Slavery in the late antique world, 150-700 CE 9781108476225, 9781108568159, 2021028677, 2021028678, 1108476228

Table of contents :
Cover
Half-title page
Title page
Copyright page
Contents
List of Figures
List of Tables
List of Contributors
Preface
A Note on Abbreviations and Sources Used
Introduction Late Antique Studies and the New Polyphony for Slave Studies
Part I Moral and Symbolic Values of Slavery
1 Masters and Slaves in Early Christian Discourse
2 Slavery and Religion in Late Antiquity: Their Relation to Asceticism and Justice in Christianity and Judaism
3 (Il)Legal Freedom: Christ as Liberator from Satanic Debt Bondage in Greek Homilies and Hymns of Late Antiquity
4 Late Roman Ideas of Ethnicity and Enslavement
Part II Slavery, Cultural Discourses, and Identity
5 Slavery in Euphemia and the Goth
6 What Was Jewish about Jewish Slavery in Late Antiquity?
7 Divining Slavery in Late Ancient Egypt: Doulology in the Monastic Works of Paul of Tamma and Shenoute
8 Rural Slavery in Late Roman Gaul: Literary Genres, Theoretical Frames, and Narratives
Part III Slavery, Social History, and the Papyrological and Epigraphical Sources
9 Slaves in Sixth-Century Palestine in the Light of Papyrological Evidence
10 Enslaved Children in Roman Egypt: Experiences from the Papyri
11 Late Antique Slavery in Epigraphic Evidence
Part IV Social and Religious Histories of Slavery on the Borders of the Empire and Beyond
12 Slavery among the Visigoths
13 Sinner, Slave, Bishop, Saint: The Social and Religious Vicissitudes of St Patrick
14 Slave Boys in Paradise? The Text of the Quran and Its Later Exegetes
Bibliography
Index

Citation preview

Slavery in the Late Antique World, 150–700 CE Slavery in the Late Antique World, 150–700 CE investigates the ideological, moral, cultural, and symbolic aspects of slavery, as well the living conditions of slaves in the Mediterranean Basin and Europe during a period of profound transformation. It focuses on socially marginal areas and individuals on an unprecedented scale. Written by an international team of scholars, the volume establishes that late ancient slavery is a complex and polymorphous phenomenon, one that was conditioned by culture and geography. Rejecting preconceived ideas about slavery as static and without regional variation, it offers focused case studies spanning the late ancient period. These case studies provide in-depth analyses of authors and works, and consider a range of factors relevant to the practice of slavery in specific geographical locations. Using comparative and methodologically innovative approaches, this book revisits and questions established assumptions about late ancient slavery. It also enables fresh insights into one of humanity’s most tragic institutions. Chris L. de Wet is Professor of New Testament and Early Christian Studies at the University of South Africa and Honorary Research Fellow at the Australian Lutheran College. He is the author of Preaching Bondage: John Chrysostom and the Discourse of Slavery in Early Christianity (2015) and The Unbound God: Slavery and the Formation of Early Christian Thought (2018). Maijastina Kahlos is Research Fellow at the Helsinki Collegium for Advanced Studies at the University of Helsinki. She is the author of Debate and Dialogue: Christian and Pagan Cultures, c. 360–430 (2007), Forbearance and Compulsion: The Rhetoric of Tolerance and Intolerance in Late Antiquity (2009), and Religious Dissent in Late Antiquity (2020). Ville Vuolanto is Senior Lecturer in the Department of History at Tampere University. He has published extensively on children and family in the Roman and Late Antique periods. His publications include Children and Asceticism in Late Antiquity (2015) and Children and Everyday Life in the Roman and Late Antique World (ed. 2017, with C. Laes).

Slavery in the Late Antique World, 150–700 CE

Edited by CHRIS L. DE WET University of South Africa

MAIJASTINA KAHLOS University of Helsinki

VILLE VUOLANTO Tampere University

University Printing House, Cambridge cb2 8bs, United Kingdom One Liberty Plaza, 20th Floor, New York, ny 10006, USA 477 Williamstown Road, Port Melbourne, vic 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781108476225 doi: 10.1017/9781108568159 © Cambridge University Press 2022 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2022 A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data names: De Wet, Chris L., 1982– editor. | Kahlos, Maijastina, editor. | Vuolanto, Ville, editor. title: Slavery in the late antique world, 150–700 CE / edited by Chris L. de Wet, University of South Africa, Maijastina Kahlos, University of Helsinki, Ville Vuolanto, Tampere University. description: 1 Edition. | New York, NY : Cambridge University Press, 2022. | Includes bibliographical references and index. identifiers: lccn 2021028677 (print) | lccn 2021028678 (ebook) | isbn 9781108476225 (hardback) | isbn 9781108568159 (ebook) subjects: lcsh: Slavery – History. | Slaves – Social conditions. classification: lcc ht863 .s557 2022 (print) | lcc ht863 (ebook) | ddc 306.3/ 6209–dc23 LC record available at https://lccn.loc.gov/2021028677 LC ebook record available at https://lccn.loc.gov/2021028678 isbn 978-1-108-47622-5 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Contents

List of Figures List of Tables

page vii viii

List of Contributors Preface

ix xiii

A Note on Abbreviations and Sources Used Introduction: Late Antique Studies and the New Polyphony for Slave Studies Chris L. de Wet

xv

1

part i moral and symbolic values of slavery 1 Masters and Slaves in Early Christian Discourse Pieter J. J. Botha

15

2 Slavery and Religion in Late Antiquity: Their Relation to Asceticism and Justice in Christianity and Judaism Ilaria L. E. Ramelli

43

3 (Il)Legal Freedom: Christ as Liberator from Satanic Debt Bondage in Greek Homilies and Hymns of Late Antiquity Arkadiy Avdokhin 4 Late Roman Ideas of Ethnicity and Enslavement Maijastina Kahlos

66 87

part ii slavery, cultural discourses, and identity 5 Slavery in Euphemia and the Goth Chris L. de Wet

v

107

vi

Contents

6 What Was Jewish about Jewish Slavery in Late Antiquity? Catherine Hezser 7 Divining Slavery in Late Ancient Egypt: Doulology in the Monastic Works of Paul of Tamma and Shenoute Christine Luckritz Marquis

129

149

8 Rural Slavery in Late Roman Gaul: Literary Genres, Theoretical Frames, and Narratives Uiran Gebara da Silva 170 part iii slavery, social history, and the papyrological and epigraphical sources 9 Slaves in Sixth-Century Palestine in the Light of Papyrological Evidence Marja Vierros 10 Enslaved Children in Roman Egypt: Experiences from the Papyri April Pudsey and Ville Vuolanto 11 Late Antique Slavery in Epigraphic Evidence Mariana Bodnaruk

191

210 224

part iv social and religious histories of slavery on the borders of the empire and beyond 12 Slavery among the Visigoths Noel Lenski 13 Sinner, Slave, Bishop, Saint: The Social and Religious Vicissitudes of St Patrick Judith Evans Grubbs 14 Slave Boys in Paradise? The Text of the Quran and Its Later Exegetes Ilkka Lindstedt Bibliography Index

251

281

298

316 355

Figures

11.1 Funeral inscription on the Christian sarcophagus of Eunuch Aedesius (late-fourth to first quarter of the fifth century ce). Catacombs of San Sebastiano, Rome. page 232 11.2 Zoninus’ slave collar (fourth to fifth century ce). From Rome. 236 11.3 Inscription on the Christian monument recording its awarder, alumnus Marcus Servilius Servilianus (312–37 ce). From Rome. 238 11.4 Slave victimarii in the suovetaurilia relief of the Decennalia base (303 ce). Forum Romanum, Rome. 247 14.1 Figure illustrating how the commentators of the Quran discussed in this chapter understood the phrase ghilmā n lahum in Quran 52:24. 314

vii

Tables

9.1 Papyri from Petra and Nessana dealing with slaves. 9.2 Names of slaves, possible slaves, and former slaves in the Petra papyri. 10.1 Recorded enslaved children in Oxyrhynchos census returns.

viii

page 194 205 213

Contributors

Arkadiy Avdokhin is Associate Professor of Late Antique and Byzantine History, HSE University, Moscow. He has published on the social and doctrinal contexts of late antique liturgy, the formation and expression of pagan and Christian identities, and inscriptions in urban and cultic spaces. Mariana Bodnaruk received her PhD in Medieval Studies at the Central European University in Budapest in 2019, and has since worked as Visiting Assistant Professor at the Al-Quds Bard College for Arts and Sciences in East Jerusalem. Her research interests centre on the social and art history of the Later Roman Empire and Early Byzantium, the history of precapitalist modes of production, epigraphy, and Marxist feminism. Pieter J. J. Botha is Professor of New Testament and Early Christian Studies, University of South Africa, and a former editor of Neotestamentica: Journal of the New Testament Society of Southern Africa. He is also the author of Orality and Literacy in Early Christianity (2012). His research publications deal with the ethnography of ancient communication, and the social history and cross-cultural comparison of historical events. Chris L. de Wet is Professor of New Testament and Early Christian Studies in the Department of Biblical and Ancient Studies, at the University of South Africa, Pretoria. He is also Honorary Research Fellow at the Australian Lutheran College, University of Divinity, Adelaide. He is the current Editor-in-Chief of the academic journals Neotestamentica and Journal of Early Christian History. He is an ix

x

List of Contributors expert on slavery in early Christianity and has written numerous articles on the topic, as well as two monographs: Preaching Bondage: John Chrysostom and the Discourse of Slavery in Early Christianity (California, 2015), and The Unbound God: Slavery and the Formation of Early Christian Thought (2018). Judith Evans Grubbs is the Betty Gage Holland Professor of Roman History at Emory University. Her research focuses on the late Roman imperial period, particularly the family, slavery, and Roman law. Her current project is ‘Children without Fathers in Roman Law: Paternity, Patrimony, and Freedom’. Uiran Gebara da Silva is Assistant Professor of Ancient History at the Universidade Federal Rural de Pernambuco (Brazil). He is interested in peasants, slaves, and the countryside of Late Roman Gaul and is the author of a book on Late Roman rural rebellions, Rebeldes contra o Mediterrâneo: revoltas rurais e a escrita da história das classes subalternas na Antiguidade Tardia (2016). Catherine Hezser is Professor of Jewish Studies at SOAS, University of London. She has published widely on the social history of Jews in Roman and early Byzantine Palestine. Her book Jewish Slavery in Antiquity (2005) was followed by the monographs Jewish Travel in Antiquity (2011), Rabbinic Body Language: NonVerbal Communication in Palestinian Rabbinic Literature of Late Antiquity (2017), and Bild und Kontext: Jüdische und christliche Ikonographie der Spätantike (2018). She is also the editor of the Oxford Handbook of Jewish Daily Life in Roman Palestine (2010). Maijastina Kahlos is a historian and classicist. She is currently working as a research fellow at the Helsinki Collegium for Advanced Studies, University of Helsinki. Her research interests broadly include migration and mobility in Roman history, religions in the Roman Empire, and Christianization of the Mediterranean regions. She is the author of Debate and Dialogue: Christian and Pagan Cultures, c. 360– 430 (2007), Forbearance and Compulsion: The Rhetoric of Tolerance and Intolerance in Late Antiquity (2009), and Religious Dissent in Late Antiquity (2020). Noel Lenski is Professor of Classics and History at Yale University. He has published extensively on slavery, including the co-edited volume What Is a Slave Society? The Practice of Slavery in Global Perspective (2018). He is currently co-authoring a translation and commentary of the Liber Iudiciorum – Leges Visigothorum.

List of Contributors

xi

Ilkka Lindstedt is University Lecturer in Islamic Theology at the University of Helsinki. He has worked on pre-Islamic Arabia, early Islam, and Arabic literature and epigraphy. Recent publications include: ‘Who Is in, Who Is out? Early Muslim Identity through Epigraphy and Theory’, Jerusalem Studies in Arabic and Islam 46 (2019): 147–246. Christine Luckritz Marquis is Associate Professor of Church History at Union Presbyterian Seminary. Her expertise covers late ancient communities, especially in Egypt, Syria, Ethiopia, and the Arabian Peninsula. Her forthcoming book, Death of the Desert, explores the nexus of places, memories, race, and violence among Egyptian monks. April Pudsey is Senior Lecturer in Roman History at Manchester Metropolitan University, where she is also Head of History and Archaeology of Childhood for the interdisciplinary Manchester Centre for Youth Studies. April has published widely around ancient childhood, youth, family, housing, and the fertility and mortality dynamics of ancient populations, particularly in Roman Egypt. She is co-editor of and contributor to two key Cambridge volumes: Demography and the Graeco-Roman World: New Insights and Approaches (2011; with C. Holleran) and Housing in the Ancient Mediterranean World: Material and Textual Approaches (2021; with J. A. Baird), and contributing co-author of an Oxford monograph, Everyday Life in Roman and Late Roman Egypt: A Social Archaeology (2021; with E. Swift and J. Stoner). Together with her co-author in this volume, Ville Vuolanto, she is working on a larger project reconstructing the lives and concerns of children and young people in the city of Oxyrhynchos, Egypt. Ilaria Ramelli has been Professor of Roman History, Senior Visiting Professor (Harvard, Columbia), Professor of Theology and Chair (Angelicum), and Senior Fellow (Durham, Princeton, Sacred Heart University, Oxford). She is also Professor of Theology (Durham, KUL), Humboldt Award Fellow (MWK), Senior Fellow (Bonn University), and Member of the Platonism Centre (Cambridge). She has authored several books, including Social Justice and the Legitimacy of Slavery: The Role of Philosophical Asceticism from Ancient Judaism to Late Antiquity (2016). Marja Vierros is Associate Professor of Classical Philology at the University of Helsinki and the PI of the project ‘Digital Grammar of Greek Documentary Papyri’. She received her PhD in 2011. She

xii

List of Contributors was a long-time member in the Finnish team publishing the carbonized papyrus dossier found in Petra, Jordan. Ville Vuolanto is Senior Lecturer in History and Latin Language at the Tampere University. He has published extensively on the history of everyday life, family, children, and old age in the Early Christian, Roman, and Late Antique contexts, including Children and Asceticism in Late Antiquity: Continuity, Family Dynamics and the Rise of Christianity (2015).

Preface

This volume has its origins in the seminar on Slavery in Late Antiquity held in the Tvärminne zoological station of the University of Helsinki in November 2016. The symposium was the twenty-fourth in the series of Finnish Symposia of Late Antiquity, and it was organized by Maijastina Kahlos, Ulla Tervahauta, and Ville Vuolanto. The theme of slavery was chosen as the research on slavery in the late Roman Empire and in the post-Roman kingdoms has been expanding and evolving in recent decades, with many of the self-evident truths cherished in previous studies having been challenged. We aimed to approach late antique slavery from a wide perspective, including social, economic, political, legal, ideological, and religious levels. With these viewpoints in mind, we had three key-note speakers in the symposium: Chris L. de Wet (University of South Africa), Marianne Bjelland Kartzow (University of Oslo), and Marja Vierros (University of Helsinki). The symposium exceeded our expectations, both intellectually and socially, and we decided to organize this edited volume on late antique slavery, with contributors to the symposium, along with other researchers of slavery in Late Antiquity, in order to enrich the volume even further. In this endeavour, our aim has been to involve and bring together contributors with different backgrounds, both geographically and intellectually. Even if in today’s global academic world one’s scholarly and national backgrounds are no longer as significant as they used to be, we hope that this variety present in our volume will be visible in a positive way and provoke new research in previously unconsidered directions.

xiii

xiv

Preface

We wish to thank the Finnish Academy Centre of Excellence in Reason and Religious Recognition, Faculty of Theology, Department of World Cultures, Faculty of Humanities, University of Helsinki, and Jaakko Frösén Fund, University of Helsinki, for supporting the symposium. Many thanks are also due to the research assistants Irinja Bickert and Riikka Juntunen at Helsinki Collegium for Advanced Studies for their meticulous work with systematizing the notes and bibliography.

A Note on Abbreviations and Sources Used

Texts in standard editions and reference works are used. Works of Latin and Greek literary authors have been abbreviated by the Latin title. The full list of these abbreviations and editions can be found in Oxford Latin Dictionary (rev. ed. Oxford: Oxford University Press 2012, www.oxford scholarlyeditions.com/page/abbreviations) and Lewis and Short Latin Dictionary (Oxford: Clarendon Press 1879) (see also https://perseus.uchi cago.edu/LewisShortAbbreviations.html), for the Latin texts, and in Liddell-Scott-James Greek Lexicon (ninth ed., Oxford: Oxford University Press 1940) (see also http://stephanus.tlg.uci.edu/lsj/01-author s_and_works.html) and Diccionario Grieco-Español (Madrid: CSCI 2019) (http://dge.cchs.csic.es/lst/2lst-int.htm), for the Greek texts. For papyri, see the Checklist of Editions of Greek, Latin, Demotic, and Coptic Papyri, Ostraca, and Tablets, founding editors John F. Oates and William H. Willis, online at https://papyri.info/docs/checklist. The following list gives the information for some lesser-known publications and editions, especially for the early medieval period and for the Arabic sources, as well as abbreviations for inscriptions, for legal material, and for some other abbreviations that may need further specification.

ACO = Acta conciliorum oecumenicorum, E. Schwartz (ed.). Berlin 1959. Acts of Thomas = A. F. J. Klijn (ed.), The Acts of Thomas: Introduction, Text, and Commentary, Supplements to Novum Testamentum 108. Leiden: Brill 2003. AE = L’Année Épigraphique. Paris (1888–).

xv

xvi

A Note on Abbreviations and Sources Used

Allen and Datema 1987 = Pauline Allen and Cornelius Datema (eds.), Leontii Presbyteri Constantinopolitani Homiliae. Turnhout: Brepols 1987. Amato 2014 = Eugenio Amato (ed.), Procopius of Gaza, Op. XI. Procope de Gaze, Discours et fragments. Paris: Les Belles Lettres 2014. al-Andalusıˉ, Bahr = al-Andalusıˉ Abuˉ Ḥ ayyā n Muḥammad ibn Yuˉ suf, alBaḥr al-muḥˉıt ̣. Ed. ʿĀ dil al-Ḥ amd ʿAbd al-Mawjuˉ d et al. Nine volumes. Beirut: Dā r al-Kutub al-ʿIlmiyya 2001. al-Bayḍā wıˉ, Tafsıˉ r = al-Bayḍā wıˉ, Nā s ̣ir al-Dıˉn, Al-Tafsıˉ r al-musammā anwā r al-tanzıˉ l wa-asrā r al-taʾwıˉ l. Two volumes. Beirut: Dā r alKutub al-ʿIlmiyya 2008. Binazzi 1989 = Gianfranco Binazzi, Inscriptiones christianae Italiae septimo saeculo antiquiores, VI. Bari: Edipuglia 1989. CE = Codex Euricianus, in Leges Visigothorum. Monumenta Germaniae Historica, Leges Nationum Germanicarum 1, K. Zeumer (ed.). Hannover 1902, 1–32. CIL = Corpus Inscriptionum Latinarum. 1863–. C. Th. = Codex Theodosianus. Cod. Iust. = Codex Iustinianus. Conc. = La colección canónica Hispana: Concilios hispanos, segunda parte. Volumes 4–6. Monumenta Hispaniae sacra. Serie canónica 4–6, G. Martínez Díez and F. Rodrígues (eds.). Madrid 1984–2002. Constas 2003 = Nicholas Constas, Proclus of Constantinople and the Cult of the Virgin in Late Antiquity: Homilies 1–5, Texts and Translations. Leiden; Boston: Brill 2003. CSLA = Cult of Saints, a database, see http://csla.history.ox.ac.uk. Dagron and Feissel 1987 = Gilbert Dagron and Denis Feissel, Inscriptions de Cilicie. Paris: De Boccard 1987. Daley 2017 = Brian E. Daley, Leontius of Byzantium: Complete Works. Oxford: Oxford University Press 2017. Datema 1978 = Conelius Datema (ed.), Amphilochii Iconiensis Opera. Turnhout: Brepols 1978. Divjak 1987 = J. Divjak (ed.), Oeuvres de Saint Augustin 46B: Lettres 1*29*, 2nd ed. Paris 1987. Drijvers 2003 = Han J. W. Drijvers, ‘The Acts of Thomas’, in W. Schneemelcher (ed.), New Testament Apocrypha Volume Two: Writings Related to the Apostles; Apocalypses and Related Subjects. Louisville, KY: Westminster John Knox 2003, 322–411. DTV = ‘The Donation and Will of Vincent of Huesca: Latin Text and English Translation’, S. Corcoran, Antiquité Tardive 11 (2003): 215–21.

A Note on Abbreviations and Sources Used

xvii

Euphemia and the Goth = F. C. Burkitt (ed.), Euphemia and the Goth, with the Acts of Martyrdom of the Confessors of Edessa. London: Williams & Norgate 1913. al-Farrā ʾ = al-Farrā ʾ, Maʿā nıˉ al-Qurʾā n. Three volumes. Beirut: ʿĀ lam alKutub 1983. Fear 2010 = Andrew T. Fear (trans.), Orosius: Seven Books of History against the Pagans, Translated Texts for Historians 54. Liverpool: Liverpool University Press 2010. Fitzgerald 1930 = Augustine Fitzgerald (trans.), The Essays and Hymns of Synesius of Cyrene Including the Address to the Emperor Arcadius and the Political Speeches. Oxford: Oxford University Press 1930. FV = Formulae Visigothicae in Miscellanea Wisigothica, 2nd ed., J. Gil (ed.). Seville: Universidad de Sevilla 1991, 71–112. GNO = Gregorii Nysseni Opera. Leiden: Brill 1958–2014. Hydatius = The Chronicle of Hydatius and the Consularia Constantinopolitana, R. W. Burgess (ed.). Oxford: Clarendon Press 1993. Ibn ʿArabıˉ, Tafsıˉ r = Ibn ʿArabıˉ, Tafsıˉ r, ed. Mus ̣tafā Ghā lib. Two volumes. Beirut: Dā r al-Andalus n.d. Ibn Saʿd, Ṭabaqā t = Ibn Saʿd, Muḥammad, al-Ṭabaqā t al-kabıˉ r, Eduard Sachau et al. (eds.) Nine volumes. Leiden: Brill 1905–28. ICVR = Inscriptiones Christianae Urbis Romae. Nova series. Rome 1922–. ILCV = E. Diehl, Inscriptiones Latinae Christianae Veteres. Berlin 1925– 67. ILS = H. Dessau, Inscriptiones Latinae Selectae, three volumes. Berlin 1892–1916. IMS = Inscriptions de la Mésie Supérieure. Belgrade 1976–. InscrAqu = J. B. Brusin, Inscriptiones Aquileiae, three volumes. Udine 1991–3. Inst. Iust. = Institutiones Iustiniani. Ioh. Bicl. Chron. = Victoris Tunnunensis chronicon cum reliquiis ex Consularibus Caesaraugustanis. Et Iohannis Biclarensis chronicon. Corpus Christianorum, Continuatio Medievalia 173A, R. Collins (ed.). Turnhout 2001. Isid. Etym. = Isidori Hispalensis episcopi etymologiarum sive originum libri XX. Two volumes, W. M. Lindsay (ed.). Oxford 1911. Isid. Hist. Goth. = Las Historias de los godos, vándalos y suevos de Isidoro de Sevilla, C. Rodrígues Alonso (ed.). Leon 1975.

xviii

A Note on Abbreviations and Sources Used

Isid. Reg. Mon. = Reglas monásticas de la España visigoda. Santos padres españoles 2, Julio Campos Ruiz, Julio and Ismael Roca Melia (eds.). Madrid 1971. I.Stratonikeia = Şahin, Mehmet Çetin (ed.), Die Inschriften von Stratonikeia, Bd. 1–3. Bonn: Habelt 1981–90. Jalā l al-Dıˉn al-Maḥallıˉ and Jalā l al-Dıˉn al-Suyuˉ tıˉ̣ , Tafsıˉ r al-Jalā layn = Jalā l al-Dıˉn al-Maḥallıˉ and Jalā l al-Dıˉn al-Suyuˉ tıˉ̣ , Tafsıˉ r al-Jalā layn, ed. ʿAlıˉ ibn Mus ̣tafā Khalluˉ f. Damascus: n.p. 2002. Jerome, Vita Malchi = Christa Gray (ed.), Jerome: Vita Malchi, Oxford Classical Monographs. Oxford: Oxford University Press 2015. LCL = Loeb Classical Library. Cambridge MA 1912–. LGPN = Lexicon of Greek Personal Names, www.lgpn.ox.ac.uk/index .html. LRV = Lex Romana Visigothorum, G. F. Haenel (ed.). Berlin 1849. LSA = ‘Last Statues of Antiquity’ – database, http://laststatues.classics.ox .ac.uk/. LSJ = Liddell–Scott–Jones Greek–English Lexicon. LTUR = M. Steinby (ed.), Lexicon Topographicum Urbis Romae. Six volumes. Rome 1993–2000. LV = Leges Visigothorum. Monumenta Germaniae Historica, Leges Nationum Germanicarum 1, ed. K. Zeumer. Hannover 1902. MAMA = Monumenta Asiae Minoris Antiquae. Manchester 1928–. Martínez Díez and Rodríguez = Gonzalo Martínez Díez and Félix Rodríguez (eds.), La colección canónica Hispana, Volumes 4–6. Madrid: Consejo Superior de Investigaciones Científicas, Instituto Enrique Flórez 1984–2002. Mujā hid, Tafsıˉ r = Mujā hid ibn Jabr, Tafsıˉ r, ed. Muḥammad ʿAbd alSalā m Abuˉ al-Nıˉl. Madıˉnat Nas ̣r. Cairo: Dā r al-Fikr al-Islā mıˉ alḤ adıˉtha 1989. Muqā til, Tafsıˉ r = Muqā til ibn Sulaymā n, Tafsıˉ r, ed. ʿAbdallā h Maḥmuˉ d al-Shahā ta. Five volumes. Beirut: Dā r Iḥyā ʾ al-Turā th al-ʿArabıˉ 2002. Neil and Allen = Bronwen Neil and Pauline Allen, The Letters of Gelasius I (492–496): Pastor and Micro-Manager of the Church of Rome, Adnotationes 1. Turnhout: Brepols 2014. NRSV = New Revised Standard Version of the Bible. Paschoud 1989 = F. Paschoud (ed.), Zosime: Histoire Nouvelle. Tome III. 2e partie. Paris: Budé 1989. Perrone 2015 = Lorenzo Perrone (ed.), Origen, New Homilies on the Psalms. Die griechischen christlichen Schriftsteller der ersten Jahrhunderte, N.F. 19. Berlin: De Gruyter 2015.

A Note on Abbreviations and Sources Used

xix

PG = Patrologia Graeca, J.–P. Migne (ed.). Paris 1857–66. Pizarra = Las pizarras visigodas: entre el latín y su disgregación, la lengua hablada en Hispania, siglos VI–VIII. Colección Beltenenebros 8, I. Velázquez (ed. and comm.). Burgos 2004. P.Nessana = Kraemer, Casper J., Excavations at Nessana (Auja Hafir, Palestine): 3, Non-literary Papyri. Princeton 1958. P.Petra I = Frösén, Jaakko, Antti Arjava, and Marjo Lehtinen (eds.), The Petra Papyri I. Amman: American Center of Oriental Research 2002. P.Petra II = Koenen, Ludwig, Jorma Kaimio, Maarit Kaimio, and Robert W. Daniel (eds.), The Petra Papyri II. Amman: American Center of Oriental Research 2013. P.Petra III = Arjava, Antti, Matias Buchholz, and Traianos Gagos (eds.), The Petra Papyri III. Amman: American Center of Oriental Research 2007. P.Petra IV = Arjava, Antti, Matias Buchholz, Traianos Gagos, and Maarit Kaimio (eds.), The Petra Papyri IV. Amman: American Center of Oriental Research 2011. P.Petra V = Arjava, Antti, Jaakko Frösén, and Jorma Kaimio (eds.), The Petra Papyri V. Amman: American Center of Oriental Research 2018. Pseudo-Joshua the Stylite, Chronicle = F. R. Trombley and J. W. Watt (eds.), The Chronicle of Pseudo-Joshua the Stylite, Translated Texts for Historians 32. Liverpool: Liverpool University Press 2000. Querolus = Les Grincheux: Comedie de la Petite Marmite. Latin text and translation by Catherine Jacquermard-Le Saos. Paris: Les Belles Lettres 1994. Al-Qurt ̣ubıˉ, Jā miʿ = al-Qurtubı ̣ ˉ, Abuˉ ʿAbdallā h Muḥammad, al-Jā miʿ liaḥkā m al-Qurʾā n, ʿAbdallā h al-Turkıˉ (ed.). Twenty-four volumes. Beirut: Muʾassasat al-Risā la 2006. Al-Rasʿanıˉ, Rumuˉ z = al-Rasʿanıˉ, ʿAbd al-Razzā q, Rumuˉ z al-kunuˉ z, ʿAbd al-Malik ibn ʿAbdallā h ibn Ruhaysh (ed.). Nine volumes. Mecca: Matktabat al-Asadıˉ 2008. Al-Rā zıˉ, Mafā tıˉ ḥ = al-Rā zıˉ, Fakhr al-Dıˉn, Mafā tıˉ ḥ al-ghayb. Six volumes. Cairo: Buˉ lā q 1861. SC = Sources Chrétiennes. Paris: Cerf 1943–. Selb and Kaufhold 2002 = Walter Selb and Hubert Kaufhold, Das Syrischrömische Rechtsbuch. Vienna: Verlag der Österreichischen Akademie der Wissenschaften 2002. SGO = Reinhold Merkelbach and Josef Stauber (eds.), Steinepigramme aus dem griechischen Osten, Bd. 1–5. München and Leipzig 1998– 2004.

xx

A Note on Abbreviations and Sources Used

S. Martín de Asan Document = ‘Cuatro documentos inéditos del monasterio visigoda de San Martín de Asán (522–586)’, G. Tomás-Faci and J. Carlos Martín-Iglesias (eds.), Mittellateinisches Jahrbuch 52 (2017): 261–86. Synod of Elvira = A. W. W. Dale (ed.), The Synod of Elvira and Christian Life in the Fourth Century: A Historical Essay. London: Macmillan 1882. Al-Ṭabarıˉ, Jā miʿ = al-Ṭabarıˉ, Jā miʿ al-bayā n, ʿAbdallā h al-Turkıˉ (ed.). Twenty-six volumes. Cairo: Dā r Hajr 2001. Al-Ṭabarsıˉ, Majmaʿ = al-Ṭabarsıˉ, Majmaʿ al-bayā n, Hā shim al-Rusuˉ lıˉ alMaḥallā ti (ed.). Ten volumes. Beirut: Dā r Iḥyā ʾ al-Turā th al-ʿArabıˉ 1979. TAM = Tituli Asiae Minoris. Vienna 1901–. Al-Thaʿlabıˉ, Kashf = al-Thaʿlabıˉ, al-Kashf wa-l-bayā n, Abuˉ Muḥammad ibn ʿĀ shuˉ r (ed.). Ten volumes. Beirut: Dā r Iḥyā ʾ al-Turā th al-ʿArabıˉ 2002. TM = Trismegistos. An interdisciplinary portal of the ancient world, www .trismegistos.org (Leuven). Vazquez de Parga 1943 = Luis Vazquez de Parga (ed.), Sancti Braulionis Caesaraugustani Episcopi Vita S. Emiliani. Madrid: Consejo Superior de Investigaciones Científicas, Instituto Jerónimo Zurita 1943. V Fruct. = La Vida de San Fructuoso de Braga. Estudio y edicion critica, M. C. Diaz y Diaz (ed.). Braga 1974. V Patrum Emeretensium = The Vitas Sanctorum Patrum Emeretensium, J. N. Garvin (ed.). Washington DC 1946. Von Gebhardt and Von Dobschütz 1922 = Oscar Von Gebhardt and Ernst Von Dobschütz (eds.), Die Akten der edessenischen Bekenner Gurjas, Samonas und Abibos, Texte und Untersuchungen zur Geschichte der altchristlichen Literatur, 37.2. Leipzig: Hinrichs 1911.

introduction Late Antique Studies and the New Polyphony for Slave Studies Chris L. de Wet

For the past few decades, studies on slavery in late antiquity have been primarily concerned with the question of whether slavery was, in fact, present during the period, or whether late antique slavery slowly declined and transformed into so-called medieval serfdom. For many years the latter proposition enjoyed favour among historians. Now, the picture is quite different. The majority of recent studies on late antique slavery confirm that slavery was alive and well during the period. Foundational studies such as Chris Wickham’s monumental analysis of the medieval period, followed more recently by Alice Rio’s focused study of early medieval slavery, Youval Rotman’s reconstruction of Byzantine slavery, and finally Kyle Harper’s extensive survey of late ancient slavery come to similar conclusions: the model or paradigm of ‘transition’, with its roots in nineteenth-century Marxist economic theory (especially from Marx and Engels), has outlived its usefulness for understanding labour and modes of production in the late antique world.1 Referring to the shift from the ‘ancient’ to the ‘medieval’ world, Wickham aptly remarks that ‘the path between them was by no means a straight one, and in the period 400–800, the actual period of the shift between ancient and medieval, the contrast is considerably less useful’.2 There is no need to rehash the arguments against the transitionary model of late antique slavery – these arguments are to be found in all of the above-mentioned studies. By now, claiming

1

2

Wickham 2005; Rio 2017; Rotman 2004; Harper 2011; see also De Wet 2015; 2018a; Lenski 2008; 2011a; 2017; MacMullen 1987; Ramelli 2016b; Vera 2007, with respective bibliographies. Wickham 2005, 259.

