if she still lived, and the empress dowager if not.³¹⁰ In Ch’angan, the capital of Han China, there were two palaces, the Eternal Palace for the emperor, and the Palace of Prolonged Joy for the dowager empress, reflecting the institutionalized nature of this position.³¹¹ At the Habsburg court in Vienna, the dowager empress had her own independent court (Hofstaat) between 1637 and 1742, unlike the empress herself, who was still at least partially dependent on the emperor’s establishment.³¹² In an extreme example, at the still-existent court of the Oba of Benin in Nigeria, the queen mother, or ‘Iyoba’, resides in a completely separate establishment from her son and his harem, which she is forbidden to enter. However, she has her own court, and ‘assumes a male gender role and functions as a senior male chief ’.³¹³ Imperial mothers were part of the court community in Rome, but there was never an equivalent to ‘queen mother’ or ‘dowager empress’, in either the Principate or Late Antiquity. Imperial mothers could be awarded the title of Augusta, which they might share with imperial wives and daughters, a situation which often led to rivalry and tension. They also

³⁰⁶ Peirce 2008. For an analysis of how architecture and ceremonial reflected the influence of royal women and the harem in this period, see Necipoğlu 1991: 22–3, 92–3, 95–6, 165–77. ³⁰⁷ Necipoğlu 1991: 164; Peirce 1993: 91–112; Peirce 2008: 87. ³⁰⁸ Nor do we wish to claim this is a universal tendency. Kaplan 2008: 124–5 notes that even today the names of the wives of the Oba of Benin are not known, and they are referred to by the names of their children within the court. ³⁰⁹ For comparative overviews, see Walthall 2008: 11–13; Duindam 2016: 58–60, 95–6, 106–7. ³¹⁰ Duindam 2016: 115. ³¹¹ Van Ess 2007: 238. ³¹² Duindam 2003: 77–9. ³¹³ Kaplan 2008: 128–31.

354

   

performed very useful function as symbols of virtue, as in the case of Constantine’s mother, Helena, who became the paradigmatic Christian empress.³¹⁴ The political influence of Roman imperial mothers depended on two factors: firstly, their continued longevity, since they had to survive to see their sons rule;³¹⁵ and secondly, whether or not their sons wished to maintain them at court once they came of age. To take one example, Nero’s mother Agrippina exerted considerable influence in the early part of his reign, even holding her own salutatio. However, there was no formalization of her position comparable to that of the Chinese empress dowager. Agrippina’s continued existence proved to be anything but joyous for the young emperor, and his decision to cut down the size of his mother’s household and remove her guard escort was a sure sign that her life would certainly not be prolonged.³¹⁶ There was never any question of imperial women in the Principate or Late Antiquity being given a formal place on the consilium or consistorium: Agrippina, for instance, had to listen to the senate from behind a curtain.³¹⁷ Nor did imperial mothers or other imperial women ever formally exercise powers as a ‘regent’ for minority emperors in the Principate or Late Antiquity, unlike many other monarchical societies in which such authority was delegated to them by law.³¹⁸ In these cases, royal mothers sat on state councils, such as Louis XIII’s mother Marie de’ Medici, who retained a place on the council even after her regency ended.³¹⁹ The formal position of regent and the creation of regency councils for minority Roman emperors did not occur until the seventh century.³²⁰ By referring to fifth-century imperial women such as Pulcheria and Galla Placidia as ‘regents’ for their brother and son, respectively, we occlude the fact that the rule of the empire was actually acknowledged as a partnership between the emperor and his magister militum, as Meaghan McEvoy demonstrated in Chapter 8. The ³¹⁴ Drijvers 1992b; Herrin 2013: 2, 139–40. ³¹⁵ On the relatively low life expectancy of imperial women, see Scheidel 1999: 257–8. ³¹⁶ Barrett 1996: 173–4. ³¹⁷ Tac. Ann. 13.5.1. The presence of women in imperial advisory councils under Claudius has sometimes been maintained (e.g. Crook 1955: 43–4; Drinkwater 2019: 68–9), but this is based on fictional evidence from the Acta Alexandrinorum. Hdn. 4.3.5–9 places Julia Domna on a council called when Caracalla and Geta wanted to divide the empire between them, but this incident is also fictional. It is telling that even when Severus Alexander’s mother and grandmother selected his advisors, they were all men (Hdn. 6.1.2). For Late Antiquity, see especially Harries 2013 on the exclusion of women from the consistorium. Smith 2007: 229–30 argues that under child emperors, ‘it was not the elite officials of the consistory who in practice ruled collectively in the emperor’s name, but usually an empress . . . or else a general’. But imperial women did not have any formal legislative power. For example, Ambrose does not claim that Justina, the mother of Valentinian II, attended meetings of the consistorium, which shows that her authority must have been exercised in a different way. For this situation, see McEvoy 2013: 124–5; 2016c. ³¹⁸ For the Principate, see Hug 2022: 73. For Late Antiquity, the arguments against ‘regents’ were originally presented by Mommsen 1903, and see also Holum 1982: 29–30; McEvoy 2013: 9–12. Those who have recently used to the term ‘regent’ to describe fifth-century imperial women such as Galla Placidia and Pulcheria include Cooper 2004: 51; Chew 2006: 212; Sivan 2011 (passim); Salisbury 2015 (passim); Leonard 2019: 341. It has even been enshrined in the Oxford Dictionary of Late Antiquity (Nathan 2018). ³¹⁹ Moote 1989: 168. ³²⁰ Herrin 2013: 230–1; Bartlett 2020: 117.

      

355

use of the term ‘regent’ prevents us from conceptualizing how these women really did exercise political influence.³²¹ Angela Hug has pointed the way to a better appreciation of such relationships in terms of ‘broader cultural expectations of respect and obedience towards one’s parents’.³²² Moreover, it is surely more accurate to say that Galla Placidia was able to play a role at the court of her son Valentinian III not because she was a ‘regent’, but through the workings of the ‘politics of intimacy’, since, by virtue of her relationship with the emperor, she was able to speak to him and influence him in private. Aristocratic women were regarded as an integral part of the social world of the Roman imperial court, though they did not have any formal positions or titles, such as ‘ladies-in-waiting’, found in some other monarchical societies.³²³ Female relatives of senators and equites did visit and develop friendships with imperial women, as shown by the relationships cultivated by Cratia, wife of Marcus Aurelius’ tutor, Cornelius Fronto. Cratia was known to have attended on both Marcus’ mother Domitia Lucilla and his wife Faustina the Younger.³²⁴ These women, as Benjamin Kelly demonstrated in Chapter 5, were enveloped into the wider emotional community of the court, bound together by expressions of love and devotion. Aristocratic women were also viewed as important figures of influence who could secure outsiders favours or preferment because of their proximity to members of the imperial family. When Jerome wrote to Salvina to console her on the loss of her husband Nebridius, he revealed his fear that some might claim ‘we are wheedling our way into the royal court’ (aulae nos insinuare regali).³²⁵ The influence of these female networks could even extend beyond the court to other institutions. For example, Pulcheria and her female companions regularly used to have dinner at the patriarch’s palace in Constantinople after Sunday communion.³²⁶ The association between aristocratic women and imperial women strengthened and validated both groups as members of the court community and, by extension, the imperial monarchy.³²⁷ And so it was that when the Augusta Sophia wished to isolate Ino, the wife of the Caesar Tiberius, she forbade senatorial women from visiting Ino in the palace of Hormisdas. Ino left Constantinople and did not return until her husband became Augustus, his accession finally validating her own status.³²⁸

³²¹ For the female regent in Medieval Europe, see Bartlett 2020: 114–19. ³²² Hug 2022: 67. ³²³ Drinkwater 2019: 60 writes anachronistically when he refers to ‘ladies-in-waiting’ at the Neronian court. For ladies-in-waiting as a cultural phenomenon in other courts, see Duindam 2003: 59–60, 77–8, 94; Akkerman and Houben 2014. At the Ptolemaic court, there does seem to have been a way of recognizing aristocratic women at court, as female suntrophoi, or ‘foster-sisters’ of the queen (Strootman 2014: 106). ³²⁴ Davenport and Manley 2014: 70–4, 99–100. Richlin 2011 treats the relationship between Cratia and Domitia Lucilla in very speculative terms, even wondering if their bond might be ‘lesbian-like’ (p. 177). It seems to us that ‘female friendship’ covers their relationship more accurately. ³²⁵ Jer. Ep. 79.1. ³²⁶ Letter to Cosmas 5 (ed. Nau 1919: 278); McLynn 2018: 17–18. ³²⁷ For the favours and protections that imperial women could bestow, see Kunst 2010: 157–60. ³²⁸ Joh. Eph. HE 3.8.

356

   

The role of women in the court community did undergo one significant change between the Principate and Late Antiquity. In the early empire, all women who occupied official positions in the imperial household were slaves or freedwomen. These included ornatrices (hairdressers), vestiplicae/vestispicae (wardrobe attendants), and unctrices (masseurs).³²⁹ From the third century onwards, there was a move to replace slave and freedmen household and administrative officials with freeborn appointees. This opened up the path for the selection of free female cubiculariae (bedroom attendants), who are first attested in the fourth century. Previously, the women of both imperial and aristocratic households had only male cubicularii.³³⁰ It has been plausibly argued that Marcella and Droseria, the cubiculariae of Pulcheria who received bribes from Cyril of Alexandria, were probably wealthy aristocratic women.³³¹ A near-contemporary of these women, a certain Euphemia, who consecrated a shrine to St Christopher in 450, is styled ‘the most dignified cubicularia’ (σεμνοπρ(επεστάτης) κουβικουλαρ(ίας)) in the dedicatory inscription.³³² Therefore, although we cannot reconstruct the precise details, or where the initiative came from, there appears to have been conscious effort in Late Antiquity to formalize the appointment of aristocratic women to the households of female members of the imperial family. In the reign of Theodosius II, his sisters Pulcheria, Arcadia, and Marina each had their own palaces in Constantinople, increasing the number opportunities for aristocratic women to become a formal part of the court as cubiculariae.³³³ As we move later into the Byzantine period, the role and place of women at court was formalized further. Aristocratic women were awarded the feminine versions of court titles granted to their male counterparts, and a separate ‘court of women’ developed by the ninth century, a natural evolution of the processes described above.³³⁴ This returns us to the point we have already made: by this period, Byzantium was truly a Medieval empire, and its court society shared many of the same aspects as other European courts of the period.³³⁵

6. Final Thoughts Rome began life as a monarchical state, but expelled its kings in the sixth century . Thereafter, the Roman Republic maintained a decidedly ambiguous attitude ³²⁹ Treggiari 1974: 78–80; Treggiari 1975: 52–3. ³³⁰ Treggiari 1974: 78, 100, citing CIL VI 9313–15 as late antique examples; Treggiari 1975: 52 for Livia’s cubicularii. ³³¹ Holum 1982: 182. Garland 1999: 242 and Herrin 2013: 226 define cubiculariae as ladies-inwaiting. ³³² PLRE II Euphemia 1; Feissel 1984: 566–71 (text published on 568). ³³³ Janin 1964: 136–7. ³³⁴ Const. Porph. de cer. 1.40 (1.49) (τὸ σέκρετον τῶν γυναικῶν); Kazhdan and McCormick 1997: 182–5; Garland 1999: 4–5. The extent to which this court of women existed earlier is debated: empresses such as Theodora had high-ranking female retinues, but it is difficult to determine whether or not this counted as a ‘court’ (sekreton) (McClanan 2002: 130–2). ³³⁵ Herrin 2013: 222.

      

357

towards kings and kingship: royal power and pomp survived in the curule magistracies, such as the praetorship and consulship, but they were term-limited and subject to election.³³⁶ When one-man rule returned to Rome under Augustus, it was in a form that likewise capitalized on the ambiguity of the princeps’ position.³³⁷ The court and court society evolved from what was in effect Augustus’ household, which was similar to, but greater than, those of his aristocratic peers.³³⁸ The transition from the Republic to the imperial monarchy played a significant role in delineating the shape of the Roman imperial court, in contrast with courts in societies that had always been monarchical states.³³⁹ From the reign of Augustus onwards, decisions had to be made, either consciously or unconsciously, about the configuration of the court: whether or not to formalize membership of the court community, and if so, in what way(s); whether or not to create specific court posts, and if so, who should be appointed to them; whether or not women should play informal or formal roles at court, and if so, how this should be regulated; and so on. The emperor could control and influence some of these aspects, but not all of them: more often than not, the change and evolution of the Roman imperial court took place in response to other forces, both inside and outside the court community. The decision to have an imperial college and multiple imperial courts in the late third and fourth centuries was made by emperors, but we should not forget that it was in reaction to the need to defend the frontiers and assure provincials that they were protected. The Tetrarchic palaces all had audience halls which magnified the emperors’ sacredness and distance from their subjects, but the foundations for this change had been laid with performances of flattery by courtiers for several centuries. Sometimes the weight of tradition played a role in maintaining the status quo: the relatively permeable nature of the court society reflected the practices and structure of Republican households, so that when there was some attempt to define membership of the court, it only came very late, in the fourth century. But tradition was a powerful force, and there had to be specific, compelling reasons to institute changes, as both Peter Bang and Olivier Hekster have shown.³⁴⁰ Hence, often there was much slower, progressive development over centuries, such as the gradual transformation of jurists from occasional consultants into bona fide courtiers. Even more dramatic changes, such as the influence of generals during the ‘age of the magister militum’ or late antique imperial women being given the power to choose the next emperor, only occurred when they were supported by the consensus of the court community at large.

³³⁶ ³³⁸ ³³⁹ ³⁴⁰

Rawson 1975. ³³⁷ Wallace-Hadrill 1982. On Republican precedents, see Rilinger 1997; Potter 2011. Winterling 1997b and Bang 2011 offer two different takes on this problem. Bang 2011; Hekster 2015.

358

   

The adoption of Christianity by the emperor and his family is the defining difference which separates the Principate from the world of Late Antiquity. Its impact on the court community and structures of patronage and access was complex and multifaceted. Bishops and other ecclesiastical figures had to compete with other non-Christian elements at court for access to, and intimacy with, the emperor. There was no attempt in our period to offer any religious figures formal positions at court, even after the embedding of the court in Constantinople in the fifth century and the beginning of the construction of chapels within the palace complex. Stories set at court featuring Christian figures, from bishops to holy men, reflect some of these frustrations. The fractured and unstable nature of the Christian community may well have prompted emperors to keep the court separate from the ecclesiastical hierarchy. Yet at the same time, Christianity made a definite impact on the daily ritual of the court, from morning prayers and the celebration of feast days, and on dynastic state ceremonial, such as imperial accessions and funerals, in which the new imperial religion became increasingly central. These conclusions do not seek to reduce the importance of the Roman emperor himself. The personality and behaviour of emperors set the emotional tone of their courts. The ages, marriages, and fertility of emperors could shape the structure of a court community and the role of individuals within it. All these factors determined whether the emperor was accessible, on which occasions, and to whom, as well as his choice of intimate confidants. Our aim, therefore, has been to set the emperor, his decisions, and his relationships in the wider context of the court community, in order to highlight the symbiotic and interconnected nature of the emperor and his court.

Bibliography Akkerman, N. and Houben, B. (eds) 2014. The Politics of Female Households: Ladies-inWaiting Across Early Modern Europe. Leiden and Boston. Alföldi, A. 1970. Die monarchische Repräsentation im römischen Kaiserreiche. Darmstadt. Andaya, B. 1993. ‘Political development between the sixteenth and eighteenth centuries’, in N. Tarling (ed.), The Cambridge History of Southeast Asia. Volume 1: From Early Times to c. 1800. Cambridge: 402–59. Anderson, G. 1986. Philostratus: Biography and Belles Lettres in the Third Century. London. Ando, C. 2000. Imperial Ideology and Provincial Loyalty in the Roman Empire. Berkeley. Andriollo, L. 2020. ‘Imperial adjudication in Late Antiquity: evolutions and perceptions in the light of documentary evidence’, Studia Historica. Historia Antigua 38: 245–72. Angelova, D. N. 2004. ‘The ivories of Ariadne and ideas about female imperial authority in Rome and early Byzantium’, Gesta 43: 1–15. Angelova, D. N. 2015. Sacred Founders. Women, Men, and Gods in the Discourse of Imperial Founding, Rome through Early Byzantium. Berkeley. Appelbaum, A. 2007. ‘I Clothed You in Purple’. The Rabbinic King-Parables of the Third Century. PhD Dissertation. Yale. Arce, J. 1984. ‘Los funerales del emperador Constancio II (a. 361 d.C.)’, in P. Bádenas de la Peña (ed.), ATHLON. SATVRA GRAMMATICA. In honorem Francisci R. Adrados, vol. 2. Madrid: 29–39. Arce, J. 1988. Funus Imperatorum. Los funerales de los emperadores romanos. Madrid. Arce, J. 2000. ‘Imperial funerals in the later Roman empire: change and continuity’, in F. Theuws and J. L. Nelson (eds), Rituals of Power. From Late Antiquity to the Early Middle Ages. Leiden and Boston: 115–29. Asch, R. G. 1991. ‘Introduction: court and household from the fifteenth to the seventeenth centuries’, in R. G. Asch and A. M. Birke (eds), Princes, Patronage, and the Nobility: The Court at the Beginning of the Modern Age, c. 1450–1650. Oxford: 1–38. Asch, R. G. 2016. ‘Patronage, friendship, and the politics of access: the role of the early modern favourite revisited’, in D. Raeymaekers and S. Derks (eds), The Key to Power? The Culture of Access in Princely Courts, 1400–1750. Leiden and Boston: 178–201. Avery, W. T. 1940. ‘The “adoratio purpurae” and the importance of the imperial purple in the fourth century of the Christian era’, MAAR 17: 66–80. Balty, J. 1982. ‘Paedagogiani—pages de Rome à Byzance’, in L. Hadermann-Misguich and G. Raepsaet (eds), Rayonnement grec. Hommages à Charles Delvoye. Brussels: 299–312. Bang, P. F. 2011. ‘Court and state in the Roman empire: domestication and tradition in comparative perspective’, in J. Duindam, T. Artan, and M. Kunt (eds), Royal Courts in Dynastic States and Empires. A Global Perspective. Leiden: 103–28. Bajramovic, G. 2011. ‘Pride, pomp and circumstances: palace, court and household in Assyria 879–612 BC’, in J. Duindam, T. Artan, and M. Kunt (eds), Royal Courts in Dynastic States and Empires. A Global Perspective. Leiden: 27–61. Barnes, T. D. 1975. ‘Constans and Gratian in Rome’, HSCP 79: 325–33. Barnes, T. D. 1976. ‘Sossianus Hierocles and the antecedents of the “Great Persecution” ’, HSCP 80: 239–52.

360



Barnes, T. D. 1981. Constantine and Eusebius. Cambridge, MA. Barnes, T. D. 1982. The New Empire of Diocletian and Constantine. Cambridge, MA. Barnes, T. D. 1986. ‘Synesius in Constantinople’, GRBS 27: 93–112. Barnes, T. D. 1989. ‘Christians and pagans in the reign of Constantius’, in A. Dihle (ed.), L’Église et l’empire au IVe siècle. Sept exposés suivis de discussions. Geneva: 301–43. Barnes, T. D. 1993. Athanasius and Constantius. Theology and Politics in the Constantinian Empire. Cambridge, MA. Barnes, T. D. 2011. Constantine. Dynasty, Religion and Power in the Later Roman Empire. Oxford. Barnes, T. D. 2016. Early Christian Hagiography and Roman History. Second edition. Tübingen. Barrett, A. 1993. Caligula: The Corruption of Power. London. Barrett, A. 1996. Agrippina: Mother of Nero. London. Bartlett, R. 2020. Blood Royal: Dynastic Politics in Medieval Europe. Cambridge. Bauer, F. A. 1996. Stadt, Platz und Denkmal in der Spätantike. Untersuchungen zur Ausstattung des öffentlichen Raums in den spätantiken Städten Rom, Konstantinopel und Ephesos. Mainz. Beard, M. 2007. The Roman Triumph. Cambridge. Beaujeu, J. 1955. La religion romaine à l’apogée de l’empire. I. La politique religieuse des Antonins (96–192). Paris. Becker, A. 2016. ‘Dieu et le couronnement des empereurs protobyzantins’, in M.-F. Baslez and C.-G. Schwentzel (eds), Les dieux et le pouvoir. Aux origines de la théocratie. Rennes: 143–56. Becker, A. 2018. ‘Verbal and nonverbal diplomatic communication at the imperial court of Constantinople (fifth-sixth centuries)’, DOP 72: 79–92. Becker, A. 2022. Dieu, le souverain et la cour. Stratégies et rituels de légitimation du pouvoir impérial et royal dans l’Antiquité tardive et au haut Moyen Âge. Bordeaux. Beeley, C. A. 2012. The Unity of Christ: Continuity and Conflict in Patristic Tradition. New Haven. Bekker, I. (ed.) 1838–9. Georgius Cedrenus. Bonn. Bell, C. 1992. Ritual Theory, Ritual Practice. Oxford and New York. Bell, H. I. 1944. ‘An Egyptian village in the age of Justinian’, JHS 64: 21–36. Bell, P. (ed.) 2009. Three Political Voices from the Age of Justinian. Liverpool. Bellen, H. 1981. Die germanische Leibwache der römischen Kaiser des julisch-claudischen Hauses. Wiesbaden. Ben Zeev, M. P. 2005. Diaspora Judaism in Turmoil, 116/117 C.E.: Ancient Sources and Modern Insights. Leuven. Benoist, S. 2005. Rome, le prince et la Cité. Pouvoir imperial et cérémonies publiques (Ier siècle av.–début du IVe siècle apr. J.-C.). Paris. Bensammar, E. 1976. ‘La titulature de l’imperatrice e sa signification’, Byzantion 46: 243–91. Bérenger, J. 1974. ‘Pour une enquête européenne: le problème du ministériat au XVIIe siècle’, Annales 29: 166–92. Berger, A. (ed.) 2013. Accounts of Medieval Constantinople. The Patria. Cambridge, MA. Bergmann, M. 1998. Die Strahlen der Herrscher. Theomorphes Herrscherbild und politische Symbolik im Hellenismus und in der römischen Kaiserzeit. Mainz. Bernard, P. 2017. ‘Rapport introductif ’, in S. Destephen (ed.), L’évêque de cour: figure politique, figure polémique. Paris: 5–34.



361

Bernstein, N. W. 2022. ‘Literary patronage and the Roman imperial court from Augustus to the Severan dynasty’, in in B. Kelly and A. Hug (eds), The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge: 438–60. Betegh, G. and Gregoric, P. 2014. ‘Multiple analogy in Ps.-Aristotle, De mundo 6’, CQ 64: 574–91. Beyeler, M. 2011. Geschenke des Kaisers. Studien zur Chronologie, zu den Empfängern und zu den Gegenständen der kaiserlichen Vergabungen im 4. Jahrhundert n. Chr. Berlin. Bileta, V. 2019. ‘The venatio in the emperor’s presence? The consistorium and the military men of the late Roman empire in the West’, in K. C. Choda, M. S. de Leeuw, and F. Schulz (eds), Gaining and Losing Imperial Favour in Late Antiquity: Representation and Reality. Leiden: 73–101. Bingham, S. 1999. ‘Security at the games in the early imperial period’, Echos du Monde Classique/Classical Views 18: 369–79. Bingham, S. 2013. The Praetorian Guard. A History of Rome’s Elite Special Forces. London and New York. Birley, A. R. 1993. Marcus Aurelius: A Biography. London. Birley, A. R. 1997. Hadrian: The Restless Emperor. London. Birley, A. R. 2020. ‘Antoninus Pius’ Guard Prefect Marcus Gavius Maximus with an Appendix on New Evidence for the Fasti of Britain under Antoninus’, in D. J. Breeze and W. S. Hanson (eds), The Antonine Wall: Papers in Honour of Professor Lawrence Keppie. Archaeopress Roman Archaeology 64. Oxford: 313–31. Bleckmann, B. 1994. ‘Constantina, Vetranio und Gallus Caesar’, Chiron 24: 29–68. Blockley, R. C. 1972. ‘Constantius Gallus and Julian as Caesars of Constantius II’, Latomus 31: 433–68. Blockley, R. C. (ed.) 1983. The Fragmentary Classicizing Historians of the Later Roman Empire. Eunapius, Olympiodorus, Priscus and Malchus. Volume 2. Translation and Historiographical Notes. Liverpool. Boak, A. E. R. 1915. ‘The Roman magistri in the civil and military service of the empire’, HSCPh 26: 73–164. Boatwright, M. T. 2021. Imperial Women of Rome. Power, Gender, Context. Oxford. Bonfil, R. 2009. History and Folklore in a Medieval Jewish Chronicle. The Family Chronicle of Ahima’az ben Paltiel. Leiden and Boston. : Boschung, D. and Eck, W. (eds) 2006. Die Tetrarchie. Ein neues Regierungssystem und seine mediale Präsentation. Wiesbaden. Bourdieu, P. 1982. ‘Les rites comme actes d’institution’, Actes de la recherche en sciences sociales 43: 58–63. Bowersock, G. W. 1969. Greek Sophists in the Roman Empire. Oxford. Bowersock, G. W. 1978. Julian the Apostate. London. Bowes, K. 2008. Private Worship, Public Values, and Religious Change in Late Antiquity. Cambridge. Bowie, E. L. 2009. ‘Philostratus: the life of a sophist’, in E. L. Bower and J. Elsner (eds), Philostratus. Cambridge: 19–32. Bravo Castañeda, G. 1997. ‘El ritual de la proskynesis y su significado político y religioso en la Roma imperial (con especial referencia a la Tetrarquía)’, Gerión 15: 177–91. Bremmer, J. N. 1988. ‘An imperial palace guard in Heaven: the date of the Vision of Dorotheus’, ZPE 75: 82–8. Bremmer, J. N. 2002. The Rise and Fall of the Afterlife. The 1995 Read-Tuckwell Lectures at the University of Bristol. London.