1

Chris L. de Wet

2

that late antique slavery slowly declined and disappeared and/or transformed into medieval serfdom is a simplistic and inaccurate scholarly narrative. Added to the premises in the works of Wickham, Rio, Rotman, Harper, and many others, in favour of a model that complexifies late antique slavery, is the point that the period and study of late antiquity in itself have experienced major changes in the second half of the twentieth century. For many years, late antiquity occupied what might be called a liminal historical space. It was considered, in itself, as a ‘non-period’ – a transitional placeholder for what some consider the ‘great ages’ of humanity. Before the 1970s, scholars interested in the period of late antiquity would have mostly been classicists, Roman historians, church historians, and, more specifically, scholars of patristics. For some classicists, late antiquity would have represented the period of early classical ‘reception’ (similarly, perhaps, to the field of biblical studies). For Roman historians, late antiquity marked the margins of the disciplinary field and the end of an era – this was especially shaped by Edward Gibbon’s paradigm of the decline and fall of the Roman Empire. For church historians and scholars of patristics, late antiquity was simply the ‘background’ or ‘stage’ on which their key figures and events, the great church fathers and councils, played out.3 After A. H. M. Jones, in 1964, laid an albeit conventional foundation for the beginnings of late ancient studies,4 much of the status quo changed when scholars such as Peter Brown5 and Averil Cameron,6 to mention only two of the more influential figures, began to promote the study of late antiquity as a discipline in and of itself. Brown devoted six centuries to what he called ‘late antiquity’, from 150 to 750 ce. Cameron had a narrower focus, from 395 to 600 ce. Most scholars now understand late antiquity to start around the third century and end somewhere between the seventh and even eighth centuries. This volume follows a similar delimitation for what is late antiquity (ca. 150–700 ce). What is significant of Brown’s early start to the period is that it includes part of the history of the High Roman Empire but excludes the rise of earliest Christianity, as embodied in the New Testament and earlier second-century Christian works; significant in the later boundary for late antiquity is the fact that it also includes the formation of early Islam.

3 4 5 6

Martin 2005. Jones 1992 [1973, 1964]. Brown 1971. Cameron 1993.

The New Polyphony for Slave Studies

3

But late antique studies did not simply entail delimiting and claiming a specific period in history. The study of late antiquity was also characterized by certain specific methodological shifts. From the perspective of religious studies, patristics slowly gave way to the study of early Christianity in late antiquity, which moved away from traditional approaches to the church fathers and councils often followed at seminaries or at some Christian (especially Catholic) universities. From its inception, late ancient studies has shown a particular affinity towards understanding (differently) the religious concepts and figures of Christianity, Judaism, Manichaeism, and Islam. This is related to the importance of cultural and social history in the study of late antiquity. Both Brown and Cameron have the term ‘world’ in the titles of their books. For both, ‘world’ meant more than geography – the term especially encompassed the social and cultural ‘worlds’ of the period.7 More recently, there is also what we might call an ‘exploded’ view of the late antique world, which decentres Europe and the Mediterranean and focuses attention on what was previously known as the ‘margins’. There is now a focus on ‘the interconnections between the Mediterranean and Africa, Iran, Arabia, the Baltic, Scandinavia, the British Isles, China, India and all of Asia, as well as disrupting the assumed connection between the late ancient/ Christian Mediterranean and modern, western Europe’.8 In its scholarly profile, this book also subscribes to this more expanded and inclusive approach to late ancient studies, and showcases authors coming not only from Europe and North America but also from Africa and South America. Essentially, then, Brown, Cameron, and others9 helped us to understand late antiquity less as a period of transition and more one of transformation and complexity in terms of geographical, anthropological, social, cultural, and religious contexts. This volume is firmly situated in this view of late antiquity, both as one with a particular interest in religious, social, and cultural transformations, and as one that does not have western Europe and the Mediterranean as its only geographical focus. Slavery in the Late Antique World, 150–700 CE presents a series of case studies that investigate the settings of slaveholding and representations of slave experience in late antiquity. The chapters in this volume scrutinize the ideological, moral, cultural, and symbolic value(s) of slavery

7 8

9

Martin 2005, 1–9. So the description of the academic journal Studies in Late Antiquity (first issue, 2017) (https://online.ucpress.edu/sla). See, for instance, the numerous articles in the two handbook volumes on late antiquity, edited by Johnson (2012) and Rousseau (2012).

4

Chris L. de Wet

as well as the status and living conditions of late antique slaves. The central argument of the case studies is to show that late ancient slavery is culturally and, often, geographically conditioned. Consequently, like late antiquity in toto, late ancient slavery is in itself a complex and polymorphous phenomenon – the volume as a whole will show that it is indeed more appropriate to speak of late ancient ‘slaveries’. Rather than painting late ancient slavery in broad generalizing strokes, this volume aims at more focused and specialized studies of the topic, looking at how slavery transformed and was transformed in itself in varying material and literary contexts. These include in-depth analyses of individual authors and works, papyrological and epigraphic evidence, specific geographical locations, and studies that are socially, culturally, and politically specific. The volume is interdisciplinary and transdisciplinary in its methodology. We therefore especially employ comparative and methodologically innovative approaches including studies revisiting and questioning traditionally held assumptions about late ancient slavery. Furthermore, in regard to geography, while several studies do focus on western Europe and the Mediterranean, including Egypt, a number of chapters focus on the nature of slavery in geographical areas on the socalled ‘margins’ of the Roman Empire, including late antique Syria and Mesopotamia, late antique Palestine, ‘Arabian’ contexts, the northern European contexts including Roman Britain and Gaul, and North African contexts, thus even beyond the borders of the Roman world. Some chapters focus on the nature of slavery in rural settings and slavery among late ancient nomadic cultures. Our aim is to include somewhat neglected areas in the study of slavery; such areas were often neglected because they have been considered geographically marginal to the Roman world (like the case of St Patrick) or, especially, because they have not been written in Greek or Latin, but in Syriac, Coptic, or Arabic. Likewise, this book includes studies on slavery in contexts of social and religious diversity and transformation, for instance slavery and gender, ethnicity, age, and embodiment. There are also chapters on the use of slavery as a metaphor on different levels of religious thought, and on the use of slavery discourse in different contexts of religious practice. A case study of slavery in earliest Islam, and the Quran, is also included, since early Islamic slavery is often discussed outside of its continuity (and discontinuity) with other late antique cultures. Furthermore, this volume also uses late ancient literary sources that have been neglected in previous studies of slavery, especially those not written in Greek or Latin, such as Syriac, Coptic, and Classical Arabic sources.

The New Polyphony for Slave Studies

5

Slavery in the Late Antique World, 150–700 CE is divided into four main thematic sections. The first three chapters explore the diverse ideological backgrounds as well as the symbolic and moral values of slavery. The second section extrapolates on intersections between slavery, cultural discourses, and identity. The chapters in the third section discuss social historical perspectives of slavery, both in the eastern and western Mediterranean. The final section examines the social history of slavery on the fringes of the Roman Empire and beyond. The question of slavery as a moral dilemma in late antiquity is a complex and contested matter. The chapters in the first section of this book will showcase this complexity, both in ancient responses to the moral problems of slavery and in the varying scholarly responses. The aim of this section is not to achieve an ancient or modern consensus, but to demonstrate and maintain the emotive tensions in approaches to slavery as a moral and symbolic issue. For instance, on the one hand, Pieter Botha’s chapter exhibits how early Christian discourse was complicit, in a very ‘Roman’ way, in the upholding of oppressive practices of slavery; on the other hand, Ilaria Ramelli illustrates that not all ancient authors were blind to the injustice of slavery, and that justice (in relation to slavery) was indeed a moot matter, especially in early Christian ascetic discourse. While both show that there was no concept of ‘abolition’ in early Christian discourse, both chapters demonstrate that the idea of the ‘naturalness’ of slavery, to late antique writers, might not be the most useful concept for understanding slavery as a moral and symbolic dilemma. Botha’s chapter asks how and what early Christian slaveholding practices communicate. Starting with Pliny’s infamous Epistula 10.96 and reviewing some of the information we can gather about Christian slaveholders in the second and third centuries, this chapter explores aspects of the impact of Christianity on Roman slaveholding. Scholars of early Christianity have, in general, told two stories about slavery and the church. The one story is one of golden beginnings among ‘equals’ with liberal relations between women and men, slaves and slaveholders, followed by a gradual adaptation to Roman hierarchies and internalizing of traditional social structures. The other story depicts a Christianity that eventually triumphs over the social values of the Roman Empire, so that the rise of Christianity leads, over a period of centuries, to the weakening and demise of slave society. What both ‘versions’ need to explain, or at least account for, is the continuity over centuries that existed with regard to attitudes towards slavery. Over the long course of antiquity, improvements in slave conditions were never made. So, an important question

6

Chris L. de Wet

must be: how does such a ‘continuity’ stay in place? Building on the work of Bourdieu, Botha proposes that the power of violence lies not in its capacity to harm, but rather in its capacity to communicate meaning, to instantiate power, legitimacy, and history, over those not physically or directly targeted in the act itself. The efficacy of violence is not measured in its casualties, but in its capacity to transform the subjectivities of those indexed in its performance. Hence, the symbolic violence of early Christian discourse facilitated slavery and for all practical purposes excluded any changes in the ‘meaning(s)’ of Roman slaveholding. In turn, Ilaria Ramelli discusses individuals or couples who emancipated their slaves and gave up their possessions in favour of the poor upon embracing the ascetic life. She also considers monastic groups who liberated all the slaves who joined their communities. She examines the link between ‘spiritual fasting’ and ‘immaterial self-restraint’ (as Gregory of Nyssa called them) based primarily on practising justice, and the renunciation of slave ownership. Asceticism thus turns out to be a matter not simply of self-restraint, but of justice. Spiritual asceticism is abstinence from oppressing others, owning other humans, and ‘robbing the poor with injustice’ (Gregory of Nyssa, Evagrius, Cassian). Eustathius of Sebaste’s and his followers’ ideas and practices, as well as monastic resistance to Gangra, clearly questioned established religious and societal norms, concepts, and institutions. The model for asceticism, indeed, was not that of a society on earth, but angelic life. The same opposition is highlighted within Judaism: regulations on slavery in Rabbinic Judaism will be contrasted with earlier Jewish ascetic groups who rejected slavery and social injustice, and embraced poverty. If we move beyond the moral dimensions of slavery, we might also enquire about the symbolic value of slavery. Late antique slavery also experienced various symbolic and discursive transformations. In this vein, Arkadiy Avdokhin’s chapter, on Christ as liberator from satanic debt in late antique Greek homilies and hymns, looks at one particular aspect of the afterlife of the apostle Paul’s conceptual framework of the faithful as ‘slaves of God’, who are at the same time freed from slavery to Satan. Therefore, he explores the discourse of setting the believer free from satanic slavery in late antique hymns and homiletic texts, as well as the interaction between the two. He suggests that in these texts, the legalistic thinking behind Paul’s original symbolism would stretch into portraying Christ as a type of ‘good brigand’, almost a Robin Hood, putting an end to the tyranny of the technically lawful but ultimately inequitable spiritual slavery of the humankind. Avdokhin discusses the homilies of Greek

The New Polyphony for Slave Studies

7

authors of the fifth and sixth centuries ce from Asia Minor – Proklos of Constantinople, Basil of Seleucia, and Pseudo-Asterios of Amasea, Pseudo-Chrysostom, and others. In them, enthralling images of Christ tearing down the legal certificate of debt leading to slavery are presented. He also discusses sixth-century hymns by Romanos the Melodist. In a number of his kontakia, Romanos emphatically presented Christ as destroying the technically correct but metaphysically transgressive documents keeping humankind enslaved to Satan. Most importantly, Avdokhin argues that the nullification of the satanic debt certificate is replaced by a new, yet equally bureaucratic, ‘heavenly’ system of paperwork that binds the believer into servitude to Christ. Maijastina Kahlos looks at how late antique writers dealt with the enslavement of foreigners, that is, non-Roman and non-Greek people outside the frontiers of the Roman Empire, conventionally called ‘barbarians’. In her analysis of the late antique authors (Christian and nonChristian alike), she considers their views on the social background of slave trade and frontier wars. Furthermore, the voices of the late Roman writers Synesius, Augustine, Gregory of Rome, Ambrose, and Basil of Caesarea are set in the context of the Graeco-Roman ideas of slaveholding. Moreover, she examines late antique Christian writers who confronted the issue of slavery by discussing the biblical exempla of Ham and Esau. Kahlos shows that ethnicity continued to play a role in the production of slave stereotypes and presuppositions in late antiquity. Kahlos’s findings are also important for the chapters in the next section in the book, which deals with slavery in relation to culture and identity. Late antique slavery was not, however, only related to moral and symbolic issues and dilemmas. As the chapters by Botha and Kahlos, in the first section, show, slavery was also a very potent cultural discourse, and one that always had concerns in the transformation of an individual’s or group’s identity. This point is especially demonstrated by Chris de Wet. His chapter analyses the Syriac (possibly) fifth-century life of Euphemia and the Goth, which recounts the tale of a young woman, Euphemia, who is deceived and manipulated into marrying an unnamed Gothic soldier. The theme of the deceived and abducted bride is a common one in the literature of the time. De Wet shows how slavery functions as a constant yet complex theme in the Syriac ascetic literature of late antiquity, and Euphemia and the Goth is no exception. In this narrative, slavery intersects with discourses of cultural identity and ethnicity (especially Syriac identity), law and warfare (and displacement), and morality and asceticism. De Wet investigates closely these discursive dynamics in Euphemia

8

Chris L. de Wet

and the Goth. The analysis gives us a glimpse not only into the complicated dynamics of slavery, displacement, and freedom on the margins of the Roman Empire, but also reveals the complicated dynamics of slavery as an ascetic discourse in the late ancient Christian cultures of Syria and Mesopotamia. While De Wet explores the dynamics of slavery in Syriac Christian discourse, Catherine Hezser’s chapter asks a basic yet fundamental question: what was Jewish about Jewish slavery in late antiquity? She explores attitudes, rules, and practices towards slavery mentioned in ancient Jewish literary sources and analyses them within the Graeco-Roman context in order to determine whether a specifically Jewish or rabbinic approach to slavery existed. A related question she ponders is whether certain attitudes expressed in the sources simply constituted an ideal or whether they can be considered to have been practised by at least some Jews in antiquity. After analysing late antique Syriac and Jewish sources, this section proceeds to late antique Egypt with Christine Luckritz Marquis’s chapter, in which she argues that the study of slavery in early Christian Egyptian monastic thought remains underdeveloped for multiple reasons. At the semantic level, the range and variety of words that were used to denote a slave both in Coptic and Greek are so numerous that no easy survey of extant literature is possible and indicates that late ancient Egyptian society did not systematize types of enslaved people in the ways modern scholars might expect. Furthermore, she demonstrates how Christian scripturalizing of slavery in Egypt in particular (where the Israelites were purportedly enslaved) further muddies access to historical slaves. Especially in the monastic writings of Shenoute (but also figures like Paul of Tamma) one finds slavery language theologically deployed, even as it seems plausible that slave labour was required of Shenoute’s church-building project. Luckritz Marquis shows that the only clear place where slaves emerge is in papyrological documentation, but that even there one finds little more than traces of names, rare glimpses of violence, and boundary policing between enslaved and free persons. Finally, this section of the book concludes with critical literary glimpses of slavery in late antique Gaul. Uiran Gebara da Silva’s chapter examines rural slavery in late Roman Gaul in the fourth and fifth centuries, as described in Gallo-Roman literary evidence. His discussion addresses the methodological and theoretical challenges offered by literary sources, discussing the way ideology and literary genre frame the representations of rural slaves. In particular, he analyses the writings by Ausonius, Salvian of Marseilles, and Paulinus of Pella, as well as Querolus, a late Roman

The New Polyphony for Slave Studies

9

comedy, and highlights the permanence of more intense forms of rural labour exploitation in the countryside. The majority of chapters up to this point focus mainly on literary texts from the late antique period. However, to get a fuller picture of the complexities of late antique slavery, one should also appreciate the papyrological and epigraphical evidence of slavery in the period. Papyri and epigraphy are especially crucial for reconstructing the social history of slavery. But as chapters in this next section show, it is by no means a simple and straightforward reconstruction. Arguing from the Petra Papyri, Marja Vierros, for instance, shows us how some of the basic terminology related to slaves is not always as clear as one would hope and imagine. Vierros provides a detailed and contextualized study of actual evidence of slaves in the documentary papyrological material, thus providing an important basis for the general socio-historical discussions (and, implicitly, some important caveats) on slaveholding in late antiquity. She also sheds light on slavery on another boundary of the Roman Empire, namely Palestine. Papyrological evidence from the area of Palestine is scarce, but the recent publication of the Petra Papyri gives us more information about slaves in this province of the eastern Roman Empire. So far, the only other papyrological dossier from the Palestine region, the Nessana Papyri, mentions slaves only in passing. Therefore, the Petra Papyri give us some long-awaited information on onomastics of slaves, their prices, their position in the elite households, and, on the whole, their existence in the Christian community of Petra. While Vierros examines slaves more generally in papyri from the area of Palestine, the chapter of Pudsey and Vuolanto takes us back to Egypt, but in this account, searching for enslaved children. They examine the visibility, activities, and agency of slave children and adolescents in the city of Oxyrhynchus, in Roman Egypt. This case study, in turn, provides us with abundant documentation of matters relating to the lives of young people, including slaves. Therefore, the chapter adds to the recent research into slave children across the Roman world based largely on literary sources and inscriptions. Among other issues, Pudsey and Vuolanto discuss household connections and their permanence; travel and agriculture; and working and learning conditions. They also take up the question of how to identify child slaves, as ‘child-vocabulary’ is also used to denote (adult) slaves. The chapter is based on the database of young persons in the city collected by Pudsey and Vuolanto, for which they have systematically examined the over 7,500 published documentary and literary papyri, material objects, inscriptions, and literary texts preserved to us

10

Chris L. de Wet

from Roman and early Christian Oxyrhynchus. Those texts that can be connected both to slavery and children, together some seventy, cover one tenth of all the cases in the database. Late antique inscriptions recording slaves have received much less scholarly attention than earlier ones. In order to address aspects of this lacuna in scholarship, Mariana Bodnaruk examines the epigraphic representation of the late Roman slavery from 260 ce until the sixth century, thus engaging in a wide-ranging analysis of late Roman epigraphic practice registering slaves both in the imperial centres and in the rural periphery, as well as on the borders of the Roman Empire. She further elaborates on what the epigraphic monuments and concomitant imagery reveal about late antique slavery and its further decline, and consequently reexamines traditional assumptions about late Roman slavery. As she brings into focus both Latin and Greek epigraphic evidence, the chapter is able to provide new insights into the persistence and spread of slavery in late antiquity. The last section of this book takes the step of crossing the boundaries and exploring slaveries beyond the Roman Empire. The section begins with Noel Lenski’s analysis of slavery in the Visigothic kingdom. The kingdom of the Visigoths endured for nearly three centuries, from 418 until the early 720s, moving its capital from Gaul to Spain after 507. It has left distinctive traces in the source record, particularly in normative sources such as the Laws of the Visigoths and the Acts of the Visigothic Councils. In addition, the Visigothic kings issued a separate code (Breviarium Alaricianum) for their Roman subjects that preserves and elaborates many Roman laws implemented for the management of slaves. Lenski examines this vast evidence synoptically in order to explore what was distinctive about Visigothic slaveholding and what represented inheritances from the Roman system that preceded it. According to Lenski, the Visigoths offer clear evidence for a robust slaveholding culture in the sixth and seventh centuries that corresponded with a period of relatively weak central state authority but relatively strong overall economic performance. Assembled from an admixture of Roman and Germanic customs and norms, Visigothic society hosted a unique but highly intensified slaveholding culture worthy of study unto itself. Judith Evans Grubbs then focuses on a somewhat unconventional yet quite rich source for the experiences of the enslaved beyond the borders of the Empire: the Roman Briton Patricius, known today as ‘St Patrick’. The freeborn son of a decurion and slaveowner, Patrick was born in Britain around the turn of the fifth century, shortly before or just around the time

The New Polyphony for Slave Studies

11

that Roman forces were withdrawing from the province forever. His family’s villula was attacked by pirates and he, along with many of his family’s slaves, was taken into slavery in barbarian Ireland. Patrick’s experience of capture and enslavement was far from unique at this troubled time, but his account of his subsequent escape from Ireland after six years of slavery and his eventual return to preach the gospel to the very people who enslaved him is unique. His two authentic writings, the Confessiones and the Epistola ad milites Corotici (Letter to the Soldiers of Coroticus), reveal how a devout Christian, thoroughly imbued with scripture, could both justify his own enslavement and condemn the lawless enslaving activities of a contemporary British warlord. Evans Grubbs shows how Patrick’s own enslavement affected his idea of himself as sinner, his sense of responsibility for the very people who had enslaved him, and his sympathy for the sufferings of slaves. Finally, it is acknowledged that the issues of slavery in classical Islamic law and thought, as well as the social reality of slaves and ex-slaves (clients) in the medieval Near East, have been gaining increased attention in recent decades. Ilkka Lindstedt’s chapter on slave boys in paradise is the first specific discussion on slavery and slaves in the Quran or Quranic environment. Lindstedt situates and analyses the Quran’s discourse on slavery, and later Islamic understandings of it, in the late ancient Arabian context. The special focus here is the Quran’s depiction of heavenly servants in verse 52:24 as well as the reception of this verse in exegetical works. The later Muslim commentators of Quran 52:24 put forward differing interpretations of the verse but, in particular in the late medieval era, understood it to refer to slaves that serve the believers in the hereafter. Intriguingly, some Quranic commentators appear to have suggested that these paradisal slaves might be eunuchs. In conclusion, as a ‘new polyphony’ for slavery studies, the aim of this book is to present a picture of the varieties and complexities of late antique slavery by means of a wide array of case studies. These case studies are themselves varied, both in terms of methodology and content. Some case studies are located in Europe and the Mediterranean, but many others push the boundaries, geographically but also culturally and discursively, by focusing on slavery in the regions of Syria, Palestine, Arabia, Roman Gaul, Britain, and Ireland. It is our intention, therefore, to show how necessary it is to do research that is embedded in the certain geographically and chronologically set contexts; only in this way can actual continuities and changes be studied, and then comparisons (between time periods, locations, or cultural spheres) can become meaningful. With this

12

Chris L. de Wet

approach, the book hopes to act as a stimulus for further research into the lesser explored regions, also geographically, culturally, and discursively, of late antiquity (such as Persia, the Balkans, and China), where slavery and its legacy might also be present. The book also hopes to contribute to the ongoing discussion in contemporary times on the oppression, marginalization, and indeed enslavement, of individuals considered to be ‘marginal’ by those in positions of power and in possession of privilege.

part i MORAL AND SYMBOLIC VALUES OF SLAVERY

1 Masters and Slaves in Early Christian Discourse Pieter J. J. Botha

transformation and survival At the heart of Christianity lies an imperative to change. Following Jesus, becoming part of the movement, is about transformation: changing oneself, changing communities, indeed, changing the world. But then, we are all also subject to the imperative to survive: to accept, to adopt and adapt, to conform and to continue, to compromise, and to let be. Nothing reveals this paradoxical nature of the early church more clearly than the history of early Christianity and slavery. Jennifer Glancy points out that scholars of early Christianity tend to have two distinct perspectives on slavery, the church, and society.1 While not contradictory, the two views, in essence, depict different trajectories of which one can be characterised as of descent and the other of ascent. According to one view, the Christian movement in the earliest years was a golden age for relations between women and men, slaves and slaveholders. This story begins with the small group of men and women who were disciples of Jesus, proclaiming that those in Christ are divided neither by gender nor by legal status, slave or free (Gal. 3:28). This version emphasises that, as the decades continue, the Christian movement accommodates itself more and more to the structures of the surrounding society, so that ultimately the Roman Empire triumphs over the social values of the first Christians. Persecution is sometimes invoked to explain this ‘surrender’. After all, violence weakens even the strong, and understandably ‘Christians seek to

1

Glancy 2002, 145.

15

16

Pieter J. J. Botha

become more like their neighbors to avoid the attention of authorities.’ The accommodation ‘is thus not so much a compromise as an evangelical strategy’.2 According to the alternative view, in which Christianity eventually triumphs over the social values of the Roman Empire, the rise of Christianity over a period of centuries leads to the weakening and demise of slave society. Although the church does not directly oppose the institution of slavery, Jesus’ teaching about the dignity of each person ultimately undermines centuries of the dehumanisation of slaves. ‘Christianity is incompatible with the institution of slavery, but it takes centuries for Christians to come to understand this and to transform the structures of their world in accordance with this moral vision.’3 As Glancy points out, both stories attempt to impose linear order on scattered and at times paradoxical references to slavery in early Christian literature. But what should be noted clearly (and fearlessly) is the apologetic and rationalising slant of such versions of early Christianity and its dealings with slavery. While Christians of the first centuries were not in a position to challenge the imperial social order, they neither condemned nor forbade slaveholding within Christian communities. Christians throughout the Empire continued to contend with slavery in an astoundingly (and disturbingly) neutral way, at best. Some Christian leaders and thinkers could be critical of slaveholding practices, but they were frequently more concerned with the moral well-being of slaveholders than with the physical and social well-being of slaves. Over the long course of antiquity, improvements in slave conditions were never made. ‘How could things have been otherwise, when slaveowners’ attitudes to slaves and slavery showed no fundamental alteration either?’4 How does such a ‘continuity’ stay in place? How does it happen that, despite an undeniable change in religious perspective and an adoption of a distinct vocabulary (emphasising love, salvation for all, freedom, and lack of discrimination) by early Christians, there is this pervasive and unchanging neglect, a blind spot obscuring the disregard experienced by enslaved humans? When discussing the history of slavery, we can only start with an acknowledgement of the complicity of early Christian discourse and 2 3 4

Glancy 2002, 145. Glancy 2002, 145. Bradley 1997, 281, discussing Garnsey 1996.

Masters and Slaves in Early Christian Discourse

17

practice in the history of slavery. Rather than another apology, we should strive for a sober evaluation of early Christianity’s slaveholding practices and attitudes. Of course one may observe from our own current moral vantage point that early Christians were creatures of their times; that the attitudes they voice towards women and slaves were more or less settled dogma, common to the point of near universality (even if there was an occasional fringe critique of the unreflective rectitude of the status quo); that Christian slavery was, from the standpoint of vicious abuse, often less disgraceful than its Roman counterparts; that some early Christians thought masters should befriend their slaves, or were indeed indulgent towards their own slaves, at times freeing them – but all of these factors do not allow one to look away from the subdued and submissive bodies, and to not notice the chains and the whips in the background. After all, even (a little) less violence is still violence, and we need to understand something of the dynamics, the process, of supporting or accepting slavery. We have to grasp how there could be a lack of awareness of the tension between freedom, equality, and slavery practices. The suggestion here is that we can understand these histories of death,5 displacement, and dispossession better by paying attention to the performative capacities of violence to generate culture. We need to note how violence (and implied violence) structures bodies and bodily practices, and how conceptions of self and other are fashioned when violent behaviour is deemed acceptable. Some time ago, Allen Feldman observed about political violence in Northern Ireland that it became ‘a residual and cultural institution . . . possessing its own symbolic and performative autonomy’.6 This can be said about other forms of violence in many locations. It is worthwhile to consider aspects of the role of violence in the formative centuries of Christianity with regard to attitudes and actions towards slaves.

5

6

Patterson describes slavery as social death, in addition to which we have to remind ourselves that this social death could and did entail physical destruction of bodies as well. The importance of Patterson’s conceptualisation lies in the focus on the social condition of slaves rather than on their status as property. Slavery is a parasitic relationship between master and slave, invariably entailing the violent domination of a natally alienated and generally dishonoured (hence socially dead) person; see Patterson 1982, 6, 13, 173, 335. Feldman 1991, 9.

18

Pieter J. J. Botha

to ascertain the truth from two slave girls . . .7 At the start of the year 110 ce, Plinius Caecilius Secundus, or as we commonly refer to him, Pliny the Younger, was sent by Emperor Trajan as his official representative to the volatile province of Pontus-Bithynia. Besides the task of curtailing the extravagant expenditure of public funds, the new governor had to cope with the strife of factions in the cities of the Roman province. One of the dangers to harmony lay in the formation of political factions (or clubs), which might resort to open violence or, at best, organise the voters against the rule of the wealthy and more conservative element favoured by Rome. Pliny, accordingly, issued an edict prohibiting the formation of so-called ‘brotherhoods’, a name that might easily be used as a cover for an association of a political character. Between 18 September 111 and 3 January 112, Pliny found himself in Sinope, a town on a peninsula about midway along the south coast of the Black Sea. Pliny had to deal with several prominent community leaders about possible dubious loyalties before arriving in Sinope.8 But here he encountered an unexpected resistance from the locals: what about the Christians? The buying, slaughter, and sale of sacrificial meat had become poor business because of the Christians, and that surely had to be proof of their lack of loyalty. Pliny responded with an immediate investigation, which led to orders for the execution of several Christians. He also prohibited the evening meetings of Christians, which were seen to be similar to the activities of outlawed associations. However, his initial actions only led to more accusations, and further investigation was necessitated. To gather reliable information, he arrested two slave girls belonging to a household or households involved in this ‘Christian faction’. Given the (apparent) gravity of the situation, there is a curious lack of specifics about persons involved in 7

8

Referring to ‘quo magis necessarium credidi ex duabus ancillis . . . et per tormenta quaerere’ (Plin. Ep. 10.96.8): ‘I became convinced that it is necessary to ascertain the truth from two slave girls . . . and under torture.’ Text consulted: Mynors 1963; my translation. It is not explicitly said where Ep. 10.96 should be situated; geography, itinerary, and letter sequence are ambiguous in Plin. Epp. 10.90–103. Ep. 92 was written from Amisus, Ep. 98 from Amastris, so commentators propose either Amastris or Amisus for Ep. 96 (e.g. Williams 1990, 139; Sherwin-White 1966, 691). In order to make sense of possibly ‘some other town between the two’ (Wilken 1984, 15), I propose Sinope. Pliny would have stayed either in Sinope or in Amastris, setting up a temporary office: Sinope was a Roman colonia, Amastris the seat of the provincial council. As Pliny preferred coastal journeys, he probably made Sinope, an important maritime centre, his base (Sherwin-White 1966, 532). For discussion see Harris 1980, 872–3; Benko 1980, 1068–76; Barnes 1968, 36–7. There was an early and influential presence of Christianity in Sinope (Robinson 1906, 268–9; Kaçar 2008, 198–201).

Masters and Slaves in Early Christian Discourse

19

Pliny’s correspondence, with no reference to pertinent leaders or organisers;9 he mentions only ‘some’ and ‘others’. Pliny then proceeded to have the girls tortured in order to ‘ascertain the truth’. The quaestio per tormentum (interrogation by torture10) to which Pliny refers invokes various social relation issues.11 More critically, torture basically entailed the death penalty. One was literally tortured until the end. It was the only way in which the torturer could be convinced that the ‘full’ truth had been extracted. Notice the taken-for-grantedness of Pliny’s actions. From the correspondence it is clear that neither he nor the emperor thought it strange or even unusual that two young women were to be arrested and tortured to death; all was coolly reported. The disregard for women’s bodies can hardly be more evident. Ancient Sinope was not a large town. The arrests and especially the incarceration of the two women would have drawn attention. Their screams, while their bodies were cut open by whips, were probably heard throughout the community. We have a good idea what (at least some of) the elders of the Christian group had to say while the two deaconesses’ joints were being torn on the rack. Slaves, accept the authority of your masters with all deference, not only those who are kind and gentle but also those who are harsh. For it is a credit to you if, being aware of God, you endure pain while suffering unjustly. If you endure when you are beaten for doing wrong, what credit is that? But if you endure when you do right and suffer for it, you have God’s approval. For to this you have been called, because Christ also suffered for you, leaving you an example, so that you should follow in his steps. (1 Pet 2:18–21)12

9

10

11

12

Keeping in mind the situation of the letters is ‘a context-specific quick fix to urban unrest in an unimportant backwater of the Empire’ and Trajan’s response is a local and limited one, ad hoc ‘to local turbulence’ (Corke-Webster 2017, 261, 249). ‘Torture, designed to force admission and recantation, made use of the rack, and of redhot metal plates laid on the skin, and of the metal “claws” which lacerated flesh and exposed bones and internal organs’ (Clark G. 1998, 99). ‘By torture we mean the infliction of anguish and agony on the body to elicit the truth. Mere interrogation or mild intimidation does not come within this edict. . . . It is when an investigation is conducted with force and bodily torment that there is said to be torture’, according to Ulpian (Dig. 47.10.15.41; cf. Watson 1998, 293). Instruments of torture used by the Romans were the eculeus (the rack, or ‘stretching horse’), fidicula (rope), ungula (iron claw), and of course, the flagellum (whip); cf. Vergote 1972; Shaw 1996, 271. Not everyone could be tortured. The proscription of torturing witnesses who were free persons was still being upheld by Hadrian (Garnsey 1970, 213–16). The testimony of slaves was allowed only within the context of torture (Jones 1972, 114–16; Thür 1972, 105–7). All Scripture translations are NRSV.

20

Pieter J. J. Botha

The next day, when some of the other slaves spoke about the blood against the wall and on the floor they had to clean, we can surmise what their comments might have been: Beloved, do not be surprised at the fiery ordeal that is taking place among you to test you, as though something strange were happening to you. But rejoice insofar as you are sharing Christ’s sufferings, so that you may also be glad and shout for joy when his glory is revealed. (1 Pet 4:12–13)

Slaves were not cremated, and burial was typically delayed or denied to the tortured. We can presume that the congregation members paid a quiet last honour to the broken and cut-up bodies when they were taken outside the town walls. Then someone probably said something like: Blessed be the God and Father of our Lord Jesus Christ! By his great mercy he has given us a new birth into a living hope through the resurrection of Jesus Christ from the dead, and into an inheritance that is imperishable, undefiled, and unfading, kept in heaven for you, who are being protected by the power of God through faith for a salvation ready to be revealed in the last time. (1 Pet 1:3–5)

After all, the First Letter of Peter was written to the faithful in PontusBithynia (1 Pet. 1:1), probably at the beginning of the second century.13 Whatever was done and said, astoundingly, Pliny leaves us with no misunderstanding: there was acceptance of his actions even among the Christians.14 Among the congregation there was a deliberate submission about the whole incident, and, as usual, the focal point of this acceptance, this obedience, was the bodies of women. It is this matter-of-factness about the suffering of the enslaved and the unconcern about slave bodies15 that requires deliberate consideration of the slave’s material body as a pressing moral concern.

writing about slaves Adolf Harnack wrote that it would be a ‘mistake to suppose that any “slave question” occupied the early church. The primitive Christians 13

14

15

For discussion of the setting and context of 1 Peter, see Downing 1988, 114–16; Gielen 2013. Gielen argues for a Hadrianic date for 1 Peter. Feldmeier 2008, 28–40, points to the period of Domitian. At no point does Pliny indicate any resistance from the Christians or their leaders. He notes the success of his measures: ‘It is at any rate certain that temples which were almost abandoned have begun to be crowded, and the solemn rites which for long had been suspended are being restored. The flesh of the sacrificial victims, for which up to now only a very occasional buyer was found, is now on sale in many places. This leads me readily to believe that if opportunity for repentance is offered, a host of people can be set right’; Plin. Ep. 10.96.10. This characteristic of early Christian texts/authors is central to Glancy’s 2002 study.

Masters and Slaves in Early Christian Discourse

21

looked on slavery with neither a more friendly nor a more hostile eye than they did upon the state and legal ties’. Early Christians simply assume that Christian masters have slaves (not merely that ‘pagan’ masters have Christian slaves), and they give no directions for any change in this relationship. On the contrary, slaves are earnestly admonished to be faithful and obedient.16 What Harnack is doing is to summarise a paraphrase of early Christian views on slavery. Noting the continuity between early Christian and traditional Roman attitudes and practices with regard to slavery – as many scholars since Harnack have done – is just reiterating the obvious. The question cannot be whether they did, but to say something about why and how the early Christians practised their slaveholding. Of course, the encouragement by some Christian writers that ‘good relationships’ between masters and slaves are important and should be noted, and the efforts towards manumission in certain cases should be acknowledged, but Roman authors also encouraged good relationships between masters and slaves. Christianity – where amelioration was attempted – simply reflected good, decent Roman behaviour. The Roman Empire did not fear Christianity as the great slave revolution. The quest for understanding involves more than conceding that early Christianity accepted slavery; it is about questioning the illusion of naturalness and transparency associated with slavery, and reviewing the common-sense normative expectations and conventions with which Christian slaveholders would have been familiar. Is it just convention when early Christian sources perpetuate the widely held Roman caricatures of slaves? So, for example, Tertullian casually notes that ‘our slaves (domestici nostri)’, by the nature of things, ‘are strangers and enemies of the truth’.17 Harrill shows that ancient Christian literature is filled with common ‘stock types’, ‘bland polarities’, and ‘clichés’ of slaves and slavery, employed by free writers (of which most were masters of slaves) for rhetorical benefit in the context of their own projects. Christian writings ‘reflect, participate in, and promote the literary imagination about slaves and the ideology of mastery widely diffuse in the ancient Mediterranean’, thereby upholding the Roman ideology of auctoritas.18

16

17 18

Harnack 1908, 167. This is in the chapter ‘The Gospel of love and charity’, subsection ‘Care for slaves’, where Harnack explains what early Christianity was about (historically speaking), according to him. Tert. Apol. 7.3 (Glover 1931, 38 [LCL]; however, my translation). Harrill 2006b, 2.

22

Pieter J. J. Botha

There were clear efforts at incorporating slaves into the Christian community,19 but far from subverting stock representations of slaves, early Christian writings, and even martyrdom accounts, reiterate conventional Roman representations, reinforcing the distinctions between the free and the enslaved. Even more to the point are the ever-present references to the menace of violence, punishment, humiliation, and extortion in early Christian discussions of slavery. The issue of a reliable historiography of early Christian slaves must be mentioned. Technically, we have only the ‘literary imaginations’ of early Christians, and one should be careful not to confuse such rhetoricised literary typologies with history.20 But when we do bring the literary and rhetorical effects of our sources to the front, such characterisations and effects are meaningful in context, in actual situations where the conditions for violence towards slaves rest on a consensus regarding the inferiority and lack of honour of slaves. Most of what the early Christian literary representations reveal about slaves can only be meaningful in a social context where violence towards slaves is always present. Tertullian implies quite clearly that harsh punishments are ‘appropriately’ common; the bodies of manumitted slaves of Christians that have been subject to whips, shackles, and brands will not suffer the same things in the resurrection.21 Origen too acknowledges that slaves were frequently trained by Christians with the fear of lashings.22

augustine At the beginning of his career as bishop of Hippo,23 Augustine, in the course of praising his mother, Monica, tells of an incident concerning 19

20

21 22 23

So, for instance, the manuals of church order from the third and fourth centuries indicate that church officials should inquire whether potential catechumens were slaves. A slave of a believing owner could only join the catechumenate when the slaveholder certified the slave’s worthiness for baptism. Slaves who were the property of non-believing slaveholders were urged to please their owners in order to serve the reputation of the gospel (Const. ap. 8.32.3; Hippol. Trad. ap. 15.3–5). Particularly interesting is Combes’s chapter on patristic liturgy, which includes an analysis of cultic actions likely to have been associated with the postures and activities of slaves; Combes 1998, 109–20. Note, for instance, Harrill’s hesitancy about a social history of early Christian slavery: ‘most slaves in the New Testament and early Christian literature are literary products, drawn from classic stock scenarios that reflected the conventional Roman value of auctoritas’; Harrill 2006b, 196. Tert. Res. 57. Or. Princ. 3.5.8. The Confessiones dates from 397–401.

Masters and Slaves in Early Christian Discourse

23

Monica and her mother-in-law. Tensions arose between the two of them because the household slaves were gossiping about Monica. To prevent further occurrences of such incidents, the mother-in-law arranged with her son to have the slaves whipped. After a severe beating, the chastened slaves, according to Augustine, caused ‘no more problems’ and harmony prevailed between mother- and daughter-in-law.24 It is worth mentioning this incident, as Augustine likes to distinguish between reliable, loyal slaves and unfaithful slaves. In many homes slaves lived in comfort and by their fidelity added to the contentment of their masters.25 Yet, many of the wealthy lived in fear in their own homes, with the constant threat of untrustworthy slaves murdering and stealing from their owners.26 So how can one ensure that one’s slave is ‘faithful’? Well, by berating them,27 and with physical punishments, such as a good scourging,28 a bit of incarceration, being put in chains, or a stint in the baker’s mill,29 slaves were kept honest and faithful. Augustine explicitly reminds us that through fear of these penalties, even vexatious slaves would strive to please their masters.30 He may not have been a social revolutionary, but the bishop of Hippo did advocate for the humane treatment of individual slaves, as slavery imposed a beneficial order on a sinful society.31 Like the Stoic thinkers, Augustine considered internal slavery to the passions to be far more dangerous than chattel slavery.32 Augustine clearly took slavery – and the ‘normal’ way of dealing with slaves – for granted. He also explained the origins of and reasons for slavery.33 My interest, however, is in Augustine’s perception and practice of this ‘normality’ of slavery in everyday life – a habitus. 24 25 26 27 28 29 30 31 32

33

Aug. Conf. 9.9.20. Aug. Serm. 159.4. Aug. Serm. 259.5, 113.4; En. Ps. 38.12. Aug. En. Ps. 73.16. Aug. Serm. 161.9. Consider the vivid description by Apuleius (Met. 9.12) of working in a baker’s mill. Aug. In Evang. Iohan. 43.7. Cf. Klein 1988, 102–9. This is standard Stoic thought on metaphorical slavery (e.g. Diog. Laert. V. phil. 7.121–2; Philo Prob. 3.17–18; cf. Long and Sedley 1987, 410–23). More detail on ‘internal’ slavery in Graeco-Roman times: Garnsey 1996, 132–3, 206–7; Harrill 1995, 193–5; Glancy 2011, 460–5; Conley 2015, 140. A key text is Civ. 19.15–16. Augustine acknowledges that slavery is not the arrangement prescribed by nature, for rational beings ought to rule lesser beings, not each other. But once sin has occurred slavery has its place, either punishing sins and leading to repentance, or, where it seems to be undeserved suffering, signifying the general human

24

Pieter J. J. Botha

Peter Brown is impressed by Augustine’s Epistula 10*, in which an elderly Augustine discusses the tragedy of bands of slave merchants and kidnappers who were ensnaring free Romans, including women and children, and shipping them as slaves to all parts of the Empire.34 As a deeply law-abiding Roman – a bishop, after all – he requires Roman law to be applied, so he writes to Alypius, his long-standing friend in Rome, to assist him to find relevant laws, but also to bring the matter to the emperor. Augustine even gives advice on how the emperor could be persuaded, namely in the name of Roman freedom. As he explains in De civitate Dei, slavery is indeed unnatural, but the result of sin.35 Sin must be kept in check, and this motif underlies Augustine’s attempts to invoke (or rely on) the state to look into unacceptable slave trading. The point of Augustine’s concern is still the good order, the ‘correct’ (if legal) way to deal with the ‘inappropriate’ slave trade.36 The older Augustine’s attitudes or convictions are still only a concern for the security of free citizens over and against enslaved ones. In Epistula 10*.3 Augustine describes the practice of slave trading as impietas, and he does encourage the good treatment of slaves. The degradation, sufferings, humiliation, and fear experienced by real slaves seem to lie completely outside of his awareness. In fact, so far as he acknowledges such things, they are simply the darker side of natural law, responses to sin authorised by God – so real slaves must simply cower down and accept the suffering. Augustine does not connect the Pauline injunction of ‘no more slave’ to anything in actuality. His social standing locates him as a person of privilege whose family had been keeping slaves for generations. Despite

34 35 36

condition of sinfulness. Only a world without sin would be a world without slavery. Slavery, additionally, serves the good purpose of keeping devastation and slaughter, which will result from vanquished peoples left alone, in check; Aug. Civ. 18.2. For good discussions of Augustine and slavery, see Chambers 2013; Conley 2015; Corcoran 1985; De Wet 2010; Garnsey 1996, 206–19, 228–35; Ramelli 2016b, 153– 63; TeSelle 1988. See especially the chapters of Kahlos and Evans Grubbs in this volume. Brown 2000, 470–1. Thus, while not natural, slavery is inevitable and unavoidable. Augustine laments emotionally that only five or six (of about 120 people) were legally sold by their parents. ‘Almost no one could refrain from crying at hearing the conditions under which the rest fell’; Aug. Ep. 10*.7 (translations from Augustine’s letters are my own). In Ep. 24* he explicitly seeks clarity about possible limitations to the extent of selling persons into slavery. Invoking Titus 2.9, he justifies Christians requiring slaves to obey their masters, but clearly stipulates that ‘we cannot impose the yoke of slavery upon free persons’.

Masters and Slaves in Early Christian Discourse

25

living at a time when Roman citizens were in more danger of being kidnapped and enslaved than for many centuries previously, as a member of the slave-owning class he was restricted in his ability to see, or at least empathise with, the moral depravity of owning slaves. Augustine never actually addresses a particular, individual enslaved human being.37 In any case, there is always the cheering thought that it is better to be the slave of a human master than the slave of lust.38 One can sense the body language of these Christians when they (slaves and masters) interacted. Taking a wider look, it is noteworthy that Augustine, in his scriptural hermeneutics, goes to some lengths to prove that violent passages in scripture, when read from a proper perspective, are not violent at all. His reasoning is that God seeks to bless, not to curse; God desires mercy, not judgement; and, when properly applied, violence belongs to God alone. It is the presence and actions of sin that require God to allow and even make use of violence. It is a divine decision that orders the universe by manifesting justice in (violent) punishment for sin. This apologetic aim of Augustine has been often noted, commented on, and commended at times.39 It is a hermeneutic built on the causal assumption of sin that needs violent response. Note that the logic of sin requiring punishment, of evil causing suffering, of disobedience generating humiliation, is a dynamic one: where there is violence there must have been sin.40 Not only does such thinking offer no comfort to those who have been on the receiving end of hostility undertaken in the name of the just God, but the collateral damage, the embodied suffering of such righteous justice, will always fall outside the scope of concern. A set of associations like these will determine how one postures and gestures and how one looks at others. It will also make one insensitive: after all, because there are guilt and obstinance (sin), violence must have a place and is reasonable. 37

38 39

40

He does, however, acknowledge some of them, such as the household slave (famulae) who cared for Monica and his father when young. This household slave-woman was held in honour by her dominis, and treated as a member of their Christian household. She remains nameless though; see Aug. Conf. 9.8. Aug. Civ. 19.15. This in summary of aspects of Augustine’s wide-ranging, complex, and dynamic process of scriptural interpretation; see, for discussion, Andrews 2012; Babcock 1993; Burns 1988; Mann 2001; Norris 2003; Otten 2010; Pool 2004; Westerholm and Westerholm 2016, 129– 61. Again, it is the wider context, the longue durée, the mentalité at stake here: ‘Physical and legal degradation corresponded in Roman society to moral degradation’; Roller 2001, 226.

Pieter J. J. Botha

26

analysing attitudes: cultural capital, symbolic violence, and orthoposture As is well known, Bourdieu proposes a theory of practice (human actions) that connects structure and agency in a dialectic relationship between culture, structure, and power. Bourdieu, extending the idea of capital beyond the economic into the symbolic realm of culture, argues that capital forms the foundation of social life and dictates one’s position within the social order. Contrary to a common (mis)reading of his work, his is not a utilitarian theory of social action in which individuals consciously strategise to accumulate wealth, status, or power. Bourdieu holds that the important source of conduct is the need for dignity, which society alone can provide. By being granted a name, a place, a function, within a group or institution, the individual can hope to escape the contingency, finitude, and ultimate absurdity of existence. Human beings become such by submitting to the ‘judgement of others, this major principle of uncertainty and insecurity but also, and without contradiction, of certainty, assurance, consecration’.41 Social existence means difference, and difference implies hierarchy, which in turn sets off individuals and groups in the endless dialectic of distinction and pretension, recognition and misrecognition, arbitrariness and necessity. ‘Capital’ functions both as resource and as stakes in the struggle to position oneself and gain dignity within cultural fields (social contexts). Capital is a key component within Bourdieu’s theory of social reproduction.42 He conceptualises capital as the legitimate, valuable, and exchangeable resources in a society that can generate forms of social advantage within specific fields (for instance, education) for those who possess it. Bourdieu identifies various types of capital that, through interactions with habitus (a person’s internalised matrix of dispositions, which guides behaviour) within fields (social contexts), produce relations of privilege or subordination in society.43 Social capital refers to networks and relations, and cultural capital refers to qualifications, dispositions, and cultural goods. Symbolic capital refers to those forms of capital that are accorded special legitimation with high social prestige and that hence may be the most powerful in accruing social advantage.

41 42 43

Bourdieu 2000, 237. Bourdieu 1977; 1984; 1986. Bourdieu 1986.

Masters and Slaves in Early Christian Discourse

27

Capital – as embodied, institutionalised, and/or objectified resources – entails diverse types of social skills and communicative utilities, somewhat like the ‘cards’ that a player possesses (and the knowledge of the ‘rules’) within a particular ‘game’, which will shape the ability to play and one’s options for surviving.44 Capital ‘does not exist and function except in relation to a field’ because field governs the ‘rule of the game’, determining the value of particular forms of capital within a given context.45 There is clearly a conceptual overlap between forms of capital, which can lead to them being treated synonymously.46 Cultural capital is both a source of social inequality – certain forms of cultural capital are valued over others, and help or hinder one’s social standing (or mobility) – and the structuring mechanism, the maintenance thereof. Cultural capital can be embodied (the long-lasting dispositions of the mind and body)47 and institutionalised. In its institutionalised form, cultural capital refers to symbolic credentials and qualifications or titles that symbolise valued competencies (and often authority), the form that confers entirely original properties on the cultural capital that it is presumed to guarantee.48 Cultural capital is not necessarily or as such positive; it is not a matter of ‘possession’. It refers to how ‘things’ function in relationships – not just the system producing cultural goods, but the instruments for appropriating such goods.49 It is in this way that actions and performances of violence, humiliation, and submission become and facilitate cultural capital. Interwoven with cultural capital are assumptions and connections – practices – that constitute the substance and limit of common sense for most people under its sway.50 This ‘common-sense’ feature of cultural practices leads Bourdieu to the notion of ‘symbolic violence’,51 the various modes of social/cultural domination. Symbolic violence is the unnoticed (partly unconscious)

44 45 46

47

48 49 50

51

‘Winning’ or ‘losing’ depending on perspective; cf. Lareau and Horvat 1999. Bourdieu and Wacquant 1992, 101. The polysemy in Bourdieu’s thinking – a deliberate relativising and interconnecting of concepts – is well summarised by Robbins 2000, 137–40. In embodied cultural capital, ‘external wealth [is] converted into an integral part of the person, into a habitus’ (Bourdieu 1986, 243). Bourdieu is critical of the chronic reduction of cultural capital to economic features, although he consistently insists that the economic base of cultural capital should not be ignored. Bourdieu 1986, 243. Cf. Bourdieu 1993b, 115. To borrow some Gramscian language, cf. Boggs 1976, 17, 39. Bourdieu’s ‘conductorless orchestration’ (Bourdieu 1990b, 59) is close to Gramsci’s exposition of hegemony. See esp. Bourdieu 2001.

28

Pieter J. J. Botha

domination that everyday social habits maintain over the subject. To formulate the point: representations of slavery in early Christian culture were a form of cultural capital used in the imposition of symbolic violence. Such institutionalised modes of ‘discipline and punishment’, as Michel Foucault noted,52 acquire a positive social value, despite the violence involved in these practices. Indeed, Bourdieu reminds us that such ‘soft’ violence has mostly been overlooked in social theories, and is subject to ‘misrecognition’ in everyday life. Misrecognition allows symbolic violence to hide itself within dominant discourses as these are spoken, and within other forms of violence as these are applied to bodies. Misrecognition is integral to the effects of symbolic violence, among which would be those forms of violence that, through their misrecognition, are applied by the subject to the subject. Much of symbolic violence is internal to the selfconsciousness of the individual, most often without telltale observable features. The various forms of symbolic violence operate upon the body and articulate a (misrecognised) relationship between the individual’s selfconsciousness and body. Symbolic violence is a way of conceptualising how the dominated takes up the point of view of the dominant both on the dominant and on themselves. This embodied violence articulates the nondiscursively available portion of social identity, which is reproduced simultaneously within the self-consciousness and also as a social practice, that is, as a shared habitus. Bourdieu argues that we make sense of human bodies – and human bodies make sense – through a ‘system of structured, structuring dispositions’, the habitus.53 Thus, habitus refers to a ‘system of acquired dispositions functioning on the practical level as categories of perception and assessment or as classificatory principles as well as being the organizing principles of action’.54 Habitus is embodied history, internalised as a second nature, its historicity ignored; it translates itself into knowledge borne in the body.55 Habitus does not exclude the notion that violent

52 53 54 55

See Foucault 1979. Bourdieu 1990b, 53, 140. Bourdieu 1990a, 12–13. Bourdieu 1990b, 56, 58. The body is ‘full of numb imperatives’ and one could ‘endlessly enumerate the values given body, made body, by the hidden persuasion of an implicit pedagogy which can instil a whole cosmology, through injunctions as insignificant as “sit up straight” or “don’t hold your knife in your left hand”, and inscribe the most fundamental principles of the arbitrary content of a culture in seemingly innocuous details of bearing or physical and verbal manners, so putting them beyond the reach of consciousness and explicit statement’; Bourdieu 1990b, 69.

Masters and Slaves in Early Christian Discourse

29

behaviour as part of a sociocultural environment is similarly internalised and embodied, affecting the construction and expression of culture. The conventional terms used to describe the ascription of correct attitude in a group or social institution (‘orthodoxy’ and ‘orthopraxy’) describe an adherence to correct speech and behaviour. In view of the Bourdieusian reflections above, it is necessary to not ignore or underplay embodiment. Orthodoxy and orthopraxy do not cover features of life such as complicity. ‘Correct’ speech and beliefs focus on ‘hard’ violence and do not deal with how hard violence follows on and is embedded in soft violence – or that the mere possibility of hard violence is the presupposition of soft violence. I would like to draw attention to a particular aspect of habitus, to embodied attitude, to posture (both physical and symbolic). The notion of a proper attitude requires a neologism: orthoposture.56 I use this term to describe the compliance of individuals to ‘willingly’ submit to and participate in relations and attitudinal conditions generated by hierarchy and authority. Orthoposture is an expression of the individual not maintaining a distance from the required attitudes, and so not engaging in any counter-reflection, avoiding the outcomes of parody, cynicism, and critical scrutiny. It is, to adapt an insight, ‘the imaginary quality of the individual’s identification with a subject position which gives it so much psychological and emotional force’.57 Orthoposture is perspectival, not just about the dominated, but dynamically also about the master submitting to the structure of violent behaviour as well. Domination is the common thread of violence and especially symbolic violence, and orthoposture is both the main outcome of domination and its main practice. The condition of orthoposture can describe the complicit agreement of the subject position in circumstances of domination. This attitude masks domination by pretending that the subject has some control. The pretence is supported by narratives (myths) that naturalise domination by a variety of daily life actions and rituals so that the outcome is a subject that is, in effect, self-dominated. Orthoposture ‘releases’ the subject from a critical perspective and creates a space that allows for spontaneous action. In interaction with 56

57

Following Bruce Caron (2010), who wrote a short piece on ‘orthoposture’ to describe ‘the willingness of the individual to spontaneously submit to a supplied attitudinal condition in a social circumstance’. Positioning theory could also be invoked (cf. the foundational study by Davies and Harré 1990), though in order to foreground embodiment, I consider orthoposture more useful. Weedon 1987, 31.

30

Pieter J. J. Botha

others dignity is involved, and to maintain spontaneity and honest solidarity the belief (conviction) of relatedness, psychic closeness, and mutual respect is required. Failure to participate with good heart will express rejection of those present or of the setting. Spontaneous involvement in the prescribed focus of attention confirms the reality of the world and its rules and the unreality of other potential worlds – and it is upon these confirmations that the stability of immediate definitions of the situation depends.58 For an ideology to be successful, ways to guarantee the orthopostural attitude of all its subjects (‘participants’) must be found. The (Roman) hierarchical, paternalistic, and exploitive ideology needs (cultural) capital for its success: a discursive and gestural set of values and actions, and at the heart of some Christian traditions were exactly such ‘sources’ of capital, ready to be implemented. In early Christianity, a distinct orthoposture acquired (powerful) cultural capital, related to a symbolic culture with a rhetoric of suffering/ asceticism as ways of disciplining the body. Combined with symbolic violence we have a habitus of ‘orthoposture’ exploiting this symbolic capital of suffering (linked to certain scriptural traditions, folklore, and entrenched religious convictions). Hierarchy, domination, and acceptance of violence continued even as it found different groundwork. All participants ‘played the game’, whether it was repenting, humbling themselves, denying their bodies, and especially ‘educating’ bodies59 so that misrecognition became standard. Finding ways to express and exploit Christian cultural capital, acceptance of ‘unjust’ suffering started to play a role and became part of Christian ways of expressing symbolic violence. Views and considerations of talk and rituals relating to suffering acquired ‘value’. We can see something of this in the role exercised by persecution within the rhetoric of the early Christian texts. Using Bourdieu’s notion of cultural capital, when one examines this literature, one can demonstrate how these authors variously present the withstanding of unjust suffering as necessary for the corroboration of the status of Christ-followers. One can also demonstrate the inverse potential of the rhetoric; the motif of suffering is used to

58 59

Cf. Goffman 1961, 36–7. Not only ‘was physical punishment of slaves considered acceptable; it was seen to have positive pedagogic value for the slave, and so a slaveholder could be seen to fall short in his duties as a householder if he failed to display adequate force in disciplining a slave’; Glancy 2010, 63.

Masters and Slaves in Early Christian Discourse

31

establish the status of outsiders and others – such as the Jews, but also unbelievers, women, and slaves. Lack of status (and here the capital is salvation, grace from God) almost requires suffering (unjustly), and so it becomes common sense that there should be bodily discipline.60 Suffering, especially considered suffering, such as ascetic practices or engaging in ‘appropriate’ physical abuse/punishments, could and did bring distinction. They were ways of becoming and being worthy. ‘The specifically symbolic logic of distinction additionally secures material and symbolic profits for the possessors of a large cultural capital.’61 Jesus’ lowly status and his corresponding corporal vulnerability to abuse offered a lens through which at least some ancient Christians viewed bodies and the abuse of bodies. This sets up a social system where slavery will not be seen for what it really is. Hence the author of 1 Peter can exhort slaves to bear brutality, especially when they are innocent, in imitation of Christ, who also suffered: When he was abused, he did not return abuse; when he suffered, he did not threaten; but he entrusted himself to the one who judges justly. He himself bore our sins in his body on the cross . . . by his bruises you have been healed. (1 Pet. 2:23–4)

This paradigmatic text encapsulates the ideological slant of cultural capital: enjoining the community to endure suffering in a Christlike manner; thus, enslaved Christians whose bodies absorbed unwarranted abuse served as a model for the entire Christian community to copy. But the modelling was very one-sided, and for the not-enslaved never real; ‘following’ is just labelling the structure of social relationships. Slaves were exhorted to tolerate abuse because Christian identity was associated with the very violation of their bodies.

suffering, violence, and attitudes: pauline beginnings The problem identified at the start of my discussion, namely that the ‘new religion of Christianity probably reinforced rather than challenged [slavery’s] existence’,62 can now be stated in Bourdeusian terms, as actors unwittingly 60

61 62

‘One of the fundamental effects of the orchestration of habitus is the production of a commonsense world endowed with the objectivity secured by consensus on the meaning (sens) of practices and the world, . . . the harmonization of agents’ experiences and the continuous reinforcement . . . from the expression . . . of . . . experiences’; Bourdieu 1977, 8. Bourdieu 1986, 245. Bradley and Cartledge 2011, frontispiece. Also: ‘Christianity did not make a difference to slavery in antiquity, and in the absence of any notion of universal freedom or of comparable rights and privileges as understood in the modern Western liberal tradition, slavery

32

Pieter J. J. Botha

reproducing the social stratification order. Paul addressed questions of slavery in a few rather ambiguous statements. Although he clearly states the unimportance of distinctions such as slave/free, male/female, and Jew/Greek, transcending such status differences was probably only a cultic ideal. Slavery was not a matter of theological indifference, but what differences slavery might make with regard to relationships, socio-economic conditions, and daily life experiences were, if not ignored, considered unimportant. The influence of Paul’s fostering the ‘normality’ of slavery comes to the fore in the writings associated with the Pauline circle, where the connections between hierarchy, values of suffering, and pursuing ‘submission’ grew in acceptance and influence. A habitus was being established in which the embodiment of slavery is seen in a particular way (with the effect that slaves were perceived, and interacted with, in much the same way any Roman slaveholder would).63 Barclay writes that ‘the Colossian household code is profoundly “Christianized” even if, to an outside observer, Christian wives, children and slaves performed no differently than their non-Christian counterparts, beyond taking the (conventional) duties with the utmost seriousness’.64 Precisely. Barclay has it spot on: this is what early Christians were telling themselves, reinterpreting their social relations, yet continuing to ignore the embodiment of living with and as slaves. No wonder they developed distinct associations with slavery metaphors. The Deutero-Pauline Epistles’ reinforcement of ancient norms for the behaviour of slaves was very useful to second-century Christian authors in their formulations of moral instruction on how to deal with slaves.65

a christian habitus The use of slavery metaphors among Christians reflects concepts that had wide currency in Graeco-Roman thinking, such as Stoic notions of

63

64 65

in the ancient Mediterranean world never became a problem’ (Bradley and Cartledge 2011, 2). ‘Slaves, obey your earthly masters (kyrioi) in everything, not only while being watched and in order to please them, but wholeheartedly, fearing the Lord (kyrios). Whatever your task, put yourselves into it, as done for the Lord and not for your masters, since you know that from the Lord you will receive the inheritance as your reward; you are slaves of the Lord Christ’; Col. 3:22–5; cf. Eph. 6:5–9; 1 Tim. 6:2; Titus 2:9–10. The continuity between traditional Roman and Christian ‘household codes’ has drawn considerable attention; see, e.g., MacDonald 2012; Osiek 2006. Barclay 2001, 47–8. E.g. Clem. Paed. 3.12.95–6; Cypr. Quir. 3.7.2.

Masters and Slaves in Early Christian Discourse

33

slavery to the passions. In Christian texts we find instances in which slavery – slavery to sin, to the law, to false gods, and to desire – characterises the unredeemed human condition. The figurative uses of slavery are not always consistent: metaphorical slavery is not intrinsically a condition to embrace or to repudiate. The ambivalence of the trope is apparent, for example, in Romans (6:15–23) where Paul contrasts the slave to sin (undesirable) to the slave to righteousness (very desirable). Christians living in the Roman Empire called Jesus of Nazareth not only the Messiah but lord or master, and referred to themselves as slaves. From this a range of uses for slavery metaphors was generated that became a distinct part of their religious discourse. There is a huge debate on what to make of these slavery metaphors in early Christianity,66 especially due to the considerable flexibility in their use. So, for example, baptism can be described in terms of manumission and of the slave market.67 Origen develops the image of a spiritual movement that begins in slavery to sin, moves to a stage of slavery to God, and ultimately reaches the status of sonship.68 Chrysostom, by contrast, emphasises the glory of being identified as a willing and obedient slave of God, which is construed as a position of true freedom.69 For this discussion I draw attention to a consideration of symbolic violence: slavery as metaphor for more than theological insight, as deflection of social tensions in cultural fields. The slavery metaphor facilitates the conversion of symbolic capital into cultural capital. We only need to ask how the slavery metaphor would have been perceived by real slaves in early Christianity. As Marianne Kartzow emphasises, for them to call God a slave owner would not (necessarily) be experienced as liberating; for a slave, ‘the slavery metaphor could not be separated from the embodied experience of punishment and penetration’.70 The realisation that such metaphors actually reinforced real slavery follows. Paul may have had in mind the possibilities of social mobility,71 but the persistent use of slavery imagery and metaphors in early Christian literature to strengthen negative stereotyping of slaves as infantile adults who require beatings to learn and

66

67 68 69 70 71

See, especially, Combes 1998; Harrill 2006b; Kartzow 2018a. Two very different, oppositional views: Martin 1990 and Byron 2003. On which see Combes 1998, 160. Cf. Garnsey 1996, 227–8, 234. On which see De Wet 2015, 22–3, 74. Kartzow 2018a, 145. Martin 1990.