362



Brenk, B. 1996. ‘Innovation im Residenzbau der Spätantike’, in B. Brenk (eds), Innovation in der Spätantike. Wiesbaden: 67–114. Brennecke, H. C. 2006. ‘Zwei Apologien des Athanasius an Kaiser Constantius II’, Zeitschrift für antikes Christentum 10: 67–85. Brett, G. 1954. ‘The automata in the Byzantine “Throne of Solomon” ’, Speculum 29: 477–87. Briant, P. 2002. From Cyrus to Alexander: A History of the Persian Empire. Trans. P. T. Daniels. Winona Lake, IN. Briant, P. 2015. Darius in the Shadow of Alexander. Trans. J. M. Todd. Cambridge. Brockliss, L. W. B. 1999. ‘Concluding remarks: the anatomy of the minister-favourite’, in J. H. Elliott and L. W. B. Brockliss (eds), The World of the Favourite. New Haven and London: 279–309. Brooks, E. W. 1893. ‘The Emperor Zenon and the Isaurians’, EHR 30: 209–38. Brooks, E. W. 1923. ‘John of Ephesus: Lives of the Eastern Saints (I)’, in R. Griffin and F. Nau (eds), Patrologia Orientalis. Tomus Decimus Septimus. Paris: 1–306. Brosius, M. 2007. ‘New out of old? Court and court ceremonies in Achaemenid Persia’, in A. J. S. Spawforth (ed.), The Court and Court Society in Ancient Monarchies. Cambridge: 17–57. Brosius, M. 2010. ‘Das Hofzeremoniell’, in B. Jacobs and R. Rollinger (eds), Der Achämenidenhof/The Achaemenid Court. Wiesbaden: 459–71. Brottier, L. 1996. ‘L’impératrice Eudoxie et ses enfants’, Revue des sciences religieuses 70: 313–32. Brown, P. 1971. ‘The rise and function of the holy man in Late Antiquity’, JRS 61: 80–121. Brown, P. 1992. Power and Persuasion in Late Antiquity: Towards a Christian Empire. Madison. Brown, P. 2002. Poverty and Leadership in the Later Roman Empire. Hanover, NH. Brown, R. 2000. Group Processes: Dynamics within and between Groups. Oxford. Brown, R. 2020. ‘The social identity approach: appraising the Tajfellian legacy’, British Journal of Social Psychology 59: 5–25. Browning, R. 1976. The Emperor Julian. London. Brunt, P. A. 1974. ‘Marcus Aurelius in his Meditations’, JRS 64: 1–20. Bühl, G. 2001. ‘Eastern or western?—that is the question. Some notes on the new evidence concerning the eastern origin of the Halberstadt diptych’, Acta ad archaeologiam et artium historiam pertinentia 15: 193–203. Burgess, R. W. 1993/1994. ‘The accession of Marcian in the light of Chalcedonian apologetic and monophysite polemic’, BZ 86/87: 47–68. Burgess, R. W. 2008. ‘The summer of blood: the “great massacre” of 337 and the promotion of the sons of Constantine’, DOP 62: 5–51. Burke, P. 1992. The Fabrication of Louis XIV. New Haven. Burke, P. 2009. Popular Culture in Early Modern Europe. Third edition. London. Burton, G. P. 1975. ‘Proconsuls, assizes and the administration of justice under the empire’, JRS 65: 92–106. Bury, J. B. 1907. ‘The ceremonial book of Constantine Porphyrogenitus’, EHR 22: 209–27 and 417–39. Bury, J. B. 1910. ‘Magistri Scriniorum, ἀντιγραϕῆς and ῥεϕερενδάριοι’, HSCP 21: 23–9. Bury, J. B. 1919. ‘Justa Grata Honoria’, JRS 9: 1–13. Busch, A. 2015. Die Frauen der theodosianischen Dynastie. Macht und Repräsentation kaiserlicher Frauen im 5. Jahrhundert. Stuttgart.



363

Busch, A. 2021. ‘Praise and Polemics. Theodosian Empresses and their Contemporaries’, in M. Vincent (ed.), Studia Patristica Vol. CXXX. Leuven, Paris, and Bristol, CT: 197–214. Butler, L. 2002. Emperor and Aristocracy in Japan, 1467–1680: Resilience and Renewal. Cambridge, MA. Callu, J.-P. 1992. ‘La dyarchie constantinide (340–350): les signes d’évolution’, in M. Chistol, S. Demougin, and Y. Duval (eds), Institutions, société et vie politique dans l’empire romain au IVᵉ siècle ap. J.-C. Paris: 39–68. Cameron, A. D. E. 1969. ‘Theodosius the Great and the regency of Stilicho’, HSCPh 73: 247–80. Cameron, A. D. E. 1970. Claudian: Poetry and Propaganda at the Court of Honorius. Oxford. Cameron, A. D. E. 1973. Porphyrius the Charioteer. Oxford. Cameron, A. D. E. 1976. Circus Factions: Blues and Greens at Rome and Byzantium. Oxford. Cameron, A. D. E. 1978. ‘The house of Anastasius’, GRBS 19: 259–76. Cameron, A. D. E. 1998. ‘Consular diptychs in their eastern context: new eastern evidence’, JRA 11: 384–403. Cameron, A. D. E. 2013. ‘The origin, context, and function of consular diptychs’, JRS 103: 174–207. Cameron, A. D. E. 2015. ‘City personifications and consular diptychs’, JRS 105: 250–87. Cameron, A. D. E. and Long, J., with a contribution by Sherry, L. 1993. Barbarians and Politics at the Court of Arcadius. Berkeley. Cameron, A. M. 1975. ‘The Empress Sophia’, Byzantion 45: 5–21. Cameron, A. M. (ed.) 1976. Flavius Cresconius Corippus. In laudem Iustini Augusti minoris. London. Cameron, A. M. 1979. ‘Images of authority: elites and icons in late sixth-century Byzantium’, P&P 84: 3–35. Cameron, A. M. 1987. ‘The construction of court ritual: the Byzantine Book of Ceremonies’, in D. Cannadine and S. Price (eds), Rituals of Royalty. Power and Ceremonial in Traditional Societies. Cambridge: 106–36. Cameron, A. M. 2000. ‘Form and meaning: the Vita Constantini and the Vita Antonii’, in T. Hägg and P. Rousseau (eds), Greek Biography and Panegyric in Late Antiquity. Berkeley: 72–88. Cameron, A. M. and Hall, S. G. (eds) 1999. Eusebius: Life of Constantine. Oxford. Campbell, J. B. 1984. The Emperor and the Roman Army, 31 BC-AD 235. Oxford. Canard, M. 1951. ‘Le cérémonial fatimite et le cérémonial byzantine: essai de comparaison’, Byzantion 21: 355–420. Canepa, M. 2009. The Two Eyes of the Earth: Art and Ritual of Kingship between Rome and Sasanian Iran. Berkeley. Cannadine, D. 1993. ‘The context, performance and meaning of ritual: the British monarchy and the “invention of tradition”, c. 1820–1977’, in E. Hobsbawn and T. Ranger (eds), The Invention of Tradition. Cambridge: 101–64. Caporusso, D., Donati, M. T., Masseroli, S., and Tibiletti, T. 2014. Immagini di Mediolanum. Archeologia e storia di Milano dal V secolo a. C. al V secolo d. C. Milan. Carile, A. 1998. ‘EUTAXIA: l’ordine divino nel cosmo e nell’impero’, in P. Catalano and P. Siniscalco (eds), Spazio e conetralizazione del potere. Roma: 131–6. Carney, E. D. 1991. ‘What’s in a name? The emergence of a title for royal women in the Hellenistic period’, in S. B. Pomeroy (ed.), Women’s History and Ancient History. Chapel Hill, NC: 154–72.

364



Carson, R. A. G. 1965. ‘Gold medallions of the reign of Maxentius’, Congresso internazionale di numismatica, Roma 11–16 settembre 1961. Atti. Vol. 2. Rome: 347–52. Champlin, E. 1980. Fronto and Antonine Rome. Cambridge, MA. Chaniotis, A. 2011. ‘Emotional community through ritual: initiates, citizens, and pilgrims as emotional cmmunities in the Greek world’, in A. Chaniotis (ed.), Ritual Dynamics in the Ancient Mediterranean: Agency, Emotion, Gender, Representation. Stuttgart: 264–90. Chaniotis, A. 2016. ‘Creating emotional community: the epigraphic evidence’, in E. Sanders, and M. Johncock (eds), Emotion and Persuasion in Classical Antiquity. Stuttgart: 93–111. Chastagnol, A. 1985. ‘Autour du thème du princeps clausus’, in J. Straub (ed.), Bonner Historia-Augusta-Colloquium, 1982/1983. Bonn: 149–61. Chastagnol, A. 1992. Le Sénat romain à l’époque impériale: recherches sur la composition de l’assemblée et la statut de ses membres. Paris. Chen, A. H. 2016. ‘Rival powers, rival images: Diocletian’s palace at Split in light of Sasanian palace design’, in D. Slootjes and M. Peachin (eds), Rome and the Worlds Beyond its Frontiers. Leiden and Boston: 213–42. Chen, A. H. 2018. ‘Omitted empresses: the (non-)role of imperial women in Tetrarchic propaganda’, JLA 11: 42–82. Chenault, R. 2012. ‘Statues of senators in the Forum of Trajan and the Roman Forum in Late Antiquity’, JRS 102: 103–32. Chew, K. 2006. ‘Virgins and eunuchs: Pulcheria, politics, and the death of emperor Theodosius II’, Historia 55: 207–27. Chrol, E. D. and Blake, S. 2022. ‘Sexuality and the court’, in B. Kelly and A. Hug (eds), The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge: 349–70. Clauss, M. 1980. Der magister officiorum in der Spätantike. Munich. Clerq, V. C. de 1954. Ossius of Cordova. A Contribution to the History of the Constantinian Period. Washington. Clover, F. M. 1971. ‘Flavius Merobaudes: a translation and historical commentary’, TAPA 61: 1–78. Clover, F. M. 1973. ‘Geiseric and Attila’, Historia 22: 104–17. Coarelli, F. 1999. ‘L’edilizia pubblica a Roma in età tetrarchica’, in W. Harris (ed.), The Transformations of Urbs Roma in Late Antiquity. Portsmouth, RI: 23–33. Coleman, K. M. (ed.) 1988. Statius: Silvae IV. Oxford. Conant, J. 2012. Staying Roman: Conquest and Identity in Africa and the Mediterranean, 439–700. Cambridge. Connolly, S. 2010. Lives Behind the Laws: The World of the Codex Hermogenianus. Bloomington. Cook, G. M. (ed.) 1942. The Life of Saint Epiphanius by Ennodius. Washington, DC. Cook, M. 2011. ‘Did the Prophet Muhammad keep court?’, in A. Fuess and J.-P. Hartung : (eds), Court Cultures in the Muslim World. London: 23–9. Cooper, K. 1998. ‘Contesting the nativity: wives, virgins, and Pulcheria’s imitatio Mariae’, Scottish Journal of Religious Studies 19: 31–43. Cooper, K. 2004. ‘Empress and Theotokos: gender and patronage in the Christological controversy’, Studies in Church History 39: 39–51. Corcoran, S. 2000. The Empire of the Tetrarchs. Imperial Pronouncements and Government, AD 283–324. Revised edition. Oxford. Corcoran, S. 2017. ‘Maxentius: a Roman emperor in Rome’, AnTard 25: 59–74. Courtney, E. (ed.) 1993. The Fragmentary Latin Poets. Oxford. Cox Miller, P. 1983. Biography in Late Antiquity: A Quest for the Holy Man. Berkeley.



365

Cox Miller, P. 2000. ‘Strategies of representation in collective biography: constructing the subject as holy’, in T. Hägg and P. Rousseau (eds), Greek Biography and Panegyric in Late Antiquity. Berkeley: 209–54. Crabbe, A. M. 1981. ‘The invitation list to the Council of Ephesus and metropolitan hierarchy in the fifth century’, JThS 32: 369–400. Creed, J. L. (ed.) 1984. Lactantius. De Mortibus Persecutorum. Oxford. Cribiore, R. 2007. The School of Libanius in Late Antique Antioch. Princeton. Croke, B. (ed.) 1995. The Chronicle of Marcellinus. Byzantina Australiensia 7. Canberra. Croke, B. 2005a. ‘Dynasty and ethnicity: Emperor Leo I and the eclipse of Aspar’, Chiron 35: 147–203. Croke, B. 2005b. ‘Leo I and the palace guard’, Byzantion 75: 117–51. Croke, B. 2006. ‘Justinian, Theodora, and the Church of Saints Sergius and Bacchus’, DOP 60: 25–63. Croke, B. 2007. ‘Justinian under Justin: reconfiguring a reign’, BZ 100: 13–56. Croke, B. 2010. ‘Reinventing Constantinople: Theodosius I’s imprint on the imperial city’, in S. McGill, C. Sogno, and E. Watts (eds), From the Tetrarchs to the Theodosians: Later Roman History and Culture, 284–250 CE. Cambridge: 241–64. Croke, B. 2015a. ‘Ariadne Augusta: shaping the identity of the early Byzantine empress’, in G. Dunn and W. Mayer (eds), Christians Shaping Identity from the Roman Empire to Byzantium. Leiden and Boston: 293–320. Croke, B. 2015b. ‘Dynasty and aristocracy in the fifth century’, in M. Maas (ed.), The Cambridge Companion to the Age of Attila. Cambridge: 98–124. Crone, P. 2015. Pre-Industrial Societies. Anatomy of the Pre-Modern World. Third edition. London. Crook, J. 1955. Consilium Principis. Imperial Councils and Counsellors from Augustus to Diocletian. Cambridge. Cullhed, M. 1994. Conservator Urbis Suae. Studies in the Politics and Propaganda of the Emperor Maxentius. Stockholm. Ćurčić, S. 1993. ‘Late-Antique palaces: the meaning of urban context’, Ars Orientalis 23: 67–89. Curran, J. 2000. Pagan City and Christian Capital. Rome in the Fourth Century. Oxford. Dagron, G. 1974. Naissance d’une capitale: Constantinople et ses institutions de 330 à 451. Paris. Dagron, G. 1996. Empereur et prêtre. Étude sur le césaropapisme byzantine. Paris. Translated into English by J. Birrell and published as Dagron, G. 2003. Emperor and Priest. The Imperial Office in Byzantium. Cambridge. Dagron, G. and Flusin, B. (eds) 2020. Constantin VII Porphyrogénète: Le livre des cérémonies. 5 vols. Paris. Davenport, C. 2010. ‘The building inscription from the fort at Udruh and Aelius Flavianus, tetrarchic praeses of Palaestina’, JRA 23: 349–57. Davenport, C. 2011. ‘Iterated consulships and the government of Severus Alexander’, ZPE 177: 281–8. Davenport, C. 2012a. ‘Cassius Dio and Caracalla’, CQ 62: 796–815. Davenport, C. 2012b. ‘The provincial appointments of the emperor Macrinus’, Antichthon 46: 184–203. Davenport, C. 2015. ‘Inscribing senatorial status and identity, A.D. 200–350’, in A. Kuhn (ed.), Social Status and Prestige in the Graeco-Roman World. Stuttgart: 269–90. Davenport, C. 2016. ‘Fashioning a soldier emperor: Maximian, Pannonia, and the panegyrics of 289 and 291’, Phoenix 70: 381–400.

366



Davenport, C. 2017. ‘Rome and the rhythms of imperial life from the Antonines to Constantine’, AnTard 25: 23–39. Davenport, C. 2019. A History of the Roman Equestrian Order. Cambridge. Davenport, C. 2020a. ‘Dying for justice: narratives of Roman judicial authority in the High Empire’, in A. König, R. Langlands, and J. Uden (eds), Literature and Culture in the Roman Empire, 96–235. Cambridge: 269–87. Davenport, C. 2020b. ‘The dynamics of imperial government: collegiality and regionalism’, in N. Baker-Brian and S. Tougher (eds), The Sons of Constantine, AD 337–361: In the Shadows of Constantine and Julian. London: 223–54. Davenport, C. 2021. ‘Giving voice to the late Roman emperor: Eumenius’s For the Restoration of the Schools (Pan. Lat. 9[4]) in context’, JLA 14: 9–28. Davenport, C. 2022. ‘The court and ceremonial’, in B. Kelly and A. Hug (eds), The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge: 288–317. Davenport, C. and Kelly, B. 2022. ‘Administration, finances, and the court’, in B. Kelly and A. Hug (eds), The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge: 115–45. Davenport, C. and Manley, J. (eds) 2014. Fronto: Selected Letters. London. Davies, P. J. E. 2000. Death and the Emperor. Roman Imperial Funerary Monuments from Augustus to Marcus Aurelius. Cambridge. de Blois, L. 2001. ‘Roman jurists and the crisis of the third century A.D. in the Roman empire’, in L. de Blois (ed.), Administration, Prosopography and Appointment Policies in the Roman Empire. Amsterdam: 136–53. de Cena, E. C. 2014. ‘Constantine in the imperial palace at Serdica’, in N. Lemajić (ed.), Constantine, Sirmium and Early Christianity. Sremska Mitrovica: 7–24. de la Bédoyère, G. 2017. Praetorian. The Rise and Fall of Rome’s Imperial Bodyguard. New Haven and London. de Leeuw, M. S. 2019. ‘Buying imperial favour: Cyril of Alexandria’s blessings’, in K. C. Choda, M. S. de Leeuw, and F. Schulz (eds), Gaining and Losing Imperial Favour in Late Antiquity. Representation and Reality. Leiden: 151–70. Déchaux, J.-H. 1993. ‘N. Elias et P. Bourdieu, analyse conceptuelle comparée’, Archives européennes de sociologie 34: 364–85. Delehaye, H. and Peeters, P. (eds) 1925. Acta Sanctorum. Novembris. Tomus IV. Quo dies nonus et decimus continentur. Brussels. Delmaire, R. 1995. Les institutions du Bas-Empire romain de Constantin à Justinien. Les institutions civiles palatines. Paris. Delmaire, R. 2008a. ‘Flavius Aetius, delatorum inimicissimus, uindex libertatis, pudoris ultor (CIL VI 41389)’, ZPE 166: 291–4. Delmaire, R. 2008b. ‘Les soldats de la garde impériale à l’époque théodosienne. Le témoignage des sources religieuses’, AnTard 16: 37–42. Demandt, A. 1970. ‘Magister Militum’, in Paulys Realencyclopädie der Classischen Altertumswissenschaft. Supplementband XII. Stuttgart: 553–789. Demandt, A. 1980. ‘Der spätrömische Militäradel’, Chiron 10: 609–36. den Boeft, J., Drijvers, J. W., Den Hengst, D. and Teitler, H. C. (eds) 1995. Philological and Historical Commentary on Ammianus Marcellinus XXII. Groningen. den Boeft, J., Drijvers, J. W., Den Hengst, D. and Teitler, H. C. (eds) 2014. Philological and Historical Commentary on Ammianus Marcellinus XXIX. Leiden. Destephen, S. 2016. Le voyage impérial dans l’Antiquité tardive. Paris.



367

Dey, H. 2015. The Afterlife of the Roman City. Architecture and Ceremony in Late Antiquity and the Early Middle Ages. Cambridge. Diefenbach, S. 1996. ‘Frömmigkeit und Kaiserakzeptanz im frühen Byzanz’, Saeculum 47: 35–66. Diefenbach, S. 2002. ‘Zwischen Liturgie und civilitas. Konstantinopel im 5. Jahrhundert und die Etablierung eines städtischen Kaisertums’, in R. Warland (ed.), Bildlichkeit und Bildorte von Liturgie: Schauplätze in Spätantike, Byzanz und Mittelalter. Wiesbaden: 21–50. Dignas, B. and Winter, E. 2007. Rome and Persia in Late Antiquity: Neighbours and Rivals. Cambridge. Dillon, J. N. 2012. The Justice of Constantine: Law, Communication, and Control. Ann Arbor. Dirschlmayer, M. 2015. Kirchenstiftungen römischer Kaiserinnen vom 4. bis zum 6. Jahrhundert—die Erschließung neuer Handlungsspielräume. Münster. Downey, G. 1963. Ancient Antioch. Princeton, NJ. Drake, H. A. 1988. ‘What Eusebius knew: the genesis of the Vita Constantini’, CPh 83: 20–38. Drake, H. A. 2000. Constantine and the Bishops. The Politics of Intolerance. Baltimore. Drake, H. A. 2011. ‘Intolerance, religious violence, and political legitimacy in Late Antiquity’, Journal of the American Academy of Religion 79: 193–235. Drijvers, J. W. 1992a. Helena Augusta. The Mother of Constantine the Great and the Legend of Her Finding of the True Cross. Leiden. Drijvers, J. W. 1992b. ‘Flavia Maxima Fausta: some remarks’, Historia 41: 500–6. Drijvers, J. W. 1993. ‘Helena Augusta: exemplary Christian empress’, Studia Patristica 24: 85–90. Drijvers, J. W. 2011. ‘Helena Augusta, the cross and the myth: some new reflections’, Millennium 8: 125–74. Drijvers, J. W. 2012. ‘Decline of political culture: Ammianus Marcellinus’ characterization of the reigns of Valentinian and Valens’, in D. Brakke, D. Deliyannis, and E. Watts (eds), Shifting Cultural Frontiers in Late Antiquity. London: 85–97. Drinkwater, J. F. 2019. Nero: Emperor and Court. Cambridge. Driver, G. R. and Hodgson, L. (eds) 1925. Nestorius. The Bazaar of Heracleides. Oxford. Duindam, J. 1995. Myths of Power. Norbert Elias and the Early Modern European Court. Amsterdam. Duindam, J. 2003. Vienna and Versailles. The Courts of Europe’s Dynastic Rivals 1550–1780. Cambridge. Duindam, J. 2004. ‘Norbert Elias and the history of the court: old questions, new perspectives’, in R. Butz, J. Hirschbiegel, and D. Willoweit (eds), Hof und Theorie: Annäherungen an ein historisches Phänomen. Cologne: 91–104. Duindam, J. 2011. ‘Royal courts in dynastic states and empires’, in J. Duindam, T. Artan, and M. Kunt (eds), Royal Courts in Dynastic States and Empires. A Global Perspective. Leiden and Boston: 1–23. Duindam, J. 2016. Dynasties: A Global History of Power, 1300–1800. Cambridge. Duindam, J., Artan, T., and Kunt, M. (eds) 2011. Royal Courts in Dynastic States and Empires. A Global Perspective. Leiden and Boston. Dumser, E. A. 2005. The Architecture of Maxentius. A Study in Architectural Design and Urban Planning in Early Fourth Century Rome. PhD dissertation, University of Pennsylvania. Duval, N. 1997. ‘Les résidences impériales: leur rapport avec les problèmes de légitimité, les partages de l’empire et la chronologie des combinaisons dynastiques’, in F. Paschoud and J. Szidat (eds), Usurpationen in der Spätantike. Stuttgart: 127–53.