34

Pieter J. J. Botha

who are so untrainable that they must be tortured to death72 problematises any such reading. The claims and implications of the (various forms of the) metaphor cannot be excised from their legitimation of the institution of slavery. Whatever theological claims were associated with the metaphor, the dayto-day, actual concrete experiences cannot be sanitised. This was a structuring, conceptual metaphor, fitting in with and elaborating hierarchy and authority. To return to the title κυριός/dominus (and extending some comments by Brown): ‘God was a dominus – a master. And a Roman dominus was a master of slaves.’ Master ‘did not imply distant majesty, muffled in fur and velvet. It conjured up life in the raw – life lived face to face in a Roman household, lived to the sound of the crack of the whip and punctuated by bursts of rage’.73 Make no mistake, to the Church Fathers and early Christian leaders, punishment was a key method for establishing the habits of obedience in slaves, and Christian slaveholders lived by the moral obligation that they had to use sufficient force to keep their slaves in line.74 De Wet writes: In the minds of the Church Fathers, the reward and punishment of slaves mirrored God’s own daily dealings with his people. In this sense, the early Church Fathers did not escape the culture of violence that defined the Roman world. Rather, they accepted and proliferated violence, but also transformed it and used it to fashion their theology and ethics.75

As a conceptual metaphor, slavery images animated thought and subtly reproduced the moral (and economic and status) vantage point of view of the master. The Church Fathers were never going to be everyone’s slave. They used a language that affirms that there are indeed masters and slaves, and that they are on the ‘right’ side. Though such values and practices are naturally associated with elites and proper slaveholders, keep in mind that in a cultural field there is no clear or sharp separation of upper-class values and perceptions from those of the lower classes. It is in any case questionable whether the humble freeborn population actually practised solidarity with the servile masses. Lower-status persons often share, if not exaggerate, the values and prejudices of their social superiors. These metaphors are shorthand for the Roman domus (household) ruled by a dominus (owner, slave master) 72 73 74 75

Harrill 2006b, 145–62; 2013, 511–13. Brown P. 2017, 45. Glancy 2010. De Wet 2016, 276.

Masters and Slaves in Early Christian Discourse

35

and they provide a generative symbolic space that animates their practices, norms, and attitudes. ‘Masters always pretend that they are not masters, insisting that they are only doing what is best for society as a whole.’76 The elements of Christian habitus shared by masters and slaves can be discerned in the writings of Lactantius. In his Institutiones Divinae he emphasises the equality of all Christians. As believers, they know one another as brothers and sisters and as fellow slaves.77 Nonetheless, strewn throughout his writings are casual references to the mundane brutalisation of slaves. Lactantius was unperturbed that Christian slaveholders wielded the whip against slaves they called brothers and sisters, and he makes frequent appeal to the customs of harsh punishment for slaves as a type of the judgement of God. God is angry and will judge men, just as the owner is angered by his slave and punishes him.78 In fact, surviving unbelievers will be in perpetual slavery in the millennial reign of Christ.79 Lactantius’ words are astoundingly insensitive to their plight.80 Archytas (the famous ancient philosopher and mathematician), for instance, neglected to strike one of his delinquent slaves. Lactantius casually points out that slaves require punishment; indeed, a worthless slave should rejoice that his master is angry. God gave anger for the sake of correcting faults.81 ‘As for the bad slave, he [the master] brings a range of punishments to bear on him: curses, lashings, nakedness, hunger, thirst, chains.’82 The reason for these extreme measures of punishment: ‘He will thus give the rest of the slaves an incentive not to misbehave, and the bad slave to behave well. Fear will restrain some of them, while others will be encouraged by the desire for honour.’83 Quite consistent and pervasive in his discourse about Christian life is precisely the distinction of free and slave; this distinction is in fact fundamental to his theology and ethics. There is righteous anger, which the wise and just master should exert in punishment so that faults are removed and licentiousness curbed: ‘For the world is, as it were, the house of God, and humanity, as it were, his slaves; and if his name is a mockery to them, what kind or amount of patience is it to give up his own honours, to see wicked

76 77 78 79 80 81 82 83

Cone 2010, 12. Lact. Inst. 7.24.4, 5.15.2–5. Lact. Inst. 5.18.14–16; Ir. 5.12; 17.16–19; 18.2, 7, 12. Lact. Mort. Pers. 15. Cf. Rupprecht 1974, 264. Lact. Ir. 18; of course, no free male requires or will benefit from punitive pedagogy. Lact. Ir. 5.12 (translations from Lactantius are my own). Lact. Ir. 5.12.

36

Pieter J. J. Botha

and unjust things done, and not to be indignant, which is fitting and natural to him who is displeased with sins.’84 Thus, as God is to the good men of the world, so should these masters be to their slaves. Notice the misdirection: the actual problem is not merely to be wise and just, but who determines what is wise and just.85 After all, Christians measure things ‘not by the body but by the spirit’, as in lowliness of mind, in humility, Christians are in equality (aequitas), according to Lactantius.86

making and marking of bodies In order to understand early Christianity and its interaction with slavery better, a mere reiteration of what early Christians said is not good enough. We need to connect and search for interconnections between actions and daily life (interpreting the rhetoric of the early Christian authors). Attitudes spring from and feed into behaviour, so interest should be not only in ‘what they say’, but to explore what these discourses do. Or, to put the point in Bourdieusian language, to explore the ways in which diverse groups of individuals, sometimes with diverging or even opposing agendas and speaking to different audiences, deployed discursive strategies to reproduce and enact versions of the same violent discourse. Violence occurs as a mode of structuring relations of power between victims, perpetrators, and witnesses, drawing from a common lexicon or vocabulary of symbolic action. In part, it is this lexicon that this contribution is searching for: the lexicon of violence that gave social meaning to much of early Christian discourse. It is this lexicon, among other factors, that determined how early Christians were dealing with equality and freedom. Slavery persisted and ‘thrived’ because violence found new labels and forms in Christian communities. Violent behaviour is more than an instrumental way of getting what one wants; it is also very much a discursive practice with symbols and rituals.87 Recognising this is 84 85

86 87

Lact. Ir. 17.11. Cf. Garnsey’s comment that ‘a good Emperor for Lactantius is above all one who leaves Christians undisturbed’; Garnsey 2003, 43. Lact. Ir. 5.15. Theoretical reflection on violence emphasises its performative quality; namely, the performative capacities of violence to generate culture, to structure bodies and bodily practices, and to refashion conceptions of self and other. Within the field of power (such as the social order encompassing masters and slaves) dominance is established by means of cultural and symbolic capital, the successful monopolisation ‘of the legitimate violence (specific authority) which is characteristic of the field in question’ (Bourdieu 1993a, 73). Bourdieu’s conceptual frames lend themselves to an exploration of the

Masters and Slaves in Early Christian Discourse

37

a step towards affirming and interpreting, and not just condemning, the presence of slavery in our history and appreciating its intrinsic meanings. Whether it is local, regional, Roman, or Mediterranean, a lexicon of symbolic action requires performance for an audience in order for it to have any social meaning. This is precisely what symbolic violence refers to.88 The power of violence lies not only in its capability to harm, but also in its capacity to communicate meaning, to instantiate power and legitimacy, especially over those not physically or directly targeted in the act itself.89 The efficacy of violence is its capacity to transform the subjectivities of those indexed in its performance; the violent imagery and metaphors in early Christian rituals and religious discourse cannot be separated from their morality and virtues.

punishing the body Part of Roman culture was flogging as a form of disciplining slaves. As Saller puts it, ‘The servile spirit was one that had to be goaded by the lash; the servile back was one marked with scars from past whippings. Precisely because verbera were fit for slaves and encouraged a servile mentality of grudging fear, such punishment was considered inappropriate and insulting for freeborn adult filiifamilias.’90 The whip was a cultural sign of servitude in Roman society, and therefore reserved for slaves. Remarkably, the whip (and flogging) acquires additional nuances in early Christianity. Cyprian, for example, alludes to the motif of the divine father who beats his children when they transgress his law, and he bases this extension of Roman custom on Psalm 88:31–4. In his catalogue of biblical testimonies regarding ‘that the faithful will be corrected and saved’,91 Cyprian notes: ‘If his children abandon my law and in my judgements they do not walk, if they violate my precepts and my

88 89 90 91

performative aspects/roles of violence, with emphasis on an ‘anthropology of experience’ over an ‘anthropology of identity’. While the former focuses on political, social, and economic transformation, the latter is more concerned with meanings, emotions, and bodily practices that can lead to a better knowledge of violence and its sources in various communities (Whitehead 2004). Meaning, emotion, and subjectivity can be studied against a range of performative bodily practices; see, e.g., Axel 2001; Butler 1993; Feldman 1991; Patraka 1999. There is an embodied and ritualised performance of values associated with violence in Christian tradition that needs to be acknowledged. Schröder and Schmidt 2001, 5–6; McDonald 2009. Whitehead 2004. Saller 1991, 165. Cypr. Quir. 3.57; cf. 2.1.

38

Pieter J. J. Botha

ordinances they do not observe, I will visit their wicked deeds with the rod (virga) and with the lash (flagellis) their iniquities; but my mercy will I not scatter away from them.’ Verses such as Jeremiah 2:30 and 5:3 often figure in Cyprian’s admonitions during and after the Decian persecution. In Epistula 11, Cyprian portrays the persecution as punishment for sins, particularly the sins of those who should know better, the presbyters, deacons, and confessors, as flogging. Cyprian actually says that ‘we are feeling these rods and lashes’ (like slaves), and quite deservedly.92 More than disciplining slaves, believers (free citizens) receive misfortune or disaster as dei flagella et verbera. Among early Christians we also find that the figure of the pater flagellans, the father who ‘whips’ the son he loves (see, for instance, Proverbs 3.12; Hebrews 12.6), becomes commonplace. The background is the conviction that it is as master that God, like the paterfamilias, chastises and punishes his sons. Lactantius links the scriptural admonitions to the patria potestas accorded the father in Roman law. Insofar as the father is master of the son, the son is slave to the father. To Lactantius, anger is appropriate to God and indeed necessary to divine providence. Lactantius asks whether those who believe God to be impassive (the Epicureans) expect a householder to tolerate insurrection and insults from his slaves. Obviously not. Failure to punish severely a rebellious slave, says Lactantius, is tantamount to cruelty rather than mercy. So too with God, and Lactantius likens the son to the slave when he defends God’s right to chastise the sinner plagis ac verberibus.93 From the fourth century we find that the appropriation of the biblical usage of scourging (flagellare) modifies the representation of discipline; a discourse in which flogging the son is the norm rather than the exception develops. The flagellum is no longer just the cultural sign of servitude, but becomes the cultural sign of being a son. Ambrose’s sermons on Psalm 118 serve to illustrate. Developing the theme that ‘the Lord corrects (castigat) the one he loves’, Ambrose comments: ‘Is not a son corrected more often than a slave? Is a father’s piety more unfair than a master’s severity? A father castigates his son more severely than a master his slave, but the father’s harsh lashings (dura flagella) are not deemed inappropriate, because the father wishes his son to be better than his slave.’94

92 93 94

Cypr. Ep. 11.2; cf. Bernard Donna 1964, 29–30. Lact. Inst. 4.3.14–17, 4.4.2–5; Ir. 17.11. Ambros. Exp. Ps. 118 14.15 (my translation); cf. Exp. Ps. 118 9.7–8, 10.2, 11.26, 14.15, 20.24.

Masters and Slaves in Early Christian Discourse

39

And when Ambrose turns to the issue of penitence, he also invokes scourging, not to obliterate Roman sensibilities but to invert them. ‘What has changed is not so much the degradation associated with whipping, as the value that is now attached to such degradation. It epitomizes humility before God.’95 In Augustine’s thought, this link is powerfully present. It is a father’s duty to apply the whip to a disobedient son. To deny the whip is cruel; to use it is an act of love and devotion.96 We find a discourse imbued with the penitential posture of the Psalms combined with Roman notions of paternal authority; and so sufferings and pain here and now are actually discipline from God, who, like a father, beats his children precisely because he loves them, and by doing so prepares his heirs for eternal blessedness. This clear shift in associations also influences other aspects of Augustine’s thinking. In Contra litteras Petiliani he responds to arguments that Christ teaches nonviolence. He points to the temple incident (John 2:13–15), where the Lord Christ drove out shameless merchants from the temple with a whip. Thus, according to Augustine, ‘we do find Christ a persecutor’ when He bodily expelled those ‘shameless merchants with whippings’.97 He notes, further on, that a boy is often beaten by the very man that is most dear to him.98 Similar sentiments are expressed in his drawn-out homiletic series on the Gospel of John, in that Jesus made a scourge of ropes and with it he scourged the undisciplined.99 In a way, early Christian thinkers had to reconcile classical connotations of bodily punishments with Christian ideals and principles. As faithful Christians, they had to strive for (or emulate) the ideal of martyrdom, supremely attested by Jesus Christ, but also by certain apostles and other Christian martyrs as well. Early Christian authors found various creative ways to reconcile or constructively affiliate the submission and humiliation traditionally associated with flogging (and torture) with Christian honour. Tertullian, for instance, finds a useful analogy in the contests and tribulations of (Graeco-Roman) religious festivals. A commonplace of classical exhortations to virtue is the Spartan youths who, cheered by family and friends, endure being whipped before the altar 95 96

97 98 99

De Bruyn 1999, 260–1. A well-researched aspect of Augustine’s thinking: Clark 1998; De Bruyn 1999; Shaw 1987. Aug. C. litt. Petil. 2.10.24 (my translation). Aug. C. litt. Petil. 2.95.216. Aug. In Evang. Iohan. 10.5.

40

Pieter J. J. Botha

of Artemis Orthia.100 So should Christians, argues Tertullian, bear their suffering.101 Note the reorientation in cultural capital: in Roman culture, wounds would be grounds for an action of iniuria,102 but for the Christian (as for the ‘pagan’ contestants when they are exhibited in their stadia and festivals) these wounds confer glory.103 Tertullian reconfigures traditional connotations of flogging into something valuable (capital) by means of preaching the flogging of Christ. In his Apologeticus 39.5–6, Tertullian reports of a monthly collection of funds for charitable causes (deposita pietatis), of which part is intended for the imprisoned and the enslaved, and those condemned to the mines, if it will be for the sake of the faith (ex causa dei sectae). Tertullian reflects a bias, one natural to Roman slaveholders. When he invokes interactions with slaves he typically writes of ‘we (the masters)’ and ‘our slaves’.104 And ‘we’ are not only slaveholders, but Christians. When ‘we’ reward slaves who are submissive and respectful, is there any risk of a different result ‘when we have a Master (dominus) more just in his judgements and powerful in his actions?’105 The threat of physical, violent punishment – as with any proper Roman master – remains constant. Slaves, after all, require taming like beasts; like beasts, they have been given to ‘us’ by God ‘for our purposes’. To escape his [the master’s/the Lord’s] severity or to invite his liberality one needs diligence in obeying, which is proportionate to the threats uttered by his severity or the promises made by his liberality. Yet, we ourselves exact obedience not only from people who are bound to us by the bonds of slavery or who, because of some other legal bond, are under obligation to us, but also from our flocks and even from wild animals. We understand that they have been provided and granted by the Lord for our purposes.106

So, when Tertullian retells the story of Vedius Pollio – a story embedded in Roman elite literature107 – who punished his slaves for simple 100

101 102 103 104

105 106 107

Cic. Tusc. 2.14.34, 5.27.77; Sen. Prov. 4.11–12. The cult of Artemis Orthia was practised at Sparta until well into the fourth century ce, and known throughout the Graeco-Roman world. On the rite, see Plut. Arist. 17.8, Inst. Lac. 239cd, Lyc. 18.1; Paus. 3.16.10–11; see Burkert 1985, 262; Lloyd-Jones 1983. Tert. Apol. 50.9; Mart. 4.8. Bagnall 1989; Lendon 2011. Tert. Scorp. 6.1–6. Tert. Ux. 2.8.3; Spect. 17.5; Paen. 4.4, 6.1; Nat. 1.7.15, 2.13.17; Apol. 27.5–7; Marc. 5.6.11; Pat. 10.5. Tert. Pat. 10.5 (my translation). Tert. Pat. 4.2–3; see Arbesmann, Daly, and Quain 1959, 198. Harris 2001, 325–6.

Masters and Slaves in Early Christian Discourse

41

household mistakes by feeding them to lampreys in a fishpond built for that very purpose, he, like any Roman moralist, expresses outrage over Vedius Pollio not because he put slaves to death for household slip-ups, but because of his vicious anger and loss of self-control.108 Tertullian could read and write in the presence of slaves, listen to them and preach to them, expect proper submission and good service from these fellow brothers and sisters (in Christ), urge punishment when deemed necessary, because they all shared the same habitus. Posturing correctly towards a master (and the Master) provided a sense of dignity, a cultural capital well invested.

conclusion ‘Christians seem to have had neither a special affinity for nor antipathy toward the manumission of slaves’,109 and it is this almost banal complacency about slavery that should, to my mind, be problematised. Hence the question as to why the early Christians failed to properly connect their preaching (and emphasis) on ἐλευθερία/libertas to actual social realities. Why the many discussions of equality, but with little (if any) effect on daily life, where gender roles, bodily disabilities, and humiliating subservience actually mattered? How could Christian teachers instruct slaves that virtue required submission to slaveholders? When the church developed imperial influence in the fourth century, why did bishops and monks not demand an end to the institution of slavery? I propose that in part the answers to these lie in the adoption of a rhetoric of suffering, punishments, and bodily discipline – which, like accent, elocution, clothing, gestures, and practices became part of a cultural capital that manifested in orthoposture. At times, this orthoposture was expressed in and experienced as vicious violence, but mostly it entailed symbolic violence. Part of this orthoposture involved a way of looking and a way of not seeing. Fitting in social settings (fields) imbued with violence, orthoposture required submissive acceptance by Christian slaves, and confident, punitive behaviour from masters while echoing or mirroring their theology of the Saviour Master. Given the diversity of ancient Christianity, the complexities of social history, and the time span involved, any explanation must be insufficient. 108 109

Tert. Pall. 5. Glancy 2002, 95–6.

42

Pieter J. J. Botha

The suggestion pursued here is that the cruel, violent, ‘traditional’ Roman culture persisted and found new expressions in early Christianity. What did it mean to inherit a structure of mastery, to come to defend it as one’s own, to maintain a position within it that precluded listening to the voices of the very ones the structure ravages? How might such habits be unmade, abandoned, refused? How does one refuse what makes one’s identity possible, organises and sustains one’s safety and power? Starting with Pliny’s Epistula 10.96, and reviewing some of the information we can gather about Christian slavery in the second and third centuries, this study notes how a traditional Roman habitus continued by adaptation. Building on the work of Bourdieu, I propose that the power of violence lies not in its capacity to harm, but rather in its capacity to communicate meaning and to institutionalise power, legitimacy, and history over those not physically or directly targeted in the act itself. The efficacy of violence is not measured in its casualties, but in its capacity to transform the subjectivities of those indexed in its performance. Tertullian and Augustine are remarkable case studies. Of course, this is not saying that their thinking actually equals early Christianity, or that either of them was a dominant influence, but it is possible to look at their discourse as representative of early Christian slaveholding. That is, from their writings we can glimpse a habitus where dealing with slaves is common, frequent, and considered self-evident. There is an expression of personal, first-hand experience of slaveholding embedded in daily life. The rhetoric of early Christianity’s images, figures, and references to slaves and slavery is deeply entwined with the traditional, hierarchical, and abusive Roman institution and ideology of slaveholding. A possible explanation for how such a ‘continuity’ stayed in place is that it is the symbolic violence of early Christian discourse that facilitated slavery and for all practical purposes excluded any changes in the ‘meaning(s)’ of Roman slaveholding. The ‘marking of bodies’ continued unabated. It happened by means of everyday, common daily practices. The history of early Christianity and slavery must make us recognise how fragile and contingent ‘ordinary’, unviolated existence really is.

2 Slavery and Religion in Late Antiquity Their Relation to Asceticism and Justice in Christianity and Judaism Ilaria L. E. Ramelli

Philosophical asceticism played an important role in the opposition to social injustice, oppression, and slavery across religious traditions in imperial and late antiquity in the Mediterranean world. A connection emerges from recent research between asceticism (or at least a strand thereof), the rejection of slavery as an institution, and the embrace of social justice in ancient philosophy, Jewish Hellenism, and especially Christianity in antiquity and late antiquity.1 When Christian ascetics chose poverty and low status in service to Christ, they were often also concerned for those who were victims of social injustice and oppression. Since at least some Christian – and Jewish, and ‘pagan’ – philosophical ascetics spoke explicitly of ‘justice’ in this connection, we can surmise that at least a part of them embraced asceticism also for the sake of justice.2 While a strand of early Christian and early Jewish asceticism was opposed to slaveholding, other strands, as will be clarified (for instance in the case of Paula’s monastery), accepted slavery as being compatible with the ascetic life, not to speak of the flourishing of slavery in early Christianity in general.3 I thank the editors for their invitation, and for the discussion from the audiences at the Universities of Cambridge, Stanford, Oxford, Chicago, LMU München, Erfurt MWK, Columbia, Harvard, UC Santa Barbara and UC Berkeley, and Holy Cross, where I presented versions of this paper and chapters of my monograph at invited lectures and seminars from 2015 to 2019. All translations of ancient texts are my own, unless otherwise indicated. 1 As thoroughly argued by Ramelli 2016b. 2 The connection between justice and asceticism I pointed out, at least in some strands of asceticism, is followed, for instance, by Graiver 2018. 3 As pointed out in Ramelli 2016b, 101–20. Even the metaphor of being a slave of God, per se very positive in ancient Judaism and Christianity, can be read negatively by scholars: Kartzow 2018a, 47–70 reads Mary, the mother of Jesus, as a powerless woman ‘enslaved’

43

44

Ilaria L. E. Ramelli

In a strain of Christian (and Hellenistic Jewish) asceticism, in voluntary poverty and the eschewal of slavery, a deeper level of asceticism was at work: the principle of renouncing the oppression of fellow humans and thereby committing injustice, either by claiming to own other people or by accumulating wealth that automatically meant, in these ascetics’ judgement, stealing necessities from the poor. Several ancient Christian ascetics support this tenet that wealth is tantamount to theft.4 Renunciation of oppression and injustice was, in the case of many ascetics, the common root of both the rejection of slavery and the rejection of social injustice by embracing voluntary poverty; hence the interrelation between the two. Other options were also available, such as almsgiving in a variety of degrees, instead of voluntary poverty.5 Almsgiving represented a milder form of reduction of socio-economic injustice. Thus, this chapter will consider late-antique and early Byzantine sources concerning individuals or couples who emancipated their slaves and gave up their possessions in favour of the poor when they embraced the ascetic life, as well as some monastic groups who liberated all the slaves who joined their communities, and the reaction of the ‘Church of the Empire’ to this destabilising practice. I will point out the link, at work in some strands of asceticism, between ‘spiritual fasting’ and ‘immaterial self-restraint’ – as Gregory of Nyssa called them – based mainly on practising justice, and the renunciation of slave ownership. Asceticism, in this case, turns out to be a matter not simply of self-restraint, but of justice. Spiritual asceticism is abstinence from oppressing others, owning other humans, and ‘robbing the poor with injustice’. We shall find this conception, which, as mentioned, equates excessive possessions with theft, in Origen, Gregory of Nyssa, and other ascetics in the Origenian line, such as Evagrius and Cassian. In this connection between excessive wealth and theft, possessions become illegitimate when one’s needs are exceeded, but owning humans for some ascetics was never admissible. Those who rejected slave ownership and possessions were going against societal norms and the institution of slavery itself, which was a pillar of ancient society and economy; hence also the harsh criticism of ‘pagan’ polemicists against Christian ascetics as destroyers of society. Eustathius of Sebaste’s practice and that of his followers, as well as

4 5

to God, who is made to reproduce by order of her master. On the slavery metaphor in antiquity and late antiquity, including that of enslavement to passions and sins in Stoics and Christians, see also, for instance, De Wet 2015; 2018a; Ramelli 2016b, 26–100 and passim; Bernhardt 2017. Thoroughly analysed in Ramelli 2016b, 190–211. On voluntary poverty and almsgiving in early Christianity, see Downs 2016.

Slavery and Religion in Late Antiquity

45

monastic resistance to the decisions of the Council of Gangra, although not by all monastic circles and by all ascetics, clearly questioned established religious and societal norms, concepts, and institutions. The model for asceticism, indeed, as clarified by Gregory of Nyssa, was not that of a society on earth, but of angelic life – and angels own neither slaves nor wealth. The same opposition will be highlighted within Judaism: regulations on slavery in Rabbinic Judaism (like, earlier, in the Hebrew Bible) will be contrasted with Jewish ascetic groups who rejected slavery and social injustice and embraced poverty. The chapter will conclude with a reflection on the meaning of asceticism and its relation to justice in an important thread of late-antique asceticism.

late-antique sources and the link between asceticism and the renunciation of possessions and oppression Between the fourth and fifth century, some married couples who embraced asceticism, for example Therasia and Paulinus of Nola, and the two Melaniae, emancipated all their slaves and renounced their wealth.6 Paulinus, born in Bordeaux from a very affluent family, in 394/95 ce, with his wife Therasia decided to pursue a life of prayer and chastity. They renounced their possessions, giving them to the poor, and also renounced slave ownership.7 Paulinus was ordained a priest, retired to Nola, became its bishop, and promoted the cult of St Felix. He publicly renounced his senatorial seat, a decision that attracted criticism from his peers, including Ausonius. According to the account of Palladius, who followed the Origenian– Evagrian line of philosophical asceticism, after the death of her husband, Melania the Elder, at the age of twenty-two, espoused asceticism, thereby renouncing her possessions and ownership of slaves. She gave up all her vast property and moved from Rome to Egypt, where she dwelt with the Origenian monks, known as the ‘Tall Brothers’, whom she followed to Palestine in 373. Qua ascetics, they kept no slaves, so Melania voluntarily served them as a slave. She was even ‘wearing the dress of a young slave’, and proudly declared to be the daughter and wife of prominent noblemen, but also ‘the slave of Christ’, like Paul.8 She founded a double monastery

6

7 8

On Melania the Elder (c. 341–410) and Younger (c. 385–439), see several essays in Chin and Schroeder 2017; on Melania the Younger, see Hunt 2016; on these and other ascetic couples and monasteries in late antiquity, Ramelli 2016b, 212–31. Ambros. Ep. 6.27. Pall. H. Laus. 46.

46

Ilaria L. E. Ramelli

in Jerusalem, which she directed and in which Rufinus lived. She supported him as a patroness. For thirty-seven years she continued to give hospitality to people who needed it and support churches and monasteries, strangers, prisoners, her family, her son, and her stewards. Finally, ‘she possessed not even a span of land’ and died ‘having got rid of her possessions’.9 Melania persuaded her own granddaughter, Melania the Younger, her husband, and her daughter-in-law to sell all their goods and become ascetics. She read virtually all of the literary production of Origen, Gregory of Nyssa, and Basil seven or eight times.10 She could absorb Origen’s ascetic ideals and Gregory’s and Basil’s views on slavery and social justice, which we shall briefly address later. Melania the Younger, aged twenty, after the death of her two children, and her husband Pinianus, decided to embrace monastic life with the divestment of their massive wealth and the liberation of their slaves: She freed 8,000 slaves who wished freedom. The others did not want, but preferred to be slaves to her brother: she allowed him to take them all for three pieces of money. But having sold her possessions in Spain, Aquitaine, the regions of Tarragona and the Gauls, she reserved for herself only those in Sicily, Campania, and Africa, and appropriated their income for the support of monasteries . . . . She ate every other day . . . and assigned to herself a part in the daily work of her own former slavewomen, whom also she made her fellow-ascetics [συνασκητρίδας]. She had with her also her mother Albina, who lived a similar ascetic life and distributed her riches for her part privately. Now these ladies are dwelling on their properties . . . with fifteen eunuchs and sixty virgins, both free and former slaves.11

This eulogistic account, intended to promote asceticism and link it with the embrace of poverty and the renunciation of slave ownership, concurs with that by Gerontius, who in his biography of Melania represents Melania and Pinianus as renouncing their whole wealth while embracing asceticism. They may have kept some resources,12 but they used them to help the poor and other ascetics. Note that a number of Melania’s slaves were sold to her brother at a symbolic price, almost given to him as a gift. This was a result of the slaves’ own choice, since they felt abandoned and did not know how to survive after their manumission.13 I emphasised this problem in my treatment of slavery and justice and the response, for instance, of John 9 10 11 12 13

Ibid. 54. Ibid. 55. Ibid. 61. Dunn 2014, 113–14. See Harper 2011, 192–5.

Slavery and Religion in Late Antiquity

47

Chrysostom to this issue. He advised people to buy slaves, teach them a trade, and manumit them as a real work of φιλανθρωπία. In this way, the slaves, after their manumission, did have a way to survive.14 The ascetic couples examined so far point to the connection between asceticism and the renunciation of slave ownership and wealth. For ancient Christian thinkers in the Origenian–Evagrian line (Melania’s own line of thinking as well), reducing one’s possessions is just in itself and to the benefit of the poor, since being wealthy was seen as an act of robbery, as will be examined in the next section. Hermits and extreme ascetics such as the stylites or dendrites, and other ascetics of Syria described by Theodoret, kept neither slaves nor possessions. Simeon the Stylite (†459), who lived for forty years atop a pillar, led a life of poverty, restraint, and renunciation of slave ownership. Theodoret portrays his asceticism as philosophical, calling it ‘philosophy’ and ‘love of labour’ (φιλοσοφία; φιλοπονία; these are also Origen’s own traits), and the monastery in which he spent ten years a ‘school of philosophy’.15 One of the first known ascetics, later regarded as the father of monasticism, Antony (†356), according to his biography by Athanasius,16 was inspired by Jesus’ exhortation to sell all possessions, by the apostles’ example of leaving everything to follow Jesus, and by the description in the Book of Acts of the first Christian community holding all possessions in common for the support of the needy. Antony renounced all his wealth when he embraced asceticism, and gave his possessions to the villagers, especially the poor. He liberated or sold the slaves who worked on his parents’ estate or gave the land he owned to them as well when he distributed it to the villagers. He kept neither slaves nor possessions. Some monastic communities may have owned slaves in common. But in some monasteries, such as that of Macrina, the sister of Basil and Gregory of Nyssa, according to Gregory’s account, former slaves were freed before entering and did not have to serve any human master. Both the ex-slaves and the free women served each other. In a work devoted to monasticism, Praevia institutio ascetica, attributed to Basil in the late fourth century, the author insists that the monk must embrace a ‘life without possessions’ (ἀκτήμων).17 The same principle is maintained in

14

15 16 17

Ioh. Chrys. Hom. 40 in I Cor. 6; see also the analysis in Ramelli 2016b, 4, 166–7, 242–3, 250–1. Thdt. H. rel. 26. Ath. V. Anton. 2–3. See PG 31.621.16.