368



Eberlein, J. J. 1982. Apparitio regis, revelatio veritatis. Studien zur Darstellung des Vorhangs in der bildenden Kunst von der Spätantike bis zum Ende des Mittelalters. Wiesbaden. Eck, W. 1992. ‘P. Aelius Apollonides, ab epistulis Graecis’, ZPE 91: 236–42. Eck, W. 1997. ‘Cum dignitate otium: senatorial domus in imperial Rome’, SCI 16: 162–90. Eck, W. 1998. ‘M. Cornelius Fronto, Lehrer Marc Aurels und consul suffectus im Jahr 142’, RhM 141: 193–6. Eck, W. 2006. ‘Der Kaiser und seine Ratgeber: Überlegungen zum inneren Zusammenhang von amici, comites und consiliarii am römischen Kaiserhof ’, in A. Kolb (ed.), Herrschaftsstrukturen und Herrschaftspraxis. Berlin: 67–77. Eck, W. 2007. ‘Eine historische Zeitenwende. Kaiser Konstantins Hinwendung zum Christentum und die gallischen Bischöfe’, in F. Schuller and H. Wolff (eds), Konstantin der Große. Kaiser einer Epochenwende. Lindenberg: 69–94. Eich, P. 2005. Zur Metamorphose des politischen Systems in der römischen Kaiserzeit: Die Entstehung einer ‘personalen Bürokratie’ im langen dritten Jahrhundert. Berlin. El Cheikh, N. M. 2004. Byzantium Viewed by the Arabs. Cambridge, MA. El Cheikh, N. M. 2014. ‘The institutionalization of ‘Abbāsid ceremonial’, in J. Hudson and A. Rodríguez (eds), Diverging Paths? The Shapes of Power and Institutions in Medieval Christendom and Islam. Leiden: 351–70. Elias, N. 1969. Die höfische Gesellschaft. Neuweid and Berlin. Elias, N. 2006. The Court Society. The Collected Works of Norbert Elias. Volume 2. Translated by E. Jephcott. Edited by S. Mennell. Dublin. Elias, N. 2012. On the Process of Cilivization. The Collected Works of Norbert Elias. Volume 3. Translated by E. Jephcott. Edited by S. Mennell. Dublin. Elkins, N. T. 2013. ‘A note on late Roman art: the provincial origins of camp gate and baldachin iconography on the late imperial coinage’, AJN 25: 283–302. Elliott, J. H. 1977. ‘Philip IV of Spain: prisoner of ceremony’, in A. G. Dickens (ed.), The Courts of Europe: Politics, Patronage and Royalty, 1400–1800. London: 169–89. Elliott, J. H. 1999. ‘Introduction’, in J. H. Elliott and L. W. B. Brockliss (eds), The World of the Favourite. New Haven and London: 1–10. Elliott, J. H. and Brockliss, L. W. B. (eds) 1999. The World of the Favourite. New Haven and London. Elliott, J. K. (ed.) 1993. The Apocryphal New Testament. Oxford. Elm, S. 2012. Sons of Hellenism, Fathers of the Church: Emperor Julian, Gregory of Nazianzus, and the Vision of Rome. Berkeley. Elman, B. A. 2000. A Cultural History of Civil Examinations in Late Imperial China. Berkeley. Elman, B. A. 2013. Civil Examinations and Meritocracy in Late Imperial China. Cambridge, MA. Elton, H. 2000. ‘Illus and the imperial aristocracy under Zeno’, Byzantion 70: 393–407. Elton, H. 2009. ‘Imperial politics at the court of Theodosius II’, in H. Cain and N. Lenski (eds), The Power of Religion in Late Antiquity. Aldershot: 133–42. Engels, D. 2017. ‘The Achaemenid and Seleucid court: continuity or change?’, in A. Erskine, L. Llewellyn-Jones, and S. Wallace (eds), The Hellenistic Court: Monarchic Power and Elite Society from Alexander to Cleopatra. Swansea: 69–100. Engemann, J. 1999. ‘Das spätantike Consulardiptychon in Halberstadt: westlich oder ostlich’, JAC 42: 158–68. Enßlin, W. 1942. ‘Zur Torqueskrönung und Schilderhebung bei der Kaiserwahl’, Klio 35: 268–98. Epstein, I. (ed.) 1932–52. The Babylonian Talmud. 35 vols. London.



369

Errington, J. 1982. ‘Speech in the royal presence: Javanese palace language’, Indonesia 34: 89–101. Errington, R. M. 2006. Roman Imperial Policy from Julian to Theodosius. Chapel Hill. Erskine, A., Llewellyn-Jones, L., and Wallace, S. 2017a. ‘Introduction’, in A. Erskine, L. Llewellyn-Jones, and S. Wallace (eds), The Hellenistic Court: Monarchic Power and Elite Society from Alexander to Cleopatra. Swansea: xv–xxx. Erskine, A., Llewellyn-Jones, L., and Wallace, S. (eds) 2017b. The Hellenistic Court: Monarchic Power and Elite Society from Alexander to Cleopatra. Swansea. Evans, J. A. S. 2001. The Age of Justinian. The Circumstances of Imperial Power. London. Evans, R. J. W. 1991. ‘The court: a protean institution and an elusive subject’, in R. G. Asch and A. M. Birke (eds), Princes, Patronage, and the Nobility: The Court at the Beginning of the Modern Age, c. 1450–1650. Oxford: 481–91. Färber, R. 2014. Römische Gerichtsorte. Räumliche Dynamiken von Jurisdiktion im Imperium Romanum. Munich. Farquharson, A. S. L. (ed.) 1944. The Meditations of the Emperor Marcus Aurelius, 2 vols. Oxford. Featherstone, J. M. 2006. ‘The Great Palace as reflected in the De Cerimoniis’, Byzas 5: 47–61. Feder, A. L. (ed.) 1916. S. Hilarii Episcopi Pictaviensis Opera, pars quarta. Vienna. Feeney, K. 2019. ‘The emperor is dead, long live the emperor: imperial interregna in the fifth century’, in J. W. Drijvers and N. Lenski (ed.), The Fifth Century: Age of Transformation. Bari: 185–96. Feissel, D. 1984. ‘Notes d’épigraphie chrétienne VII’, BCH 108: 545–79. Feissel, D. 2010. ‘Trois notes sur l’empereur Maurice’, TM 16: 253–72. Fitzgerald, A. (ed.) 1930. The Essays and Hymns of Synesius of Cyrene, Including the Address to the Emperor Arcadius and the Political Speeches. Oxford. Flaig, E. 1992. Den Kaiser herausfordern: Die Usurpation im Römischen Reich. Frankfurt. Flaig, E. 2019. Den Kaiser herausfordern: Die Usurpation im Römischen Reich. Second edition. Frankfurt. Fleury, P. 2006. Lectures de Fronton: Un rhéteur latin à l’époque de la Seconde Sophistique. Paris. Flinterman, J.-J. 2004. ‘Sophists and emperors: a reconnaissance of sophistic attitudes’, in B. E. Borg (ed.), Paideia: The World of the Second Sophistic. Berlin: 357–76. Flower, R. 2013. Emperors and Bishops in Late Roman Invective. Cambridge. Flower, R. 2015. ‘Tamquam figmentum hominis: Ammianus, Constantius II and the portrayal of imperial ritual’, CQ 65: 822–35. Flower, R. (ed.) 2016. Imperial Invectives against Constantius II. Liverpool. Fontaine, J. (ed.) 1967. Sulpice Sévère: Vie de Saint Martin. 3 vols. Paris. Fontaine, J. 1976. ‘Hagiographie et politique, de Sulpice Sévère à Venace Fortunat’, Revue d’histoire de l’Église de France 62: 113–40. Fontaine, T. 2003. ‘Ein letzter Abglanz vergangener kaiserlicher Pracht. Zu ausgewählten archäologischen Befunden aus dem Areal der römischen Kaiserresidenz in Trier’, in M. König (ed.), PALATIA. Kaiserpaläste in Konstantinopel, Ravenna und Trier. Trier: 130–73. Fournet, J.-L. 2010. ‘Les tribulations d’un pétitionnaire égyptien à Constantinople. Révision de P. Cair. Masp. III 67352’, in T. Gagos (ed.), Proceedings of the Twenty-Fith International Congress of Papyrology. Ann Arbor: 243–52. Fraade, S. D. 2007. ‘Midrash and ancient Jewish biblical interpretation’, in C. E. Fonrobert and M. S. Jaffee (eds), The Cambridge Companion to the Talmud and Rabbinic Literature. Cambridge: 99–120.

370



Frank, R. I. 1969. Scholae Palatinae. The Palace Guards of the Later Roman Empire. Rome. Franke, H. 1994. ‘The Chin dynasty’, in H. Franke and D. C. Twitchett (eds), The Cambridge History of China. Volume 6. Alien Regimes and Border States, 907–1368. Cambridge: 215–320. Franke, H. and Twitchett, D. C. (eds) 1994. The Cambridge History of China. Volume 6. Alien Regimes and Border States, 907–1368. Cambridge. Frazer, A. 1966. ‘The iconography of the emperor Maxentius’ buildings in Via Appia’, Art Bulletin 48: 385–92. Freedman, H. and Simon, M. (eds) 1939. Midrash Rabbah. London. Fredrick, D. (ed.) 2002. The Roman Gaze: Vision, Power, and the Body. Baltimore: Johns Hopkins University Press. Frend, W. H. C. 1972. ‘Popular religion and Christological controversy in the fifth century’, in G. J. Cuming and D. Baker (eds), Popular Belief and Practice. Cambridge: 19–29. Friedländer, L. 1862–71. Darstellungen aus der Sittengeschichte Roms in der Zeit von August bis zum Ausgang der Antonine. First edition. Leipzig. Friedländer, L. 1919–21. Darstellungen aus der Sittengeschichte Roms in der Zeit von August bis zum Ausgang der Antonine. Ninth edition. Leipzig. Friedländer, L. 1908–13. Roman Life and Manners under the Early Empire. Translated by L. A. Magnus from the German seventh edition. London. Fuess, A. 2011. ‘Between Dihlīz and Dār al-‘Adl: forms of outdoor and indoor royal representation at the Mamluk Court in Egypt’, in A. Fuess and J.-P. Hartung (eds), Court Cultures in the Muslim World. London: 149–67. Fuess, A. and Hartung, J.-P. 2011. ‘Introduction’, in A. Fuess and J.-P. Hartung (eds), Court Cultures in the Muslim World. London: 1–17. Fujitani, T. 1996. Splendid Monarchy: Power and Pageantry in Modern Japan. Berkeley and Los Angeles. Fumaroli, M. 1994. L’Âge de l’éloquence. Rhétorique et ‘res literaria’ de la Renaissance au seuil de l’époque classique. Paris. Fündling, J. 2006. Kommentar zur Vita Hadriani der Historia Augusta. 2 vols. Bonn. Fündling, J. 2017. ‘Anima anceps. Bewertung und dynastische Funktion Hadrians zwischen 138 und 180’, in C. Michels and P. F. Mittag (eds), Jenseits des Narrativs. Antoninus Pius in den nicht literarischen Quellen. Stuttgart: 31–51. Gambato, M. 2000. ‘The female-kings: some aspects of the representation of eastern kings in the Deipnosophistae’, in D. Braund and J. Wilkins (eds), Athenaeus and His World: Reading Greek Culture in the Roman Empire. Exeter: 227–30. Gangloff, A. 2009. ‘Le sophiste Dion de Pruse, le bon roi et l’empereur’, Revue historique 649: 3–38. Gangloff, A. 2019. Pouvoir impérial et vertus philosophiques: l’évolution de la figure du bon prince sous le Haut-Empire. Leiden and Boston. Gaposchkin, M. C. 2008. The Making of Saint Louis. Kingship, Sanctity, and Crusade in the Later Middle Ages. Ithaca and London. Garland, L. 1995. ‘Conformity and licence at the Byzantine court in the eleventh and twelfth centuries: the case of imperial women’, in S. Efthymiadis, C. Rapp, and D. Tsougaraakis (eds), Bosphorus. Essays in Honour of Cyril Mango. Amsterdam: 101–15. Garland, L. 1999. Byzantine Empresses: Women and Power in Byzantium, 527–1204. London. Geary, P., Ali, D., Atkins, P. S., Cooperson, M., Gomes, R. C., Dutton, P., Melville, G., Rapp, C., Speiß, K.-H., West, S., and Yu, P. 2015. ‘Courtly cultures: western Europe, Byzantium, the Islamic world, India, China, and Japan’, in B. Z. Kedar and M. E. Wiesner-Hanks



371

(eds), The Cambridge World History. Volume 5: Expanding Webs of Conflict and Exchange, 500 CE–1500 CE. Cambridge: 179–205. George, M. 2022. ‘Imperial villas’, in B. Kelly and A. Hug (eds), The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge: 239–66. Gering, J. 2012. Domitian, dominus et deus? Herrschafts- und Machtstrukturen im Römischen Reich zur Zeit des letzten Flaviers. Rahden. Giesey, R. E. 1987. ‘The king imagined’, in K. M. Baker (ed.), The Political Culture of the Old Regime. Princeton, NJ: 41–59. Gillett, A. 1995. ‘The birth of Ricimer’, Historia 44: 380–4. Gillett, A. 2001. ‘Rome, Ravenna and the last western emperors’, PBSR 59: 131–67. Gillett, A. 2003. Envoys and Political Communication in the Late Antique West, 411–533. Cambridge. Gillett, A. 2012. ‘Epic panegyric and political communication in the fifth century West’, in L. Grig and G. Kelly (eds), Two Romes: Rome and Constantinople in Late Antiquity. Oxford: 265–90. Goffart, W. 1989. Rome’s Fall and After. London. Goldbeck, F. 2010. Salutationes. Die Morgenbegrüßungen in Rom in der Republik und der frühen Kaiserzeit. Berlin. Goldbeck, F. and Wienand, J. (eds) 2017. Der römische Triumph in Prinzipat und Spätantike. Berlin and New York. Goldhill, S. 2009. ‘The anecdote: exploring the boundaries between oral and literate performance in the Second Sophistic’, in W. A. Johnson and H. N. Parker (eds), Ancient Literacies: The Culture of Reading in Greece and Rome. Oxford and New York: 96–103. Goodrich, R. J. (ed.) 2015. Sulpicius Severus: The Complete Works. New York and Mahwah, NJ. Gorce, D. (ed.) 1962. Vie de Sainte Mélanie: Texte Grec, Introduction, Traduction et Notes. Paris. Gordon, C. D. (ed.) 1960. The Age of Attila. Fifth-Century Byzantium and the Barbarians. Michigan. Gotter, U. 2008. ‘Die Nemesis des Allgemein-Gültigen: Max Webers Charisma-Konzept und die antiken Monarchien’, in P. Rychterová et al. (eds), Das Charisma. Funktionen und symbolische Repräsentationen, Beiträge zu den Historischen Kulturwissenschaften 2. Berlin: 173–86. Gotter, U. 2015. ‘Penelope’s web, or: How to become a bad emperor post mortem’, in H. Börm (ed.), Antimonarchic Discourse in Antiquity. Stuttgart: 215–33. Graff, D. 2002. Medieval Chinese Warfare, 300–900. London. Grant, R. M. 1975. ‘Religion and politics at the Council of Nicaea’, Journal of Religion 55: 1–12. Graumann, T. 2013. ‘Theodosius II and the politics of the first council of Ephesus’, in C. Kelly (ed.), Theodosius II. Rethinking the Roman Empire in Late Antiquity. Cambridge: 109–29. Greatrex, G. 1996. ‘Flavius Hypatius, quem vidit validum Parthus sensitque timendum. An investigation of his career’, Byzantion 66: 120–42. Greatrex, G. 2007. ‘The early years of Justin I’s reign in the sources’, Electrum 12: 99–113. Greatrex, G., Phenix, R. R., Horn, C. B., Brock, S. and Witakowski, W. (eds) 2011. The Chronicle of Pseudo-Zachariah Rhetor. Church and War in Late Antiquity. Translated Texts for Historians 55. Liverpool. Gregory, T. E. 1979. Vox populi: Popular Opinion and Violence in the Religious Controversies of the Fifth Century. Columbus, OH.

372



Grig, L. 2012. ‘Competing capitals, competing representations’, in L. Grig and G. Kelly (eds), Two Romes. Rome and Constantinople in Late Antiquity. Oxford: 31–52. Grig, L. and Kelly, G. 2012. ‘Introduction: from Rome to Constantinople’, in L. Grig and G. Kelly (eds), Two Romes. Rome and Constantinople in Late Antiquity. Oxford: 3–30. Grosrichard, A. 1979. Structure du sérail. La fiction du despotisme asiatique dans l’Occident classique. Paris. Gruen, E. S. 2002. Diaspora: Jews Amidst Greeks and Romans. Cambridge, MA. Gruen, E. S. 2011. Rethinking the Other in Antiquity. Princeton, NJ. Gruen, S. E. 2016. The Construct of Identity in Hellenistic Judaism: Essays on Early Jewish Literature and History. Berlin. Gualandi, M. L. 2010. ‘Due imperatori per un trionfo. La matrice di Olbia: un hapax “furoi contesto” ’, in M. Milanese, P. Ruggeri, and C. Vismara (eds), L’Africa romana; XVIII: I luoghi e le forme dei mestieri e della produzione nelle province africane. Vol. 3, Rome: 1915–33. Guidetti, F. 2023. ‘Public rituals and performance: the ceremonial staging of imperial authority under Diocletian’, in F. Carlà-Uhink and C. Rollinger (eds), The Tetrarchy as Ideology: Reconfigurations and Representations of an Imperial Power. Stuttgart: 119–40. Gutsfeld, A. 1998. ‘Der Prätorianerpräfekt und der kaiserliche Hof im 4. Jahrhundert n. Chr.’, in A. Winterling (ed.), Comitatus: Beiträge zur Erforschung des spätantiken Kaiserhofes. Berlin: 75–102. Gwynn, D. M. 2007. The Eusebians. The Polemic of Athanasius of Alexandria and the Construction of the ‘Arian Controversy’. Oxford. Haake, M. 2017. ‘Zwischen Severus Alexanders Triumph über die Sāsāniden im Jahre 233 und den Triumphfeierlichkeiten Diocletians und Maximians im Jahre 303. Zum römischen Triumph im dritten Jahrhundert n. Chr.’, in F. Goldbeck and J. Wienand (eds), Der römische Triumph in Prinzipat und Spätantike. Berlin and Boston: 357–95. Haarer, F. K. 2006. Anastasius I: Politics and Empire in the Late Roman World. Leeds. Habberstad, L. 2017. Forming the Early Chinese Court: Rituals, Spaces, Roles. Seattle and London. Hacì Zeyrek, T. and Özbay, I. 2006. ‘Statuen und Reliefs aus Nikomedia’, Ist. Mitt. 56: 273–316. Hadijitryphonos, E. 2011. ‘The Palace of Galerius in Thessalonike: its place in the modern city and an account of the state of research’, in G. Bülow and H. Zabehlicky (eds), Bruckneudorf und Gamzigrad. Spätantike Paläste und Großvillen im Donau-BalkanRaum. Bonn: 203–17. Haensch, R. 2006. ‘Von den Augusti liberti zu den Caesariani’, in A. Kolb (ed.), Herrschaftsstrukturen und Herrschaftspraxis. Berlin: 153–64. Haensch, R. 2007. ‘Die Rolle der Bischöfe im 4. Jahrhundert. Neue Anforderungen und neue Antworten’, Chiron 37: 153–81. Hägg, T. 2012. The Art of Biography in Antiquity. Cambridge. Hagl, W. 1997. Arcadius Apis Imperator. Synesios von Kyrene und sein Beitrag zum Herrscherideal der Spätantike. Stuttgart. Hagen, J. 2017. Die Tränen der Mächtigen und die Macht der Tränen: Eine emotionsgeschichtliche Untersuchung des Weinens in der kaiserzeitlichen Historiographie. Stuttgart. Haines, C. R. (ed.) 1919–20. Fronto: Correspondence. 2 vols. London and Cambridge, MA. Haldon, J. 1984. Byzantine Praetorians: An Administrative, Institutional and Social Survey of the Opsikion and Tagmata c. 580–900. Bonn. Haldon, J. 1990a. Byzantium in the Seventh Century. The Transformation of a Culture. Cambridge.



373

Haldon, J. (ed.) 1990b. Constantine Porphyrogenitus. Three Treatises on Imperial Military Expeditions. Vienna. Haldon, J. 2002. ‘Humour and the everyday in Byzantium’, in G. Halsall (ed.), Humour, History and Politics in Late Antiquity and the Early Middle Ages. Cambridge: 48–71. Haldon, J. 2016. The Empire That Would Not Die. The Paradox of Eastern Roman Survival, 640–740. Cambridge, MA. Haley, E. W. 1994. ‘A palace of Maximianus Herculius at Corduba?’, ZPE 101: 208–14. Halfmann, H. 1986. Itinera principum. Geschichte und Typologie der Kaiserreisen im Römischen Reich. Stuttgart. Hall, E. 1989. Inventing the Barbarian: Greek Self-Definition through Tragedy. Oxford. Hall, J. W. 1990. ‘The Muromachi bakufu’, in K. Yamamura (ed.), The Cambridge History of Japan. Volume 3: Medieval Japan. Cambridge: 175–230. Hall, J. W. 1991. ‘The bakuhan system’, in J. W. Hall (ed.), The Cambridge History of Japan. Volume 4: Early Modern Japan. Cambridge: 128–82. Halsall, G. 2007. Barbarian Migrations and the Roman West, 376–568. Cambridge. Hammond, M. (ed.) 2006. Marcus Aurelius: Meditations. Harmondsworth. Hanne, E. J. 2007. Putting the Caliph in His Place. Power, Authority, and the Late Abbasid Caliphate. Cranbury, NJ. Harbus, A. 2002. Helena of Britain in Medieval Legend. Suffolk. Harker, A. 2008. Loyalty and Dissidence in Roman Egypt: The Case of the Acta Alexandrinorum. Cambridge. Harries, J. 1988. ‘The Roman imperial quaestor from Constantine to Theodosius II’, JRS 78: 148–72. Harries, J. 1994a. ‘Pius Princeps: Theodosius II and fifth-century Constantinople’, in P. Magdalino (ed.), New Constantines. The Rhythm of Imperial Renewal in Byzantium, 4th to 13th Centuries. Aldershot: 35–44. Harries, J. 1994b. Sidonius Apollinaris and the Fall of Rome, AD 407–485. Oxford. Harries, J. 1999. Law and Empire in Late Antiquity. Cambridge. Harries, J. 2012. Imperial Rome, AD 284–363: The New Empire. Edinburgh. Harries, J. 2013. ‘Men without women: Theodosius’ consistory and the businenss of government’, in C. Kelly (ed.), Theodosius II. Rethinking the Roman Empire in Late Antiquity. Cambridge: 67–89. Harries, J. 2014. ‘The empresses’ tale, AD 300–360’, in C. Harrison, C. Humfress, and I. Sandwell (eds), Being Christian in Late Antiquity. A Festschrift for Gillian Clark. Oxford: 197–214. Harris, W. V. 2001. Restraining Rage: The Ideology of Anger Control in Classical Antiquity. Cambridge, Mass. and London. Hartmann, E. 2016. Ordnung in Unordnung. Kommunikation, Konsum und Konkurrenz in der stadtrömischen Gesellschaft der frühen Kaiserzeit. Stuttgart. Hartmann, F. 1982. Herrscherwechsel und Reichskrise: Untersuchungen zu den Ursachen und Konsequenzen der Herrscherwechsel im Imperium Romanum der Soldatenkaiserzeit (3. Jahrhundert n. Chr.). Frankfurt and Bern. Hasse-Ungeheuer, A. 2016. Das Mönchtum in der Religionspolitik Kaiser Justinians I. Die Engel des Himmels und der Stellvertreter Gottes auf Erden. Berlin. Haug, A. 2012. ‘Die Stadt als Repräsentationsraum. Rom und Mailand im 4. Jh. n. Chr.’, in T. Fuhrer (ed.), Rom und Mailand in der Spätantike. Repräsentation städtischer Räume in Literatur, Architektur, und Kunst. Berlin: 111–36. Heather, P. 1999. ‘Ammianus on Jovian’, in J. W. Drijvers and D. Hunt (eds), The Late Roman World and its Historian: Interpreting Ammianus Marcellinus. London: 105–16.