48

Ilaria L. E. Ramelli

another work attributed to Basil: a monk must enter an ἀκτήμων βίος (a life without possessions).18 In these proto-monastic and monastic communities or individuals, a striking discrepancy can be observed regarding the retention or rejection of slave ownership and social hierarchy. For instance, Paula, the ascetic, wealthy friend of Jerome’s, in her house-monastery in Bethlehem, at least according to the representation offered by Jerome himself, maintained all social distinctions, even emphasising them spatially.19 The virgins were divided into three classes: noble, middle, and lowest. They worked and ate separately. Macrina, instead, lived together with her former slaves in her monastery, sharing her ascetic life with them. She and her siblings encouraged their household slaves to enter their proto-monastic community. Gregory himself, full of admiration, in his bio-hagiography of his sister,20 recounts how Macrina convinced her mother Emmelia to join her ascetic community and live together with their own former slaves, now made ‘of equal dignity’ (ὁμότιμοι) with their former masters. Emmelia made ‘all the slaves and servants she had, her sisters and equals’ (ὁμοτίμους), because she ‘gave up the services performed by her slaves’.21 In De vita Macrinae 11 (SC 178.176.1), the keyword ὁμότιμος in reference to Emmelia and her former slaves is repeated: Emmelia shared the same table and same kind of bed with them. Gregory emphasises that for Emmelia, Macrina, and their family, embracing ascetic life coincided with renouncing slave ownership. Thus, Gregory draws a close connection between asceticism – the ‘angelic life’ – and the rejection of slavery. Gregory also extols his brother Naucratius for adopting ‘a life without possessions’ (again, ἀκτήμονα βίον)22 when he embraced an ascetic lifestyle, and for ‘renunciation of possessions’ (ἀκτημοσύνη),23 ‘bringing with himself nothing else than himself’.24 Naucratius accepted that one of his former slaves followed him just to share ‘the same life choice with him’.25 Naucratius made himself a servant of the poor, old, and ill.26 Gregory praises 18 19 20 21 22 23 24 25 26

De ascetica disciplina sermo 1; see PG 31.648.42–3. Hier. Ep. 108. Gr. Nyss. V. Macr. 7; SC 178.164. Gr. Nyss. V. Macr. 7; SC 178.164. Gr. Nyss. V. Macr. 8; SC 178.166. Gr. Nyss. V. Macr. 11; SC 178.176. Gr. Nyss. V. Macr. 8; SC 178.166. Ibid. Gr. Nyss. V. Macr. 8; SC 178.168. On the issue of illness in Gregory of Nyssa, note the importance of the soul–body relation in De anima and Gregory’s interest in science and medicine (Ramelli 2018a; 2020); on Origen, see Ramelli 2012.

Slavery and Religion in Late Antiquity

49

Naucratius for refusing ecclesiastical positions and privileging asceticism and poverty. Gregory identifies asceticism as total renunciation of slave ownership, voluntary self-assimilation to slaves, and voluntary poverty.

gregory of nyssa’s theological arguments against social injustice and oppression: ‘spiritual asceticism’ and ‘immaterial self-restraint’ Deeply influenced by Origen’s philosophical theology, Gregory of Nyssa argued for the rejection of slavery and social injustice leading to the dire poverty of some, especially through his ‘theology of the image’. Gregory rejected the legitimacy of the institution of slavery; his theological arguments against slavery parallel his theological arguments against social injustice resulting in the dire poverty of many.27 This line partially continues the earlier philosophy of justice by Epiphanes. In a fragment concerning the justice of God, cited by Clement, Epiphanes adduces the distribution of light by God to everyone, without distinction between ‘rich or poor’ or ‘males and females’: ‘no one can have a greater degree, or take away a part from his fellow-human, to possess himself a double quantity of light’.28 So should the possessions of all, given by God, belong to all equally. Indeed, it is ‘the violation of the principles of community [κοινωνία] and equality [ἰσότης] that has given rise to the thief’.29 I think Epiphanes was inspired by Plato’s Respublica about the commonality of both property and wives (an aspect that already influenced Zeno’s Respublica30). The same principle was enforced by the Didache 4.8: ‘You will never turn away from the indigent, but will put your possessions in common with your brother, and will not state that these are your own [ἴδια].’

Gregory’s Arguments against Slavery and Social Injustice Gregory’s main argument against slavery is found in In S. Pascha (379) and his fourth homily on Ecclesiastes.31 Gregory presents Easter as the feast of liberation. He describes the manumissio in ecclesia as ‘good and humane’ because it frees slaves in a dignified way.32 Gregory assimilates 27 28 29 30 31 32

Ramelli 2016b, 172–211. Clem. Str. 3.6.1–2. Clem. Str. 3.7.4. On Epiphanes, see Ramelli 2016b, 124–9; Jurasz 2017. As demonstrated by Ramelli 2011b. See GNO 5.334–52. Gr. Nyss. In S. Pascha; GNO 9.1.250.15–20.

50

Ilaria L. E. Ramelli

slave owners to Pharaoh,33 who elsewhere in his works symbolises evil and the devil. Gregory uses arguments concerning spiritual slavery to conclude that legal slavery is illegitimate and against God, and therefore has to be abolished. He urges masters to release their slaves immediately, on that very day, and restore their confidence. He qualifies manumission as ‘good’ (ἀγαθόν),34 as opposed to slavery, which is evil. Gregory urges not only ascetics but all the heads of households in his audience to manumit all their slaves. Slaves must be freed on the day of Christ’s victory over death and his resurrection. Gregory’s plea extends to all those afflicted, such as the indigent or beggars (πτωχοί) and the sick. The association of slaves and the poor is no accident in Gregory, who used the same theological arguments to denounce as illegitimate both the institution of slavery and social injustice leading to poverty. Gregory regarded the misfortunes of the poor and the sick as totally undeserved and produced by other people’s greed. All masters should free their slaves. These are human beings, and thanks to Christ’s paschal work, all humans ‘inherit God’, the supreme Good.35 Humans inherit God fully in the end, but they are already heirs of God, because Christ’s resurrection contains and prefigures the final resurrection/restoration of all humanity.36 Slavery is illegitimate and against God because every human being is free, being created in God’s image. This demonstration from the ‘theology of the image’ is expounded especially in Gregory’s fourth homily on Ecclesiastes. From the beginning,37 Gregory criticises the person who presumes to be the ‘master of fellow humans’. This is ‘outlandish presumption’. No one can claim to possess another person, which would be ‘against God’, since all humans belong only to God. Those who claim to own other persons are stealing God’s possession and go against God and God’s decree,38 which made all humans free and endowed them with free will: ‘You condemn to slavery the human being, whose nature is free and self-determining, and so you make laws that are contrary to that of God.’ Anyone who dares enslave any human ‘fights against God’s own ordinance’.39 Slavery is against God’s will. 33 34 35 36 37 38 39

Ibid. 250.24. Ibid. 251.2. Ibid. 251.21. Ramelli 2013, 372–440. Gr. Nyss. Hom. 4 in Eccl.; GNO 5.334. Ibid. 335. Gr. Nyss. Hom. 4 in Eccl.; GNO 5.335.

Slavery and Religion in Late Antiquity

51

Gregory’s condemnation of the institution of slavery emerges throughout his works and his injunction to all masters to free all their slaves is rooted in his theology (primarily the ‘theology of the image’), on which his anthropology depends. In homily 4 on Ecclesiastes Gregory claims that a human being, God’s image, as a rational creature endowed with logos (the Logos being Christ), cannot be bought at any price. Every human being is free, qua image of God, who is free and powerful par excellence. Only an insane person, deceived by the devil, could presume to be the owner of God’s image: ‘Has the devil tricked you into believing that you are the master of God’s image? O what foolishness!’40 In this homily, Gregory echoes 1 Tim. 6:10, which states that the root of all sins is the love of money, thereby linking his condemnation of oppression through slavery to that of oppression through greed and economic injustice.41 Thus, Gregory grounded what I call his ‘theology of freedom’ in the theology of the image. Like Origen, Gregory stressed individual free will. Origen inspired Gregory’s ‘theology of freedom’42 and made free will a core feature of his anti-Valentinian polemic – the propulsory of his doctrine of apokatastasis. Indeed, in the main exponents of Christian philosophical asceticism, from Origen (and Bardaisan), to Gregory and Macrina, to Cassian and Evagrius, it is possible to detect a theorisation of apokatastasis.43 Besides the ‘theology of the image’, Gregory used another theological argument against both slavery and social injustice: that from ‘social analogy’, which argues for equality within the Trinity and within humanity. Freedom and equality within the Trinity44 are reflected in freedom and equality among all humans. In his treatise against the ‘neo-Arian’ Eunomius, Gregory attacks Eunomius on the grounds that the tenet that a human being has no master (ἀδέσποτον) reflects the fact that God has no master. By nature, only irrational creatures are slaves of rational ones.45 As the divine nature cannot be divided into slavery and mastery, so too is the case with human nature. The whole of creation is a slave, but of God

40 41 42

43 44

45

Ibid. 337. Ibid. 339. On Gregory’s theology of freedom, Ramelli 2016b, 179–87; on that of Origen, see Lekkas 2001; Fürst 2018. A monograph on Origen’s philosophical theology, in preparation, deals with Origen’s notion of free will, in relation to divine providence as well as to ethics. Ramelli 2013. Gregory supported it in his anti-subordinationism, drawn from Origen (Ramelli 2011c; forthcoming a). Gr. Nyss. Or.dom.; GNO 7.2.70–1.

52

Ilaria L. E. Ramelli

alone.46 The analogy of the unity of nature in both divinity and humanity substantiates the claim that slavery cannot divide a nature – divine or human – that is one. Gregory’s views on social justice inspired the later ‘social doctrine of the (Catholic) Church’.47 Rachel Moriarty assimilated some of Gregory of Nyssa’s statements on slavery in homily 4 on Ecclesiastes to Seneca’s Epistula 47, concluding that Gregory’s polemic against slavery is only apparent and drawn from rhetorical commonplaces, just as Seneca’s advocacy of a humane treatment of slaves did not entail a condemnation of slavery as an institution.48 But Gregory’s theological arguments against slavery, based on (1) the ‘theology of the image’ and (2) ‘social analogy’, are grounded in the Bible and its philosophical interpretation, mainly in light of Platonism. Gregory also adduces (3) an eschatological argument: the eventual elimination of slavery will result from Christ’s liberation of humanity from slavery by voluntarily taking slavery upon himself.49 Because slavery cannot abide in the end, since it is against God’s will and utterly impious and evil, it must be eradicated now, by all people, not only ascetics. The theological arguments that underpin Gregory of Nyssa’s condemnation of legal slavery are the same as those that support his condemnation of social injustice and usury. Gregory joined Origen’s, John Chrysostom’s, and Evagrius’ conviction that excessive wealth means depriving other people of what they need.50 Gregory’s attitudes towards poverty and slavery are grounded in the practice of Gregory and other members of his family, such as Naucratius and Macrina, who rejected both riches and slave ownership altogether. This choice was admired by Gregory, who, in the footsteps of Origen, Pierius, Pamphilus, Eusebius, and other Origenians, identified the philosophical life with a life of asceticism and voluntary poverty. He identifies the highest kind of life as the life ‘without possessions’.51 Portraying model ascetics in the bio-hagiography of his sister Macrina, Gregory paralleled his theological arguments and exhortations against slavery and social injustice with his depictions of Macrina, Naucratius, and other ascetics, who embraced poverty and voluntary service and rejected slavery, as a model for his readers to call for imitation. 46 47 48 49 50 51

Gr. Nyss. Eun. 1.1.526, 3.1.15, 3.3.54–5, 3.4.37–8, 3.5.12, 3.8.44, 50, 53, 54–8. See Widow 2017. Moriarty 1993. Gr. Nyss. Eun. 3.8. Ramelli 2016b, 199, 204–7. Gr. Nyss. Fat. 34.3.

Slavery and Religion in Late Antiquity

53

The ‘theology of the image’ is the main basis for Gregory’s condemnation not only of slavery, but also of social injustice. Gregory maintains that no one can be considered to be less worthy than others, since all humans are the image of Christ, who is the image of God.52 Christ has given his very countenance to all humans alike, because he has taken up all humanity. Gregory refers to Matthew 25:35–45, where Jesus identifies himself with the poor: ‘I was hungry and you gave me food, I was thirsty and you gave me drink . . . as you did it to one of the least of these brethren, you did it to me.’53

‘Spiritual Asceticism’ and ‘Immaterial Self-Restraint’ God has given the goods of this world to all humanity equally; therefore, those who possess more than they need are in fact depriving other people of what they need.54 Wealth is equated with theft; thus, ‘stay away from iniquitous gain, starve your idolatrous greed for riches: let nothing be stored up in your house that comes from violence and robbery’.55 Gregory is following Origen, who insisted that if one accumulates wealth, it is necessarily acquired through injustice: ‘One of these alternatives must necessarily be the case: either to gain a lot through injustice, or only a little, but with justice . . . abundant riches are tantamount to iniquity’.56 Ascetic practices, such as fasting and abstinence, are not appreciated by God if one is oppressing one’s brothers. Taking away necessities from the poor means ‘biting your brother with wickedness’ and ‘drinking their blood out of evilness’. Bodily fasting is useless ‘if the intellect is not purified. Self-restraint is useless, unless it includes all the other aspects of justice [δικαιοσύνην] as inseparable and consequent . . . Isaiah asks: To what end do you fast, while you strike the poor with your fists?’57 The bread to be asked for in the Lord’s Prayer, therefore, is the bread of justice, as detailed in homily 4 In orationem dominicam. This is a bread that comes from justice, from one’s own sweat and labour (see Genesis

52 53 54

55 56 57

Gr. Nyss. Benef.; GNO 9.1.93–108. Author’s italics. See also Bas. Hom. 6 (cf. 7, 8) and Gr. Naz. Or. 14.24–6. The principle that one’s possessions should not exceed one’s needs was supported by Epicureanism in antiquity; see Morel 2016. Gr. Nyss. Benef.; GNO 9.1.94. Or. Hom. 3 in Ps. 36.6; ed. Perrone 2015, 146. Gr. Nyss. Benef.; GNO 9.1.94–5.

54

Ilaria L. E. Ramelli

3:19), not from oppression. ‘Lord, give me bread, that is, let my bread come from just toils. For if God is Justice,58 he who procures food by greed does not gain his bread from God.’59 Again, a theological tenet founds Gregory’s social thought: If your abundance does not come from what belongs to others; if your income is not derived from tears; if no one goes hungry because of your satiety; if no one groans on account of your fullness . . . He who is attentive to justice derives his bread from God. But he who cultivates injustice is fed by the evil contriver of injustice . . . if you offer gifts acquired by injustice, your gift amounts to the yelp of a dog or the pay of a harlot . . . But if you have your bread from God, that is, from just toils, then you are permitted to offer to Him of the fruits of justice.

John Chrysostom, although he did not wholly oppose slaveholding (but agreed that life without slaves was ideal),60 likewise warned that alms deriving from the degrading of people are not welcome to God. He exhorts his audience ‘to abstain from greed first’, and then show mercy through alms. He agrees that too much wealth is tantamount to robbing others: Give so that others need not beg . . .. I cannot, you say, endure hunger. For that reason do you expose others to hunger? Do you know what a dreadful thing it is to beg, how dreadful to be perishing by hunger? Spare also your brethren! . . . to be hungry is neither a disgrace nor a crime; but to cast others into such a state brings not only disgrace, but extreme punishment [in the other world].61

According to Gregory, as well as to Origen, Chrysostom, and other ascetics, asceticism must pair acts of self-restraint with justice. Gregory invites his flock to share their food and houses with the poor: the fact that they are poor is a grave ‘injustice’ (ἀδικία).62 Again, he employs a theological argument: fear of God should eliminate social and economic inequalities, since it is ‘a just balancer who makes people equal’ (δίκαιος ἐπανισωτής). The poor and sick are salvation for the rich, ‘the salvation of your wife and your own children’.63 If the rich do not share their wealth with the poor, their own health and salvation, and those of their families, will be in danger. ‘Do not despise the poor . . . Consider who they are, and you will find out their value/dignity [ἀξίωμα]: they have put on the very

58 59 60 61 62 63

Origen’s point (Cels. 6.64), appropriated by Gregory. Gr. Nyss. Or.dom. 55.20. See the discussion in Ramelli 2016b, 152–71; De Wet 2015. Ioh. Chrys. Hom. 4 in II Cor. 13; PG 51.300. Gr. Nyss. Benef.; GNO 9.1.96–7. Ibid. 97–8.

Slavery and Religion in Late Antiquity

55

countenance of our Saviour. Ηe, who loves humanity, lent them his own face.’64 To the argument from the theology of the image – the poor have Christ’s countenance (Matthew 25:35–45) – Gregory adds another: God made creation for all humans equally. All should limit themselves to the possession of what they need: ‘For all goods belong to God . . . it is better and more just if siblings participate in their heritage in equal parts.’65 Gregory accuses those who refuse to give one third or fifth of their goods.66 He also alludes to the parable of Dives and Lazarus, on which he expands also in De anima, and warns the rich who give nothing to the poor that they may die soon, and the abyss awaits them.67 He condemns injustice due to covetousness (ἐπιθυμία τῆς πλεονεξίας),68 which is ‘madness incapable of controlling itself’, inspired by the devil.69 The usurer’s alms are unwelcome to God, since they come from his siblings’ tears, flesh, and blood.70 For Gregory of Nyssa, pursuing justice is a matter of asceticism: renouncing oppressing others by slavery or robbery (that is, wealth, which is robbery). Asceticism cannot do without the renunciation of any form of oppression. This connection was already made in the Christianised Pythagorean ascetic work, the Sententiae Sexti.71 Gregory reinforces this link: no Christian can practise asceticism, even in the moderate form of fasting or almsgiving, and own slaves or keep riches that impoverish other people. Basil mentions Gregory’s ‘οἰκέτης Asterios’, the carrier of a letter by Gregory.72 Klein identified Asterios as a ‘slave’.73 However, οἰκέτης does not necessarily mean slave.74 What is more, Gregory’s letter carried by Asterios is from 371, when Gregory was not yet a bishop and had not yet pronounced his stark condemnations of slavery. He was a married ecclesiastical reader and rhetorician; every married couple of that rank had slaves75 – apart from some ascetic couples who at some point renounced

64 65 66 67 68 69 70 71 72 73 74

75

Ibid. 98–9. Ibid. 101–13 [author’s italics]. Ibid. 104. Ibid. 106–8. Cf. Gr. Nyss. Orat. 284.6–9. Gr. Nyss. Orat. 286.1–9. Gr. Nyss. Usur. 9.201.25–9. Ramelli 2016a; 2016b, 1–25. Bas. Ep. 58.1. Klein 2000, 22 n34, 72 n89. The term οἰκέτης is not the same as δοῦλος (slave). The former can mean ‘servant’, a free person, or a freedman; see Lampe 939; LSJ 1202; see also Vierros, in this volume. On the household in late antiquity, see Sessa 2018, 84–124.

56

Ilaria L. E. Ramelli

slavery and riches, as seen at the beginning of this chapter. The mention of Gregory’s οἰκέτης during his married life is not necessarily in glaring contradiction of Gregory’s own later condemnation of slavery during his ascetic life and his bishopric.76 Gregory called for ‘spiritual fasting’ and ‘immaterial self-restraint’ based on practising justice, since ascetic practices such as fasting and almsgiving are not acceptable to God if one oppresses people, keeps slaves and excessive wealth, and does not practise justice. Gregory calls abstinence from oppressing others ‘spiritual asceticism’: There exists a spiritual fasting [νηστεία], an immaterial self-restraint [ἐγκράτεια]. Τhis is abstinence from sin, which pertains to the soul. Fast now from evil! Control yourself in your greed for what belongs to others! Give up dishonest gains! Starve your idolatrous greed for money to death! Let nothing be stored up in your house that comes from violence and robbery . . . you keep strict frugality by yourself to no avail, if you rob the poor with injustice.77

Ascetic practices are in vain, if one does not renounce oppressing others through slavery, usury, and robbery (even just by possessing more than one needs), which Gregory calls ‘injustice’ and ‘evil’. This position was inherited by the Origenian–Evagrian thinker, Cassian. Fasting is not only abstaining from material food but fasting spiritually.78 This means abstaining from greed, amassing wealth, owning slaves, and so on. Abba Theodore is reported to have deemed it best to possess nothing, not even books.79 Therefore, we can see in Gregory the concept of a deeper level of asceticism, related to justice, poverty, renunciation of slavery, and service to others, for which Gregory argues on the basis of theological arguments, and which he attributed to the ascetics in his own family: his sister Macrina, his mother Emmelia once she entered the house-monastery, and his brother Naucratius, an ascetic-hermit.

eustathius and gangra: movements for the liberation of slaves in ascetic contexts and reactions Eustathius of Sebaste was a philosophical ascetic who supervised a hospice for the poor in the 350s ce and was committed to social justice and the 76 77 78 79

Ramelli 2018a. Gr. Nyss. Benef.; GNO 9.1.94. Ioh. Cass. Inst. 5.21.1. Apophthegmata Patrum (alphabetic series), Theodore of Pherme 1; PG 65.188.

Slavery and Religion in Late Antiquity

57

liberation of slaves independently of their owners’ consent. Eustathius was the son of Eulalius, the bishop of Caesarea in Cappadocia, the same seat as Basil had.80 Basil testifies that he upheld Eustathius as a model since his young age.81 Eustathius, who organised almshouses, hospitals, and refuges,82 indeed inspired Macrina and Basil, who also built hospitals and kept an emphasis on asceticism and the rejection of slavery and social injustice, although Basil did not wholly oppose slaveholding as his brother Gregory of Nyssa did.83 Socrates relates Eustathius’ teaching and activity, full of ascetic traits, and attests to his commitment to the liberation of slaves independently of their owners’ consent: He forbade marriage and taught as a dogma that it was necessary to abstain from foods. This is why he separated many people who were married from their partners and persuaded those who were averse to churches to make communion in private houses. He also snatched slaves from their masters under the pretext of piety. He wore philosophical garb himself, and had his followers wear a strange outfit, and women cut their hair. He also taught to neglect the prescribed fasting days and rather fast on Sundays, and prohibited praying in the houses of married people. He prescribed to decline as a defilement the blessing and communion of a presbyter who had a wife, even if he had married her according to the law while he was still a layman.84

The Cappadocians, who knew Eustathius and took up some aspects of his thought, nevertheless did not embrace the prohibition of marriage for non-ascetics. The Synod of Gangra, in Paphlagonia, in the 340s, prohibited the liberation of slaves in monasteries against their owners’ consent. Canon 12 of the Synod forbade the use of the philosopher’s mantle for the sake of asceticism. Indeed, Eustathius is presented by Socrates as a philosopher who wore philosophical garb even while a bishop, like Heraclas of Alexandria according to Origen, who adduced the example of Heraclas as a Christian philosopher:85 Heraclas, the future bishop of Alexandria, ‘who now sits in the πρεσβυτέριον [namely, among the presbyters] of Alexandria’, still dresses as a philosopher and studies the ‘books of the Greeks’.86 There was a connection between Eustathius’ being a philosopher and his asceticism,

80 81 82 83 84 85 86

Socr. H.E. 2.43. Bas. Ep. 244. Epiph. Haer. 75.1; Soz. H.E. 3.14.36. See the discussion in Ramelli 2016b, 152–71. Socr. H.E. 2.43 (author’s italics); SC 493.226. In a letter reported by Eusebius; see H.E. 6.19.12–14. Ibid., as analysed by Ramelli 2011b; 2019.

58

Ilaria L. E. Ramelli

which had an impact on his rejection of slavery. Asceticism and philosophy also went together in Origen’s life according to Pamphilus and Eusebius, and in Philo’s life according to Eusebius, who modelled his portrait of Philo on that of Origen. The resemblance of these two portraits has been highlighted by Ramelli and then Rogers.87 Eustathius and his followers were condemned at Gangra. Eustathius submitted, but an Antiochene synod condemned him again for ‘perjury’.88 The official teaching of the third canon of Gangra was that slavery had to be retained as an institution. The canon condemned to excommunication those who exhorted slaves to stop serving their masters, while they should have done so with respect and goodwill.89 The fourth canon of the Council of Chalcedon (451) again forbade monasteries to offer refuge to runaway slaves without the permission of their masters. The reiteration of the prohibition suggests that ascetic groups had meanwhile continued to free slaves against their masters’ will.90 This confirms that many ascetics within the church rejected slavery (as did some Jewish and perhaps some ‘pagan’ philosophical ascetics, as we shall see91), although not all of them, and some ascetics were certainly present at many of the councils of the ‘Church of the Empire’, and in much of the hagiographic works ascetics are associated with the church (even if they sometimes keep their distance). Zeno in 484 decreed that slaves, provided that their masters agreed, should be allowed to participate in the monastic life. But if they left monasticism, they would have to return to the condition of slavery.92 This points again to a strong connection between ascetic life and the rejection of slavery.93 However, the need for the slave owner’s permission limited the effects of Zeno’s decree, whereas monks such as those condemned at Gangra and at Chalcedon acted independently of the consent of the slaves’ masters. Eustathius and his followers were undermining the institution of slavery.

87

88 89

90 91

92 93

Ramelli 2011a; 2016b, 295; Rogers 2017 follows Ramelli’s remarks (1, 8, 10). By presenting him as a ‘Hebrew’ and not a ‘Jew’, Eusebius shows respect for Philo and locates him in a middle position between Judaism and Christianity (albeit never asserting Philo’s conversion), since he connected Christianity with the older Hebrew race in H.E. 1.4.4–5. Soz. H.E. 3.14.36; 4.24.9. As the Deutero- and Pseudo-Pauline household codes commanded; see Ramelli 2016b, 108–13. On monasteries as places of asylum for runaway slaves, see Rotman 2009, 144. For the Jewish ascetics, see the following section of this chapter; for the ‘pagan’ ascetics, see Ramelli 2016b, 26–100. Cod. Iust. 1.3.37. As argued thoroughly in Ramelli 2016b.

Slavery and Religion in Late Antiquity

59

The connection between asceticism, monasticism, and the rejection of slavery, although traceable from earlier ages,94 becomes especially clear in the age of Justinian, when monasteries were allowed to receive fugitive slaves who intended to become monks, but the legal owner of these slaves could still reclaim their slaves within three years of their embracing the monastic life.95 Those who rejected slave ownership and possessions were going against societal norms and even the institution of slavery, a pillar of ancient society and economy. Thus, we also find the harsh criticism of ‘pagan’ polemicists against Christian ascetics as destroyers of society.96 Celsus criticised Jesus’ claims that the rich have no access to God and people should not be concerned with food and wealth.97 Eustathius’ and his followers’ practice, as well as monastic resistance to Gangra, clearly questioned established religious and societal norms, concepts, and institutions.

jewish asceticism and renunciation of oppression The same contrast can be highlighted in Judaism. Regulations on slavery as an accepted institution in Rabbinic Judaism98 contrast with (earlier) Jewish ascetic groups, Essenes and Therapeutae, who rejected slavery and social injustice de jure and de facto, and embraced poverty and an ascetic lifestyle. The Essenes had neither money nor possessions.99 This ideal of voluntary poverty, as well as of rejection of slavery, was later exalted by Gregory of Nyssa, as we have seen, and became a feature in Christian monasticism. The link between asceticism, poverty, and philosophy – taken over in the tradition of Origen, Pamphilus, Eusebius, Gregory of Nyssa, and Evagrius – is particularly clear in Philo’s description of the Therapeutae in De vita contemplativa. Philo also celebrated voluntary poverty in Quod omnis probus liber sit, on Diogenes and other Cynics who have acquired true freedom.100 Philo did not share the Essenes’ and Therapeutae’s ideals and did not reject slave ownership. But the Essenes and Therapeutae themselves, like 94

95 96 97 98 99

100

On the origins of monasticism (present at least in the time of Antony in Egypt), see Ramelli 2016b, 20–4 and passim, which underlines the continuity between ancient philosophical asceticism and Christian asceticism and monasticism; see also Davis 2018. Rotman 2009, 144–5. Ramelli 2016b, 219–20. Or. Cels. 7.18. Hezser 2005; Ramelli 2016b, 96–100. Philo Prob. 1.77: ἀχρήματοι καὶ ἀκτήμονες [‘those without money and possessions’]. Bernhardt 2017, on the Essenes alongside the Pharisees and the Hasmoneans. On Cynicism and social justice, see Ramelli 2016b, 45–9 and passim.

60

Ilaria L. E. Ramelli

Gregory of Nyssa, did, and motivated with justice their rejection of slavery and social iniquity. Jodi Magness endorsed the historicity of the Essenes’ and Therapeutae’s rejection of slavery and possessions, identifying their reasons with ritual purity.101 Philo and Josephus translated these reasons into more Hellenised, philosophical terms. The description of the Essenes resembles that of the Pythagoreans,102 due to the attribution of philosophical asceticism to both. Richard Finn also deems the Therapeutae closer to the Levites, while Philo depicted them in philosophical terms.103 The ascetic groups of the Essenes and Therapeutae are the only ones, in pre-Christian antiquity, credited with unequivocally refusing both to keep slaves and to recognise slavery as an institution, on the grounds of natural equality and kinship among human beings and of an evaluation of slavery as unjust. This representation of their practice and motivations presupposes a strong link between equality and justice. The Essenes, ‘Jewish sages’, entertained ‘love for God, for virtue, and for humans’,104 and avoided what can arouse cupidity; shared their possessions, meals, and homes; and renounced ownership.105 What inspired their ascetic practices was ‘freedom, which escapes every slavery’.106 This is why they rejected slavery theoretically and practically, at least according to the sources, mainly Philo and Josephus. The Essenes’ ascetic lifestyle is described by Josephus in the following terms: ‘holiness/purity’, rejection of pleasure as a vice, and embracing ‘temperance and control of passions’ and sexual renunciation.107 All ‘despise riches’ and practise ‘communality of goods’, so among them ‘one will nowhere see either abject poverty or inordinate wealth’108 – as in the Jesus community in Jerusalem described in Acts 2:42–7, which became a model for Christian socio-economic egalitarianism.109 The community of goods was represented as typical of Pythagorean communities.110 The episode of 101 102 103 104 105 106 107 108 109

110

In a conversation in summer 2013. According to Taylor 2004. Finn 2015. Philo Prob. 83. Philo Prob. 78, 86, 87. Philo Prob. 88. Joseph. B.J. 2.120–61; quotation from 120. Joseph. B.J. 122. See Ramelli 2016b, 26–100. Marguerat 2013, 151, argues that the author of Acts wanted to indicate the realisation of Deuteronomy 15:4: ‘there will be no needy among you’, but also to show that the first church represented the golden age of Christianity. Iambl. V. Pyth. 30.

Slavery and Religion in Late Antiquity

61

Ananias and Sapphira, who died as a result of having kept some possessions for themselves instead of giving everything to Peter for the community (Acts 5:1–11), must be inscribed in this perspective and was debated in antiquity between critics of the severity of the punishment, such as Macarius of Magnesia’s Hellene, and Christian exegetes, from Origen to late ancient Christian authors.111 The link between the excessive wealth of some and the poverty of many was highlighted by early Christian thinkers such as Origen, Gregory of Nyssa, Evagrius, and Chrysostom, as pointed out above. When new members entered the Essene community, ‘all the resources of the community are put at their disposal, as though they were their own’; they change their garments and shoes only when worn out and ‘there is no buying or selling among them, but each gives what he has to any in need, and receives from him in exchange what is useful to himself. They are freely permitted to take anything from any of their siblings without returning it.’112 The Essenes ‘help those deserving, when in need, and supply food to the destitute’; their vow consisted in piety towards the deity, justice towards humans, wronging nobody, hating injustice, keeping faith with all people, loving truth, and abstaining from stealing, unholy gain, and robbery.113 Note the insistence on justice and avoiding oppression of humans. Therefore, they kept no slaves and rejected the institution of slavery. According to Philo, it is said that they denounced slave owners not only because of their injustice in violating the law of equality, but also due to their impiety in infringing the statute of Nature, who, like a mother, bore and reared all humans alike, and created them genuine siblings, not simply in name, but in reality.114

The Essenes refused to keep slaves: as the same passage by Philo confirms, ‘It is impossible to find even just one slave among them, but all of them are free and serve each other’, while most Stoics did not refrain from owning some or many slaves. This was also the Therapeutae’s lifestyle according to the sources. Josephus also reports that the Essenes kept no slaves because slavery is tantamount to injustice.115 Philo and Josephus therefore imply that the 111 112 113 114 115

See Ramelli 2016b, 6–7, 119–21; Van der Bergh 2017; Montero 2017. Joseph. B.J. 124, 127. Joseph. B.J. 133–5, 139–42. Philo Prob. 79. AI 18.21: ἀδικία.

62

Ilaria L. E. Ramelli

Essenes did not simply reject slave ownership as part of their vow of renouncing ownership of property (ἀκτημοσύνη) and self-sufficiency (αὐτάρκεια), but because they realised the intrinsic injustice of that institution. This hypothesis is supported by Philo: the Essenes qualified the ‘injustice’ and ‘impiety’ that brought about the institution of slavery in terms of ‘arrogance and avarice’.116 The natural kinship and equality of all humans was blurred by ‘the triumph of the vicious arrogance and avarice’ of some who began to oppress others. The same position, that slavery and social injustice must be rejected because they imply oppression, will be held by Christian ascetics such as Gregory of Nyssa, but the latter thought that not only ascetics but all people should reject slavery, since this is against God, and all should limit social injustice since God has distributed goods equally to all. The rejection of slavery is prominent for the ascetic Therapeutae, morally117 and literally.118 Philo describes them as ‘no longer slaves of anyone’, having renounced their possessions, slaves, and relatives.119 In De vita contemplativa 70–2 Philo is not applying the Stoic metaphor of moral slavery,120 but testifies to the Therapeutae’s radical rejection of the institution of slavery and relates this to their asceticism (in particular, that junior Therapeutae served the elders): They receive service, but not by slaves, because they deem the possession of servants altogether against nature. For nature has generated all humans free; it is rather the acts of injustice and arrogance of some people who pursue inequality, the principle of all evils, that, accumulating upon one another, conferred to the stronger power over the weaker. Now, in this holy community, as I said, nobody is a slave, but it is free people who serve other people, performing the necessary services not by force, nor waiting for orders, but anticipating the requests with zeal and willingness, voluntarily. These services are not performed by any free person, no matter which, but it is rather the young of the group . . . who do so with every solicitude . . . as legitimate children, diligently and happily serve fathers and mothers, deeming them their own common parents, closer to themselves than their biological parents. For nothing is closer and more familiar to the wise than excellence in virtue. And while they perform these services, they wear no belt and let their short frocks hang down free, to avoid bringing even just a shade of slavish appearance.

The Therapeutae’s reasons for their rejection of slavery are similar to those of the Essenes, a contemporary Jewish ascetic group. The tenet that

116 117 118 119 120

Philo Prob. 79. Philo, Cont. 18–20. Ibid. 70–2. Cont. 18–20. On which see Ramelli 2016b, Ch. 1.

Slavery and Religion in Late Antiquity

63

by nature all humans are free fits well with a Stoic framework. However, the Therapeutae did not simply reject slavery in principle, but refused to own slaves. And they did so not merely because they renounced all possessions, among which slaves were counted, but because they deemed the institution of slavery against nature and an example of inequality, a fruit of injustice and arrogance, resulting in the oppression of others. Also, the dire poverty of some and the exaggerated wealth of others are instances of inequality among human beings, which Philo declared the principle of all evils. Arguments against slavery and against social inequalities among Jewish ascetics are similar to the later arguments of Gregory of Nyssa, who knew Philo’s writings.121 In Judaism, as in Christianity, there was a tradition of organised charity, even if those practising it did not adhere to the Essenes’ and Therapeutae’s line. Emperor Julian, wishing to indicate an example for virtuous behaviour, stated that ‘no Jew ever has to beg’.122 This points to organised charity in Jewish communities in the fourth century, although it seems that in Julian’s day there were in fact some Jewish beggars, as known from sources such as the Talmud. Philo also insists on the rejection of injustice by the Therapeutae in De decalogo 2. Cities, from which the Therapeutae flee, are full of all sorts of injustice (ἀδικίαι). The Therapeutae, like the Essenes, counted slavery among these. Some Christians, such as Eusebius, regarded the Therapeutae described by Philo as Christians,123 thus idealising ascetics who refused slavery and embraced voluntary poverty and (some) gender equality. Indeed, the few condemnations of slavery in ancient Judaism and Christianity arose from ascetics.

conclusions: slavery and religion, ancient philosophical asceticism, and justice In antiquity and late antiquity, asceticism was not hatred of the body, not even in the Platonic tradition, as recent scholarship is discovering and arguing.124 Asceticism was also not simply about the refinement of the body. In both ‘pagan’ and Christian Platonism, the ascent through the hierarchy of bodies reflects a purification and progress of the soul. Origen

121 122 123 124

Ramelli 2008; 2018b. Julian Ep. 22.430c–d. Ramelli 2011a. See, for instance, Ramelli 2021 and Griffin and Ramelli forthcoming.

64

Ilaria L. E. Ramelli

spelt out this soul–body correlation in the best way: each soul must be in a corporeal state that is appropriate to its rank or order, referring to its moral development.125 For several ascetics, in particular for the philosophical ascetics of Hellenistic Judaism and especially late-antique Christianity, the progress of the soul involved justice, solidarity with other people, and renouncing the oppression of fellow human beings through slave ownership and the impoverishment of others as a result of one’s own excessive wealth.126 Ancient and late-antique philosophical asceticism involved not only dietary or sexual restrictions, or the mortification of the body, but also voluntary poverty, voluntary service to others, and refraining from oppressing others, as part and parcel of piety towards the divine. This is not common to all strands of asceticism, but is stressed in the Pythagorean tradition, Christianised through the Sententiae Sexti, and in Hellenistic Jewish and patristic philosophy, especially ‘Origenian’ philosophical asceticism.127 The ascetic concern for justice is paramount both in the harmony of the single person’s faculties, including the control of passions (central for instance in Evagrius128), and in the harmony within humanity as a whole, with an equal distribution of the goods and without one oppressing another. It is correct that ‘asceticism emerges as a means of legitimating authority, rather than simply a set of religious practices’.129 This may have been the case, for instance, with bishops or other male authorities claiming control over female ascetics. But asceticism could also be subversive vis-à-vis authority, both ecclesiastical and social. Thus, as seen, the ‘official’ church had to intervene repeatedly to curb some ascetics’ revolutionary practice of freeing slaves against their masters’ will when they entered the monastic life. Peter Brown concentrated on one facet of asceticism, sexual restraint in the first five centuries of Christianity,130 but there were many kinds of selfcontrol. Virginity itself was conceived broadly by ascetics such as Origen, Methodius, and Gregory of Nyssa, to the point of embracing the whole of the self-discipline that aimed at the eradication of passions (ἀπάθεια). In late-antique Christianity, monasticism, anchoritic and cenobitic, was the 125 126 127

128 129 130

Ramelli forthcoming b. The link between asceticism and justice is now also underlined by Westergren 2018. See Ramelli 2016b; on theological perspectives on slavery see also Priesching and Grieser 2016. Sorabji 2000; Ramelli 2017. Krawiec 2008, 5. Brown 2008 [1988].

Slavery and Religion in Late Antiquity

65

privileged environment for the application of asceticism. William Harmless rightly insists not only on the renunciation of marriage, family, and any sexual activity as marks of asceticism, but also on the renunciation of wealth, property, and career, performing manual labour, as well as the deprivation of food and sleep.131 Cenobitic asceticism was already practised by Essenes and Therapeutae, and such ascetic circles, especially what I have called ‘philosophical asceticism’, both Jewish and Christian, provide the most remarkable instances of rejection of slavery, social injustice, and oppression, and the embracement of a principle of justice, at least on a theoretical plane. These thinkers also read this ideal in practical instantiations, such as Eusebius in the Therapeutae and Gregory of Nyssa in the ascetics of his family, but also the sources on ascetic couples who renounced slavery and embraced poverty and on Eustathius and his followers. Sometimes, they argued for the practical application of these ideals, such as Gregory of Nyssa in his homilies and in the general arguments he offers against both slavery and social injustice. Philosophical asceticism and its conceptions of slavery and social injustice did not obviously represent a major trend in late-antique Christianity, but it certainly contributed to the complexity and diversity of slavery, and theories of slavery, in late antiquity.

131

Harmless 2008.

3 (Il)Legal Freedom: Christ as Liberator from Satanic Debt Bondage in Greek Homilies and Hymns of Late Antiquity Arkadiy Avdokhin

At some point in the fifth century ce, a series of honorific statues were erected in Stratonikeia, a major city in Caria, Asia Minor. The statues were set up to commemorate the benefactions that one Maximos did for the local community, primarily for the sake of the dispossessed (ἀκτέανοι). While the majority of the inscribed praises that accompanied the statues heaped conventionally generic praise on Maximos for his euergetism, one of the inscriptions gave a rather precise reason for the high esteem that the benefactor enjoyed in the city. When the poor (τῶν πενήτων) of Stratonikeia were hard-pressed to pay the tax of chrysargyron, Maximos stepped in three times and paid the tax on their behalf from his own resources.1 The social dynamics reflected in these inscriptions seem emblematic of late antique economic and power relationships between the destitute and their benefactors, as well as the Christian ideology of patronage, earthly and divine, that I discuss below.2 The chrysargyron was a notoriously harsh tax, and almost inevitably engendered massive fiscal arrears. As The results of the project ‘Models of Representation of the Past in the Middle Ages and Early Modern Period’ carried out within the framework of the Basic Research Program at the National Research University Higher School of Economics (HSE) in 2019 are presented in this work. 1 The inscriptions and commentary are accessible in I.Stratonikeia 1204 (= SGO 2/6/15 = LSA 657), I.Stratonikeia 1387 (= LSA 1200), I.Stratonikeia 1521 (= LSA 1201), I. Stratonikeia 1530 (= LSA 1202). For a recent discussion and further bibliography, see Ward-Perkins 2015. 2 For the relationship between the (super) rich and the poor in late antiquity, most centrally in Christian contexts and discourses, see e.g. Allen, Neil, and Mayer 2009; López 2013; as well as the magisterial, if controversial, Brown 2012.

66

(Il)Legal Freedom: Christ as Liberator

67

a result, large numbers of people were suppressed by hopeless debts, and could even be brought to the state of debt bondage. The predicament could be resolved, almost exclusively, only through the intercession of powerful patrons like Maximos. Men and women were therefore effectively ‘bought out’ from the incessant poverty and debt to the state, as well as to lenders, to whom they were forced to apply in an effort to remedy their failed budgets.3 As I will discuss in what follows, the complex financial, social, and symbolic interactions between financially struggling individuals and their benefactors inspired metaphors, and wider theological thinking, about God and the fallen humanity, that were widespread in late antique liturgically set texts in Greek (homilies and hymns). The scope of modes of this struggling could be wide, and its specific character could vary, as the pertinent recapitulation on recent academic discussions of ‘poverty’ in antiquity has shown.4 In the discussion below, I will mostly speak of the so-called ‘conjunctural’ (episodic/epidemic) poverty incurred by people who would struggle for sustenance under dire circumstances, including tax imposition and episodes when they were pursued by debtors. The poor, in these terms, regularly found themselves in danger of, or already immersed in, incessant indebtedness that led many of them to debt bondage, and would be occasionally rescued from this bleak state by private patrons like Maximos or emperors annulling tax arrears and debt records. Likewise, Christ as the saviour of humanity from the primordial debt of sin was imagined, and portrayed, through socially grounded metaphors and imagery. By selectively exploring late antique homiletic and hymnic writings in Greek, I will highlight the role of metaphors of sin as debt leading to enslavement, and of Christ as liberator from debt enslavement to Satan.5 I will also discuss the social underpinnings of these religious mindsets. I will suggest that Christian authors of homilies and hymns, and their audiences, imagined their liberty from sin in terms of ideological matrices 3

4

5

For the tax (collatio lustralis), see Jones 1964, 431–2, 858–9; Bagnall 1993a, 153–4 (with further bibliography). For late antique attitudes to it, see Amato 2014, 19–30, 295, 321. For the periodic annulment of the tax, see p. 73. See the stimulating overview of ancient terminology of poverty, modern bibliography, and a sketch of potential new takes on the subject in Chapter 1 of Allen, Neil, and Mayer 2009 (alongside case studies in the subsequent chapters). For an earlier collection of studies on various aspects of ancient poverty that follow a range of methodologies, see Atkins and Osborne 2006. For a stimulating and methodologically advanced study of metaphors of slavery in early Christian writings, see Kartzow 2018a.

68

Arkadiy Avdokhin

that moved between the two frames of thinking. One was a radical vision of abolishing the ostensibly legal but essentially unjust debt slavery to Satan, alongside the entire paradigm of orderly legalistic proceedings that underpinned it. The other relied on conjuring up a whole new universe that functioned through the celestial bureaucracy of salvific bondage to Christ.

metaphors of debt and salvation: late antique christian contexts Texts that Christians wrote in late antiquity, both those intended for a more literary and solitary use and those that would be performed in the cultic context of the Christian service, were permeated by language and imagery drawing on concepts, and practices, of debt as sin, repayment and remission, debt bondage and slavery, manumission and freedom. For one of the most popular late antique homilists John Chrysostom, for example, as Chris de Wet has convincingly shown, these and connected themes were central.6 For John, thinking in terms of slavery and bondage to either the devil and sin or God was one of the fundamental premises of his preaching. For late antique pious audiences, the modes of imagining what it meant to be a virtuous Christian or, on the contrary, an outcast sinner relied on a profound understanding – not only abstract but socially and bodily embedded – of the socio-economic realities of master–slave relationships, indebtedness, debt bondage, and salvation from these states.7 As recent scholarship has increasingly suggested, various texts circulating in Christian communities in late antiquity, including the widely accessible genre of the homily, mediated these and similar frames of thinking and worked towards instilling them into broad Christian audiences. Oriented towards oral performance and wide accessibility, homilies would be a natural part of (para)liturgical performances of the Christian piety and devotion that shaped the audiences’ identities.8 Hymnic texts, 6 7

8

De Wet 2015; 2018a. For slavery in early Christianity (including economic aspects), see e.g. Harrill 1995; Glancy 2002. For a wider discussion of slavery in late antiquity, see Harper 2011. For the ritual setting of late antique homiletics and its engagement with the audience, see Cunningham and Allen 1998; Dupont, Boodts, Partiens, and Leemans 2018 (Latin West); Maxwell 2018 (an up-to-date overview with further bibliography). Recent studies of individual homiletic authors in the Christian East include Constas 2003 (Proklos of Constantinople); Forness 2018 (Jacob of Serugh); Leemans, Roskam, and Segers 2019 (Severian of Gabala and John Chrysostom). For Chrysostom’s output, the literature is immense; see the recent critical overview in De Wet and Mayer 2019. For the term ‘(para)

(Il)Legal Freedom: Christ as Liberator

69

with their special place in Christian celebrations and services, as well as the impact of the poetic form on the believer, became natural conduits of various forms of devotional sentiments, including those that focus on images of sin as debt and Christ as liberator from it.9 In what follows, I will continue in the path of analysis of late antique homiletics and hymnody in Greek pioneered in earlier studies that unpack the texts’ ability to mediate compellingly complex metaphors and messages to broad audiences. I will look at homilies and hymns that drew on the vocabulary and thinking about such key Christian topics as sin and redemption in terms of gaining freedom from slavery and/or debt bondage while entering new, allegedly different, enslavement relationships with the new master – God. Before delving into the homiletic and hymnic output of late antique Christian writers, however, I will briefly review the New Testament foundations of their symbolism of debt and remission as well as the social context of debt slavery in the later Empire.

the new testament’s ‘economic theology’ of debt and sin The vocabulary of indebtedness and episodes that thematize it loom prominently across different books of the New Testament. As has been well acknowledged in scholarship, the tropes of debt, repaying and remitting it (or failure to do so) are crucial in constructing what may be called the economic theology of indebtedness. Within it, debt and repayment are concepts used to encode sin, remittance, and salvation by Christ.10 The Gospel of Matthew is particularly rich in images and episodes built around economic transactions and realities (for instance, hidden treasure in Matt. 13:44; the precious pearl in Matt. 13:45–6), including loans and their repayment – as for instance the parable of the unforgiving servant (Matt. 18:23–35), or the parable of the workers in the vineyard (Matt. 20:1–16).11 Importantly, the Lord’s prayer, both in Matthew’s rendering and in other Gospels, revolves around the idiom and notions of debt as sin liturgical’ as referring to various enactments performed during and in broad connection with liturgical services in churches, see e.g. De Bruyn 2015. 9 For mediation of ideas and devotional sensibilities through hymnody, see Krueger 2014. 10 Important recent publications (where further scholarship is discussed) on this topic, which attract much scholarly attention, include Martin 1990 and Byron 2003. 11 For Matthew’s use of the trope of debt as sin, and the intricate theological framework he builds on it, see especially Eubank 2013 (with further bibliography).

70

Arkadiy Avdokhin

(ὀφείλημα and ἁμάρτημα) and the request to God for remittance.12 In the New Testament epistles, this symbolism also features significantly.13 These emphases of New Testament authors tap into earlier traditions and ideologies of sin as ‘debt’ to God that is to be re-paid, while good deeds are ‘treasures’ accumulated in heaven in expectation of the final reckoning in apocalyptic times. Alongside the previously paved paths of the religious conceptualization of sin as debt in Old Testament and broader Jewish traditions, New Testament writers, it has been argued, were heavily invested in economic and social imagery (while spiritually reconceptualizing it) in an act of implicit, or even sonorously explicit, social criticism of the contemporary practices. The economically and legally specific petition phraseology of the Lord’s prayer, in its reliance on the vocabulary of forgiveness (ἀφίεμι) of sins/debts (ὀφείλημα), evokes the idiom of royal edicts of remittance of tax arrears and other debts that had been a significant aspect of the imperial propaganda of the sovereign’s benevolence (φιλανθρωπία) since at least the Hellenistic era.14 One of the key scriptural passages that thematizes humanity’s debt and deliverance from it is Colossians 2:14. This passage expressly connects indebtedness and slavery to Satan and casts the link in powerful images that were to have a lasting impact on subsequent Christian thinking about salvation: Having wiped away the certificate of debt . . . which is adverse to us (ἐξαλείψας τὸ καθ’ ἡμῶν χειρόγραφον), he [Christ] destroyed it completely, having nailed it to the cross.15

This passage has been heavily interpreted both by early Christian exegetes and modern scholars. The particular point of the academic discussion has been the somewhat enigmatic cheirographon. While early Christian exegeses of the debt symbolism and the meaning of this term varied significantly (including readings in which it is Christ himself), it is today’s academic consensus that the Greek term derived from legal and economic technical idioms of the epoch contemporary to the composition of the Epistle to Colossians, and would clearly evoke relevant images in the audiences of

12 13 14

15

See e.g. Oakman 2014 for a recent synthesis on the much-studied theme. For legal thinking in Pauline writings, see e.g. Martin 1989. For Jewish concepts of sin as debt, see Anderson 2009 (with further extensive literature). For a Hellenic background, see Bazzana 2011. Translations are mine throughout. In my interpretation of the notoriously loose χειρόγραφον τοῖς δόγμασι, I follow Carr 1973.

(Il)Legal Freedom: Christ as Liberator

71

the text. The cheirographon denotes a record of debt signed by the debtor. In this sense, the word is firmly attested in papyrological record.16 For the author of Colossians 2:14, Christ performs a drastic redemptive act of destruction of the debt record of the entire humanity. Within the broader New Testament paradigm, where debt is a typical metaphor to conceptualize sin, the soteriological message of the passage is clear: through his death and resurrection, Christ cancelled humans’ indebtedness to Satan incurred through their sins, and therefore saved them. The cosmological and millenarian implications of the episode are also recognizable. Since remittance of ‘debts’ as sins is consistently portrayed in the gospel narratives as the ultimate prerequisite of salvation, and is the object of everyday request in the Lord’s prayer, the abolition of the enslaving debt record is Christ’s greatest gift to humanity. It seems to be only a natural corollary that the debt was due to Satan – a line of exegesis that would develop and spread in early Christian thinking at least by the fourth century ad.17

debt bondage: late antique social contexts Alongside the scriptural background, the close tie between slavery proper and debt bondage as causing it would have significant currency in late antiquity for a number of socio-economic and ideological reasons. Heavy indebtedness and the patterns of legal enforcement connected with it often brought the impoverished to the brink of losing their own, as well as their families’, freedom. As the evidence of documentary papyri attests, children could be used as security for loans.18 They would be taken away from insolvent borrowers by creditors and therefore lose their freedom.19 There is the dramatic story of one Pamonthios in fourth-century Egypt who took out a loan to cover his tax arrears, failed to pay it back, and lost his children to the creditor; or the situation of a fishmonger mortgaging his teenage daughter and having her 16

17

18 19

Modern academic interpreters vary in their understanding of the original meaning of the intriguing ‘certificate of debt’ in Colossians 2:14; see e.g. Blanchette 1961; Yates 1990; Lang 2016. The classic, and still fullest, discussion of the patristic reception and exegesis of this passage is Best 1956, who explores the origin of understanding the cheirographon as Adam’s certificate of debt to Satan. For wide-ranging creative reworkings of the trope of Adam’s debt certificate in the subsequent Christian thinking, particularly apocryphal writings and folk traditions, see Stone 2002. See the chapters of Vierros, and Pudsey and Vuolanto, in this volume. My discussion in this paragraph relies on to the data gathered and analysed in Vuolanto 2003, esp. 189–97 (followed by Samellas 2017, 152, with further bibliography).

72

Arkadiy Avdokhin

seized as securing property. These are but tiny glimpses into the broad, and consistent, social pattern of heavy indebtedness leading to loss of freedom.20 Christian writers attacked similar practices across the Empire in purposefully moving portrayals (which, however, can hardly be read as independent accounts, relying as they do on Basil of Caesarea’s seminal emphases in one of his homilies on the Psalms).21 Augustine speaks of instances of selling children as labourers – a practice coming close to the paramoneˉ contract that has been a subject of a lively scholarly debate in recent studies.22 Self-sale under the pressure of tax arrears and debts was a well-attested, although a much criticized, practice in late antiquity.23 The legal prohibitions of the practice that span from at least the third century ce on only serve to highlight its lasting character.24 The phenomenon continued to loom large, at least as a literary frame of reference, and to be eagerly presented both in popular narratives and in liturgically set homilies of prominent episcopal authors well into in the middle Byzantine epoch.25 For late antique audiences of homilies and hymns, enslavement and heavy debt would be inextricably linked as interlocking social phenomena, as the pressure to keep up with soaring tax payments and personal loans was constant. For the less privileged, both were sources of significant financial tension resulting from the need to prevent the extreme indebtedness that potentially led to debt bondage and even enslavement. If not the most widespread phenomenon statistically, the horrifying expectation and dread of such developments would be a significant psychological factor that contributed to the social climate.26 One of the ways to tackle the social tension, or rather to harness it towards maintaining power, would be actions by state administration, and primarily by emperors, to alleviate tax and debt pressure.

20

21

22

23

24

25

26

P.Lond. VI 1915 (ca .340 ce) for Pamonthios; see Laes 2008, 269; P.Coll.Youtie II 92 (Antinoopolis, 569 ce) for the fishmonger; see Kreuzsaler 2014: 271–2. E.g. Bas. Hom. 14 in Ps. 4 (PG 29.277); Sev. Ant. Hom. 13.20; see the discussion of these and other sources in Vuolanto 2003, 173–5, with additions in Samellas 2017, 152. Aug. Ep. 10*; see also the chapter by Kahlos in this volume. For paramoneˉ, see Brass 2005. A good example is Palladios’ account in his H. Laus. 37.6–8. For voluntary bondage and self-sale in late antiquity, see Silver 2018 (with further literature). The legal prohibitions of the practice that span from at least the third century on only serve to highlight its lasting character; see Cod. Iust. 8.16.6 (293 ce). For popular accounts, see e.g. Paul of Monembasia, Narrationes 16; for the importance of the topic in liturgical homilies, see Vuolanto 2003. Ilian 2014, 130–2.

(Il)Legal Freedom: Christ as Liberator

73

Imperial remittance of debts on a massive scale, although implemented somewhat sparingly, was perceived as a constantly present possibility. It clearly predates the epoch of the Christian emperors and goes back at least to Hellenistic kings, whose practices had also inspired the religious conceptualization of remittance of sins as debts that had influenced New Testament authors.27 Contexts that are more immediately relevant to late antiquity are those of the Roman imperial era. Famously, in 118 ce, Hadrian issued an edict ruling that debt records be collectively burned – an act spectacularly inscribed into the communal memory through such visual means as the so-called Anaglypha Hadriani (which may indeed be Traianic) reliefs at the Forum of Trajan.28 In the later Empire, public acts of emperor-instigated abolition of debts became a widespread tactic to keep the populace’s discontent at bay in the context of contested successions, raging economic crises, and the everrising level of economic austerity. A flurry of edicts sanctioning the remittance of debt and tax arrears are recorded in the fifth and sixth centuries (in 414, 433, 438, 450 – accession of Marcian, 527 – accession of Justinian).29 Ironically, these acts quite often cured the harm inflicted by imperial regulations themselves, as the tax arrears would accrue in no small part because the taxes levied were unbearable for a significant portion of the population. Anastasios’ abolition of the obnoxious chrysargyron in 498 ce is illustrative in this respect – it was the very tax that Maximos, the benevolent patron of Stratonikeia’s poor, paid on behalf of his fellow citizens.30 Late antique Christians would perceive New Testament symbolism of sin as debt, and Christ’s salvific abolition of humanity’s enslaving debt to Satan, against the ever-present background of massive poverty, debt, and the often-ensuing debt bondage – a sense to which the enforcement and annulment of imperial taxation contributed significantly.

debts and sins: harmonization of various metaphors of salvation in liturgical writings In good accord with the religious sensibilities expressed in the New Testament, and in constant reference to contemporary social realities,

27 28 29 30

See note 14. For the monument, including its contested dating, see Rüdiger 1973. Jones 1964, 467. See note 1.

74

Arkadiy Avdokhin

enslavement to sin and humanity’s indebtedness, through sin, to the devil were typically brought together within a single thematic framework in late antique Christian writings. A literary and ideological process of ‘harmonization’ of a number of scriptural emphases was underway as the message of salvation from debt enslavement to Satan (Colossians 2:14) was merging with wider New Testament ideas revolving around slavery to sin and the freedom granted by Christ. In homilies of late antique preachers, images and vocabulary of satanic debt bondage typically interlaced with those of slavery to sin. This combination demonstrably arose from the exegesis of the Pauline passages engaging with these themes, primarily Romans 6:15–23 (slavery to sin) and 1 Corinthians 7:21–3 (the metaphor of ‘buying out’ – τιμῆς ἠγοράσθητε – from slavery). The theological frameworks that drew on the New Testament metaphors of debt, bondage, freedom, and their economic dimensions contributed to shaping this unified vision of humanity’s freedom from a host of oppressive satanic influences. This exegetical fusing of originally distinct scriptural messages is consistently present in late antique (para)liturgical texts that focus on soteriology and, quite often, on Christ’s salvific passion and resurrection. I will illustrate this through a few examples. Gregory of Nyssa in his homily In sanctum Pascha strings together metaphors of setting free the prisoner, remittance of borrower’s debts, and manumission: ‘Now the prisoner (δεσμώτης) is released, the debtor’s (χρεώστης) debt is remitted, the slave is set free (ἐλευθεροῦται) by the kind and humane proclamation of the church.’31In these metaphors, the freed prisoner, debtor, and former slave equally encode humanity, which is now restored to its original nature, free from sin and slavery to Satan. The author of the widely popular pseudo-Chrysostomic homily In adorationem venerandae crucis, writing probably in the fifth or sixth century, follows a similar line of liturgical exegesis, and rhetorically equates humanity’s liberation from captivity and enslavement to Beliar by the tearing up of Adam’s certificate of debt to Satan: Let us sing along with David the prophet: ‘The light was shown to us of your face, O Lord’. Which light is it, if not the cross of our Lord and God and Saviour Jesus Christ, through which the world was saved, and those who have believed in it were liberated from Beliar’s captivity, and the document listing our misdeeds was torn down (ὁ χάρτης τῶν ἀνομιῶν ἡμῶν διερράγη)?32 31 32

Gr. Nyss. In S. Pasch, PG 9.250, 14; see also the chapter by Ramelli in this volume. Ps.-Chrys. In adorationem venerandae crucis, PG 62.747, 33.

(Il)Legal Freedom: Christ as Liberator

75

Leontios of Constantinople writing in the late sixth century embellished his homily on Pentecost with a litany of rhetorical questions that seamlessly link the ideas of slavery (to sin) and debt bondage to devil: How is a sinner possible that would destroy sins? When does a slave set another slave free? When does someone who is held criminally liable buy out another one equally liable? When does a debtor tear down the debt certificate (τὸ χειρόγραφον διαρρήσσει) of another debtor? It is a free person who sets a slave free, not a slave.33

Resulting from such literary portrayals is an amalgamation of metaphors and vocabulary of freedom from, on the one hand, slavery, and debt on the other. In the context of (para)liturgical celebration of Christ’s victory over death and the devil, therefore, the distinctions between the metaphors and messages of debt bondage and slavery were increasingly irrelevant. The tearing up of the ‘certificate of debt’ (cheirographon) at the cross is the end of debt bondage, through sin, to Satan, and the ultimate liberation of humankind. As in socio-economic realities of late antiquity, when indebtedness would often be practically, and causally, linked to bondage and loss of freedom, they would also be brought together in liturgical genres. What, however, did this annihilation of the primordial debt record mean, and how did late antique audiences of homiletic and hymnic texts imagine and interpret it? Did the act imply a forceful destruction by Christ as liberator of the demonically bureaucratic paperwork through which humankind used to be enslaved, and the inauguration of the epoch when people would be eternally free from the power of damning records of their sins? In other words, did late antique homiletic writers suggest to their audiences that the legalistic framework, within which the metaphor of debt record as the embodiment of ultimate evil functioned, had been made redundant as intrinsically void of grace and satanic? Or did they suggest, rather, that Christ came to introduce a new, rectified model of cosmic regulation that would supplant the devil’s unjust arrangement with a godly one? The two options can be seen as poles of an interpretative spectrum within which homiletic authors and hymn-writers of Christian late antiquity oscillated. Sketched below is the plurality of their takes on the problem of what happens after Christ performs the tearing up of the debt certificate, and of the soteriological and anthropological implications of this act. As I will suggest, in some texts the nature (and consequences) of

33

Leont. B. Mesopent.; in Allen and Datema 1987, 314. The best introduction to Leontios is now Daley 2017.

76

Arkadiy Avdokhin

Christ’s destruction of humankind’s debt bondage to Satan is an irreversible abolition of the very legalistic principle of human salvation. For the authors of other texts, the end of the illegal and unjust debt is only a dawn of a new era of eternal rule of Christ-inaugurated cosmic regularity that works through the power – and paperwork – of the divine law of salvific dispensation.

destroying the debt record, challenging legalistic ontology? An interesting case in point is the nativity homily of Theodotos of Ancyra, which was pronounced as part of the deliberations of the council of Ephesus.34 In the homily, Theodotos gives a rather atypical turn to the New Testament image and wording of destroying legal paperwork in application to his Christological message. As he discusses the subtle dialectics of the two natures in Christ and their interaction in his physical body, Theodotos appropriates the image of an imperial edict as logos that is put into material flesh of the document’s physical support (charteˉ ): Suppose the Emperor makes a legal pronouncement and puts it into words on a written document, the so-called sacra, which is sent out to various cities. This pronouncement, clothed both in the paper of the document and into words, grants freedom, or other imperial benefits, to those who have been requesting them. Now what if there is a disobedient unbeliever, who also should happen to be inimical towards the city and an enemy of the Emperor, who gains access to this document, the sacra, as they call it in Latin, and, snatching it, tears it down (διαρρηγνύτω)? Tell me, what has been torn down (ἐρράγη) in this case? Is it only the paper, or also the Emperor’s pronouncement? So, given that it was only the paper that was torn up (χάρτης διερράγη), that person who did the tearing up is not to be held liable, or is he liable only to pay five obols, as the material is cheap? He is given, however, the ultimate punishment, and is put to death as someone who not only erased the text written on paper, but tore down an imperial pronouncement (λόγον διαρρήξας βασιλικόν).35

In Theodotos’s homily, the standard Christological implications of the ‘tearing up’ (διαρρήξας) are reversed. Rather than destroying a piece of harmful satanic writing, the act of violating the sacred wholeness of an imperial edict and its physical embodiment is in focus. The usual emphases of the portrayal of bureaucratic writing as formalistic, and conflicting with the power of God’s grace, are turned around. Attacking 34 35

For Theodotos’s homiletic input at the council of Ephesos, see Frenkel 2016. Thdot. Anc. Hom.; ACO 1.1.2, 80.