374



Heather, P. 2006. The Fall of the Roman Empire. A New History of Rome and the Barbarians. Oxford. Heather, P. and Moncur, D. (eds) 2001. Politics, Philosophy, and Empire in the Fourth Century: Select Orations of Themistius. Liverpool. Hebblewhite, M. and Whately, C. (eds) 2023. Brill’s Companion to Bodyguards in the Ancient Mediterranean. Leiden and Boston. Hekster, O. J. 1999. ‘The city of Rome in late imperial ideology: the Tetrarchs, Maxentius, and Constantine’, Mediterraneo Antico 2: 717–48. Hekster, O. J. 2001. ‘All in the family. The appointment of emperors designate in the second century AD’, in L. de Blois (ed.), Administration, Prosopography and Appointment Politics in the Roman Empire. Amsterdam: 35–49. Hekster, O. J. 2002. Commodus: An Emperor at the Crossroads. Amsterdam. Hekster, O. J. 2005. ‘Captured in the gaze of power: visibility, games and Roman imperial representation’, in O. J. Hekster and R. Fowler (eds), Imaginary Kings: Royal Images in the Ancient Near East, Greece and Rome. Stuttgart: 157–76. Hekster, O. J. 2015. Emperors and Ancestors. Roman Rulers and the Constraints of Tradition. Oxford. Hekster, O. J. 2020. ‘Imperial justice? The absence of images of Roman emperors in a legal role’, CQ 70: 247–60. Hekster, O. J. 2022. ‘Epilogue: continuity and change at the Roman imperial court’, in B. Kelly and A. Hug (eds), The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge: 479–97. Hellenkemper, H. 2015. ‘Politische Orte? Kaiserliche Sommerpaläste in Konstantinopel’, in M. Featherstone, J.-M. Spieser, G. Tanman, and U. Wulf-Rheidt (eds), The Emperor’s House. Palaces from Augustus to the Age of Absolutism. Berlin: 243. Hen, Y. 2007. Roman Barbarians: The Royal Court and Culture in the Early Medieval West. Basingstoke. Hen, Y. 2020. ‘The Merovingian polity: a network of courts and courtiers’, in B. Effors and I. Moreira (eds), The Oxford Handbook of the Merovingian World. Oxford: 217–37. Hendy, M. 1972. ‘Mint and fiscal administration under Diocletian, his colleagues, and his successors’, JRS 62: 75–82. Hengerer, M. 2016. ‘Access at the court of the Austrian Habsburg dynasty (mid-sixteenth to mid-eighteenth century): a highway from presence to politics?’, in D. Raeymaekers and S. Derks (eds), The Key to Power? The Culture of Access in Princely Courts, 1400–1750. Leiden and Boston: 124–52. Herr, M. D. 1971. ‘The historical significance of the dialogues between Jewish sages and Roman dignitaries’, Scripta Hierosolymitana 22: 123–50. Herrmann-Otto, E. 1980. Ecclesia in Re Publica. Die Entwicklung der Kirche von pseudostaatlicher zu staatlich inkorporierter Existenz. Frankfurt. Herrin, J. 2000. ‘The imperial feminine in Byzantium’, P&P 169: 3–35. Herrin, J. 2013. Unrivalled Influence. Women and Empire in Byzantium. Princeton, NJ. Herrin, J. 2016. ‘Late antique origin of the ‘Imperial Feminine’: western and eastern empresses compared’, ByzSlav 74: 5–25. Hess, H. 2002. The Early Development of Canon Law and the Council of Serdica. Oxford. Heucke, C. 1994. Circus und Hippodrom als politischer Raum. Hildesheim. Hillner, J. 2017. ‘A woman’s place: imperial women in late antique Rome’, AnTard 25: 75–94. Hogg, M. A. and Tindale, R. S. (eds) 2001. Blackwell Handbook of Social Psychology: Group Processes. Oxford.



375

Holloway, S. and Worsley, L. 2017. ‘ “Every body took notice of the scene of the drawing room”: Performing emotions at the early Georgian court, 1714–60’, Journal for Eighteenth-Century Studies 40: 434–64. Holum, K. G. 1982. Theodosian Empresses. Women and Imperial Dominion in Late Antiquity. Berkeley. Holum, K. G. and Vikan, G. 1979. ‘The Trier Ivoy, adventus ceremonial and the relics of St. Stephen’, DOP 33: 113–33. Honoré, T. 1978. Tribonian. London. Honoré, T. 1987. ‘Scriptor Historiae Augustae’, JRS 77: 156–76. Honoré, T. 1994. Emperors and Lawyers. London. Honoré, T. 2002. Ulpian, Pioneer of Human Rights. Oxford. Horbury, W. 2014. Jewish War under Trajan and Hadrian. Cambridge. Horn, C. B. and Phenix Jr, R. R. (eds) 2008. John Rufus. The Lives of Peter the Iberian, Theodosius of Jerusalem, and the Monk Romanus. Atlanta. Hug, A. 2022. ‘The imperial family’, in B. Kelly and A. Hug (eds), The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge: 60–84. Humphrey, J. H. 1986. Roman Circuses. Arenas for Chariot Racing. London. Humphries, M. 2003. ‘Roman senators and absent emperors in Late Antiquity’, Acta ad archaeologiam et artium historiam pertinentia 17: 27–46. Humphries, M. 2012. ‘Valentinian III and the city of Rome (425–55): patronage, politics, power’, in L. Grig and G. Kelly (eds), Two Romes: Rome and Constantinople in Late Antiquity. Oxford: 161–82. Hunsucker, R. 2018. ‘Maxentius and the aeternae urbis suae conditores: Rome and its founders from Maximian to Constantine (289–313)’, in D. P. W. Burgersdijk and A. J. Ross (eds), Imagining Emperors in the Later Roman Empire. Leiden: 83–112. Hunt, E. D. 1989. ‘Did Constantius II have “court bishops”?’, Studia Patristica 19: 86–90. Hunt, E. D. 2007. ‘Valentinian and the bishops: Ammianus 30.9.5 in context’, in J. den Boeft, J. W. Drijvers, D. den Hengst, and H. C. Teitler (eds), Ammianus After Julian: The Reign of Valentinian and Valens in Books 26–31 of the Res Gestae. Leiden: 71–93. Hurlet, F. 2018. ‘Interdicta aula. Cour impériale et hiérarchie spatiale (Haut-Empire romain)’, in C. Moatti and C. Müller (eds), Statuts personnels et espaces sociaux. Questions grecques et romaines. Paris: 271–86. Hurst, G. C. 1999. ‘Insei’, in D. H. Shively and W. H. McCullough (eds), The Cambridge History of Japan. Volume 2: Heian Japan. Cambridge: 576–643. Icks, M. 2017. ‘Of lizards and peacocks. Criticism of the princeps clausus in fourth- and fifth-century sources’, Mediterraneo antico 20: 467–94. Icks, M. 2020a. ‘Keeping up appearances: evaluations of imperial (in)visibility in Late Antiquity’, in E. Manders and D. Slootjes (eds), Leadership, Ideology and Crowds in the Roman Empire of the Fourth Century AD. Stuttgart: 163–79. Icks, M. 2020b. ‘Kept in the dark: narratives of imperial seclusion in Late Antiquity’, in K. C. Choda, M. S. de Leeuw, and F. Schulz (eds), Gaining and Losing Imperial Favour in Late Antiquity: Representation and Reality. Leiden and Boston: 173–92. Imber, C. 2012. ‘Government, administration, and law’, in S. N. Faroqhi and K. Fleet (eds), The Cambridge History of Turkey. Volume 2: The Ottoman Empire as a World Power, 1453–1603. Cambridge: 205–40. Irshai, O. 2013. ‘Christian historiographers’ reflections on Jewish-Christian violence in fifth-century Alexandria’, in N. B. Dohrmann and A. Y. Reed (eds), Jews, Christians, and the Roman Empire. Philadelphia: 137–53. Isaac, B., 2004. The Invention of Racism in Classical Antiquity. Princeton, NJ.

376



Jacobs, B. 2010. ‘Höfischer Lebensstil und materielle Prachtentfaltung’, in B. Jacobs and R. Rollinger (eds), Der Achämenidenhof/The Achaemenid Court. Akten des 2. Internationalen Kolloquiums zum Thema ‘Vorderasien im Spannungsfeld klassischer und altorientalischer Überlieferungen’. Wiesbaden: 377–409. Jacobs, M. 1995. Die Institution des jüdischen Patriarchen. Tübingen. Jaeschke, V. 2017a. ‘Sedes imperii. Architektur und Herrschaftsrepräsentation in den tetrarchischen Residenzstädten’. PhD Dissertation, European University Viadrina, Frankfurt an der Oder. Jaeschke, V. 2017b. ‘The Roman civic center under Maxentius (AD 306–312). Buildings for a new concept of sovereignty’, in A. Hofmann and M. Zimmermann (eds), History Takes Place: Rome. Dynamics of Urban Change. Berlin: 177–87. Jaeschke, V. 2020. ‘Architecture and power—defining Tetrarchic imperial residences’, in E. Manders and D. Slootjes (eds), Leadership, Ideology and Crowds in the Roman Empire of the Fourth Century AD. Stuttgart: 19–34. Jaffee, M. S. 2001. Torah in the Mouth: Writing and Oral Tradition in Palestinian Judaism. Oxford. James, E. L. 1972. The Origins of France: From Clovis to the Capetians, 500–1000. Basingstoke and London. James, E. 2001. Empresses and Power in Early Byzantium. Leicester. Janin, R. 1964. Constantinople byzantine. Développement urbain et repertoire topographique. Second edition. Paris. Janin, R. 1966. ‘Les processions religieuses à Byzance’, REB 24: 69–88. Janniard, S. 2020. ‘Les scholes palatines (IVe–VIe siècles)’, in C. Wolff and P. Faure (eds), Corps du chef et gardes du corps dans l’armée romaine: Actes du septième congrès de Lyon sur l’armée romaine (25–27 octobre 2018). Paris: 497–519. Jarvis, P. 2015. ‘Avidius Cassius and Maecianus in the Historia Augusta’, Mnemosyne 68: 666–76. Jeffreys, E., Jeffreys, M., and Scott, R. (eds) 1986. The Chronicle of John Malalas. Byzantina Australiensia 4. Canberra. Jeličić-Radonić, J. 2009. ‘Diocletian and the Salona urbs orientalis’, in N. Cambi, J. Belamarić, and T. Marasović (eds), Diocletian, Tetrarchy and Diocletian’s Palace on the 1700th Anniversary of Existence. Split: 307–33. Jenkins, R. J. H. and Mango, C. A. 1956. ‘The date and significance of the Tenth Homily of Photius’, DOP 9/10: 123–40. Jeremić, M. 2009. ‘The Sirmium imperial palace complex, in light of the recent archaeological investigations’, in N. Cambi, J. Belamarić, and T. Marasović (eds), Diocletian, Tetrarchy and Diocletian’s Palace on the 1700th Anniversary of Existence. Split: 471–99. Johnson, J. H. 1983. ‘The Demotic Chronicle as a statement of a theory of kingship’, The Society for the Study of Egyptian Antiquities Journal 13: 61–72. Johnson, A. P. 2014. Eusebius. London and New York. Johnson, M. J. 1992. ‘Where were Constantius I and Helena buried?’, Latomus 51: 145–50. Johnson, M. J. 2009. The Roman Imperial Mausoleum in Late Antiquity. Cambridge. Jones, A. H. M. 1964. The Later Roman Empire 284–602: A Social, Economic and Administrative Survey. 3 vols. Oxford. Jones, B. W. 1992. The Emperor Domitian. London. Jones, C. 2010. ‘Themistius after the death of Julian’, Historia 59: 501–6. Jones, M. and McFadden, S. (eds) 2015. Art of Empire. The Roman Frescoes and Imperial Cult Chamber in Luxor Temple. New Haven and London.



377

Junod, E. 1992. ‘ “Apocryphes du Nouveau Testament”: Un appellation erronée et une collection artificielle’, Apocrypha 3: 17–46. Junod, E. and Kaestli, J.-D. (eds) 1983. Acta Iohannis. 2 vols. Turnholt. Just, P. 2003. Imperator et episcopus. Zum Verhältnis von Staatsgewalt und christlicher Kirche zwischen dem 1. Konzil von Nicaea (325) und dem 1. Konzil von Konstantinopel (381). Stuttgart. Kalavrezou-Maxeiner, I. 1975. ‘The imperial chamber at Luxor’, DOP 29: 225–51. Kaldellis, A. 2015. The Byzantine Republic. People and Power in New Rome. Cambridge, MA. Kalmin, R. 2006. ‘The formation and character of the Babylonian Talmud’, in S. T. Katz (ed.), The Cambridge History of Judaism. Cambridge: 840–76. Kantorowicz, E. H. 1957. The King’s Two Bodies. A Study in Medieval Political Theology, Princeton. Kaplan, F. E. S. 2008. ‘Politics in an African royal harem: women and seclusion at the royal court of Benin, Nigeria’, in A. Walthall (ed.), Servants of the Dynasty: Palace Women in World History. Berkeley: 115–36. Karlin-Hayter, P. 1991. ‘L’adieu à l’empereur’, Byzantion 61: 112–55. Kaster, R. A. 2005. Emotion, Restraint, and Community in Ancient Rome. Oxford. Kazhdan, A. and McCormick, M. 1997. ‘The social world of the Byzantine court’, in H. Maguire (ed.), Byzantine Court Culture from 829 to 1204. Washington: 167–97. Keegan, P. 2013. ‘Reading the “pages” of the domus Caesaris: pueri delicati, slave education, and the graffiti of the Palatine paedagogium’, in M. George (ed.), Roman Slavery and Roman Material Culture. Toronto: 69–98. Kelly, B. 2022a. ‘Introduction’, in B. Kelly and A. Hug (eds), The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge: 1–15. Kelly, B. 2022b. ‘Violence and security at court’, in B. Kelly and A. Hug (eds), The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge: 371–94. Kelly, B. and Hug, A. (eds) 2022. The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge. Kelly, C. 1998. ‘Emperors, government, and bureaucracy’, in A. Cameron and P. D. A. Garnsey (ed.), The Cambridge Ancient History. Volume XIII: The Late Empire, AD 337–425. Cambridge: 138–82. Kelly, C. 2004. Ruling the Later Roman Empire. Cambridge, MA. Kelly, C. 2013a. ‘Rethinking Theodosius’, C. Kelly (ed.), Theodosius II. Rethinking the Roman Empire in Late Antiquity. Cambridge: 3–64. Kelly, C. 2013b. ‘Stooping to conquer: the power of imperial humility’, in C. Kelly (ed.), Theodosius II. Rethinking the Roman Empire in Late Antiquity. Cambridge: 221–43. Kelly, G. 2003. ‘The new Rome and the old: Ammianus Marcellinus’ silences on Constantinople’, CQ 53: 588–607. Kelly, G. 2008. Ammianus Marcellinus: The Allusive Historian. Cambridge. Kelly, G. 2016. ‘Claudian’s last panegyric and imperial visits to Rome’, CQ 66: 336–57. Kemezis, A. 2014. Greek Narratives of the Romn Empire under the Severans. Cassius Dio, Philostratus, and Herodian. Cambridge. Kempf, T. K. 1950. ‘Konstantinische Deckenmalereien aus dem Trierer Dom’, Trierer Zeitschrift 19: 45–1. Kempf, T. K. 1977. ‘Die konstantinische Deckenmalereien aus dem Trierer Dom’, AK 7: 147–59.

378



Kennedy, H. 2005. When Baghdad Ruled the Muslim World. The Rise and Fall of Islam’s Greatest Dynasty. Cambridge, MA. Kennedy, H. 2016. The Prophet and the Age of the Caliphates: The Islamic Near East from the Sixth to the Eleventh Century. Third edition. London. Kessels, A. H. M. and Van der Horst, P. W. 1987. ‘The Vision of Dorotheus (Pap. Bodmer 29): Edited with Introduction, Translation, and Notes’, Vigiliae Christianae 41: 313–59. Kiessel, M. 2012/13. ‘Das spätantike Palastareal nordöstlich und östlich der “Basilika” in Trier’, Trierer Zeitschrift 75/76: 85–197. Kim, S. 2017. ‘Severian of Gabala as a witness to life at the imperial court in fifth century Constantinople’, Studia Patristica 96: 189–204. Kirin, A. 2000. The Rotunda of St. George and Late Antique Serdica: From Imperial Palace to Episcopal Center. PhD Dissertation, Princeton University. Klein, H. A. 2006. ‘Sacred relics and imperial ceremonies at the Great Palace of Constantinople’, Byzas 5: 79–99. Knecht, R. J. 2014. Hero or Tyrant? Henry III, King of France, 1574–89. Farnham. Knickrehm, W. 2014. Offizielle und locale Münzprägestätten des Gallischen Sonderreichs in und um Trier. Trier. Knight, J. E. 2015. The Politics of Anger in Roman Society: A Study of Orators and Emperors, 70 BCE–68 CE. PhD Dissertation, University of British Columbia. Knight, J. E. 2016. ‘Anger as a mechanism for social control in imperial Rome’, in E. Sanders and M. Johncock (eds), Emotion and Persuasion in Classical Antiquity. Stuttgart: 183–97. Kolb, F. 2001. Herrscherideologie in der Spätantike. Berlin. Kōjirō, N. and Bock, F. G. 1993. ‘The Nara state’, in D. M. Brown (ed.), The Cambridge History of Japan. Volume 1: Ancient Japan. Cambridge: 221–67. Konstan, D. 1997. ‘Friendship and monarchy: Dio of Prusa’s Third Oration on kingship’, Symbolae Osloenses 72: 124–43. Konstan, D. 2014. Beauty: The Fortunes of an Ancient Greek Idea. New York. Kosiński, R. 2010. The Emperor Zeno: Religion and Politics. Cracow. Kostenec, J. 2005. ‘Observations on the Great Palace at Constantinople’, Reading Medieval Studies 31: 27–56. Kropp, M. 2011. ‘Notes on preparing a critical edition of the Śər‘atä mängəśt’, Northeast African Studies 11: 111–40. Kühn, W. 1987. ‘Der Kuss der Kaisers: Plinius paneg. 24.2’, WJA 13: 263–71. Kuhoff, W. 2001. Diokletian und die Epoche der Tetrarchie. Frankfurt. Kuhoff, W. 2009. ‘Das tetrarchische Herrschaftssystem und seine Darstellung in der Architektur: Herrscherresidenzen und Altersruhesitze als Ausdruck kaiserlicher Regierung und Repräsentation’, in N. Cambi, J. Belamarić, and T. Marasović (eds), Diocletian, Tetrarchy and Diocletian’s Palace on the 1700th Anniversary of Existence. Split: 95–116. Kulikowski, M. 2015. ‘Regional dynasties and imperial court’, in J. Wienand (ed.), Contested Monarchy: Integrating the Roman Empire in the Fourth Century AD. Oxford: 135–48. Kumar, S. 2011. ‘Courts, capitals and kingship: Delhi and its sultans in the thirteen and fourteenth centuries CE’, in A. Fuess and J.-P. Hartung (eds), Court Cultures in the Muslim World: Seventh to Nineteenth Centuries. London: 123–48. Kunst, C. 2010. ‘Patronage/Matronage der Augustae’, in A. Kolb (ed.), Augustae. Machtbewusste Frauen am römischen Kaiserhof? Berlin: 11–35. Labarre, S. 2004. ‘La composition de la Vita Martini de Sulpice Sévère’, Vita Latina 171: 102–20.



379

Laeben-Rosén, V. 2005. Age of Rust. Court and Power in the Severan Age (188–239 AD). Uppsala. Laes, C. 2009. ‘What could Marcus Aurelius feel for Fronto?’ Studia Humaniora Tartuensia 10: 10.A.3. Laniado, A. 1997. ‘Un fragment peu connu de Pierre le Patrice’, BZ 90: 405–12. Lamoureux, J. (ed.) and Ajoulat, N. (trans. and comm.) 2008. Synésios de Cyrène. Tome V. Opuscules II. Paris. Lampe, G. W. H. (ed.) 1969. A Patristic Greek Lexicon. Oxford. Landelle, M. 2014. ‘La titulature des magistri militum au IVe siècle’, AnTard 22: 195–221. Lanfranchi, G. 2010. ‘Greek historians and the memory of the Assyrian court’, in B. Jacobs and R. Rollinger (eds), Der Achämenidenhof/The Achaemenid Court. Wiesbaden: 39–65. Langlands, R. 2018. ‘Extratextuality’, in A. König and C. Whitton (eds), Roman Literature under Nerva, Trajan and Hadrian: Literary Interactions, AD 96–138. Cambridge: 330–46. Lauterbach, J. Z. (ed.) 2004. Mekhilta de-Rabbi Ishmael. Philadelphia. Laurence, R. and Paterson, J. 1999. ‘Power and laughter: imperial dicta’, PBSR 67: 183–97. Lavan, M. 2018. ‘Pliny, Epistles 10 and imperial correspondence: the empire of letters’, in A. König, and C. Whitton (eds), Roman Literature under Nerva, Trajan and Hadrian: Literary Interactions AD 96–138. Cambridge: 280–301. Lavin, I. 1962. ‘The house of the lord: aspects of the role of palace triclinia in the architecture of late antiquity and the early middle ages’, The Art Bulletin 44: 1–27. Leadbetter, B. 2009. Galerius and the Will of Diocletian. London. Leberl, J. 2004. Domitian und die Dichter. Poesie als Medium der Herrschaftsdarstellung. Göttingen. Lee, A. D. 2007. War in Late Antiquity: A Social History. Oxford and Malden, MA. Lee, A. D. 2013a. ‘Theodosius and his generals’, in C. Kelly (ed.), Theodosius II: Rethinking the Roman Empire in Late Antiquity. Cambridge: 90–108. Lee, A. D. 2013b. From Rome to Byzantium AD 363 to 565. Edinburgh. Lee, A. D. 2014. ‘Emperors and generals in the fourth century’, in J. Wienand (ed.), Contested Monarchy: Integrating the Roman Empire in the Fourth Century AD. Oxford: 100–18. Lehnen, J. 1997. Adventus principis. Untersuchungen zu Sinngehalt und Zeremoniell der Kaiserankunft in den Städten des Imperium Romanum. Frankfurt. Lendon, J. E. 1997. Empire of Honour: The Art of Government in the Roman World. Oxford. Lenel, O. (ed.) 1889. Palingenesia Iuris Civilis. Leipzig. Lenfant, D. (ed.) 2004. Ctésias de Cnide. La Perse. L’Inde. Autres fragments. Paris. Lenfant, D. 2007. ‘On Persian tryphē in Athenaeus’, in C. Tuplin (ed.), Persian Responses: Political and Cultural Interaction with(in) the Achaemenid Empire. Swansea: 51–65. Lenfant, D. (ed.) 2011. Les Perses vus par les Grecs. Lire les sources classiques sur l’empire Achéménide. Paris. Lenfant, D. 2012. ‘Ctesias and his eunuchs: a challenge for modern historians’, Histos 6: 257–97. Lenski, N. 1998. ‘Review: W. Hagl, Arcadius Apis Imperator’. Bryn Mawr Classical Review 98.3.08, http://bmcr.brynmawr.edu/1998/98.3.08.html. Lenski, N. 2000. ‘The election of Jovian and the role of the late imperial guards’, Klio 82: 492–515. Lenski, N. 2002. Failure of Empire. Valens and the Roman State in the Fourth Century A.D. Berkeley.