(Il)Legal Freedom: Christ as Liberator

77

and tearing up a document is portrayed as an action that renders the perpetrator liable to capital punishment. In a remarkable contrast with the image of a daring liberator who destroys a satanic written record, and with the broad tradition of exegesis and (para)liturgical compositions that I discuss in this study, Theodotos’ suggestively Christological wording works expressly on the premise that to physically destroy a document of significance – which humanity’s debt certificate certainly is – in itself amounts to ‘going rogue’ and performing an act of an ultimately illegal nature. In this perspective, Christ’s salvific tearing up of the cheirographon is, in hindsight, seen as a daring act of subverting the legal, done in an attempt to remedy the injustice of humanity’s enslavement to the devil through Adam’s fall. It would probably be wrong, and futile, to argue that Theodotos of Ankyra in his conciliar homily – a carefully crafted and doctrinally charged text – was intentionally offering a subversive reading of Paul’s wording in Colossians 2:14. The homilist was not likely promoting a soteriology that would be self-consciously premised on the understanding of Christ as a daring outlaw. Yet this is what followed from the logic of his text and the implications of his phrasing, which evokes the relevant Pauline idiom – emphases that may not have been lost on Theodotos’ audience. The briefness of Paul’s description of Christ’s tearing up of the cheirographon, and most centrally his vocabulary of forceful destruction, would remain enigmatic enough for fifth-century homilists (quite apart from modern biblical scholars) to explore the semantically ambiguous implications of the scriptural episode. While the subversive emphases of Theodotos’ idiom are implicit, other late antique homiletic writers composed more straightforward accounts of Christ’s aggressive remittance of sins. For instance, Amphilochios of Iconium, writing in mid-fourth-century Asia Minor, paints a dramatic picture of a thorough destruction of the cheirographon and, more significantly, of the abolition of the very principle of enslaving lending practices. In his homily In mulierem peccatricem, Amphilochios weaves a daring rethinking of the key concept of Colossians 2:14 into his exegesis of the parable of the two debtors in the Gospel of Luke 7:36–50 (which is foundational for the subsequent Christian exegesis that relies on the metaphor of loan and debt as symbols of sin and remittance): He says that one of them owed fifty denarii, and the other owed fifty as well. The story takes a terrifying turn here. Our life is an official document (γραμματεῖον) that invisibly writes down both our thoughts, and deeds, and the wanderings of

78

Arkadiy Avdokhin

our eyes, and the movements of our souls. The philanthropic creditor (φιλάνθρωπος δανειστής), however, dispels the terror, the one who tears down the certificates of sin (χειρόγραφα διαρρήσσων τῆς ἁμαρτίας), and not only does he tear them down, but also erases them clean (λεαίνων) with the waters of baptism, so that not a trace of a letter or a syllable is left to become a record of the evil things that have passed. He says that as they were not able to return the debt, he gave a gift to both of them. Have you seen this philanthropic creditor who offers loans but does not demand the money back? How is it that his hand, taken for granted as it is, remains extended towards those begging him rather than going numb of the effort? . . . He remits the debts to those who had been unable, rather than unwilling, to repay . . . That is to say, God asks nothing of us but confession of sins, as he wants us to be ever joyous and address him as we confess our wrongdoings.36

Amphilochios lays particular emphasis on the irreversibility of Christ’s destruction of the perennial debt record: not only did he tear it down, but he also made sure that not a letter or syllable of the original record survives by washing it away through baptism. The technical thoroughness of the destruction of the cheirographon and its almost violent decisiveness are, however, not the central point in Amphilochios’ rhetoric on the salvific nature of Christ’s act. More significant is the fact that the very principle of the economic imagery of lending to the poor – as a metaphor for the selfperpetuating mechanism of committing sins and receiving due, and infinite, punishment for them – is abolished. Rather than continuing to offer the ultimately enslaving loan to the destitute as an image that encodes human nature susceptible to sin, Christ establishes a totally different framework of ‘economic’, that is, mystically salvific, dispensation. He acts not as a lender who expects the money back, thereby plunging the impoverished lender into the constant danger of going back to the enslaved state. Instead, in Amphilochios, Christ is portrayed as possessing immeasurable wealth that he uses towards a never-ending largesse that absolves the also never-ending expenditure of the poor (as a metaphor for sinners). Seen in the social terms that are clearly implied in Amphilochios’ idiom rich in economic terms, Christ stands behind the image of a fabulously wealthy elite member, or the emperor, who acts as a patron of the poor. As the instances of Christian-minded euergetes like Maximos from Stratonikeia suggest, rich civic patrons could be at least expected to cover the otherwise hopeless debts incurred through tax arrears of the dispossessed lower classes in late antique cities. Both in popular imagination and in practice (however irregular), emperors would also be seen as figures of 36

Amph. Or. 4 in mul. pecc. 2; Datema 1978, 111.

(Il)Legal Freedom: Christ as Liberator

79

ultimate power who could affect the remittance of the accumulated mass debt of entire communities of cities or whole regions. Unlike his earthly counterparts, who were barely expected to extend their financial subsistence ad infinitum, however, Christ’s philanthroˉ pia is portrayed as literally limitless. As such, it transcends, and therefore subverts, the very principle of giving loans, and supplants it with the idealized – and admittedly much coveted – framework of endless largesse and patronization. From this vantage, offered in Amphilochios’ homily, the credit system is described as intrinsically evil and enslaving, and Christ’s innovation lies in a selfless procreation of income that covers the ever-accruing ‘debt’ (sin) – on the sole condition, of course, of the eagerness to repent. Theodotos’ and Amphilochios’ instances, however, remain to be seen as unusual rather than typical in the context of late antique homiletic output. Amphilochios, who has the damning and enslaving recordkeeping forcefully abolished as the underlying principle of both monetary loans and the divine economy of the remittance of sins, is far from typical in his portrayal of the figure of Christ as the celestial basileus. More often, both the earthly workings of the imperial power and its heavenly counterparts were pictured differently. Christ as emperor would be imagined as introducing a new legalistic framework of salvation for which bureaucratic record-keeping would be fundamental.

the legalistic frame: breaking and remaking the bond In a significant portion of late antique homiletic output, as well as in the (para)liturgical hymns of Romanos the Melodist I discuss below, Christ’s destruction of the debt certificate is not portrayed as the obliteration of the intrinsically deleterious inscription into a binding spiritual framework of obligations and reciprocations. Quite apart from the act of annulling Satan’s cheirographon, there are salvation-bringing obligations and mystical contracts, according to late antique Christian homilists and hymn-writers. Within this mode of thinking in terms of legalistic metaphors, legal and bureaucratic procedures can be presented as embodying spiritual realities of salvation and redemption. Properly drawn-up certificates would be portrayed as securing Christ’s ensured presence rather than eternal damnation.37

37

For the imagery of written documents as symbols of spiritual entities and spiritual states in late antiquity, see Rapp 2009; 2015.

80

Arkadiy Avdokhin

A significant number of texts and authors that harbour such, or similar, sentiments ultimately draw on the set of ideas and images that portray Christ as a triumphant emperor whose victory over Satan the usurper harbingers the inception of a new era of heavenly dispensation. The complex patterns of interaction between imperial imagery and the ideas about Christ, his salvific activities on earth, and the second coming have been differently approached in scholarship. Alongside the now traditional interpretations along the lines of a profound synthesis between the two, a more recent analytical lens has been developed that brings out the tension-riddled hybridity and emulation of imperial forms by Christian church structures, doctrinal frameworks, and modes of visual depiction.38 However we conceive of the dynamics between the Empire and the nascent Christianity, by the fourth century onwards conceiving of Christ through metaphors of imperial power and clemency had become a natural part of the literary tradition on which late antique Christian writers would rely. In this context, Christ’s abolition of Adam’s (that is, the entire humanity’s) enslaving debt bondage to Satan is often implied, or explicitly described, as one of the acts of the true, and triumphant, emperor (basileus) who puts an end to the unlawful reign of the usurper (Satan), thereby setting humans free.39 This imperial frame of reference arguably stands behind the legalistic metaphors of abandoning the essentially unlawful, if superficially licit, enslavement to the devil incurred through debt/sin bondage, and of supplanting it with a justly arranged, and eagerly embraced, servitude to Christ. He as emperor trumps Satan and abolishes the ancient debt bondage – all to inaugurate an entirely new kingdom of ontologically just relationships between the emperor (God) and his subjects (Christians).40 38

39

40

The bibliography for this is extensive: for the dynamics between subversion, of and reliance on the ideology of imperial power in apologists’ thinking about Christ, see e.g. Rankin 2001. For the close modelling of Christ’s image, including artistic depictions, on the Roman Emperor in late antiquity, see e.g. the classic Grabar 1936. The ‘emperor mystique’ in early Christian art and elsewhere is challenged in Mathews 1995; see Jefferson and Jensen 2015 for further discussion and fresh insights. For the postcolonial methodologies that approach early Christian engagement with imperial ideological frameworks in a complex manner, including rejection and subversive re-creation, see e.g. Peppard 2011; Leander 2013. For a selection of patristic contexts that portray Christ as a lawful emperor who conquers Satan the usurper, see Petruccione 1991, 75n15. For the trope of Christ the emperor in early Christian writings, see Petruccione 1991. Chrysostom (in Hom. 18 in 1 Cor. 3) develops a broadly relevant metaphor of Christ buying out from the devil (portrayed as a pimp) a young woman who had been sold as a prostitute in order to raise money for her father’s debts; see De Wet 2015, 240.

(Il)Legal Freedom: Christ as Liberator

81

Within the newly established modes of relationships, salvation is imagined through metaphors of legal procedures and paperwork. For instance, the popular pseudo-Chrysostomic homily In sanctum Pascha (which relied on, and combined, multiple late antique pieces by Severian of Gabala and Leontios of Constantinople, quite likely accessed by the author through florilegia) paints a picture of humanity newly liberated from devil’s tyranny (διαβόλου τυραννίδος ἐλευθερώσας) by Christ, who restored humans to their original noble freedom (προτέραν εὐγένειαν), acting quite like the emperor who suppressed usurpers and benevolently reinstated the rights and dignity (φιλανθρωπίαν κατηξίωσε) of those wronged.41 This portrayal resembles Eusebios of Caesarea’s influential depiction of Constantine’s benefactions after his victory over the usurping Licinius – a link that needs further exploration.42 Similarly, Basil of Seleucia, writing in the mid-fifth century in Asia Minor, elaborates a nuanced image of the victorious emperor-Christ, who eagerly and with benevolence accepts the former servants and soldiers of the defeated usurper’s (Satan’s) army, therefore giving a new start to the universe, seen as a state rearranged according to lawful and humane order.43 Cyril of Alexandria, in his Epistula paschalis 10, makes the emphasis on inclusion even stronger, as he explicitly pictures both himself and his audience at the Easter celebration as the forgiven and assimilated ex-servants of Satan who are redeemed through Christ’s benevolence and re-introduced into the now-perennial kingdom of God on earth.44 Ps.Asterios the Sophist, probably writing in fifth-century Syria, specifically focuses on military and adventus imagery of the victorious Christ, who is eagerly acknowledged as the liberating true emperor by those who were enslaved by Satan; Satan’s soldiers flee at Christ’s adventus.45 A pseudoChrysostomic text, although difficult to date precisely or argue for a safe

41

42 43

44

45

Ps.-Chrys. In sanctum Pascha, PG 52.768, 36. For the text and its literary history, see Rambault 2013, 11–14. Eus. V. Const. 2.20–2. Bas. Sel. Hom. 28: ‘Christ as the true emperor . . . having overcome the tyranny of the devil (τοῦ διαβόλου τυραννίδα κρατήσας), makes an appeal to seek freedom to those who, prompted by the devil, used to be hostile towards the Lord, and summons them to himself so that they could enjoy the kingdom’ (PG 85.325). Cyr. Ep. pasch. 10.1: ‘the Saviour condemned the devil’s tyranny (τοῦ διαβόλου τυραννίδος κατεδίκασεν), set us free, and made us his own’ (SC 392.190). Ps.-Ast. Hom. 18.9: ‘Those abiding there, in the subterranean regions, will no more worship you the tyrant (τυράννῳ), but will rather worship Christ the emperor (βασιλεῖ), and they will bend their knees in their free will’; in Richard 1962, 130. For Ps.-Asterios and his identity, see Kinzig 1990.

82

Arkadiy Avdokhin

attribution, lucidly combines the ideas of liberation from slavery to Satan the usurper with an explicit imagery of tearing up Adam’s primeval debt certificate (cheirographon). Christ is therefore presented as a victorious emperor, the destroyer of the unjust debt obligation and liberator of the humankind in his kingdom of eternal justice.46 Within this narrative, the abolition of the bondage to Satan does not amount, however, to more fundamentally challenging the essential principle of meticulously recorded relationships between humans and the superhuman. On the contrary, the victorious Christ as emperor seeks to re-establish properly documented bonds – this time lawful and just, of course. One of the most striking instances of this kind of theological thinking is found in the oeuvre of Proklos of Constantinople, who uses spectacular metaphors of Christ destroying Satan’s injustice embodied in the cheirographon only to re-introduce the paperwork-based mystical principle of divine justice. This indicates that, for the immensely popular and influential fifth-century homilist and theologian, thinking in terms of sin as debt was an important aspect of developing the doctrine of salvation.47 In the famous Marian homily probably composed and delivered briefly before the Council of Ephesos (431 ce), Proklos, addressing his audience in the Constantinopolitan Church of the Hagia Sophia in the presence of his doctrinal rival Nestorius, develops a sophisticated Christological exegesis of the Virgin Mary’s salvific activity.48 In the homily, a remarkable conceptual framework of thinking about the abolition of Adam’s original sin and enslavement to the devil is introduced, in which the divine intervention is not only – indeed not so much – destructive for the satanic injustice as it is re-formative for the entire universe: The human race was deep in debt, and had great difficulty repaying it. We all had signed the debt record (ἐχειρογραφήσαμεν) through Adam. The devil held us as slaves (δούλους). He brought forth the purchase contracts concerning us, using the much-suffering body as writing paper. There he stood, the wicked forger of documents, waving about the record of our debt (χρέος) and forcing us to repay.49

46

47

48 49

Ps.-Chrys. In triduanam resurrectionem domini: ‘Those in Hades are being set free of devil’s chains, they are rushing towards Christ . . . He cancelled the devil’s tyranny (τοῦ διαβόλου τυραννίδος), he made us free, us who had been slaves to sin, by erasing the adverse cheirographon’ (PG 60.821). For Proklos’ thought as expressed in homilies, their significance and impact, see Marx 1940; Leroy 1967; Constas 2003. See Constas 2003, 56–70 for the historical context and theology of the homily. Prokl. Hom. 1 (Constas 2003, 141).

(Il)Legal Freedom: Christ as Liberator

83

As Satan falls and the enslaving debt bondage is abandoned, a new pact between humanity and their true saviour and master is made. The novel cosmic arrangement is portrayed majestically through the images of a manuscript certificate that supplants the diabolical one. Instead of the written record of sinful debt, a certificate of liberty (ἐλευθερίας γραμματεῖον) is proclaimed.50 Addressing his fifth-century audiences, Proklos celebrates the salvific bureaucracy of God’s dispensation through the Theotokos’ agency rather than abolishing the principle of legalistic record-keeping. In sixth-century Constantinople, metaphors of sin as debt and of tearing up the cheirographon are developed with a remarkable consistency in the hymns of Romanos the Melodist. His extensive, florid, yet highly dynamic hymns rise out of an idiosyncratic synthesis of the Syriac liturgical hymnody with the vibrant tradition of late antique homiletics in Greek.51 Romanos’ kontakia are (para)liturgical compositions intended most likely for the rather popular, and well-attended, urban vespers in large churches of the capital (many of them clearly written specifically to be performed within the stunning space of Hagia Sophia). The kontakia were the perfect media through which to engage broad, and mixed, audiences, and to suggest frames of Christian sensibilities to them.52 Thinking about sin and salvation in terms of debt-as-sin imagery would be one. While the homiletic output of earlier authors is probably the source of Romanos’ keen engagement with the imagery of tearing up the primordial debt certificate as Christ’s salvific act, he approaches these tropes through his own particular vision. He adds peculiar twists and shades to what I have termed legalistic emphases in his appropriation of Colossians 2:14. For the understanding of the Christ-centred soteriology that is developed in Romanos’ hymns, tropes of Adam’s torn-up debt certificate are fundamental. In one of his kontakia on the resurrection, Adam, on hearing that Christ has risen, begins an animated dialogue with Hades. When Adam joyously teases Hades, telling him that his power is soon to be over, Hades retorts by saying that he still holds the debt certificate and therefore retains control, through Adam, over the entire human race: There is no-one to protect you and to destroy your certificate (ῥῆξαι τὸ χειρόγραφόν σου) 50 51

52

Prokl. Hom. 1 (Constas 2003, 138). For Romanos fusing his Syrian and Greek backgrounds, see van Rompay 1993; Brock 1999. On Romanos in general, de Matons 1977 is still unsurpassed. For the performative context of Romanos’ hymns, see de Matons 1980–1; Lingas 1995.

84

Arkadiy Avdokhin

For I am king of even the one who you call your helper . . .53

Hades’ claims prove vain, however, as Christ crushes down Hell and sets free Adam and Eve. His salvific mission is accomplished, and humanity is freed from slavery to Satan. The cosmic perspective of many of Romanos’ hymns that portray the climatic destruction of Satan’s power is supplemented with a much more intimate treatment of the topic of salvation in the kontakion that is almost entirely structured as a personal petition by the author. In a heartfelt and even heartbroken tone, the poet laments his fallen nature and sinfulness, asking for an individual act of remittance of debts/sins and the destruction of the enslaving power of the cheirographon over his soul: I know that you are the deliverer, o saviour, Who is full of sorrow about all my wicked deeds, Erase (ἐξάλειψον) the wrongdoings, sign the remission certificate (ὑπόγραψον ἄφεσιν), o the one who remembers no evil, Tear up (σχίσον) my cheirographon and set me free (ἐλευθέρωσον), For you are alone, o Lord, my emperor and God.54

The deeply intimate tone of the petition is reinforced by the striking personal appropriation of the abstract doctrinal language of the destruction of the certificate of debts: ‘tear up my debt certificate and set me free’. The universal significance of Christ’s act of salvation is zoomed in on the persona of the hymnographer, and would likely be individually appropriated by people praying at the vespers and joining in with the soloist and the church choir.55 It is also important to note that the act of liberation in Romanos is portrayed in bureaucratic terms beyond its destructive stage of nullifying the debt records. A written certificate of remittance is to be signed by Christ. The profoundly personal dimension of the soteriology of Romanos’ poetic petition is an aspect of a wider mystical space within which God’s dispensation works. To a significant degree, it hinges on the combination of divine bureaucracy and legalistic workings of salvation, in which freedom from sin and slavery to Satan are attained by buying into other, thoroughly documented, frameworks of power. Romanos’ fascination with the idea of the redemption-rooted cosmic order and its mystical mechanics of divine paperwork underpins the 53 54 55

Rom. Mel. Cant. 43.5 (‘on the resurrection’), stanza 4 (SC 128.554). Rom. Mel. Cant. 55 (‘penitential hymn’), stanza 9 (SC 283.520). For the likely lay participation in the performance of Romanos’ hymns, particularly in singing the refrains, see Arentzen 2017 (with further literature).

(Il)Legal Freedom: Christ as Liberator

85

imagery of divine bureaucracy, with its salvific paperwork, that surfaces consistently in his hymns. In his kontakion on Elijah, God retorts the prophet who urges him to exercise utmost austerity on the sinful humankind, with these words: Listen, therefore, o prophet, with daring, for I seek your full understanding: All the humans hold me as a certificate of merciful love (χειρόγραφον εὐσπλαγχνίας), In which I expressed in writing (συνεθέμην) that I do not want to see Death of those who do wrong, but rather their life.56 In what may be a direct echo of Proklos’ homily discussed above (who speaks of the salvific ‘certificate of freedom’), Romanos boldly re-casts the image of Satan’s enslaving cheirographon into the divine document that ensures salvation. In the hands of the faithful, the new cheirographon is the ultimate certificate of the universal salvation that mediates between humans’ unavoidable sinfulness and God’s salvific dispensation. The originally Pauline metaphor is thus inverted to produce a potent picture of a cosmologically imagined, salvation-bringing divine bureaucracy.

conclusions For late antique Greek-speaking audiences of homilies and hymns, the economic, the social, and the spiritual may not have been quite so different from one another. When hearing of things elevated and godly, an attendee of a vesper or a eucharistic service would be instantly reminded of tangibly earthly processes and frames of feeling. These were centred around solidly human, ultimately down-to-earth – merciless and hope-giving sometimes – situations of entering into the state of debt bondage and being set free from it. Since the metaphors of remittance of debt and salvation from sin were clad in virtually identical vocabulary, Christ was, with his celestial retinue, being transformed into a system of celestial bureaucracy engaged in the complex paperwork of granting salvation. Economic processes and practices of tax and debt levying, or their cancellation, were majestically upgraded on the ontological scale. At the same time, the uppermost, mystical metamorphoses of human salvation were conceived as palpably human and therefore manageable, if only through the intercession of powerful patrons. Heaven was not entirely opposed to earth, and Christ-given freedom was not altogether 56

Rom. Mel. Cant. 7 (‘on Elijah’), stanza 25 (SC 99.332).

86

Arkadiy Avdokhin

dissimilar, in its feel, from the relief from the imminent danger of the endless row of tax arrears, indebtedness, and the ensuing loss of freedom. The dynamics of this liberation are twofold. Christ could be pictured, in the mental eye of the believer, almost as a good brigand who is willing to destroy, forcefully and daringly, the much-hated, if technically valid, debt record. This soteriological violence could have wider, potentially subversive implications for the popular sentiment about formal law and order. At the same time, Christ could also be imagined and compellingly portrayed as inaugurating a whole new universe based on a mystically reformed system of divine bureaucracy that gave freedom by inscribing one into its newly drawn dispensation. The tension between the two poles that, as I have suggested, existed in late antique homiletics and hymnody may be revealing of broader conflicts in how people conceived of their human – emotional, financial, and bodily – freedom in the face of the allembracing powers of divinity and state.

4 Late Roman Ideas of Ethnicity and Enslavement Maijastina Kahlos

All this is in the face that every house, however humble, has a Scythian for slave. The butler, the cook, the water-carrier, all are Scythians, and as to retinue, the slaves who bend under the burden of the low couches on their shoulders that their masters may recline in the streets, these are all Scythians also; for it has been proved of old that theirs is the most useful genos, and the fittest to serve the Romans.1

In this way, the fourth-century philosopher Bishop Synesius of Cyrene argued that every Roman household, even the most modest, had Gothic slaves.2 In this chapter, I examine how late antique writers, Synesius among them, dealt with the enslavement of foreigners. Foreigners here refer to non-Roman and non-Greek people outside the frontiers of the Roman Empire, conventionally called ‘barbarians’. War on the frontiers stimulated commerce in humans – namely, slave trade – and vice versa: the activities of slave merchants at least partly motivated warfare in the frontier regions. Non-Roman groups took captives, Romans among them, and made a profit selling them as slaves or returning them for ransom. For their part, Romans took captives and sold them into slavery. We also have several attestations of kidnappers who abducted people during peacetime and even within the Roman Empire. Late antique bishops complained about the slave trade of Roman citizens. Augustine, for example, condemned the business of so-called ‘Galatian’ slave traders.3 In this chapter, I will analyse a few fourth- to sixth-century texts that discuss non-Roman and non-Greek slaves and captives of war. I will ask how late Roman authors (Christian and non-Christian alike) treated foreign slaves and captives in their writings. In order to set late antique writers’ views in their context, it is necessary not only to look at the social

1 2 3

Synes. Regn. 20 (trans. Fitzgerald 1930, 136). Late antique writers used the antiquarian term ‘Scythians’ to refer to contemporary Goths. Aug. Ep. 10*. See my discussion in the section ‘The Limits of the Slave Trade’.

87

88

Maijastina Kahlos

circumstances of the slave trade, but also the earlier deliberations of slaveholding in Graeco-Roman literature. Were some individuals or groups regarded as more suitable to be enslaved because of their characteristics – innate or acquired? Or was slavery understood only as a social institution, and did some individuals just fall prey to it through the whims of fate? Who was ‘appropriate’ to be a slave and who was not? Consequently, were certain ethnic groups considered to be better suited for some tasks and other groups for others? I will discuss attitudes to ‘barbarians’, as late antique sources often mention them in connection with the slave trade in the frontier areas. The voices of the late Roman writers Synesius, Augustine, Gregory the Great, Ambrose, and Basil of Caesarea are set in the context of the Graeco-Roman ideas of slaveholding. Furthermore, I will examine late antique Christian writers who confronted the issue of slavery by discussing the biblical exempla of Ham and Esau.

raiding and trading: the late antique slave trade First, we need to consider the institution of slavery, which continued to be a significant and unquestioned part of Roman society in late antiquity. As Kyle Harper has shown with his abundant material on late antique slavery, writers from that period simply considered the ownership of slaves as a standard element of adult life. Many households, sometimes even modest ones, had slaves.4 In addition to internal sources of slaves – such as slaves born in Roman households, slaves acquired through self-sale, the sale of freeborn children, abduction, and child exposure – there were external sources of slavery, such as frontier trading and raiding.5 Several wars in the frontier areas fed the slave trade with streams of captives. It is probable that the pursuits of slave traders at least partly motivated conflicts in these zones. We have evidence of several nonRoman groups who took captives, both Romans and non-Romans, and then sold them to the slave markets or freed them for ransom. Likewise, Romans took non-Romans as captives and sold them into slavery.6 One of the most well-known ‘barbarian’ captives was the Suebian girl, Bissula, 4

5 6

See Harper 2011, passim, esp. 51, 65–6, on the continuity of slavery in the Mediterranean world, even after the end of Roman rule in the western provinces. Harper writes how the slave society was replaced by a series of societies with slaves in the centuries of the postRoman kingdoms and the early Byzantine Empire. On slavery’s continuity, see also Samson 1992, 218–27 and McCormick 2017, 252. Harper 2011, 79–83. Lenski 2011b, 186–94; Lee 2007, 138.

Late Roman Ideas of Ethnicity and Enslavement

89

whom the poet Ausonius depicted in a small set of poems. She was probably taken prisoner during Valentinian I’s expedition against the Alemanni in the Rhine region in 368. According to Ausonius’ poems, Bissula is his ‘delight, darling, toy, love, and joy’; after she is freed, she becomes an exemplum of successful Romanization.7 At the same time, Ausonius praises her exotic foreignness and Germanic appearance, having the stereotypical features of blue eyes and blond hair.8 Bissula’s case was probably exceptional (one could even doubt if she was a historical person), but there were markets for luxury chattel like her. For example, John Chrysostom warned his listeners against excessive admiration of luxury items such as gold-harnessed horses, barbarian slaves, eunuchs, and expensive dresses in the marketplaces. He also scolded young men who paid courtesies to barbarian slave girls in the markets.9 Cross-border slave trading was important business – probably even the most significant – and it was controlled and supported by the imperial administration. The words of the fourth-century rhetorician Menander of Laodicea reveal what was expected from the imperial power, namely the protection of citizens and a steady flow of slaves: We fear neither barbarians nor enemies. The emperor’s weapons are a safer fortress for us than our cities’ walls. We acquire prisoners as slaves, not by going to war ourselves, but by receiving them from the emperor’s victorious hand.10

Kyle Harper describes the situation as the ‘triangulation of power between the Roman state, slave-traders, and the enemy’.11 The regions north of the Danube were a significant source of slaves for the Romans. According to Ammianus Marcellinus, Emperor Julian was convinced when fortifying the Thracian border that slave traders would be a sufficient match against the Goths. Julian remarked: ‘For the Goths the Galatian traders were enough, by whom they were offered for sale everywhere without distinction of rank.’12 7

8

9

10 11 12

Aus. Biss. 4.1: delicium, blanditiae, ludus, amor, voluptas. For Ausonius and Bissula, see Dräger 2001, 187–207 and Heinen 1995. Aus. Biss. 3.9–10: Germana maneret / ut facies, oculos caerula, flava comas. On Ausonius’ colonial gaze, see Kahlos, forthcoming. Ioh. Chrys. Hom. 4 in Rom. 4; Hom. 78 in Jo. 5. For a discussion of Chrysostom’s views about luxury items such as barbarian slaves, see De Wet 2015, 116. See also Sulp. Sev. Dial. 3.15 on barbarian slave boys and pretty slave girls for sale. On the slave trade in the later periods in medieval northern Europe, see Korpela 2014. Men. Rhet. Epid. 2.377; trans. Russell and Wilson 1981, 92–3; see also Ando 2000, 235. Harper 2011, 85. Amm. Marc. 22.7.8: illis enim sufficere mercatores Galatas, per quos ubique sine condicionis discrimine venundantur. On the importance of regions for slave trade, see also Symm. Ep. 2.78.

90

Maijastina Kahlos

To put it bluntly, it is conceivable that the Romans’ relations with Goths in the frontier regions shifted according to the needs of the slave trade. Consequently, Emperor Valens’ peace treaty with the Goths in 368 demarcated the trade centres on the borders under imperial control. In his speech to Valens, the rhetorician Themistius depicted how, ‘taking pity on the barbarians’, in the peace treaty the emperor allowed them to maintain markets and trade posts. Themistius mentions that ‘our fort commanders and garrison leaders were actually merchants and slave-dealers, since this was essentially their only occupation, to buy and to sell as much as they could’.13 In a law of 374, the imperial government was concerned with limiting the flow of gold in the trade of barbarian slaves beyond the borders of the Empire. By no means did the decree intend to restrain the import of slaves; rather, its goal was only to keep gold within the Empire.14 The crisis that led to the Battle of Adrianople and the Gothic War was the conclusion of all the speculation and profiteering connected to the slave trade on the borders.15 The business in captives meant, in practice, the break-up of families, since family members were sold to different customers around and beyond the Empire. We read of these kinds of separations from the letters of contemporary bishops. Several bishops engaged in crisis management by collecting resources to pay the ransom of Roman prisoners of war, for example by selling church property.16 We do not have evidence, to my knowledge, of the payment of ransom for non-Romans (such as Alemanni). The sixth-century historian Zosimus mentions that Alaric, the leader of the Goths, demanded all (non-Roman) enslaved captives to be returned, and only after that would he end the siege of Rome in 408.17 13 14 15

16

17

Them. Or. 10.136b; trans. Harper 2011, 85; see also Heather and Matthews 1991, 41. Cod. Iust. 4.63.2 (in 374). Giving a vivid description of the profiteering during the crisis, Ammianus (31.4.9–11) argues that the slave trade was the major cause for the skirmishes that eventually led to the battle at Adrianople; Heather and Matthews 1991, 44–5. Ambros. Off. 2.70 writes about ‘the highest kind of liberality’, that is, ‘to redeem captives’. Crisis management is the term used by Allen and Neil 2013; on ransoming of prisoners of war, see Allen and Neil 2013, 38–44; Neil and Allen 2014, 22. On collecting money for ransoms, see Klingshirn 1985, 184–7. Zos. 5.40. Another issue is how reliable this mention is, since it comes from Zosimus. For a discussion, see Lenski 2011b, 190. According to Zosimus (5.42), large numbers of slaves joined Alaric’s troops and increased his following to 40,000 people in 408–9. It is by no means certain whether these slaves were all Goths; see Liebeschuetz 1990, 69–70, 75–6. Heather 1991, 213–14, argues that most of the slaves were ex-followers of Radagaisus. However, if we are to believe Orosius (see the next footnote), many of Radagaisus’ followers, enslaved after his defeat, soon died in slavery.

Late Roman Ideas of Ethnicity and Enslavement

91

Late antique sources speak about hundreds, thousands, and tens of thousands taken as captives and either ending up in servitude or being ransomed. When Roman troops defeated the Goths led by Radagaisus in Italy, the number of captives sold into slavery was so great that the overall price of slaves fell. Orosius writes, ‘it is said that there were so many Gothic prisoners that everywhere herds of humans were bought for a single gold coin, just like the poorest sort of cattle’.18 Even though what Orosius refers to as hearsay (fertur, ‘it is said’) may be an exaggeration, it gives an idea of the number of humans that could be trafficked after intensive warfare. In addition to Goths, other ethnic groups were enslaved as the result of war. Sozomen mentions the ethnic group called Sciri, who in 409 were either made coloni or sold into slavery.19 Augustine speaks of ‘innumerable barbarian peoples’ (barbarae innumerabiles gentes) being brought to North Africa as captives and becoming slaves for Romans; this was something he could ‘see every day before our own eyes’. For Augustine as a bishop, these barbarians were the obvious target of Christian missionary work.20

the variety of ancient views on slavery A number of ancient authors argued that certain people, for some reason or another, lacked adequate mental proficiencies and therefore were incapable of living an autonomous life by themselves. Consequently, their subordination by morally and intellectually superior men was held to be beneficial for them. Plato had already expressed how enslavement of a person with deficient reason by the ‘best man’ is not only necessary but even advantageous.21 For modern readers, Aristotle has been the most renowned ancient author to outline the basic elements of what is usually 18

19 20

21

Oros. Hist. 7.37.16: Tanta vero multitudo captivorum Gothorum fuisse fertur ut vilissimorum pecudum modo singulis aureis passim greges hominum venderentur; trans. Fear 2010, 399 (modified). Orosius adds that none of those people survived. All those who were bought soon died, and those who had bought them at such a shameful price had to spend their money on burials. Soz. Eccl. 9.5.5. Aug. Ep. 199.12: hoc est in Africa barbarae innumerabiles gentes, in quibus nondum esse predicatum evangelium ex his, qui ducuntur inde captivi et Romanorum servitiis iam miscentur, cotidie nobis addiscere in promptu est; see also Harper 2011, 83–4. Pl. Resp. 590c–d. The term ho beltistos, the ‘best man’, also refers to an aristocrat; cf. Pl. Leg. 966b. Plato (Resp. 469b–c) disapproved of the subordination of Greeks (by other Greeks) and recommended that Greeks ‘turn against the barbarians’ in order to get slaves and ‘keep their hands off each other’. On the theory of natural slavery, see Garnsey 1996, 13–14, 38–9; Isaac 2004, 171–2.