380



Lenski, N. 2016. Constantine and the Cities. Imperial Authority and Civic Politics. Philadelphia. Leon, D. W. 2016. ‘The face of the emperor in Philo’s Embassy to Gaius’, CW 110: 43–60. Leonard, V. 2019. ‘Galla Placidia as “human gold”: consent and autonomy in the sack of Rome, CE 410’, Gender and History 31: 334–52. Leppin, H. 2000. ‘Das Bild der kaiserlichen Frauen bei Gregor von Nyssa’, in H. B. Drobner and A. Viciano (eds), Gregory of Nyssa: Homilies on the Beatitudes. Leiden: 487–506. Leppin, H. 2009. ‘Power from humility: Justinian and the religious authority of monks’, in A. Cain and N. Lenski (eds), The Power of Religion in Late Antiquity. Farnham: 155–64. Leppin, H. 2011. Justinian. Das christliche Experiment. Stuttgart. Leppin, H. 2012. ‘Christianisierung im Römischen Reich: Überlegungen zum Begriff und zur Phasenbildung’, ZAC 16: 247–78. Leppin, H. 2015a. ‘Coping with the tyrant’s faction’, in J. Wienand (ed.), Contested Monarchy: Integrating the Roman Empire in the Fourth Century AD. Oxford: 198–214. Leppin, H. 2015b. ‘Zwischen Kirche und Circus—der Palast von Konstantinopel und die religiöse Repräsentation Constantins des Großen’, in J. Ganzert and I. Nielsen (eds), Herrschaftsverhältnisse und Herrschaftslegitimation. Bau und Gartenkultur als historische Quellengattung hinsichtlich Manifestation und Legitimation von Herrschaft. Berlin: 129–40. Leppin, H. 2018a. Die frühen Christen. Von den Anfängen bis Konstantin. München. Leppin, H. 2018b. ‘Imperiale Eliten um Justinian’, in W. Drews (ed.), Die Interkation von Herrschen und Eliten in imperialen ordnungen des Mittelalters. Berlin: 43–62. Leszka M. J. 1998. ‘Empress-Widow Verina’s Political Activity during the Reign of Emperor Zeno’, in W. Ceran (ed.), Mélanges d’histoire byzantine offerts à Oktawiusz Jurewicz à l’occasion de son soixante-dixième Anniversaire. Łódź: 128–136. Levene, D. S. 1997. ‘Pity, fear and the historical audience: Tacitus on the fall of Vitellius’, in S. M. Braund and C. Gill (eds), The Passions in Roman Thought and Literature. Cambridge: 128–49. Lieberman, S. 1944. ‘Roman legal institutions in early Rabbinics and in the Acta Martyrum’, JQR 35: 1–57. Liebeschuetz, J. H. W. G. 1972. Antioch: City and Imperial Administration in the later Roman Empire. Oxford. Liebeschuetz, J. H. W. G. 2001. The Decline and Fall of the Roman City. Oxford. Liebeschuetz, J. H. W. G. (ed.) 2005. Ambrose of Milan. Political Letters and Speeches. Liverpool. Lillehoj, E. 2011. Art and Palace Politics in Early Modern Japan, 1580s–1680s. Leiden and Boston. Lin, S. 2021. ‘Justin under Justinian: the rise of emperor Justin II revisited’, DOP 75: 121–45. Lipsius, R. A. (ed.) 1891. Acta Apostolorum Apochrypha. Pars Prior: Acta Petri, Acta Pauli, Acta Petri et Pauli, Acta Pauli et Theclae, Acta Thaddaei. Leipzig. Littlewood, A. R. 1997. ‘Gardens of the palaces’, in H. Maguire (ed.), Byzantine Court Culture from 829 to 1204. Washington, DC: 13–38. Llewellyn-Jones, L. 2013a. King and Court in Ancient Persia, 559 to 331 BCE. Edinburgh. Llewellyn-Jones, L. 2013b. ‘“Empire of the gaze”: despotism and seraglio fantasies à la grecque in Chariton’s Callirhoe’, Helios 40: 167–91. Llewellyn-Jones, L. and Robson, J. (eds) 2013. Ctesias’ History of Persia: Tales of the Orient. London. Low, M. 2006. Japan on Display: Photography and the Emperor. London. Luibhéid, C. 1982. The Council of Nicaea. Galway.



381

Lütkenhaus, W. 1998. Constantius III: Studien zu seiner Tätigkeit und Stellung im Westreich, 411–421. Bonn. Ma’oz, Z. U. 2006. ‘The civil reform of Diocletian in the southern Levant’, SCI 25: 105–19. MacCormack, S. G. 1981. Art and Ceremony in Late Antiquity. Berkeley. MacGeorge, P. 2002. Late Roman Warlords. Oxford. Machado, C. 2006. ‘Building the past: monuments and memory in the Forum Romanum’, in W. Bowden, A. Gutteridge, and C. Machado (eds), Social and Political Life in Late Antiquity. Late Antique Archaeology 3. Leiden: 157–92. Machado, C. 2012. ‘Aristocratic houses and the making of late antique Rome and Constantinople’, in L. Grig and G. Kelly (eds), Two Romes: Rome and Constantinople in Late Antiquity. Oxford: 136–58. Machado, C. 2019. Urban Space and Aristocratic Power in Late Antique Rome, AD 270–535. Oxford. Mackay, C. S. 1999. ‘Lactantius and the succession to Diocletion’, CPh 94: 198–209. MacMullen, R. 1962. ‘The emperor’s largesses’, Latomus 21: 159–66. MacMullen, R. 1964. ‘Imperial bureaucrats in the Roman provinces’, HSCP 68: 305–16. MacMullen, R. 1988. Corruption and the Decline of Rome. New Haven. Macrides, R. 2004. ‘The ritual of petition’, in D. Yatromanolakis and P. Roilos (eds), Greek Ritual Poetics. Cambridge, MA: 356–70. Macrides, R. 2011. ‘Ceremonies and the city: the court in fourteenth-century Constantinople’, in J. Duindam, T. Artan, and M. Kunt (eds), Royal Courts in Dynastic States and Empires: A Global Perspective. Leiden: 217–35. Macrides, R., Munitiz, J. A., and Angelov, D. (eds) 2013. Pseudo-Kodinos and the Constantinopolitan Court: Offices and Ceremonies. London. Magdalino, P. 2001. ‘Aristocratic Oikoi in the tenth and eleventh regions of Constantinople’, in N. Necipolglu (ed.), Byzantine Constantinople: Monuments, Topography and Everyday Life. Leiden: 53–69. Magdalino, P. 2011. ‘Court and capital in Byzantium’, in J. Duindam, T. Artan, and M. Kunt (eds), Royal Courts in Dynastic States and Empires: A Global Perspective. Leiden: 129–44. Magdalino, P. 2015. ‘The people and the palace’, in M. Featherstone, J.-M. Spieser, G. Tanman, and U. Wulf-Rheidt (eds), The Emperor’s House. Palaces from Augustus to the Age of Absolutism. Berlin: 169–80. Magdalino, P. 2022. ‘Imperial Constantinople’, in S. Bassett (ed.), The Cambridge Companion to Constantinople. Cambridge: 135–49. Malamud, M. 2001. ‘That’s entertainment! Dining with Domitian in Statius’ Silvae’, Ramus 30: 23–45. Malik, S. 2020. The Nero Anti-Christ. Founding and Fashioning a Paradigm. Cambridge. Malmberg, S. 2007. ‘Dazzling dining: banquets as an expression of imperial legitimacy’, in L. Brubaker and K. Linardou, Eat, Drink and Be Merry (Luke 12: 91): Food and Wine in Byzantium. Aldershot: 75–91. Malone, C. 2010. ‘The virtue of rage in the fourth century’, in B. Sidwell and D. Dzino (eds), Studies in Emotions and Power in the Late Roman World: Papers in Honour of Ron Newbold. Piscataway, NJ: 59–85. Mango, C. 1976. Byzantine Architecture. New York. Mansel, P. 1984. Pillars of Monarchy. An Outline of the Political and Social History of Royal Guards, 1400–1984. London. Mantas, A. G. 2000. ‘Die Schilderhebung in Byzanz—historische und ikonographische Bemerkungen’, Byzantina 21: 537–82.

382



Mariev, S. (ed.) 2008. Ioannis Antiocheni fragmenta quae supersunt omnia. Berlin. Marincola, J. 1999. ‘Genre, convention and innovation in Greco-Roman historiography’, in J. Marincola (ed.), The Limits of Historiography. Leiden: 281–324. Marotta, V. 1991. Mandata principum. Turin. Marsham, A. 2009. Rituals of Islamic Monarchy. Accession and Succession in the First Muslim Empire. Edinburgh. Martin, A. 1996. Athanase d’Alexandrie et l’église d’Égypte au IV. siècle (328–373). Rome. Martin, A. 2017. ‘Athanase d’Alexandrie évêque de cour? Paradoxe et réalité’, in S. Destephen (ed.), L’évêque de cour: figure politique, figure polémique. Paris: 37–64. Maslev, S. 1966. ‘Die staatsrechtliche Stellung der byzantinischen Kaiserin’, ByzSlav 27: 308–43. Mass, J. P. 1990. ‘The Kamakura bakufu’, in K. Yamamura (ed.), The Cambridge History of Japan. Volume 3: Medieval Japan. Cambridge: 46–88. Mathisen, R. 2009. ‘Ricimer’s church in Rome: how an Arian barbarian prospered in a Nicene world’, in A. Cain and N. Lenski (eds), The Power of Religion in Late Antiquity. London: 307–25. Matthews, J. 1975. Western Aristocracies and Imperial Court A.D. 364–425. Oxford. Matthews, J. 1989. The Roman Empire of Ammianus. London: Duckworth. Matthews, J. 2000a. Laying Down the Law. Studies in the Theodosian Code. New Haven. Matthews, J. 2000b. ‘The Roman empire and the proliferation of elites’, Arethusa 33: 429–46. Matthews, J. 2012. ‘The Notitia Urbis Constantinopolitanae’, in L. Grig and G. Kelly (eds), Two Romes: Rome and Constantinople in Late Antiquity. Oxford: 81–115. Mayer, E. 2002. Rom ist dort, wo der Kaiser ist. Untersuchungen zu den Staatsdenkmälern des dezentralisierten Reiches von Diocletian bis zu Theodosius II. Mainz. Mayer, E. 2014. ‘The architecture of Tetrarchy’, in R. B. Ulrich and C. K. Quenemoen (eds), A Companion to Roman Architecture. Oxford and Malden, MA: 106–26. Mazzarino, S. 1942. Stilicone: la crisi imperiale dopo Teodosio. Rome. McClanan, A. 2002. Representations of Early Byzantine Empresses: Image and Empire. Basingstoke and New York. McCormick, M. 1985. ‘Analyzing imperial ceremonies’, Jahrbuch der österreichischen Byzantinistik 35: 1–20. McCormick, M. 1986. Eternal Victory. Triumphal Rulership in Late Antiquity, Byzantium, and the Early Medieval West. Cambridge. McCormick, M. 2000. ‘Emperor and court’, in A. Cameron, B. Ward-Perkins, and M. Whitby (eds), The Cambridge Ancient History, Volume XIV: Late Antiquity: Empire and Successors, A.D. 425–600. Cambridge: 135–63. McDougall, S. 2017. Royal Bastards: The Birth of Illegitimacy, 800–1230. Oxford. McEnerney, J. I. 1987. The Fathers of the Church. St. Cyril of Alexandria. 2 vols. Washington, DC. McEvoy, M. 2010. ‘Rome and the transformation of the imperial office in the late fourth– mid-fifth centuries AD’, PBSR 78: 151–92. McEvoy, M. 2013. Child Emperor Rule in the Late Roman West, AD 367–455. Oxford. McEvoy, M. 2016a. ‘Becoming Roman? The not-so-curious case of Aspar and the Ardaburii’, JLA 9: 483–511. McEvoy, M. 2016b. ‘Constantia: the last Constantinian’, Antichthon 50: 154–79. McEvoy, M. 2016c. ‘Justina, wife of the emperor Valentinian I and mother of the emperor Valentinian II’, in S. Goldberg (ed.), Oxford Classical Dictionary. Oxford. Fourth edition.



383

McEvoy, M. 2017. ‘Shadow emperors and the choice of Rome (455–476 AD)’, AnTard 25: 95–112. McEvoy, M. 2018. ‘Dynastic dreams and visions of early Byzantine emperors (ca. 518–565)’, in E. Anagnostou-Laoutides and B. Neil (eds), Dreams, Memory and Imagination in Byzantium. Leiden and Boston: 99–117. McEvoy, M. 2019a. ‘Celibacy and survival in court politics in the fifth century AD’, in S. Tougher (ed.), The Emperor in the Byzantine World. London: 115–34. McEvoy, M. 2019b. ‘Leo II, Zeno, and the transfer of power from a son to his father in 474 CE’, in J. W. Drijvers and N. Lenski (eds), The Fifth Century: Age of Transformation. Bari: 197–208. McEvoy, M. 2020a. ‘The jelly-fish emperor: the emperor Arcadius and imperial leadership in the late fourth century, 395–408 AD’, in E. Manders and D. Slootjes (eds), Leadership, Ideology and Crowds in the Later Roman Empire. Stuttgart: 181–200. McEvoy, M. 2020b. ‘Imperial cities under the sons of Constantine’, in S. Tougher and N. Baker-Brian (eds), The Sons of Constantine, AD 337–361: In the Shadows of Constantine and Julian. London: 275–307. McEvoy, M. 2021. ‘Orations for the first generation of Theodosian imperial women’, JLA 14: 117–41. McEvoy, M. 2022. ‘Emperors, aristocrats and colums: Christian imperial politics in the late fifth century CE as reflected in the Life of Daniel the Stylite’, JLA 15: 428–61. McEvoy, M. and Moser, M. 2017. ‘Introducing imperial presence in late antique Rome (2nd–7th centuries AD)’, AnTard 25: 15–21. McFadden, S. 2015a. ‘The Luxor temple paintings in context: Roman visual culture in Late Antiquity’, in M. Jones and S. McFadden (eds), Art of Empire: The Roman Frescoes and Imperial Cult Chamber in Luxor Temple. New Haven: 105–34. McFadden, S. 2015b. ‘Picturing power in late Roman Egypt: the imperial cult, imperial portraits, and a visual panegyric for Diocletian’, in M. Jones and S. McFadden (eds), Art of Empire: The Roman Frescoes and Imperial Cult Chamber in Luxor Temple. New Haven: 135–53. McGuckin, J. A. 1996. ‘Nestorius and the political factions of fifth-century Byzantium: factors in his personal downfall’, Bulletin of the John Rylands Library 78: 7–21. McGuckin, J. A. 2004. St. Cyril of Alexandria: The Christological Controversy, Its History, Theology and Texts. Crestwood. McLynn, N. B. 1994. Ambrose of Milan. Church and Court in a Christian Capital. Berkeley. McLynn, N. B. 2004. ‘The transformation of imperial churchgoing in the fourth century’, in S. Swain and M. Edwards (eds), Approaching Late Antiquity. The Transformation from Early to Late Empire. Oxford: 235–70. McLynn, N. B. 2005. ‘Genere Hispanus: Theodosius, Spain and Nicene orthodoxy’, in K. Bowes and M. Kulikowski (eds), Hispania in Late Antiquity: Current Perspectives. Leiden: 77–120. McLynn, N. B. 2018. ‘Imperial piety in action: the Theodosians in church’, in S. Destephen, B. Dumézil, and H. Inglebert (eds), Le Prince chrétien de Constantin aux royautés barbares (ive–viiie siècle). Paris: 1–25. Meier, M. 2007. ‘Die Demut des Kaisers. Aspekte der religiösen Selbstinszenierung bei Theodosius II. (408–450 n. Chr.)’, in A. Pečar and K. Trampedach (eds), Die Bibel als politisches Argument. Munich: 135–58. Meier, M. 2009. Anastasios I. Die Entstehung des byzantinischen Reiches. Stuttgart. Meier, M. 2010. ‘Ariadne. Der rote Faden des Kaisertums’, in A. Kolb (ed.), Augustae. Machtbewußte Frauen am römischen Kaiserhof? Berlin: 277–91.

384



Meier, M. 2012. ‘Ostrom-Byzanz, Spätantike-Mittelalter. Überlegungen zum “Ende” der Antike im Osten des Römischen Reiches’, Millennium 9: 187–253. Meier, M. 2017a. ‘A contest of interpretation: Roman policy toward the Huns as reflected in the “Honoria Affair” (448/50)’, JLA 10: 42–61. Meier, M. 2017b. ‘Der Monarch auf der Suche nach seinem Platz: Kaiserherrschaft im frühen Byzanz (5. bis 7. Jahrhundert n. Chr.)’, in S. Rebenich (ed.), Monarchische Herrschaft im Altertum. Berlin: Menéndez Argüín, A. R. 2011. ‘La guardia pretoriana en combate. II: Rutinas de entrenamiento, operaciones, tácticas y despliegues’, HABIS 42: 229–52. Mennell, S. 2006. ‘Note on the text’, in S. Mennell (ed.), The Court Society. The Collected Works of Norbert Elias, Volume 1. Dublin: xi–xv. Merrills, A. and Miles, R. 2014. The Vandals. Oxford. Messerschmidt, W. 2004. ‘Die statuarische Repräsentation der theodosianischen Kaiserhauses in Rome’, Röm. Mitt. 111: 555–68. Meurer, T. 2019. Vergangenes verhandeln. Spätantike Statusdiskurse senatorischer Eliten in Gallien und Italien. Berlin and Boston. Meyer, E. 1917. ‘Apollonios von Tyana and die Biographie des Philostratos’, Hermes 52: 371–424. Mian, G. 2012/13. ‘Il palazzo imperiale tardo-antico ad Aquileia. Note sullo stato della questione’, Aquileia Nostra 83–4: 89–95. Michel, A.-C. 2015. La cour sous l’empereur Claude: Les enjeux d’un lieu de pouvoir. Rennes. Michels, C. 2018. Antoninus Pius und die Rollenbilder des römischen Princeps. Herrscherliches Handeln und seine Repräsentation in der Hohen Kaiserzeit. Berlin. Michels, C. 2022. ‘Frontos gratiarum actio. Ein politisches Ritual zwischen Schmeichelei und “affirmativem Fordern” ’, Hermes 150: 443–66. Michels, C. 2023. ‘Die Briefe Frontos und senatorische Interaktion mit dem Princeps in der Hohen Kaiserzeit’, Hermes 151: 50–70. Millar, F. G. B. 1971. ‘Paul of Samosata, Zenobia and Aurelian: the church, local culture and political allegiance in third-century Syria’, JRS 61: 1–17. Millar, F. G. B. 1977. The Emperor in the Roman World, 31 BC–AD 337. London. Millar, F. G. B. 1999. ‘The Greek East and Roman law: the dossier of M. Cn. Licinius Rufus’, JRS 89: 90–108. Millar, F. G. B. 2006. A Greek Roman Empire. Power and Belief under Theodosius II (408–450). Berkeley. Miron, D. 2000. ‘Transmitters and representatives of power: royal women in ancient Macedonia’, AncSoc 30: 35–52. Missiou, D. 1982. ‘Über die institutionelle Rolle der byzantinischen Kaiserin’, Jahrbuch der österreichischen Byzantinistik 32: 489–98. Moeser, M. C. 2002. The Anecdote in Mark, the Classical World, and the Rabbis. Sheffield. Moffatt, A. and Tall, M. (eds) 2012. Constantine Porphyrogennetos: The Book of Ceremonies. Byzantina Australiensia 18. Leiden and Boston. Moles, J. 1983. ‘The date and purpose of the Fourth Kingship Oration of Dio Chrysostom’, Classical Antiquity 2: 251–78. Momigliano, A. 1973. ‘Freedom of speech in antiquity’, in P. P. Wiener (ed.), Dictionary of the History of Ideas, Volume 2. New York: 252–63. Mommsen, T. 1887–8. Römisches Staatsrecht. Leipzig. Mommsen, T. 1903. ‘Stilicho und Alarich’, Hermes 38: 101–15. Moormann, E. M. and Stocks, C. 2021. ‘Identifying demigods: Augustus, Domitian and Hercules’, in R. Marks and M. Mogetta (eds), Domitian’s Rome and the Augustan Legacy. Ann Arbor: 79–101.



385

Moote, A. L. 1989. Louis XIII, The Just. Berkeley. Moote, A. L. 1992. ‘Richelieu as Chief Minister. A comparative study of the favorite in early seventeenth-century politics’, in J. Bergin and L. W. B. Brockliss (ed.), Richelieu and His Age. Oxford: 13–43. Morgan, T. 2007. Popular Morality in the Early Roman Empire. Cambridge. Morris, R. 2004. ‘What did the epi tôn deêseôn actually do?’ in D. Feisel and J. Gascou (eds), La pétition à Byzance. Paris: 125–40. Moser, M. 2018. Emperors and Senators in the Reign of Constantius II. Maintaining Imperial Rule between Rome and Constantinople in the Fourth Century AD. Cambridge. Motley, M. E. 1990. Becoming a French Aristocrat: The Education of the Court Nobility, 1580–1715. Princeton, NJ. Muir, E. 1997. Ritual in Early Modern Europe. Cambridge. Mullen, A. 2015. ‘In both our languages: Greek-Latin code-switching in Roman literature’, Language and Literature 24: 213–32. Müller, C. 2011. ‘Redressing injustice: Maẓālim jurisdictions at the Umayyad Court of Córdoba (eighth-eleventh centuries CE)’, in A. Fuess and J.-P. Hartung (eds), Court Cultures in the Muslim World: Seventh to Nineteenth Centuries. London: 93–104. Müller, G. (ed.) 1952. Lexicon Athanasianum. Berlin. Mullett, M. and Ousterhout, R. (eds) 2020. The Holy Apostles. A Lost Monument, a Forgotten Project, and the Presentness of the Past. With Appendixes prepared by Fani Gargova. Washington, DC. Murphy, N. 2016. ‘The court on the move: ceremonial entries, gift-giving and access to the monarch in France, c. 1440–c.1570’, in D. Raeymaekers and S. Derks (eds), The Key to Power? The Culture of Access in Princely Courts, 1400–1750. Leiden and Boston: 40–64. Musurillo, H. (ed.) 1954. The Acts of the Pagan Martyrs. Oxford. Nader, H. 1988. ‘Habsburg ceremony in Spain: the reality of the myth’, Historical Reflections/Réflexions Historiques 15: 293–309. Nathan, G. 2018. ‘Galla Placidia, Aelia’, in O. Nicholson (ed.), The Oxford Dictionary of Late Antiquity. Oxford: 637. Nau, F. 1919. ‘La letter à Cosme’, Patrologia Orientalis 13: 273–86. Necipoğlu, G. 1991. Architecture, Ceremonial, and Power. The Topkapı Palace in the Fifteenth and Sixteenth Centuries. New York. Nelson, J. L. 1976. ‘Symbols in context: rulers’ inauguration rituals in Byzantium and the west in the early middle ages’, Studies in Church History 13: 97–119. Nelson, J. L. 2019. King and Emperor. A New Life of Charlemagne. London. Nesselhauf, H. 1964. ‘Patrimonium und res privata des römischen Kaisers’, in J. Straub (ed.), Bonner Historia-Augusta-Colloquium 1963. Bonn: 73–93. Newlands, C. E. 2002. Statius’ Silvae and the Poetics of Empire. Cambridge. Newlands, C. E. 2012. Statius, Poet between Rome and Naples. London. Niquet, H. 2000. Monumenta virtutum titulique. Senatorische Selbstdarstellung im spätantiken Rom im Spiegel der epigraphischen Denkmäler. Stuttgart. Nissen, T. (ed.) 1912. S. Abercii Vita. Leipzig. Nixon, C. E. V. and Rodgers, B. S. 1994. In Praise of Later Roman Emperors: The Panegyrici Latini. Berkeley. Noel, J. G. 1995. ‘Peripheral ingroup membership, status and public negativity toward outgroups’, Journal of Personality and Social Psychology 68: 127–37. Noethlichs, K. L. 1998. ‘Strukturen und Funktionen des spätantiken Kaiserhofes’, in A. Winterling (ed.), Comitatus: Beiträge zur Erforschung des spätantiken Kaiserhofes. Berlin: 13–49.

386



Noreña, C. F. 2007a. ‘Hadrian’s chastity’, Phoenix 61: 296–317. Noreña, C. F. 2007b. ‘The social economy of Pliny’s correspondence with Trajan’, AJPh 128: 239–77. Noreña, C. F. 2009. ‘The ethics of autocracy in the Roman world’, in R. K. Balot, A Companion to Greek and Roman Political Thought. Oxford and Malden, MA: 266–79. Norton, P. 2007. Episcopal Elections 250–600. Hierarchy and Popular Will in Late Antiquity. Oxford. O’Flynn, J. M. 1983. Generalissimos of the Western Roman Empire. Alberta. Olariu, C. 2005. ‘Datianus, Valentinian, and the rise of the Pannonian faction’, Historia 54: 351–4. Olden-Jørgensen, S. 2002. ‘State ceremonial, court culture, and political power in early modern Denmark, 1536–1746’, Scandinavian Journal of History 27: 65–76. Olovsdotter, C. 2005. The Consular Image: An Iconological Study of the Consular Diptychs. BAR International Series 1376. Oxford. Olson, K. 2022. ‘Dress, adornment, and self-presentation’, in B. Kelly and A. Hug (eds), The Roman Emperor and his Court, c. 30 BC–c. AD 300. Volume I: Historical Essays. Cambridge: 461–78. Olszaniec, S. 2013. Prosopographical Studies on the Court Elite in the Roman Empire (4th century AD). Toruń. Omissi, A. and Ross, A. (eds) 2020. Imperial Panegyric from Diocletian to Honorius. Liverpool. Oost, S. I. 1968. Galla Placidia Augusta. A Biographical Essay. Chicago and London. Östenberg, I. 2015. ‘Power walks: aristocratic escorted movements in Republican Rome’, in I. Östenberg, S. Malmberg, and J. Bjørnebye (eds), The Moving City: Processions, Passages and Promenades in Ancient Rome. London: 13–23. Ostrogorsky, G. 1955. ‘Zur Kaisersalbung und Schilderhebung im spätbyzantinischen Krönungszeremoniell’, Historia 4: 246–56. O’Sullivan, T. 2011. Walking in Roman Culture. Cambridge. Pani, M. 2014. La corte dei Cesari fra Augusto e Nerone. Bari. Paravicini, W. 1991. ‘The court of the Dukes of Burgundy: a model for Europe?’, in R. G. Asch and A. M. Birke (eds), Princes, Patronage, and the Nobility: The Court at the Beginning of the Modern Age, c. 1450–1650. Oxford: 69–102. Parker, H. 1999. ‘The observed of all observers: spectacle, applause, and cultural poetics in the Roman theater audience’, in B. Bergmann and C. Kondoleon (eds), The Art of Ancient Spectacle. New Haven: 163–79. Parnell, D. A. 2017. Justinian’s Men. Careers and Relationships of Byzantine Army Officers, 518–610. London. Parvis, S. 2006. Marcellus of Ancyra and the Lost Years of the Arian Controversy, 325–45. Oxford. Paterson, J. 2007. ‘Friends in high places: the creation of the court of the Roman emperor’, in A. J. S. Spawforth (ed.), The Court and Court Society in Ancient Monarchies. Cambridge: 121–56. Pazdernik, C. F. 2010. ‘Paying attention to the man behind the curtain: disclosing and withholding the imperial presence in Justinianic Constantinople’, in T. Fögen and M. M. Lee (eds), Bodies and Boundaries in Graeco-Roman Antiquity. Berlin: 63–86. Pazdernik, C. F. 2012. ‘ “How then is it not better to prefer quiet, than the dangers of conflict?” The imperial court as the site of shifting cultural frontiers’, in D. Brakke, D. Deliyannis, and E. Watts (eds), Shifting Cultural Frontiers in Late Antiquity. London: 99–111.



387

Pedersen, F. S. 1976. Late Roman Public Professionalism. Odense. Peirce, L. P. 1993. The Imperial Harem: Women and Sovereignty in the Ottoman Empire. Oxford. Peirce, L. P. 2008. ‘Beyond harem walls: Ottoman royal women and the exercise of power’, in A. Walthall (ed.), Servants of the Dynasty: Palace Women in World History. Berkeley: 81–95. Penella, R. J. 1990. Greek Philosophers and Sophists in the Fourth Century AD. Studies in Eunapius of Sardis. Leeds. Perkins, J. 1995. The Suffering Self: Pain and Narrative Representation in the Early Christian Era. London. Perkins, J. 2009. Roman Imperial Identities in the Early Christian Era. London. Pernot, L. 2001. ‘Trois styles de portrait impérial au IIe siècle après J.-C.: Antonin chez Aelius Aristide, Pausanias, Marc Aurèle’, in A. Billaut (ed.), ὈΠΩΡΑ. La belle saison de l’hellénisme. Études de littérature antique offertes au Recteur Jacques Bompaire. Paris: 103–13. Perrie, M. 1987. The Image of Ivan the Terrible in Russian Folklore. Cambridge. Persson, F. 2015. ‘ “So that we Swedes are not more swine or goats than they are”: space and ceremony at the Swedish court’, in B. B. Johannsen and K. Ottenheym (eds), Beyond Scylla and Charybdis: European Courts and Court Residences outside Habsburg and Valois/Bourbon Territories, 1500–1700. Copenhagen: 121–33. Pervo, R. I. 2005. ‘Narratives about the apostles: non-canonical Acts and related literature’, in A. Gregory and C. Tuckett (eds), The Oxford Handbook of Early Christian Apocrypha. Oxford: 65–89. Petitjean, M. 2020. ‘Les unités de garde du comitatus impérial (193–312)’, in C. Wolff and P. Faure (eds), Corps du chef et gardes du corps dans l’armée romaine: Actes du septième congrès de Lyon sur l’armée romaine (25–27 octobre 2018). Paris: 433–49. Pfeilschifter, R. 2013. Der Kaiser und Konstantinopel. Kommunikation und Konfliktaustrag in einer spätantiken Metropole. Berlin and Boston. Pflaum, H.-G. 1961. ‘Les gendres de Marc-Aurèle’, Journal des Savants 1961.1: 28–41. Pflug, J. 2014. ‘Der Weg zum Kaiser. Wege durch den Kaiserpalast auf dem Palatin in Rom’, in D. Kurapkat, P. I. Schneider, and U. Wulf-Rheidt (eds), Die Architektur des Weges: Gestaltete Bewegung im gebauten Raum. Regensburg: 360–81. Pflug, J. and Wulf-Rheidt, U. 2022. ‘The imperial palaces on the Palatine hill: architecture as a reflection of social practices and imperial authority’, in B. Kelly and A. Hug (eds), The Roman Emperor and his Court, c. 30 BC–c. AD 300. Volume I: Historical Essays. Cambridge: 204–38. Pharr, C. (ed.) 1952. The Theodosian Code and Novels, and the Sirmondian Constitutions. New York. Pietras, H. 2016. Council of Nicaea (325): Religious and Political Context, Documents, Commentaries. Rome. Piganiol, A. 1947. L’Empire Chrétien (235–295). Paris. Pohl, W. 2013. ‘Ritualized encounters: late Roman diplomacy and the barbarians, fifth-sixth century’, in A. Beihammer, S. Constantinou, and M. G. Parani (eds), Court Ceremonies and Rituals of Power in Byzantium and the Medieval Mediterranean: Comparative Perspectives. Leiden and Boston: 67–86. Pond Rothman, M. S. 1977. ‘The thematic organization of the panel reliefs on the Arch of Galerius’, AJA 81: 427–54. Popović, I. 2007a. ‘Šarkamen. Eine Residenz- und Begräbnisstätte aus der Zeit des Maximins Daia’, in U. Brandl and M. Vasić (eds), Roms Erbe auf dem Balkan. Spätantike Kaiservillen und Stadtanlagen in Serbien. Mainz: 80–95.

388



Popović, I. 2007b. ‘Sirmium (Sremska Mitrovica). Residenzstadt der Römischen Kaiser und Stätte der Frühen Christen’, in U. Brandl and M. Vasić (eds), Roms Erbe auf dem Balkan. Spätantike Kaiservillen und Stadtanlagen in Serbien. Mainz: 17–32. Popović, I. 2011. ‘A residential complex in the south-eastern part of Late Antique Sirmium: written sources and archaeological evidence’, in G. Bülow and H. Zabehlicky (eds), Bruckneudorf und Gamzigrad. Spätantike Paläste und Großvillen im Donau-BalkanRaum. Akten des Internationalen Kolloquiums in Bruckneudorf vom 15. bis 18. Oktober 2008. Bonn: 177–85. Popović, I. 2016. ‘Porphyry sculptures from Sirmium’, AnTard 24: 371–90. Popović, I. 2019. ‘Lead seals with Tetrarchic busts from the imperial palace in Sirmium’, Starinar 69: 273–85. Porena, P. 2012. ‘I dignitari di Costantino: dinamiche di selezione e di ascesa durante la crisi del Sistema tetrarchio’, in G. Bonamente, N. Lenski, and R. L. Testa (eds), Costantino prima e dopo Costantino. Bari: 293–320. Porena, P. 2018. ‘Organizzare il silenzio nel palazzo imperiale: i silentarii’, Politica Antica 8: 165–87. Portmann, W. 1999. ‘Die politische Krise zwischen den Kaisern Constantius II. und Constans’, Historia 48: 301–30. Potter, D. 2010. ‘The unity of the Roman empire’, in S. McGill, C. Sogno, and E. Watts (eds), From the Tetrarchs to the Theodosians: Later Roman History and Culture, 284–250 CE. Cambridge: 13–32. Potter, D. 2011. ‘Holding court in Republican Rome (100–44)’, AJPh 132: 59–80. Potter, D. and Talbert, R. (eds) 2011. Classical Courts and Courtiers. American Journal of Philology 132.1. Baltimore. Price, R. (ed.) 1991. Lives of the Monks of Palestine. Kalamazoo, MI. Price, R. and M. Gaddis (eds) 2005. The Acts of the Council of Chalcedon. Liverpool. Price, R. and Graumann, T. (eds) 2020. The Council of Ephesus of 431: Documents and Proceedings. Liverpool. Price, S. 1987. ‘From noble funerals to divine cult: the consecration of Roman emperors’, in D. Cannadine and S. Price (eds), Rituals of Royalty: Power and Ceremonial in Traditional Societies. Cambridge: 56–105. Priwitzer, S. 2010. ‘Dynastisches Potential von Kaiserfrauen im Prinzipat am Beispiel der Faustina minor. Tochter, Ehefrau und Mutter’, in A. Kolb (ed.), Augustae. Machtbewußte Frauen am römischen Kaiserhof? Berlin: 237–51. Purcell, N. 1986. ‘Livia and the womanhood of Rome’, PCPhS 32: 78–105. Raeymaekers, D. and Derks, S. 2016. ‘Repertoires of access in princely courts’, in D. Raeymaekers and S. Derks (eds), The Key to Power? The Culture of Access in Princely Courts, 1400–1750. Leiden and Boston: 1–15. Rankov, B. 1994. The Praetorian Guard. London. Rapp, C. 1998. ‘Imperial ideology in the making: Eusebius of Caesarea on Constantine as “bishop” ’, JThS 49: 685–95. Rapp, C. 2005. Holy Bishops in Late Antiquity. The Nature of Christian Leadership in an Age of Tradition. Berkeley. Rapp, C. 2012. ‘Death at the Byzantine court: the emperor and his family’, in K.-H. Spieß and I. Warntjes (eds), Death at Court. Wiesbaden: 267–86. Rawski, E. S. 1998. The Last Emperors: A Social History of Qing Imperial Institutions. Berkeley. Rawson, E. 1975. ‘Caesar’s heritage: Hellenistic kings and their Roman equals’, JRS 65: 148–59.



389

Rebenich, S. 2000. ‘Vom dreizehnten Gott zum dreizehnten Apostel. Der tote Kaiser in der Spätantike’, Zeitschrift für Antikes Christentum 4: 300–24. Rebenich, S. 2008. ‘Jones and continental scholarship’, in D. Gwynn (ed.), A. H. M. Jones and the Later Roman Empire. Leiden: 43–62. Rees, R. 1993. ‘Images and image: a re-examination of Tetrarchic iconography’, G&R 40: 181–200. Rees, R. 2001. ‘To be and not to be: Pliny’s paradoxical Trajan’, Bulletin of the Institute of Classical Studies 45: 149–68. Rees, R. 2002. Layers of Loyalty in Latin Panegyric. Oxford. Rees, R. 2012. ‘Bright lights, big city: Pacatus and the Panegyrici Latini’, in L. Grig and G. Kelly (eds), Two Romes: Rome and Constantinople in Late Antiquity. Oxford: 203–22. Rees, R. 2018. ‘Authorising freedom of speech under Theodosius’, in D. Burgersdijk and A. Ross (eds), Imagining Emperors in the Later Roman Empire. Leiden and Boston: 289–309. Reinhardt, N. 2007. ‘The king’s confessor: changing images’, in M. Schaich (ed.), Monarchy and Religion. The Transformation of Royal Culture in Eighteenth-Century Europe. Oxford: 153–85. Reinhardt, N. 2009. ‘Spin doctor of conscience? The royal confessor and the Christian prince’, Renaissance Studies 23: 568–90. Reiske, J. J. 1829 (ed.) Constantini Porphyrogeniti imperatoris De cerimoniis aulae Byzantinae libri duo. Bonn. Rémy, B. 2005. Antonin le Pieux, 138–161. Le siècle dʼor de Rome. Paris. Ricci, C. 2004. ‘Il principe in villa. Residenze imperiali in Italia e servizi di sicurezza’, Cahiers du Centre Gustave Glotz 15: 317–41. Ricci, C. 2017. Security in Roman Times: Rome, Italy and the Emperors. London. Richard, J.-C. 1966. ‘Les aspects militaires des funérailles imperiales’, Mélanges d’archéologie et d’histoire 78: 313–25. Richlin, A. 2006a. ‘Fronto + Marcus: Love, Friendship, Letters’, in M. Kuefler (ed.), The Boswell Thesis: Essays on Christianity, Social Tolerance, and Homosexuality. Chicago: 111–29. Richlin, A. (ed.) 2006b. Marcus Aurelius in Love. Chicago. Richlin, A. 2011. ‘Parallel Lives: Domitia Lucilla and Cratia, Fronto and Marcus’, Eugesta 1: 163–203. Ridley, R. (ed.) 1982. Zosimus. New History. A Translation with Commentary. Sydney. Rilinger, R. 1997. ‘Domus und res publica. Die politisch-soziale Bedeutung des aristokratischen “Hauses” in der späten römischen Republik’, in A. Winterling (ed.), Zwischen ‘Haus’ und ‘Staat’. Antike Höfe im Vergelich. Historische Zeitschrift. Beihefte 23. Munich: 73–90. Ringrose, K. M. 2003. The Perfect Servant. Eunuchs and the Social Construction of Gender in Byzantium. Chicago. Ringrose, K. M. 2008. ‘Women and power and the Byzantine court’, in A. Walthall (ed.), Servants of the Dynasty: Palace Women in World History. Berkeley, CA. Rist, J. 1965. ‘Hypatia’, Phoenix 19: 214–25. Rogers, R. S. 1959. ‘The emperor’s displeasure—amicitiam renuntiare’, TAPA 90: 224–37. Rodgers, B. S. 1989. ‘Eumenius of Augustodunum’, AncSoc 20: 249–66. Rohrbacher, D. 2016. The Play of Illusion in the Historia Augusta. Madison. Roller, M. 2022. ‘Dining and hunting as courtly activities’, in B. Kelly and A. Hug (eds), The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge: 318–48.

390



Rollinger, C. 2021. Gesten der Macht. Das Zeremoniell des Kaiserhofs in Konstantinopel (4.– 7. Jh.). Habilitationsschrift (unpublished). Trier. Rollinger, C. 2023a. ‘Specie dominationis: The “ceremonial” uses of imperial bodyguards under the Principate’, in M. Hebblewhite and C. Whately (eds), Brill’s Companion to Bodyguards in the Ancient Mediterranean. Leiden and Boston: 223–48. Rollinger, C. 2023b. ‘These boots aren’t made for walking: Tetrarchic court ceremonial as a language of authority’, in F. Carlà-Uhink and C. Rollinger (eds), The Tetrarchy as Ideology: Reconfigurations and Representations of an Imperial Power. Stuttgart: 93–118. Rollinger, C. forthcoming. Put a Crown on It: Imperial Coronations in the Seventh Century, in B. Bleckmann, N. Viermann, and J. Wienand (eds), Reading the Late Roman Monarchy. Stuttgart. Rordorf, W. 1982. ‘Die neronische Christenverfolgung im Spiegel der apokryphen Paulusakten’, New Testament Studies 28: 365–74. Rose, M. E. 2006. ‘The Trier ceiling: power and status on display in Late Antiquity’, G&R 53: 92–109. Rosen, K. 1996. ‘Marc Aurel und das Ideal des civilis princeps’, in G. Schöllgen and C. Scholten (eds), Stimuli. Exegese und ihre Hermeneutik in Antike und Christentum. Münster: 154–60. Rosenmeyer, P. A. 2018. The Language of Ruins: Greek and Latin Inscriptions on the Memnon Colossus. Oxford. Rosenwein, B. H. 2006. Emotional Communities in the Early Middle Ages. Ithaca, NY and London. Ross, A. 2021. ‘Envisioning adventus: Ammianus between panegyric and polemic’, JLA 14: 97–116. Rougé, J. (ed.) 1966. Expositio totius mundi et gentium. Introduction, texte critique, traduction, notes et commentaire. Paris. Roullier, P.-H. 2010. ‘Du sage au disciple: autorité charismatique, domination romaine et identité grecque dans la Vie d’Apollonios de Tyane de Philostrate’, Pallas 83: 359–75. Ruffini, G. R. 2018. Life in an Egyptian Village in Late Antiquity. Aphrodito Before and After the Islamic Conquest. Cambridge. Russell, N. 2000. Cyril of Alexandria. London. Sakurai, M. 1984. ‘On the vocabulary and kinship terminology of the imperial family of Japan’, Language Sciences 6.1: 53–72. Salisbury, J. 2015. Rome’s Christian Empress. Galla Placidia Rules at the Twilight of the Empire. Baltimore. Saller, R. P. 1980. ‘Anecdotes as historical evidence for the Principate’, G&R 27: 69–83. Salzman, M. R. 1990. On Roman Time. The Codex-Calendar of 354 and the Rhythms of Urban Life in Late Antiquity. Berkeley. Salzman, M. R. 2002. The Making of a Christian Aristocracy: Social and Religious Change in the Western Roman Empire. Cambridge, MA. Sanders, P. A. 1994. Ritual, Politics, and the City in Fatimid Cairo. Albany, NY. Şare Ağtürk, T. 2018. ‘A new Tetrarchic relief from Nicomedia: embracing emperors”, AJA 122: 411–26. Şare Ağtürk, T. 2020a. ‘Myth and eponymy on the Tetrarchic frieze from Nicomedia’, JRA 33: 417–31. Şare Ağtürk, T. 2020b. ‘The self-image of a new imperial capital city on the Tetrarchic reliefs of Nicomedia’, in A. Lichtenberger, T. Şare Ağtürk, E. Winter, and K. Zimmermann (eds), Imperial Residence and Site of Councils. The Metropolitan Region of Nicaea/Nicomedia. Bonn: 107–16.



391

Savalli-Lestrade, I. 2017. ‘Βίος αὐλικός: the multiple ways of life of courtiers in the Hellenistic age’, in A. Erskine, L. Llewellyn-Jones, and S. Wallace (eds), The Hellenistic Court: Monarchic Power and Elite Society from Alexander to Cleopatra. Swansea: 101–20. Schalk, E. 1991. ‘The court as “civilizer” of the nobility: noble attitudes and the court in France in the late sixteenth and early seventeenth centuries’, in R. G. Asch and A. M. Birke (eds), Princes, Patronage, and the Nobility: The Court at the Beginning of the Modern Age, c. 1450–1650. Oxford: 245–63. Scharf, R. 1994. Comites und comitiva primi ordinis. Stuttgart. Scheidel, W. 1999. ‘Emperors, aristocrats, and the grim reaper: towards a demographic profile of the Roman élite’, CQ 49: 254–81. Scheidel, W. 2009a. ‘Introduction’, in W. Scheidel (ed.), Rome and China: Comparative Perspectives on Ancient World Empires. Oxford: 3–10. Scheidel, W. 2009b. ‘Sex and empire: a Darwinian perspective’, in I. Morris and W. Scheidel (eds), The Dynamics of Ancient Empires: State Power from Assyria to Byzantium. Oxford: 255–324. Schlinkert, D. 1996. ‘Vom Haus zum Hof. Aspkete höfische Herrschaft in der Spätantike’, Klio 78: 452–82. Schlinkert, D. 1998. ‘Dem Kaiser folgen: Kaiser, Senatsadel und höfische Funktionselite (comites consistoriani) von der “Tetrarchie” Diokletians bis zum Ende der konstantinischen Dynastie’, in A. Winterling (ed.), Comitatus: Beiträge zur Erforschung des spätantiken Kaiserhofes. Berlin: 133–59. Schlinkert, D. 2002. ‘Between emperor, court, and senatorial order: the codification of the Codex Theodosianus’, AncSoc 32: 283–94. Schmidt-Hofner, S. 2008. Reagieren und Gestalten. Der Regierungsstil des spätrömischen Kaisers am Beispiel der Gesetzgebung Valentinians I. Munich. Schmidt-Hofner, S. 2012. ‘Trajan und die symbolische Kommunikation bei kaiserlichen Rombesuchen in der Spätantike’, in R. Behrwald and C. Witschel (eds), Rom in der Spätantike. Historische Erinnerung im städtischen Raum. Stuttgart: 33–59. Schmidt-Hofner, S. 2015. ‘Ostentatious legislation: law and dynastic change, AD 364–365’, in J. Wienand (ed.), Contested Monarchy: Integrating the Roman Empire in the Fourth Century AD. Oxford: 67–99. Schmitz, T. A. 2009. ‘Narrator and audience in Philostratus’ Lives of the Sophists’, in E. L. Bowie and J. Elsner (eds), Philostratus. Cambridge: 69–99. Schnurbusch, D. 2011. Convivium. Form und Bedeutung aristokratischer Geselligkeit in der römischen Antike. Stuttgart. Schrijver, F. 2018. ‘The court of women in early Palaiologan Byzantium (ca. 1260–1350)’, BMGS 42: 219–36. Schoell, R. and W. Kroll (eds) 1895. Corpus Iuris Civilis. Volumen Tertium. Novellae. Berlin. Scholten, H. 1995. Der Eunuch in Kaisernähe. Zur politischen und sozialen Bedeutung des ‘praepositus sacri cubiculi’ im 4. und 5. Jahrhundert n. Chr. Frankfurt am Main. Schöpe, B. 2014. Der römische Kaiserhof in severischer Zeit (193–235 n. Chr.). Stuttgart. Schulz, F. 2014. ‘Ambrosius, die Kaiser und das Ideal des christlichen Ratgebers’, Historia 63: 214–42. Schulz, F. 2019. ‘Jovian, an emperor who did not bow to heretics and infidels? A critical reading of the Petitiones Arianorum’, in K. C. Choda, M. S. de Leeuw, and F. Schulz (eds), Gaining and Losing Imperial Favour in Late Antiquity. Representation and Reality. Leiden: 193–218.

392



Schumacher, F. 2001. ‘Hausgesinde—Hofgesinde: Terminologische Überlegungen zur Funktion der familia Caesaris im 1. Jh. n. Chr’, in H. Bellen and H. Heinen (eds), Fünfzig Jahre Forschungen zur antiken Sklaverei an der Mainzer Akademie 1950–2000: Miscellenea zum Jubiläum. Stuttgart: 331–52. Schumacher, F. 2013. ‘Norbert Elias’ “sozialer Habitus” als Vorläufer des Bourdieu’schen Habitus? Eine vergleichende Analyse’, in A. Lenger, C. Schneickert, and F. Schumacher (eds), Pierre Bourdieus Konzeption des Habitus. Grundlagen, Zugänge, Forschungsperspektiven. Wiesbaden: 131–45. Schwartz, D. R. 2009. ‘Philo, his family, and his times’, in A. Kamesar (ed.), The Cambridge Companion to Philo. Cambridge: 9–31. Scott, J. C. 1977. ‘Protest and profanation: agrarian revolt and the little tradition, parts 1 and 2’, Theory and Society 4: 1–38, 211–46. Scott, J. C. 1990. Domination and the Arts of Resistance: Hidden Transcripts. New Haven and London. Seeck, O. 1919. Regesten der Kaiser und Päpste für die Jahre 311 bis 476 n. Chr. Stuttgart. Seelentag, G. 2004. Taten und Tugenden Traians. Herrschaftsdarstellung im Principat. Stuttgart. Seelentag, G. 2011. ‘Trajan, Hadrian und Antoninus Pius. Deutungsmuster und Perspektiven’, in A. Winterling (ed.), Zwischen Strukturgeschichte und Biographie. Probleme und Perspektiven einer römischen Kaisergeschichte 31 v.Chr.–192 n. Chr. Munich: 295–315. Seelentag, G. 2017. ‘Antoninus Pius und die Herrschaftsdarstellung des 2. Jhs.’, in C. Michels and P. F. Mittag (eds), Jenseits des Narrativs. Antoninus Pius in den nicht literarischen Quellen. Stuttgart: 19–30. Seymour, M. 2012. ‘Emotional arenas: from provincial circus to national courtroom in late nineteenth-century Italy’, Rethinking History 16: 177–97. Shaw, B. 2011. Sacred Violence. African Christians and Sectarian Hatred in the Age of Augustine. Cambridge. Sigmund, C. 2014. ‘Königtum’ in der politischen Kultur des spätrepublikanischen Rom. Berlin. Silver, L. 2008. Marketing Maximilian. The Visual Ideology of a Holy Roman Emperor. Princeton, NJ and London. Sivan, H. 1993. Ausonius of Bordeaux: Genesis of a Gallic Aristocracy. London and New York. Sivan, H. 2011. Galla Placidia. The Last Roman Empress. Oxford. Slootjes, D. 2004. ‘Governor trumped by bishop: shifting boundaries in Roman religious and public life’, in L. de Blois, P. Funke, and J. Hahn (eds), The Impact of Imperial Rome on Religions, Ritual and Religious Life in the Roman Empire. Leiden and Boston: 219–31. Slootjes, D. 2011. ‘Bishops and their position of power in the late third century CE: the cases of Gregory Thaumaturgus and Paul of Samosata’, JLA 4: 100–15. Smallwood, E. M. 1959. ‘The legislation of Hadrian and Antoninus Pius against circumcision’, Latomus 18: 334–47. Smallwood, E. M. (ed.) 1970. Philonis Alexandrini Legatio ad Gaium. Leiden. Smallwood, E. M. 1981. The Jews under Roman Rule: From Pompey to Diocletian. A Study in Political Relations. Leiden. Smith, R. 2007. ‘The imperial court of the late Roman Empire, c. AD 300–c. AD 450’, in A. J. S. Spawforth (ed.), The Court and Court Society in Ancient Monarchies. Cambridge: 157–232.



393

Smith, R. 2011. ‘Measures of difference: the fourth-century transformation of the Roman imperial court’, AJP 132: 125–51. Smith, R. R. R. 2021. ‘Maiestas serena: Roman court cameos and early imperial poetry and panegyric’, JRS 111: 75–152. Snee, R. 1998. ‘Gregory Nazianzen’s Anastasia church: Arianism, the Goths, and hagiography’, DOP 52: 157–86. Sode, C. 2009. ‘Die Krönung des Kaisers Justin I. im Zeremonienbuch Konstantins VII. Porphyrogennetos’, Mediterraneo Antico 12: 429–48. Sode, C. 2011. ‘Sammeln und Exzerpieren in der Zeit Konstantins VII. Porphyrogennetos: Zu den Fragmenten des Petros Patrikios im Sogenannten Zeremonienbuch’, in P. van Deun and C. Macé (eds), Encyclopedic Trends in Byzantium? Leuven: 161–76. Sogno, C. 2006. Q. Aurelius Symmachus. A Political Biography. Ann Arbor. Southern, P. 1997. Domitian: Tragic Tyrant. London. Spangler, J. 2016. ‘Holders of the keys: the grand chamberlain, the grand equerry and monopolies of access at the early modern French court’, in D. Raeymaekers and S. Derks (eds), The Key to Power? The Culture of Access in Princely Courts, 1400–1750. Leiden and Boston: 155–77. Spawforth, A. J. S. (ed.) 2007a. The Court and Court Society in Ancient Monarchies. Cambridge. Spawforth, A. J. S. 2007b. ‘Introduction’, in A. J. S. Spawforth (ed.), The Court and Court Society in Ancient Monarchies. Cambridge: 1–16. Speidel, M. P. 1994. Riding for Caesar. The Roman Emperor’s Horse Guard. London. Speidel, M. A. 2017. ‘Der Philosoph als Imperator. Marc Aurel und das Militär’, in V. Grieb (ed.), Marc Aurel—Wege zu seiner Herrschaft. Gutenberg: 49–74. Spence, K. 2007. ‘Court and palace in ancient Egypt: the Amarna period and later Eighteenth Dynsty’, in A. J. S. Spawforth (ed.), The Court and Court Society in Ancient Monarchies. Cambridge: 267–328. Spieser, J.-M. 2015. ‘Réflexions sur le Palais de Galère à Thessalonique’, in M. Featherstone, J.-M. Spieser, G. Tanman, and U. Wulf-Rheidt (eds), The Emperor’s House. Palaces from Augustus to the Age of Absolutism. Berlin: 19–30. Squatriti, P. (ed.) 2007. The Complete Works of Liudprand of Cremona. Washington, DC. Starkey, D. 1987a. ‘Introduction: court history in perspective’, in D. Starkey (ed.), The English Court from the Wars of the Roses to the Civil War. London and New York: 1–24. Starkey, D. 1987b. ‘Intimacy and innovation: the rise of the Privy Chamber, 1485–1547’, in D. Starkey (ed.), The English Court from the Wars of the Roses to the Civil War. London and New York: 71–118. Steininger, B. 2017. Chinese Literary Form in Heian Japan: Poetics and Practice. Cambridge, MA. Stephens, C. W. B. 2015. Canon Law and Episcopal Authority. The Canons of Antioch and Serdica. Oxford. Stern, H. 1954. ‘Remarks on the “Adoratio” under Diocletian’, Journal of the Warburg and Courtauld Institutes 17: 184–9. Stevenson, W. 1995. ‘The rise of eunuchs in Greco-Roman antiquity’, Journal of the History of Sexuality 5: 495–511. Stickler, T. 2002. Aetius: Gestaltungsspielräume eines Heermeisters im ausgehenden Weströmisches Reich. Munich. Stradling, R. A. 1988. Philip IV and the Government of Spain, 1621–1665. Cambridge. Strathern, A. 2019. Unearthly Powers. Religious and Political Change in World History. Cambridge.

394



Straub, J. 1952. ‘Parens principum: Stilichos Realpolitik und das Testament des Kaisers Theodosius’, La Nouvelle Clio 4: 94–112. Stroheker, F. K. 1970. ‘Princeps clausus. Zu einigen Berührungen der Literatur des fünften Jahrhunderts mit der Historia Augusta’, in J. Straub (ed.), Bonner Historia-AugustaColloquium 1968/1969. Bonn: 273–83. Strootman, R. 2011. ‘Hellenistic court society: the Seleukid imperial court under Antiochos the Great, 223–187 BCE’, in J. Duindam, T. Artan, and M. Kunt (eds), Royal Courts in Dynastic States and Empires: A Global Perspective. Leiden and Boston: 63–89. Strootman, R. 2014. Courts and Elites in the Hellenistic Empires. The Near East after the Achaemenids, c. 330–30 BCE. Edinburgh. Strootman, R. and Versluys, M. (eds) 2017. Persianism in Antiquity. Stuttgart. Swain, S. 1996. Hellenism and Empire: Language, Classicism and Power in the Greek World, AD 50–250. Oxford. Swain, S. 1997. ‘Biography and biographic in the literature of the Roman empire’, in M. J. Edwards and S. Swain (eds), Portraits: Biographical Representation in the Greek and Latin Literature of the Roman Empire. Oxford: 1–37. Syme, R. 1972. ‘Lawyers in government: the case of Ulpian’, Proc. Am. Phil. Soc. 116: 406–9. Täckholm, U. 1969. ‘Aetius and the battle on the Catalaunian Fields’, Opus. Rom. 7: 259–76. Tajfel, H. 1982. ‘Social psychology of intergroup relations’, Annual Review of Psychology 33: 1–39. Tajfel, H. and Turner, J. C. 1986. ‘The social identity theory of intergroup behaviour’, in S. Worchel and W. G. Austin (eds), Psychology of Intergroup Relations. Chicago: 7–24. Talbot, M. 2016. ‘Accessing the shadow of God: spatial and performative ceremonial at the Ottoman court’, in D. Raeymaekers and S. Derks (eds), The Key to Power? The Culture of Access in Princely Courts, 1400–1750. Leiden and Boston: 103–23. Tamiolaki, M. 2017. ‘Xenophon’s Cyropaedia: tentative answers to an enigma’, in M. A. Flower (ed.), The Cambridge Companion to Xenophon. Cambridge: 174–94. Tanner, N. 1990. Decrees of the Ecumenical Councils. 2 vols. London. Tantillo, I. 2015. ‘I cerimoniali di corte in età tardoromana (284–395 d.C.)’, in Le corti nell’alto Medioevo. Spoleto, 24–29 aprile 2014. Spoleto: 543–84. Taoka, Y. 2013a. ‘Liminal women in Fronto’s letters’, CJ 108: 419–45. Taoka, Y. 2013b. ‘The correspondence of Fronto and Marcus Aurelius: love, letters, metaphor’, ClAnt 32: 406–38. Teitler, H. 2002. ‘Raising on a shield: origin and afterlife of a coronation ceremony’, International Journal of the Classical Tradition 8: 501–21. Thelander, D. R. 1982. ‘Mother Goose and her goslings: the France of Louis XIV as seen through the fairy tale’, Journal of Modern History 54: 467–96. Thomson, R. W., Howard-Johnston, J., and Greenwood, T. (eds) 1999. The Armenian History Attributed to Sebeos. Liverpool. Thompson, E. A. 1947. The Historical Work of Ammianus Marcellinus. Cambridge. Thompson, I. A. A. 1999. ‘The institutional background to the rise of the ministerfavourite’, in J. H. Elliott and L. W. B. Brockliss (eds), The World of the Favourite. New Haven and London: 13–25. Thonemann, P. 2012. ‘Abercius of Hierapolis: Christianization and social memory in late antique Asia Minor’, in B. Dignas and R. R. R. Smith (eds), Historical and Religious Memory in the Ancient World. Oxford: 257–82. Thraede, K. 2008. ‘Kuss’, RAC 22: cols. 545–76. Timpanaro, S. 1987. ‘Il “ius osculi” e Frontone’, Maia 39: 201–11. Tougher, S. 1998. ‘The advocacy of an empress: Julian and Eusebia’, CQ 48: 595–9.



395

Tougher, S. 2008. The Eunuch in Byzantine History and Society. Abingdon and New York. Trampedach, K. 2005. ‘Kaiserwechsel und Krönungsritual in Konstantinopel des 5. bis 6. Jahrhunderts’, in M. Steinicke and S. Weinfurter (eds), Investitur- und Krönungsrituale. Herrschaftseinsetzungen im kulturellen Vergleich. Cologne: 275–90. Treggiari, S. 1974. ‘Jobs for women’, AJAH 1: 76–104. Treggiari, S. 1975. ‘Jobs in the household of Livia’, PBSR 43: 48–77. Treggiari, S. 1981. ‘Concubinae’, PBSR 49: 59–81. Tritle, L. A. 1994. ‘Whose tool? Ammianus Marcellinus on the emperor Valens’, AHB 8: 141–53. Tuckett, C. M. 2015. ‘Introduction: what is early Christian apocrypha?’ in A. Gregory and C. Tuckett (eds), The Oxford Handbook of Early Christian Apocrypha. Oxford: 3–12. Tuori, K. 2016. The Emperor of Law. The Emergence of Roman Imperial Adjudication. Oxford. Turcan, R. 1987. Vivre à la cour des Césars. D’Auguste à Dioclétien (Ier–IIIe siècles ap. J.-C.). Paris. Twitchett, D. C. and Tietze, K.-P. 1994. ‘The Liao’, in H. Franke and D. C. Twitchett (eds), The Cambridge History of China. Volume 6. Alien Regimes and Border States, 907–1368. Cambridge: 43–153. Ulrich, J. 1997. ‘Nicaea and the West’, Vigiliae Christianae 51: 10–24. Urrisk-Obertyński, R. M. 2004. Die k.u.k. Leibgarden am österreichisch-ungarischen Hof 1518–1918. Gnas. Vale, M. 2001. The Princely Court: Medieval Courts and Culture in North-West Europe, 1270–1380. Oxford. Valero Tévar, M. 2013. ‘The late-antique villa at Noheda (Villar de Domingo García) near Cuenca and its mosaics’, JRA 26: 307–30. van Berkel, M. 2019. ‘Politics of access at the court of the Caliph’, in D. Slootjes and H. Kaal (eds), New Perspectives on Power and Political Representation from Ancient History to the Present Day. Leiden and Boston: 26–36. van den Hout, M. P. J. (ed.) 1988. M. Cornelii Frontonis Epistulae. Second edition. Leipzig. van den Hout, M. P. J. 1999. A Commentary on the Letters of M. Cornelius Fronto. Leiden and Boston. van Ess, H. 2007. ‘The imperial court in Han China’, in A. J. S. Spawforth (ed.), The Court and Court Society in Ancient Monarchies. Cambridge: 233–66. van Leeuwen, R. 2017. Narratives of Kingship in Eurasian Empires, 1300–1800. Leiden and Boston. van Nuffelen, P. 2010a. ‘Beyond bureaucracy: ritual mediation in in Late Antiquity’, in M. Kitts, B. Schneidmüller, G. Schwedler, E. Tounta, H. Kulke, and U. Skoda (eds), State, Power, and Violence. Ritual Dynamics and the Science of Ritual III. Wiesbaden: 231–46. van Nuffelen, P. 2010b. ‘Episcopal succession in Constantinople (381–450 C.E.): the local dynamics of power’, JECS 18: 425–51. van Nuffelen, P. 2012. ‘Playing the ritual game in Constantinople’, in L. Grig and G. Kelly (eds), Two Romes: Rome and Constantinople in Late Antiquity. Oxford: 183–201. van Uytfanghe, M. 1993. ‘L’hagiographie: un “genre” chrétien ou antique tardif ’, Analecta Bollandiana 111: 135–88. Vanderspoel, J. 1995. Themistius and the Imperial Court: Oratory, Civic Duty, and Paideia from Constantius to Theodosius. Ann Arbor. Viggiani, M. C. 2009. ‘La Vita di S. Melania iuniore di Geronzio. Tra storia e agiografia’, Maia 61: 324–44.

396



Vlassopoulos, K. 2017. ‘Xenophon on Persia’, in M. A. Flower (ed.), The Cambridge Companion to Xenophon. Cambridge: 360–75. Vogel, L. 1973. The Column of Antoninus Pius. Cambridge, MA. von Hesberg, H. 2006. ‘Residenzstädte und ihre höfische Infrastruktur—traditionelle und neue Raumkonzepte’, in D. Boschung and W. Eck (eds), Die Tetrarchie. Ein neues Regierungssystem und seine mediale Präsentation. Wiesbaden: 133–67. Vössing, K. 2004. Mensa Regia. Das Bankett beim hellenistischen König und beim römischen Kaiser. Munich. Vout, C. 2007. Power and Eroticism in Imperial Rome. Cambridge. Wakabayashi, B. T. 1991. ‘In name only: imperial sovereignty in early modern Japan’, Journal of Japanese Studies 17: 25–57. Walentowski, S. (ed.) 1998. Kommentar zur Vita Antoninus Pius der Historia Augusta. Bonn. Walford, E. (ed.) 1853. The Ecclesiastical History of Socrates, surnamed Scholasticus. London. Wallace-Hadrill, A. 1982. ‘Civilis princeps: between citizen and king’, JRS 72: 32–48. Wallace-Hadrill, A. 1995. Suetonius: The Scholar and His Caesars. Second edition. London. Wallace-Hadrill, A. 1996. ‘The imperial court’, in A. K. Bowman, E. Champlin, and A. Lintott (eds), The Cambridge Ancient History. Second Edition. Volume X: The Augustan Empire: 43 BC—AD 69. Cambridge: 283–308. Wallace-Hadrill, A. 2011. ‘The Roman imperial court: seen and unseen in the performance of power’, in J. Duindam, T. Artan, and M. Kunt (eds), Royal Courts in Dynastic States and Empires: A Global Perspective. Leiden and Boston: 91–102. Wallach. L. 1941. ‘The colloquy of Marcus Aurelius with the patriarch Judah I’, JQR 41: 259–86. Wallraff, M. 1997. Der Kirchenhistoriker Sokrates. Untersuchungen zu Geschichtsdarstellung, Methode und Person. Göttingen. Walter, C. 1975. ‘Raising on a shield in Byzantine iconography’, REB 33: 133–76. Walthall, A. 2008. ‘Introducing palace women’, in A. Walthall (ed.), Servants of the Dynasty: Palace Women in World History. Berkeley: 1–21. Wankerl, V. 2009. Appello ad principem: Urteilsstil und Urteilstechnik in kaiserlichen Berufungsentscheidungen (Augustus bis Caracalla). Munich. Wardle, D. (ed.) 2014. Suetonius. Life of Augustus. Oxford. Ware, C. 2013. ‘Learning from Pliny: Claudian’s advice to the emperor Honorius’, Arethusa 46: 313–31. Washington, B. 2020. ‘Playing in conventions in Julian’s Encomium to Eusebia: does gender make a difference?’, in A. Omissi and A. Ross (eds), Imperial Panegyric from Diocletian to Honorius. Liverpool: 93–116. Watson, L. and Watson, P. (eds) 2003. Martial. Select Epigrams. Cambridge. Watts, E. 2005. ‘Orality and communal identity in Eunapius’ Lives of the Sophists and Philosophers’, Byzantion 75: 334–61. Watts, E. 2010. Riot in Alexandria. Tradition and Group Dynamics in Late Antique Pagan and Christian Communities. Berkeley. Watts, E. 2013. ‘Theodosius II and anti-Chalcedonian communal memory’, in C. Kelly (ed.), Theodosius II. Rethinking the Roman Empire in Late Antiquity. Cambridge: 269–84. Weber, S. 1997. Die Chronik des Sulpicius Severus. Charakteristika und Intentionen. Trier. Wei, R. 2013. ‘Fronto and the rhetoric of friendship’, CEA 50: 67–93.



397

Wei, R. and Kelly, B. 2022. ‘The Roman aristocracy at court’, in B. Kelly and A. Hug (eds), The Roman Emperor and His Court c. 30 BC–c. AD 300. Volume 1: Historical Essays. Cambridge: 85–114. Weiser, B. 2003. Charles II and the Politics of Access. Woodbridge. Weiss, P. B. 1975. Consistorium und comites consistoriani. Untersuchungen zur Hofbeamtenschaft des 4. Jh. n. Chr. auf prosopographischer Grundlage. Würzburg. Weisweiler, J. 2012. ‘From equality to asymmetry: honorific statues, imperial power, and senatorial identity in late antique Rome’, JRA 25: 319–50. Wensinck, A. J. (ed.) 1913. Legends of Eastern Saints Chiefly from Syriac Sources. Volume II: The Legend of Hilaria. Leiden. Wenskus, O. 2001. ‘Wie schreibt man einer Dame?’, Wiener Studien 2001: 215–32. Wessel, S. 2001. ‘The ecclesiastical policy of Theodosius II’, Annuarium Historiae Conciliorum 33: 285–308. Wessel, S. 2004. Cyril of Alexandria and the Nestorian Controversy: the Making of a Saint and of a Heretic. Oxford. Wheeler, E. 1996. ‘The laxity of Syrian legions’, in D. Kennedy (ed.), The Roman Army in the East. Ann Arbor: 229–76. Whitby, L. M. 2000a. ‘The army, c. 420–602’, in A. Cameron, B. Ward-Perkins, and M. Whitby (eds), The Cambridge Ancient History, Volume XIV: Late Antiquity: Empire and Successors, A.D. 425–600. Cambridge: 288–314. Whitby, L. M. 2000b. ‘Armies and society in the later Roman world’, in A. Cameron, B. Ward-Perkins, and M. Whitby (eds), The Cambridge Ancient History, Volume XIV: Late Antiquity: Empire and Successors, A.D. 425–600. Cambridge: 469–95. Whitby, L. M. 2008. ‘The role of the emperor’, in D. M. Gwynn (ed.), A. H. M. Jones and the Later Roman Empire. Leiden: 63–96. Whitby, M. 1987. ‘On the omission of a ceremony in mid-sixth century Constantinople: Candidati, curopalatus, silentiarii, excubitores and others’, Historia 36: 462–88. Whitby, M. 2006. ‘The St Polyeuktos epigram (AP 1.10): a literary perspective’, in S. Johnson (ed.), Greek Literature in Late Antiquity: Dynamism, Didacticism, Classicism. Aldershot: 159–87. White, C. (trans.) 1998. Early Christian Lives. Harmondsworth. Whitmarsh, T. 2007. ‘Prose literature and the Severan dynasty’, in S. Swain, S. Harrison, and J. Elsner (eds), Severan Culture: Cambridge: 29–51. Wieber, A. 2010. ‘Eine Kaiserin von Gewicht? Julians Rede auf Eusebia zwischen Geschlechtsspezifik, höfischer Repräsentation und kaiserlicher Matronage’, in A. Kolb (ed.), Augustae. Machtbewusste Frauen am römischen Kaiserhof? Berlin: 253–75. Wiesehöfer, J. 1996. Ancient Persia from 550 BC to 650 AD. Translated by A. Azodi. London. Wiesehöfer, J. and Huyse, P. (eds) 2006. Ērān und Anērān. Studien zu den Beziehungen zwischen dem Sasanidenreich und der Mittelmeerwelt. Munich. Wightman, E. M. 1970. Roman Trier and the Treviri. New York. Wijnendaele, J. W. P. 2015. The Last of the Romans. Bonifatius—Warlord and Comes Africae. London. Wijnendaele, J. W. P. 2017a. ‘Generalissimos and warlords in the late Roman West’, in T. Ñaco del Hoyo and F. López Sánchez (eds), War, Warlords, and Interstate Relations in the Ancient Mediterranean. Leiden: 427–51. Wijnendaele, J. W. P. 2017b. ‘The early career of Aëtius and the murder of Felix (c. 425–340 CE)’, Historia 66: 468–82.

398



Williams, C. A. 2012. Reading Roman Friendship. Cambridge. Winterling, A. 1986. Der Hof der Kurfürsten von Köln 1688–1794. Eine Fall-studie zur Bedeutung ‘absolutischer’ Hofhaltung. Bonn. Winterling, A. 1997a. ‘ “Hof ”. Versuch einer idealtypischen Bestimmung anhand der mittelalterlichen und frühneuzeitlichen Geschichte’, in A. Winterling (ed.), Zwischen ‘Haus’ und ‘Staat’. Antike Höfe im Vergelich. Historische Zeitschrift. Beihefte 23. Munich: 11–26. Winterling, A. 1997b. ‘Hof ohne “Staat”. Die aula Caesaris im 1. und 2. Jahrhunder n. Chr.’, in A. Winterling (ed.), Zwischen ‘Haus’ und ‘Staat’. Antike Höfe im Vergelich. Historische Zeitschrift. Beihefte 23. Munich: 91–112. Winterling. A. 1997c. ‘Vergleichende Perspektiven’, in A. Winterling (ed.), Zwischen ‘Haus’ und ‘Staat’. Antike Höfe im Vergelich. Historische Zeitschrift. Beihefte 23. Munich: 151–69. Winterling, A. (ed.) 1997d. Zwischen ‘Haus’ und ‘Staat’. Antike Höfe im Vergelich. Historische Zeitschrift. Beihefte 23. Munich. Winterling, A. (ed.) 1998. Comitatus. Beiträge zur Erforschung des spätantiken Kaiserhofes. Berlin: Akademie Verlag. Winterling, A. 1999. Aula Caesaris: Studien zur Institutionalisierung des römischen Kaiserhofes in der Zeit von Augustus bis Commodus (31 v. Chr.–192 n. Chr.). Munich. Winterling, A. 2009. Politics and Society in Imperial Rome. Oxford and Malden, MA. Winterling, A. 2011a. Caligula: A Biography. Berkeley. Winterling, A. 2011b. ‘Die Freundschaft der römischen Kaiser’, in A. Winterling (ed.), Zwischen Strukturgeschichte und Biographie. Probleme und Perspektiven einer neuen römischen Kaisergeschichte zur Zeit von Augustus bis Commodus. Munich: 207–32. Winterling, A. 2017. ‘Das römische Kaisertum des 1. und 2. Jahrhunderts n. Chr.’, in S. Rebenich (ed.), Monarchische Herrschaft im Altertum. Munich: 413–32. Witschel, C. 2004/05. ‘Trier und das spätantike Städtewesen im Westen des römischen Reiches’, Trierer Zeitschrift 67/68: 223–72. Wolff, C. and P. Faure (eds) 2020. Corps du chef et gardes du corps dans l’armée romaine: Actes du septième congrès de Lyon sur l’armée romaine (25–27 octobre 2018). Paris. Woods, D. 2002. ‘A misunderstood monogram: Ricimer or Severus?’, Hermathena 170/172: 5–21. Woods, D. 2017. ‘The late Roman “camp gate” reverse type and the sidus salutare’, NC 177: 159–74. Wortman, R. S. 2006. Scenarios of Power: Myth and Ceremony in Russian Monarchy from Peter the Great to the Abdication of Nicholas II. Princeton, NJ. Wrede, H. 1981. ‘Der Genius Populi Romani und das Fünfsäulendenkmal der Tetrarchen aufdem Forum Romanum’, BJ 181: 111–42. Wulf-Rheidt, U. 2007. ‘Residieren in Rom oder in der Provinz? Der Kaiserpalast Felix Romuliana im Spiegel der tetrarchischen Residenzbaukunst’, in U. Brandl and M. Vasić (eds), Roms Erbe auf dem Balkan. Spätantike Kaiservillen und Stadtanlagen in Serbien. Mainz: 59–79. Wulf-Rheidt, U. 2011. ‘Die Entwicklung der Residenz der römischen Kaiser auf dem Palatin vom aristokratischen Wohnhaus zum Palast’, in G. von Bülow and H. Zabehlicky (eds), Bruckneudorf und Gamzigrad. Spätantike Paläste und Großvillen im Donau-BalkanRaum. Akten des Internationalen Kolloquiums in Bruckneudorf 2008. Bonn: 1–18. Wulf-Rheidt, U. 2012. ‘Nutzungsbereiche des flavischen Palastes auf dem Palatin in Rom’, in F. Arnold, A. Busch, R. Haensch, and U. Wulf-Rheidt (eds), Orte der Herrschaft Charakteristika von antiken Machtzentren. Rahden: 97–112.



399

Wulf-Rheidt, U. 2013. ‘Die Bedeutung der severischen Paläste für spätere Residenzbaute’, in N. Sojc, A. Winterling, and U. Wulf-Rheidt (eds), Palast und Stadt im severischen Rom. Stuttgart: 287–306. Wulf-Rheidt, U. 2015. ‘The palace of the Roman emperors on the Palatine in Rome’, in M. Featherstone, J.-M. Spieser, G. Tanman, and U. Wulf-Rheidt (eds), The Emperor’s House. Palaces from Augustus to the Age of Absolutism. Berlin: 3–18. Wulf-Rheidt, U. 2017. ‘Die schwierige Frage der Nutzung des Römischen Kaiserpalastes auf dem Palatin in Rome in der Spätantike’, AnTard 25: 127–48. Zaccagnino, C., Bevan, G., and Gabov, A. 2012. ‘The missorium of Ardaburius Aspar: new considerations on its archaeological and historical contents’, Archeologia Classica 63: 419–54. Zanker, P. 2002. ‘Domitian’s palace on the Palatine and the imperial image’, in A. K. Bowman, H. M. Cotton, M. Goodman, and S. Price (eds), Representations of Empire: Rome and the Mediterranean World. Oxford: 105–30. Ziemssen, H. 2006. Das Rom des Maxentius. Städtebau und Herrscherbild zu Beginn des 4. Jh. n. Chr. PhD Dissertation. Hamburg. Ziemssen, H. 2012. ‘Der Herrscher im Tempel. Bild und Inszenierung im kaiserzeitlichen Rom’, in O. Dally, S. Moraw, and H. Ziemssen (eds), Bild—Raum—Handlung. Perspektiven der Archäologie. Berlin: 137–63. Zuckerman, C. 1994. ‘L’empire d’orient et les Huns: notes sur Priscus’, TM 12: 160–82.

Index For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on occasion, appear on only one of those pages. A cubiculo 296–7, 299 A libellis 28–9, 148 A rationibus 28–9 Ab epistulis 8–9, 20, 28–9, 148 ‘Abbasid Caliphate 325–6, 332, 336, 347–8 Access, politics of 31–8, 43–4, 47–8, 178, 191, 197, 212, 226–30, 256–60, 265, 278–9, 285–6, 294, 303–4, 320–2, 325–9 Acta Alexandrinorum 280, 284, 292, 296, 301 Adoratio 1–2, 14–15, 50–1, 95–7, 230–4, 294, 302–3 Adventus 12–13, 64–9, 78, 176–7, 182–3 Aetius, Fl. 174–8, 181–2, 184–91, 194–7, 335–6 Agrippina (wife of Claudius, mother of Nero) 125–6, 347–8, 354 Alypia (daughter of Anthemius) 187, 194 Amici 7–8, 23–4, 45, 47, 126–8, 145, 147–8, 151, 156–7, 299–300, 338–9, 345–6 Ammianus Marcellinus 67–9, 133–40, 163, 230–4 Anastasius I (emperor) 18–19, 111–14, 200, 319, 328–9 Anthemius (emperor) 35–6, 172, 187 Anthemius (praetorian prefect) 174–6, 194–5 Antioch 80–1, 90–2, 239, 320 Synod of 223, 228–9 Antiochus Chuzon (jurist) 152–3 Antoninus Pius (emperor) 13–14, 41–55, 129, 153, 268, 285–6, 289–90, 347–8 Apocrypha, Christian 282–5, 290–1, 293 Apollonius of Tyana 284–6, 291–3, 301–2, 304 Aquileia 81, 225–8, 236 Arcadia (sister of Theodosius II) 208–11, 244–5, 250, 314 Arcadius (emperor) 194–5, 206–9, 262–3, 274–5, 291, 312–13, 328–9 Archontes (leaders at court) 19, 111–14 Ardaburius, Fl. 172 Areobindus, Fl. 174–5, 190–1 Ariadne (wife of Zeno and Anastasius I) 111–14, 208–9, 217, 350–2 Aspar, Fl. Ardaburius 172, 174–8, 183–5, 189–90, 194–7, 199–200, 318, 335–6

Astyrius, Fl. 185 Athanasius (bishop of Alexandria) 31–2, 36, 221–39 Attila the Hun 298–9 Augusta, title of 29–30, 204–6, 208–9, 347, 353–4 Augustus (emperor) 4–5, 16–17, 24, 29, 42–3, 69–70, 144–5, 149, 268, 309–12, 324, 356–7 Aula 5–7 Aurelia Prisca (wife of Diocletian) 88–9, 204–5 Ausonius 21–2, 132–3, 137, 139–40, 152, 329–30 Banquets 51–4, 297–8 Bishop(s) 19, 29–35, 221–39, 241–4, 293–4, 320–2, 326–7 Concept of ‘court bishop’ 224–5, 339–40 Of Constantinople 19, 35, 116–17, 240–61, 314–15, 322 Book of Ceremonies 13–14, 63, 72–3, 111–14, 183, 232, 330–1 Caesarion (son of Julius Caesar) 349 Caligula (emperor) 14–15, 18–19, 124, 286, 296, 302–3, 317–18, 348–9 Cameron, Alan 179–80 Candidianus (son of Galerius) 348–9 Caracalla (emperor) 148–9 Carinus (emperor) 262–5 Celestine (bishop of Rome) 248–9, 254–6 Ceremonial 13–20, 57–9, 101–2, 178–88, 230–4, 291, 293–4, 298–9, 302–3, 323–37 Imperial Accession 18–19, 59–63, 89, 105–7, 111, 113, 115–17, 183, 315–16, 329, 334–5 Chinese imperial courts 23, 317, 334, 344–5, 353–4 Christianity Before Constantine 282–5, 290–1 Effect on politics 33–5, 159–60, 166, 213–17, 224–35, 241–4, 249–52, 294, 299–300, 322 Effect on ceremonial 71–3, 107–9, 112, 117–18, 329–30 Chrysaphius (praepositus sacri cubiculi) 214–15 Chryseros (praepositus sacri cubiculi) 258–9

402



Civilitas 42–6, 126, 131–2, 161, 268, 273–4, 293 Claudian 179–83, 186, 275–6 Claudius (emperor) 64, 126–7, 292, 296, 346 Comes 5–6, 25–6, 28–9, 344–5 Comes domesticorum 24–5, 162, 168–9, 252 Comes excubitorum 63, 115–16, 319, 352 Comes Orientis 163–5 Comes rei militaris 168–9 Comes rei privatae / rerum privatarum 24–5, 158–9, 161–2, 164, 167–8, 236–7 Comes sacrarum largitionum 24–5, 158–9, 161–4 Comites consistoriani 24–5, 28–9, 156–7, 161–2, 176–7, 258, 337–8 Comitatus 5–7, 85–90, 93, 317–18, 338 Commodus (emperor) 66, 148, 292 Consilium 23–5, 144–8, 301, 337–8 Consistorium 24–9, 156–72, 178, 187–8, 213, 230–1, 256–7, 337–8 Constans (emperor) 36, 108–11, 221–39 Constantine I (emperor) 25–6, 62–3, 67–8, 71–2, 93, 107–11, 159–60, 174, 216, 287, 297–300, 312, 324–5, 328–9, 338, 349 Constantine II (emperor) 110–11 Constantine VII Porphyrogenitus (emperor) 310, 349–50 Constantinople 5–7, 16–19, 34–5, 102–4, 107–11, 243–4, 255, 299–300, 312–19, 326–31, 337, 354–5 Constantius I (emperor) 89–90, 205 Constantius II (emperor) 31–3, 67–9, 71–3, 103–4, 110–11, 134, 156–60, 167, 169, 221–39, 334 Constantius III (Flavius Constantius) 174–5, 184–5, 189, 197–8 Constantius Gallus (emperor) 136 Consulship 183–6, 356–7 Corippus 31, 72–3, 116–17, 319–20 Council of Chalcedon (AD 451) 215–17 Council of Ephesus (AD 431) 35, 213–14, 240–61 Council of Ephesus (AD 449, the ‘robber synod’) 214–15 Council of Nicaea (AD 325) 222–3, 230 Council of Serdica (AD 343) 222–3, 226–9, 232–5 Court As a community 20–31, 122, 139–41, 306, 319, 329–32, 337, 342–5, 355–6 Consensus of 110–11, 114–17, 195–8, 319, 343–4, 357 Definition of 4–7, 57–8 Emotions at 121–41, 343 Hierarchy and Precedence at 25–6, 47–8, 232–4, 344–5 Negotiation at 13–15, 19, 29–30, 33, 111–17, 142–3, 304–6, 311, 343–4

Courtiers 13–14, 19–31, 110–15, 142–3, 281–2, 288–91, 299–300, 302–3, 332–3 Aristocratic women as 355–6 Brokering access 32–3, 35, 144, 256–60, 278–9, 335–6, 347–8 Definition of 5, 22–4, 47, 337, 344–5 Jurists as 142–55 Relationships with emperor 41–55, 123–4, 126–32, 135, 299–300, 320, 325, 330–1 Courtliness 341–2 Cratia (wife of Fronto) 355 Cubicularii / Cubiculariae 35, 115, 256–8, 314, 341–2, 356 Cyril (bishop of Alexandria) 29–30, 35, 213–14, 240–61, 356 Dalmatius (archimandrite) 255 Dio Chrysostom 270–2 Diocletian (emperor) 1–2, 11–12, 14–16, 77–81, 89–90, 272 Dioscorus of Aphrodito 321–2 Domitia Lucilla (mother of Marcus Aurelius) 129–30, 355 Domitian (emperor) 23–4, 36–7, 121, 123–4, 127–8, 269–70, 284–7, 290–3, 296–7, 304, 323–4 Domus Augustana 54 Domus Tiberiana 46–8 ‘Dowager empress’, concept of 353–4 Droseria (member of Pulcheria’s household) 258, 356 Duindam, Jeroen 3, 15–18, 58–9, 75–6, 316–17, 340–1, 347–8 Elagabalus (emperor) 149–50, 272, 348–9 Elias, Norbert 4–5, 8–11, 13–16, 76, 105–6, 121–2, 240–2, 340–2 Elton, Hugh 242 Emperors, types of Child 15–16, 27, 36–7, 172–202, 274, 312–13, 333, 335–7 ‘Soldier emperors’ 21–2, 100, 312, 337 Eudocia (wife of Theodosius II) 210, 213–14, 249–51, 289 Eudoxia (wife of Arcadius) 206–9, 212, 300–1, 328–9 Eunuchs 24–5, 31–3, 88–9, 106, 134, 230–1, 238–9, 265, 341–2, 349–50 Eusebia (wife of Constantius II) 137, 205–6 Eusebius (bishop of Caesarea) 107–11, 282–3, 287, 297–8 Eutropia (wife of Maximian) 88–9, 99–100 Excubitores 29, 33, 63, 114–15, 319

 Fausta (wife of Constantine) 205, 210 Faustina the Younger (wife of Marcus Aurelius) 300, 355 Favourites, at court 144, 242, 335–6 Felix, Fl. 185 Flaccilla (wife of Theodosius I) 206, 209, 211–12, 345–6 Freedmen/women 7–8, 22–5, 299, 338, 347–9, 356 French royal courts 8, 32–3, 325–7, 331–2, 335, 338–43, 345–6 Friedländer, Ludwig 7–8 Fronto, M. Cornelius 13–14, 21, 24, 41–55, 127–32, 268, 302–3, 332–3, 355 Funerals, imperial 29–30, 69–73, 105–18, 315–16 Galeria Valeria (daughter of Galerius) 88–9, 204–5, 349 Galerius (emperor) 82, 89, 204–5, 348–9 Galla Placidia (mother of Valentinian III) 193, 198, 210, 212, 314, 354–5 Gallienus (emperor) 66, 299–300 Gillett, Andrew 181–2, 195–6 Gordian I (emperor) 61–2 Gratian (emperor) 21–2, 132–3, 137–8, 274, 317–18, 329–30 Greeks, Alexandrian 281, 284–5, 292 Guard units 16–20, 22–3, 33, 56–74, 107–8, 269, 296, 327–8 Hadrian (emperor) 20, 45–8, 51, 130, 146 Hagiography 282–3, 287–8, 293, 297–8, 300–1 Harries, Jill 213, 244 Helena (mother of Constantine) 205, 211–12, 215–17, 353–4 Herodes Atticus 300 Holy Men 280–3, 290–1, 293, 326–7, 340–2 Honoria (sister of Valentinian III) 206–7, 349 Honorius (emperor) 174–5, 179–83, 197–8, 210, 274–6, 312–13, 346 Hunting 341–2 Illegitimate children 348–50 Ino (wife of Tiberius II Constantine) 355 Intimacy, politics of 32–3, 197, 212, 242, 256–60, 285–6, 301–4, 326–7, 354–5 Irene (empress regnant) 350–1 Isolation, imperial 36–7, 262–77, 322–4, 326–7, 336 Japanese imperial courts 323–4, 336, 342–3, 345 Jews 280–2, 285–6, 288, 301 John (bishop of Antioch) 252–7 John Chrysostom (bishop) 207–9, 329–30

403

Jones, A. H. M. 10, 174 Jovian (emperor) 28–9, 71, 165–70, 239 Julia Soaemias (mother of Elagabalus) 348–9 Julian (emperor) 21, 28–9, 62–3, 68–9, 134–5, 137–9, 160–5, 167–8, 170, 334 Jurists 27–8, 142–55, 338–9 Justice at court 144–6, 153–4, 284–8, 292–3, 300–1, 304–5, 325–6 Justin I (emperor) 63, 154, 199–200, 319 Justin II (emperor) 116–17, 319–20, 329, 350, 352 Justina (wife of Valentinian I) 300–1 Justinian (emperor) 14–15, 19, 30–1, 72–3, 116–17, 144, 152–4, 200, 293, 302–3, 314–15, 319–22, 326–7, 334–5, 337–40 Kissing at court 50–1, 128, 130–2, 302–3, 324–5, 344–5 Leo I (emperor) 15–16, 111–13, 172, 176–7, 183, 196–7, 199, 208–9, 324–6, 329, 338 Leo I (bishop of Rome) 214–15 Leo II (emperor) 199–200, 329 Leo VI (emperor) 350 Leontia (daughter of Leo I) 208–9 Licinius Rufus, M. Cn. 151–2 Liudprand of Cremona 309–10 Livia (wife of Augustus) 347 Lucius Verus (emperor) 49–50, 129–30, 132 Luxor 86–7, 94–5 Luxury 52, 265–7 MacCormack, Sabine 9–10, 78 Magister epistularum 86, 93 Magister libellorum 86, 93 Magister memoriae 86 Magister militum 15–16, 26–7, 112, 158, 162, 169, 314–16 In late fourth–late fifth century 172–202, 335–7 Magister officiorum 16, 28–9, 106, 111–12, 114–15, 163–4, 174, 229–30, 256–8, 329–31 Magnus Maximus (emperor) 288–9 Majorian (emperor) 192 Marcella (member of Pulcheria’s household) 258–9, 356 Marcian (emperor) 30–1, 178, 198–9, 215–16, 318 Marcus Aurelius (emperor) 2–3, 44–6, 48, 50–1, 127–32, 147, 300, 311–12, 355 Maria (wife of Honorius) 186, 193, 208–9 Marina (sister of Theodosius II) 208–11, 244–5, 250, 314 Martial 6–7, 296–7 Martin of Tours 288–9, 293–4, 297–8, 300–2, 326–7

404



Matthews, John 10, 15–16, 320 Maurice (emperor) 313, 319–20, 324–5 Maxentius (emperor) 97–102 Maximian (emperor) 81–3, 87–8, 99–100, 205 Maximianus (bishop of Ravenna) 340 Maximinus Daza (emperor) 88–9, 204–5, 349 Melania the Younger 287–8, 300–3 Merobaudes, Fl. 181–2, 185–8 Milan 81–3, 179, 225–6, 228, 320 Millar, Fergus 8, 240–1, 249–50, 305–6 Mistresses, imperial 347–9 Mommsen, Theodor 7–8, 10 Nero (emperor) 20, 60–1, 284–6, 292–3, 347–8, 354 Nerva (emperor) 291–2 Nestorius (bishop of Constantinople) 35, 212, 240–61 Nicomedia 78–81, 88–9 Olybrius (emperor) 314 Ossius of Cordoba (bishop) 221, 225–9 Otho (emperor) 20 Ottoman Empire 316–17, 325–6, 332, 342–3, 345–8, 352–4 Pages at court 345–6 Palaces 6–7, 79, 90–7, 296–7 Churches in 339–40 In Constantinople 107–9, 115, 324–5, 330, 334–5, 339–40, 356 In Rome 4–5, 46, 54, 61–2, 92–5, 97–8, 101, 296–9 Palatini 24–5, 338, 344–5 Palatium 6–7 Panegyric 9–10, 13, 35–7, 49–51, 77–8, 99, 116, 135, 179–82, 185, 319–20, 338–9 Papinian (jurist) 146, 148–9 Parrhesia 291–3 Parthenius (a cubiculo) 296–7 Paul (praepositus sacri cubiculi) 258–9 Paulus (jurist) 143, 146, 151–2 Pertinax (emperor) 70 Persia 36–7, 41, 265–7, 272 Pescennius Niger (emperor) 61–2 Peter the Patrician 13–14, 19, 111–14, 183, 321, 330–1, 334–5 Petitions/Petitioners 35–6, 212, 226–7, 230–1, 234, 236–9, 249–52, 281, 286–7, 321, 325–6 Philo of Alexandria 281–2, 286, 296 Philosophers 285–6, 291–2, 299–300 Philostratus 281, 284–7 Phocas (emperor) 18–19, 320, 332–3 Plinta, Fl. 174–6

Pliny the Younger 16, 36–7, 46–7, 49–51, 123–7, 131, 145–6 Plotina (wife of Trajan) 124–5, 301 Plotinus 299–300 Poets, court 296–7 Poetovio 234–6 Praepositus sacri cubiculi 35, 114–15, 164, 256–8, 319–20 Praetorian guard 16–19, 56–74, 317–18 Praetorian prefect 18–19, 26–7, 48–9, 85, 158, 214–15, 314–15, 317–18, 325–6, 335–6, 348–9 Princeps clausus 36–7, 105–6, 131, 262–77, 312–13 Priscus of Panium 298–9 Procopius (historian) 153–4 Procopius (usurper) 133–4, 162 Pulcheria (sister of Theodosius II) 35, 208–17, 244–5, 247, 250–2, 255, 314–15, 318, 351–2, 355 Quaestor sacri palatii 16, 24–5, 27–8, 106, 111, 152–4, 157, 159, 161–3, 168–9, 258–9 Referendarii 325–6 ‘Regent’, concept of 29, 197–8, 354–5 Relics 314–15 Ricimer, Fl. 174–6, 178, 180, 187, 192, 194–5, 200–1, 335–6 Rome, city of 11–13, 68, 80–3, 179–80, 311–13, 317, 344–5 Under Maxentius 97–102 Romanus I (emperor) 350 Rosenwein, Barbara 121–4, 131–2, 139–41 Russian royal court 323–4, 327–8, 331–2 Salutatio 1–2, 16, 24, 31–2, 94–7, 121, 324 Scholae palatinae 63, 114–15 Sejanus, L. Aelius 125–6, 335–6 Senators 5–7, 13–15, 21–6, 65, 70, 92–3, 139–40, 195–7, 264–5, 273, 296, 301, 317–18 At Constantinople 111–16 Under Antoninus Pius 42–4, 47–8, 50–3, 268 Under Domitian 121, 127, 271 Under Maxentius 99–102 Under the Tetrarchs 87–9 Under Trajan 121, 127–8, 270–1 Septimius Severus (emperor) 66, 145, 296, 318 Serdica 82 Serena (wife of Stilicho) 194–5, 287–8, 300–3 Severus Alexander (emperor) 149–50, 272 Sidonius Apollinaris 21–3, 35–6, 180, 187 Sirmium 80–2 Slaves 7–8, 290–1, 299, 338, 345–6, 348–9, 356 Sopater 299–300

 Sophia (wife of Justin II) 329, 350, 352, 355 Sophists 281, 286–7, 296, 298, 300 Starkey, David 32–3, 285–6, 326–7 Statius 23–4, 297–9 Statues 188–90, 325–6 Stilicho, Fl. 174–6, 179–86, 188–9, 193, 195–8, 274, 288, 335–6, 343–4 Suetonius 5, 8–9, 20, 22–3, 123–7 Synesius of Cyrene 262–5, 274–5, 291, 321–2 Tacitus 123–7 Tetrarchs, the (emperors) 11–13, 62–3, 66–7, 75–104, 312, 317 Themistius 133, 135–40, 166–7, 169 Theodora (wife of Justinian) 30–1, 293, 302–3, 339–40 Theodosius I (emperor) 133, 158–9, 193, 274–6, 312–13, 320 Theodosius II (emperor) 35, 152–3, 190–1, 199, 209, 240–61, 312–15, 318, 321, 329–30, 332–3, 339–44, 346, 356 Thermantia (wife of Honorius) 208–9 Thessalonica 80–3, 91–2 Tiberius (emperor) 16–17, 59–60, 64–5, 126, 268–9, 335–6 Tiberius II Constantine (emperor) 319, 334–5, 352 Titus (emperor) 301–2, 346 Trajan (emperor) 49–51, 65–6, 124–8, 270–2, 301 Tribonian 152–4 Trier 78, 80–1, 89–90, 93–4, 133, 225–6, 228, 288–9, 293–4, 312, 320, 329–30 Triumph, the 64–9

405

Ulpian 146, 149–51 Ursacius (bishop of Singidunum) 221–2, 224, 237–8 Valens (emperor) 133–4, 136–8, 312, 329–30, 345 Valentinian I (emperor) 62–3, 104, 132–3, 136–8, 170, 293–4, 300–1, 312, 317–18, 320, 329–30, 343, 345 Valentinian II (emperor) 27, 274 Valentinian III (emperor) 27, 172–8, 181–2, 192, 196–7, 318, 354–5 Valeria Maximilla (wife of Maxentius) 88, 99–100 Valerius Romulus (son of Maxentius) 99, 101–2 Veil, ceremonial uses of 231–2, 294 Verina (wife of Leo I) 217 Versailles 13–14, 33, 35–6, 76, 92–3, 331–2 Vespasian (emperor) 24, 285–6, 347–8 Villas, imperial 78–9, 99, 311–12 Vincentius (bishop of Capua) 227–9, 236–7 Volusius Maecianus, L. 147–8 Wallace-Hadrill, Andrew 8–9, 264 Winterling, Aloys 8–11, 42–3, 47, 52–3, 338–9, 341–2 Women, imperial 29–31, 88–9, 111–14, 124–5, 186–7, 193–5, 203–18, 300–1, 303–4, 314, 347–56 Zeno (emperor) 111–14, 174–5, 199–200, 217 Zeno, Fl. (general) 191 Zoora 293, 326–7