92

Maijastina Kahlos

called the theory of natural slavery. He famously argued that ‘some are free men and others slaves by nature’. Natural slaves are deficient in reason and, therefore, it is better for them to be ruled by others. Furthermore, Aristotle asserted that non-Greeks (barbarians) shared those common characteristics that made them well-suited to slavery.22 However, as Chris de Wet has shown, Aristotle did not have a significant influence on later ancient thought, especially in Roman views on slaveholding. Instead, it was Xenophon’s Oeconomicus that was important in shaping Roman and late Roman ideas of slavery. Cicero (106–43 bce) even translated the work into Latin.23 Xenophon advocated ideas of social subjugation according to which a good master beneficially controlled the passions of his slaves.24 In this sense, Xenophon’s views on slaveholding and estate management were not based on nature as such, but were based on the prevailing social circumstances that had shaped the subjugated individual. His ideas influenced Roman landlords and slave-owners via Roman agricultural writers, such as Cato the Elder and Varro.25 The subjugation of slaves was also associated with the Roman conquest of peoples. Cicero stated that some groups of people did service to other groups as their masters. He justified the Roman subordination of the conquered regions as being only for the good of the inhabitants of those provinces.26 Likewise, the well-known legitimation of Roman rule in Virgil’s Aeneid followed the idea that some were fit to rule and others to be ruled: parcere subiectis et debellare superbos, ‘to show mercy to the conquered and to subdue the proud’.27 The self-understanding of Romans as the rulers of the world was continually repeated in Roman literature, as well as in late antique self-assurances, as illustrated by the words of Rutilius Namatianus in De reditu suo: certandi bona parcendique voluptas: / quos timuit superat, quos superavit amat, ‘and thus you are delighted

22

23 24

25 26 27

Arist. Pol. 1253a–60a. Aristotle defends the idea of natural slavery against some critics, whom he does not specify in his text. As Garnsey 1996, 13, points out Aristotle was not the first to advance the theory of natural slavery, nor was he the only one to develop it. Isaac 2004, 46, regards the claim that humans are born to be slaves as the ultimate form of proto-racism, as the Aristotelian idea of natural slaves corresponds with what is believed to be a race (imagined common characteristics, be they physical, mental, or moral); see also Bodel 2017, 82, and Rosivach 1999, 142–4. De Wet 2015, 28–32. Xen. Oec. 3.4; 5.16; 9.5; 12.19; 13.9; 14.4; 21.12 on slaveholding. For further discussion, see De Wet 2015, 28–32. De Wet 2015, 32–4. Cic. Rep. 3. For a discussion, see Isaac 2004, 183–5. Verg. Aen. 6.853.

Late Roman Ideas of Ethnicity and Enslavement

93

by both war and mercy, / you conquer those you fear and love the conquered’.28 The claim that certain groups were prone to be masters and others to be enslaved had both abstract and tangible aspects. The abstract political justification of Roman rule was applied to non-Roman peoples collectively. The more tangible application was the idea that certain groups or individuals were ideally suited, generally, to be slaves in society. The sentiment that certain ethnic groups were fit to be slaves (to Romans) is revealed, for instance, in Cicero’s lamentation of miserable Romans enslaved by Jews and Syrians, who themselves ‘were born to be slaves’ (Iudaeis et Syris, nationibus natis servituti). Ideas of natural slavery were applied to justify both individual slavery by slave-owners and collective subjugation by Roman imperial rule.29 Likewise, the idea of the mollifying effect of slavery had a more important role in Roman society than the (Aristotelian) theory of natural slavery. Enslavement – both as an individual condition and as collective subordination – was thought to have an enfeebling effect. For example, Tacitus described the provincials as softened and emasculated by Roman rule.30 Thus, slavery corrupted persons or peoples into servile beings.31 Therefore, Synesius’ statement that the genos of Scythians (that is, Goths) was the most suitable to serve as slaves of Romans can be set in this context of the Roman social subjugation (especially the idea that some peoples were suitable to be enslaved and other peoples – the Romans, first and foremost – were fit to be rulers).32 As a skilful rhetorician, Synesius was capable of using his argumentative toolkit in a manner most appropriate for the situation. Synesius also reminds his audience that every slave was the enemy of his master, thus appealing to an old theme of imperial literature in which Roman elite writers were concerned about the mischief of their slaves.33 Consequently, he argues, it is possible and expected that Gothic slaves will rise up against their owners, just like any slave who has hopes of defeating 28 29

30

31 32 33

Rut. Namat. 1.71–2; trans. Malamud 2016, 114. Cic. Prov. cons. 5.10; similarly on Syrians: Livy 35.49.8; see also Isaac 2004, 46; Mouritsen 2011, 24 See Tac. Agr. 11.4 on how the provincials were mollified (emollierit) and lost their manhood (amissa virtute); Tac. Agr. 13.1 on servility as the result of Roman rule; Lavan 2013, 133–4. Mouritsen 2011, 18–25. Synes. Regn. 20. For the common theme of slaves as domestic enemies in Roman literature, see, e.g., McKeown 2007b and Harrill 2006b, 147–53.

94

Maijastina Kahlos

his master. Synesius’ warnings were part of his overall claims in his speech, On Imperial Rule, about the infiltration of Goths into the Roman Empire. He associates the Gothic slaves with Goths who served in the Roman army, as if they were the same entity with similar interests, and he sees them as a common threat to the Romans. Therefore, Synesius demands that the emperor ‘remove the foreign cause of the disease’ from the Empire.34 He reminds his listeners that this was the genos that the Romans had subdued. Furthermore, he uses the conventional trope of connecting past peoples depicted in classical literature with peoples encountered in the present. He straightforwardly identifies the Scythians described by Herodotus with the Goths of his own time, using the association to debase the present-day Goths. Therefore, he writes, Scythians ‘are all tainted with a feminine malady’ and consequently they are the people ‘from whose ranks slaves were recruited everywhere’.35 Here Synesius employs a recurring way of associating slaves with femininity, that is, in eyes of Roman male beholders, weakness and passivity.36 Synesius construes an unending historical necessity that Goths have always been slaves, implying that this is their inherent position.37 He insists on anger (thumos) against Goths, who should either subject themselves in obedience to the Romans in the manner of the ancient Messenians – that is, in slavery, similar to the Helots who subordinated themselves before the Spartans – or simply flee back beyond the Danube.38

34

35

36

37

38

Synesius compares the situation with Gothic slaves with the famous slave rebellions of Crixus and Spartacus. He adds that in the slave rebellions of ancient times, the insurgent slaves were not even the same race as their leaders, nor the same race as each other (homophuloi), as was now the case with the Goths; Synes. Regn. 20. Synes. Regn. 20–1. For Herodotus and his views of Scythians’ endemic slaving, see Isaac 2004, 200. and Bodel 2017, 93. For femininity understood as passivity and submissiveness in social power relations in antiquity, see Swancutt 2007. Synesius (Regn. 20–1) continues that in his day, Goths ‘have come to us, not to challenge us in battle, but as suppliants’ (hiketeusontes). The Romans had been too soft on Goths and they had become disrespectful, ignoring the kindness (euergesia) of Romans. Consequently, the emperor (Theodosius I) punished them but also pitied them, making them his allies. Synesius’ account goes on by recounting how the barbarians did not appreciate the virtuous conduct of the Romans but treated them with derision. For the dating and historical circumstances of Synesius’ Regn., see Cameron and Long 1993, 109–21; Goths as suppliants refers to the peace treaty of 376; Goths as being disrespectful refers to the crisis and Battle of Adrianople in 378; punishment by the emperor refers to the war in 379–81; and his pity refers to the treaty in 382. Heather 1988 argues that Synesius’ warnings refer to the threat of Alaric in Illyricum. Synes. Regn. 21.

Late Roman Ideas of Ethnicity and Enslavement

95

Ancient thinking about enslavement also included pragmatic argumentation, which in modern scholarship is usually called legal slavery. In this sense, a slave was thought to have become a slave only due to unfortunate external circumstances, such as being captured in war. In the Roman legal tradition, the notion of slavery as a mere human institution was acknowledged, as the famous example of the second-century ce jurisprudent Florentinus indicates. He writes that ‘slavery is an institution of the law of nations (ius gentium), whereby someone, against nature (contra naturam), is made subject to the ownership of another’.39 Likewise, the earlythird-century jurist Ulpian stated that ‘with respect to the civil law (ius civile) slaves are held to have no standing. But as regards natural law (ius naturale) that is not the case – as according to natural law, all humans are equal.’40 Thus, Roman jurisprudents regarded slavery as an institution based on human social conventions, and Ulpian even recognized the equal status of slaves as humans, despite their suppressed status in society. Similarly, Stoic philosophers such as Seneca were ready to acknowledge the common humanity of the free and slaves respectively, and they reminded their readers that the only difference was their condition as enslaved persons (which a whim of fate could change at one stroke).41 Parallel to the ‘liberal’ Stoic tradition were strands of early Christianity. Most Christian writers saw humans as equal in the sight of God. The Apostle Paul’s maxim, ‘There is no longer Jew or Greek, there is no longer slave or free, there is no longer male and female; for all of you are one in Christ Jesus’ (Gal. 3:28)42 was quoted abundantly by late antique ecclesiastical writers. Early Christian writers accepted slavery as a necessary part of earthly Roman society while, at the same time, recognizing the common humanity of all.43 From time to time in their sermons, John Chrysostom, Augustine, and others felt it necessary to remind the slave-owners in their audiences that slaves were as human as they were, and that slaves had 39

40

41

42 43

Dig. 1.5.4.2: Servitus est constitutio iuris gentium, qua quis dominio alieno contra naturam subicitur; trans. Garnsey 1996, 48. For a discussion, see Bodel 2017, 82; Mouritsen 2011, 14–17; Garnsey 1996, 48. Dig. 50.17.32: Quod attinet ad ius civile, servi pro nullis habentur: non tamen et iure naturali, quia, quod ad ius naturale attinet, omnes homines aequales sunt; trans. Garnsey 1996, 64 (modified). Sen. Ben. 3.28.1; Ep. 31.11; 47.10. On the Stoic attitudes to legal slavery, see Garnsey 1996, 134–8. NRSV. Glancy 2002, 58–69. Garnsey 1996, 188, even infers that the Apostle Paul’s and other early Christian leaders’ instructions on maintaining slavery within the existing social structure ‘contained the recipe for its survival and future strength’.

96

Maijastina Kahlos

immortal souls just like their masters. When discussing the disciplining of slaves, Augustine exhorts the slave-owners to remember that the slave and master were both human.44 The most eloquent appeals to the commonality of humanity were made by Gregory of Nyssa and Salvian of Marseille. Gregory of Nyssa reminded his listeners that masters and slaves originated from the same ancestors, had the same pains and pleasures, and shared sickness and death. Salvian declared that even though slavery made people inferior (in status), humanity made them equal. However, most likely neither of them aspired to the abolition of slavery. Both writers were more concerned about the morals and self-control of the masters than the status of the slaves. Salvian, for example, chastised Roman slaveowners for embracing servile vices and becoming worse than their slaves.45

god’s order and human institutions In fourth- and fifth-century Christian writers’ views of slavery, we can perceive competing elements – notions of legal slavery and ideas of common humanity – as well as views that see some persons as fit for slavery and others as masters. Augustine formulated his famous defence of slavery in close connection with his vision of the two civitates: civitas Dei and terrena civitas. Slavery resulted from the fall of humankind. The cause of slavery was human sin, and, in that sense, humans were responsible for it.46 Therefore, slavery was a social phenomenon and characteristic of the terrena civitas. There was no slavery before the Fall. Augustine admits that slavery is not natural. However, slavery resulted from God’s judgement of humans for their sins; therefore, it belongs to God’s order. Thus, it was more than an ephemeral socio-historical institution. In Augustine’s 44

45

46

Aug. Serm. 159/B.4 = 21.4 (Dolbeau): Non quidem ita debes, si christianus es; non ita debes, si hominem te esse consideras; non ita debes, si consideras diversa quidem esse nomina ‘servus’ et ‘dominus’, sed non diversa ‘homo’ et ‘homo’. Ioh. Chrys. Hom. 8 in Jo. 1: immortal souls and the same gifts. Greg. Nyss. Hom. 4 in Eccl. 2.7; Salv. Gub. 3.28; see also De Wet 2018b, 413–14; Klein 2000, 205–15; Garnsey 1996, 71, 74, 80–5. Aug. Civ. 19.15: Prima ergo servitutis causa peccatum est, ut homo homini condicionis vinculo subderetur, quod non fit nisi Deo iudicante, apud quem non est iniquitas et novit diversas poenas meritis distribuere delinquentium. ‘The first cause of slavery, then, is sin, whereby human was subjected to human in the condition of bondage; and this can only happen by the judgment of God, with whom there is no injustice, and who knows how to allot different punishments according to the deserts of the offenders’; trans. Bettenson 2003, 875 (modified). On Augustine’s theory of slavery, see Garnsey 1996, 47; Klein 1988, 87–93.

Late Roman Ideas of Ethnicity and Enslavement

97

theory, every individual had a place in the hierarchical order of things (ordo rerum) in the postlapsarian world. Some were subordinate to others: in the family, children were subordinate to their parents; wives were subordinate to their husbands; and slaves were subordinate to their masters. This order was maintained to the benefit of the subordinate ones. They were in a happier state than those who ruled, who were harmed by arrogance (superbia). Augustine thus admonishes slaves to serve ‘not with the slyness of fear, but with the fidelity of affection, until all injustice disappears’ at the end of this world. In his defence of the earthly order and slavery, Augustine leaves it open as to why slavery, as God’s punishment, is the lot of some people and not others (since, according to him, all humans are guilty of sin). He states vaguely that all humans are enslaved to something (for example, some to lust), so physical slavery is only one type of subordination that all human beings suffer as punishment.47 From Augustine’s argumentation, we can uncover voices that are critical of natural slavery (or slavery in the postlapsarian world order). Against these, Augustine appears to defend his views. For example, Steven Epstein reconstructs several points of criticism to which Augustine seems to reply. First, it was not God’s intention that humans would subordinate each other. Augustine agrees but explains domination and slavery as resulting from human sin. The second objection is that slavery is unjustly imposed on innocent people. In reply, Augustine seems to infer that no human is innocent, and there is the implication that a person who becomes enslaved must be guilty of some sin. The third point of criticism is that there is no justice in the enslavement of the defeated in war. Augustine’s response is that it must be the just consequence of sin. The fourth objection to slavery is that it is not the judgement of God but only a human institution. As we saw above, Augustine tries to balance between human responsibility and God’s providence. As everything must be happening according to God’s providential plan (even the Fall), slavery must also be part of the divinely ordained hierarchy. Augustine could have criticized slavery as one of the selfish and sinful inventions of humanity, but instead he legitimates it as an institution ultimately coming from God and as beneficial to the subordinated. As to the question why good Christians must be slaves for sinners, Augustine 47

Aug. Civ. 19.15. Augustine resembles the Roman tradition, exemplified by Cicero (Rep. 3.35), who argued that slavery was for the good of some people. For Augustine’s views of slavery, see also Casias 2018.

98

Maijastina Kahlos

gives an answer that he means to be comforting, namely, that it is easier to be a slave to a person than to lust.48 Another example of balancing between the traditions of natural slavery and legal slavery is found with Gregory, the bishop of Rome in the late sixth century. He wrote that from the beginning, nature (natura) allowed humans to be free, but the law of nations (ius gentium) had subjected them to the yoke of slavery. Gregory echoes the Roman legal tradition that, as we saw above, did not regard slavery as natural but as a legal condition.49 Gregory stressed the fundamental equality of all humans. However, while all were equal by nature (omnes homines natura aequales genuit), by God’s secret dispensation (dispensatio occulta) they were set in relation to an order of merits, and some were subordinated to others. The hierarchy was ordained according to each individual’s sin (ex vitio).50 Gregory’s balancing between the principle of natural equality and the inequality in the worldly order goes back to Augustine. Both thinkers justify slavery and hierarchy by means of sins and merits and by the idea that slavery is for the good of the subordinated. Both Augustine and Gregory emphasize good master–slave relations as the basis of peace and order in the household at the micro level and in society at the macro level, and both see domestic stability and civic tranquillity as going hand in hand.51 Gregory stresses the spiritual benefits for slaves and the responsibilities of masters for their spiritual well-being.52

ham and esau: criteria for enslavement? In their paternalistic discourse, Augustine of Hippo and Gregory of Rome made it clear that for some people being a slave was only beneficial, while being a master was a much heavier task to fulfil. Likewise, in his De republica, Cicero stated that slavery is beneficial for some people. Weakness of intellect or other weaknesses had also led some Stoics to

48

49

50 51

52

Aug. Civ. 19.15: Et utique felicius servitur homini quam libidini; see Epstein 2001, 141– 3. As Epstein points out, Augustine is ‘profoundly conservative’ and ‘accepts everything that exists in human history, including slavery, as the just working out of God’s plan’. Greg. Mag. Ep. 6.12: . . . homines, quos ab initio natura liberos protulit et ius gentium iugo substituit servitutis. On Gregory’s views on slavery, see Serfass 2006. Greg. Mag. Mor. 21. The interdependence of domestic stability and civic tranquillity was also in the interest of the Roman philosopher Seneca, who for all his humanitarianism was keen to maintain the social order in his own household as much as in Roman society. Serfass 2006, 78–85.

Late Roman Ideas of Ethnicity and Enslavement

99

justify hierarchies in which more intelligent persons ruled those perceived to be weaker.53 By commenting on the biblical tales of the curse of Ham and the brothers Jacob and Esau, a few late antique Christian writers grappled with similar questions of who was particularly fit to be a slave and why.54 The curse of Ham originates from the account of his father Noah’s drunkenness (Gen. 9:18–27). Noah’s sons panic when they see their father lying drunk and naked, and Shem tries to resolve the embarrassing situation by covering him up. Afterwards, Noah curses Ham’s son Canaan: ‘Cursed be Canaan; lowest of slaves shall he be to his brothers.’55 The argument concerning Esau comes from the tale of the twin brothers Jacob and Esau. Esau sold his birth right to Jacob, and Jacob also received the blessing of the firstborn from their father. The result was that Esau and his descendants were destined to serve Jacob and his descendants (Gen. 25:29–34; 27:29). When commenting on the tales of Ham or Esau, Christian writers acknowledged that there were many reasons for slavery. By contrast, the tale of Joseph and the enslavement of the Hebrews functioned as examples of unjust (non-natural) slavery. Origen argued that some, like the Hebrews, had faced slavery out of necessity (for example, because of war), but others were slaves by nature because they were touched by the curse of Ham. Consequently, the Egyptians, whom Origen regarded as Ham’s descendants, were prone to slavery. Origen justified their slavery through their inclination towards the immoral life, and he explained that they deserved ‘a judgment of this kind, that his son Canaan should be a servant to his brothers’.56 Basil of Caesarea explained slavery as resulting from different circumstances: one became captive or one ended up in slavery because of poverty. Basil also insisted that no one was a slave by nature. However, he gives yet one more reason for slavery, namely a lack of rational capabilities, which he explicates with the examples of Esau and Ham. Basil explains that this is only for the benefit of a person lacking in intelligence and self-control. This is why Jacob became Esau’s master. In Canaan’s case, the deficiency

53

54

55 56

Cicero, Rep. 3.35; M. Aur. Med. 5.30; Posidonius in Ath. Deipnosoph. 263c–d; see Garnsey 1996, 146–50, for other examples. For Philo of Alexandria, the case of Esau functioned as an exemplum of naturally suitable slaves; see Prob. 57; Leg. 3.88–104. Philo influenced both Basil of Caesarea and Ambrose of Milan; see Klein 2000, 50; 1988, 44–9; Garnsey 1996, 163–6. NRSV. Orig. Hom. 16 in Gen. 1; trans. Garnsey 1996, 44. On the early Christian discussion of the curse of Ham, see De Wet 2018a, 104–13.

100

Maijastina Kahlos

in intelligence (and, consequently, his subordinate condition) is considered hereditary: Ham lacked understanding, and this is why Canaan must become ‘the lowest of slaves to his brothers’.57 Likewise, Ambrose of Milan used the stories of the cursed Ham and the imprudent Esau as examples that justified the subordination of less rational individuals. Noah’s curse made Ham a domestic slave (servus domesticus) and Ham’s brothers the masters (fratres eius dominos), as Ham had acted foolishly (insipienter) and his brothers wisely (sapienter).58 Ambrose also describes Esau’s subordination to Jacob as the result of his folly (insipientia), while Jacob in his wisdom was worthy (sapientiae merito) of being his brother’s master.59 Ambrose explains that it is not nature that makes a human slave but foolishness. With the biblical imagery of Esau and Joseph, he distinguishes between genuine slavery and genuine freedom: Esau was born free but ended up a slave, while Joseph was sold into slavery but was able to remain virtuous, that is, genuinely free. Again, the subordination of the foolish brother under the wise one is framed as beneficial for the intellectually weaker of the two.60 In De Jacob et vita beata, Ambrose repeats the idea of irrationality as a reason for subordination and the advantage that the hierarchy accords to intellectually weaker humans.61 According to David Goldenberg, in a number of later Syriac Christian writings (falsely attributed to Ephrem the Syrian), the curse of Ham and the subordination of Canaan were associated with black Africans. The curse of Ham was quoted in the form: ‘Cursed be Canaan and may God make his face black.’ This tradition continued in later Islamic literature. The arguments for and the legitimation of slavery by means of biblical examples already had its beginning in late antiquity.62 57

58

59

60 61

62

Bas. Caes. Spir. 20; see also Bas. Caes. Hom. jej. 1.6 on Esau; Klein 2000, 49; Combes 1998, 63–4; Garnsey 1996, 45. Ambros. Ep. 7.6: Nonne primus Noe quam advertisset quia insipienter filius suus Cham riserat nudatum patrem, maledixit ei dicens: ‘Maledictus Cham, servus domesticus erit fratribus suis, et praeposuit ei fratres eius dominos, qui sapienter patris senectutem honorandam putarunt; see also Klein 1988, 17–21, 94–7; 2000, 50; Garnsey 1996, 193–7. Ambros. Ep. 7.7: Nonne etiam fons ille disciplinarum omnium Iacob ubertatem disputationis huius omnium pectoribus infudit, qui sapientiae merito praelatus est fratri seniori? also Ambros. Ep. 7.8: Diligens ergo pater et consulens, servum eum fratri suo fecit, ut eius regeretur consilio. Ambros. Ep. 7.9. Ambros. Jacob 2.3.11: Recte ergo eum subdidit, ut regentis imperio suum melioraret affectum; see also Ambros. Ep. 20.3–8. Goldenberg 2005, 99–101, 134–5. Harper 2011, 90–1, argues that the Arab slave trade in Africa had its precursor in Roman trade routes, which were established in antiquity and survived into the Islamic period; on Roman trade routes in Africa, see Bradley 2004, 312.

Late Roman Ideas of Ethnicity and Enslavement

101

how to recognize a slave? late antique physiognomy We discussed above the exoticism linked to some groups of slaves, as in the case of the fair-haired Germani. However, slaves in Roman and late antiquity were not intrinsically connected with a particular marker, such as skin colour. No one would have distinguished a slave from a free person only by bodily markers.63 That being said, ancient literature, including late antique writings, is nonetheless filled with servile characterizations.64 On the one hand, Greek and Roman authors, especially drama writers, used stereotypes when depicting slaves or introducing slave characters in their plays, for example, representing them as weakly figures, or with ruddy faces and red hair. On the other hand, ancient writers frequently connected certain physiognomic features to the slavish character in general – such as thick and heavy legs, small nose, small blue eyes, red hair, or pale skin – features that, as J. Albert Harrill remarks, the Mediterranean Greeks and Romans traditionally regarded as deficient and associated with Northern barbarians (like Celts, Scythians, Thracians, and Germani).65 Thus, in polemical contexts, free persons were vilified with physiognomic traits that hinted at a servile disposition, and all this polemic was aimed at exposing an opponent or other as someone who lacked proper manhood, morals, and status. Several physiognomic treatises introduced theories in which certain somatic features indicated permanent psychic and intellectual qualities, and a ‘slavish disposition’ with lack of intelligence and selfcontrol was one of them.66

the limits of the slave trade Ecclesiastical writers took slavery as an inevitable yet lamentable part of contemporary society, pertaining to the earthly order of things. The imperial government protected the slave trade in general but severely prohibited its trafficking of free Romans. In a similar vein, Augustine 63

64

65

66

Harper 2011, 26; Gruen 2011, 210, also stresses that freed slaves were admitted to the citizenry. However, the possibility of emancipation only became a reality for a few slaves. Furthermore, ancient Greeks and Romans also had negative connotations of black skin colour, e.g. as the metaphor of darkness and as an ill omen; see Goldenberg 2009, 91– 102; McCoskey 2012, 23–4. Anon. Physiognomonia 24; 51 (ed. Förster 1893, 39, 70); for a discussion, see Harrill 2006b, 41–2. For the physiognomic literature, see Goldman 2015; Ziegler 2009, 181–4.

102

Maijastina Kahlos

simply stated that slaves should obey their masters, but free Romans should not be enslaved.67 Late antique bishops were against the enslavement of freeborn Romans and actively took part in ransoming Roman captives of war. Likewise, they vehemently condemned slave traders who kidnapped people in the provinces within the Empire. In a letter to Alypius in 428, Augustine complained about the ‘Galatian’ slavers who harassed the North African countryside, snatching free Romans.68 He describes how a multitude of mangones (slave traders) is ‘exhausting the greater part of humankind and exporting what they sell into transmarine provinces’. What worried Augustine was not the slave trade as such – that was the norm – but rather the fact that ‘almost all of these are free’ (paene omnes liberos).69 That slave-raiding and trading had extended into the Roman territory of the freeborn was entirely unacceptable and illegitimate in the eyes of the imperial government, bishops, and the general public. Kidnappers and merchants organized slave-hunting and raiding in close cooperation. The mangones’ predatory activities against provincial Romans were illegal, as Augustine stresses. This is reminiscent of Emperor Honorius’ nearly contemporaneous edict in which impious merchants (talisque impietatis negotiatores) were sentenced to flogging and banishment into exile. The ‘Galatian’ slave merchants nonetheless enjoyed strong protection at high levels, and Augustine points out bitterly that the ‘Galatian’ traders did not lack patrons.70 Augustine speaks of ‘real slaves’ (veros autem servos) in distinction to the ‘free’.71 Some are real slaves and others are really free persons. For Augustine, the ‘free’ that should be left untouched are Romans in the

67

68

69

70 71

Aug. Ep. 24.1* to Eustochius: quod possumus secundum apostolicam disciplinam, ut dominis suis sint subditi, servis praecipere, non autem liberis iugum servitutis imponere. In the letter, Augustine is concerned about children from mixed unions between freeborn persons and slaves. Aug. Ep. 10*. For discussions, see Garnsey 1996, 63; Glancy 2002, 71–3; Harper 2011, 92–4; Vuolanto 2003, 178–9; Conant 2012, 299–300. Aug. Ep. 10.2*: Tanta est eorum qui vulgo ‘mangones’ vocantur in Africa multitudo, ut eam ex magna parte humano genere exhauriant, transferendo quos mercantur in provincias transmarinas et paene omnes liberos. For another possible interpretation, reading liberi as children, see Evans Grubbs’s chapter in this volume. Aug. Ep. 10.2–3*; see also Aug. Ep. 10.8*: non enim desunt patroni Galatis. Aug. Ep. 10.2*: Nam vix pauci reperiuntur a parentibus venditi quos tamen non ut leges Romanae sinunt ad operas viginti quinque annorum emunt isti, sed prorsus sic emunt ut servos et vendunt trans mare ut servos; veros autem servos a dominis omnino rarissime. As Harper 2011, 355, remarks, Augustine had ‘an internalized habit of thought, a belief in the legitimacy of the status system’.

Late Roman Ideas of Ethnicity and Enslavement

103

Roman provinces. He makes a clear distinction between provincial Romans and barbarians: The barbarians are resisted when the Roman army is in good condition for fear that Romans will be held in barbarian captivity. But who resists these traders who are found everywhere, who traffic not in animals but in human beings, not in barbarians, but in Romans from the provinces (non quorumcumque animalium sed hominum, nec quorumcumque barbarorum sed provincialium Romanorum)? Who will resist in the name of Roman freedom – I shall not say the common freedom but their very own.72

Augustine stresses his interest in Roman freedom (libertas Romana), not freedom in general. He also draws an ambiguous parallel between trading animals and humans and trading barbarians and Romans. Whether Augustine’s deliberate intention was to bestialize barbarians, we cannot know; in any case, a late antique reader could get that impression. Animalization of barbarians indeed was not rare in ancient and late antique literature. For example, the fourth-century orator Symmachus articulately compared the Alamanni in the battle to ‘gazelles swiftly fleeing in the arena’ and ‘stags hunted out from their hidings’. As Cristiana Sogno shows, in his panegyric Symmachus depicts the barbarians as if in a venatio, the hunting spectacle of the amphitheatre.73

conclusion This article discussed how late antique writers viewed the enslavement of non-Romans and non-Greeks (in the ancient parlance, ‘barbarians’). The discussion was started with a quotation from Synesius of Cyrene, who judged the Goths to be the fittest and most suitable genos to serve the Romans. It is difficult to determine how representative his claim was in the later Roman Empire, but at least he expected his audience to understand his polemical procedures and to find them convincing. He also appealed to the view that regarded some people as suited to slavery and others suited to act as masters. Synesius was able to articulate ideas that would discredit the Goths and best persuade his audience. 72

73

Aug. Ep. 10.5*: Et barbaris resistitur, cum bene et prospere geritur Romana militia, ne barbarica Romani captivitate teneantur; his vero negotiatoribus non quorumcumque animalium sed hominum, nec quorumcumque barbarorum sed provincialium Romanorum usquequaque dispersis, ut in eorum manus pretia pollicentium vel violenter rapti vel insidiis decepti ubicumque et undecumque ducantur, pro libertate Romana (non dicam communi sed ipsa propria) quis resistit? trans. Garnsey and Humfress 2001, 96. Symm. Or. 2.10–12; see also Sogno 2011.

104

Maijastina Kahlos

The Roman legal tradition recognized slavery as a human institution based on social conventions. The Roman jurisprudents even acknowledged the equal status of slaves as human beings although, in their social status as slaves, they were subordinate to their masters. Human beings became slaves by the whims of changing fortune. Many Christian writers embraced this ‘liberal’ tradition, as their empathic exhortations to slaveowners demonstrate. They appealed to the common humanity of all when advising listeners to treat their slaves in a Christian manner. The same Christian writers tried to address the problem of the institution of slavery in relation to the Fall. Augustine was the most influential ideologue to formulate ideas of enslavement. On the one hand, slavery was a social and earthly phenomenon that did not exist in paradise before the Fall, but on the other hand, slavery also represented divine punishment for sins, and thus it was part of the divine order. Therefore, for Augustine slavery was more than a socio-historical institution. In his theory, the issue of why some were slaves and others were masters remained vague, as he explained that all suffer from enslavement as punishment for sin. Late antique Christian writers wrestled with the issue of slavery by discussing the biblical exempla of Ham and his son Canaan, as well as the tale of Jacob and Esau. In their commentaries on these narratives, Basil of Caesarea and Ambrose of Milan argued that Ham and his son Canaan, along with Esau, were enslaved because of their foolishness. They were subordinated to their intellectual superiors, and this arrangement was for the sake of their own good. These commentaries came very near to the theories of natural slavery in which the nature of being a slave was hereditary (for example, the error and curse of Ham were transferred to Canaan). Exempla from stories of the early phases of humankind could thus be used to justify slavery in the present. In later historical periods, these Christian paternalistic arguments and the authority of church fathers provided justification for slavery, since bondage was claimed to be beneficial to the slaves themselves.

part ii SLAVERY, CULTURAL DISCOURSES, AND IDENTITY

5 Slavery in Euphemia and the Goth Chris L. de Wet

(a) ‘Trafficking in persons’ shall mean the recruitment, transportation, transfer, harbouring or receipt of persons, by means of the threat or use of force or other forms of coercion, of abduction, of fraud, of deception, of the abuse of power or of a position of vulnerability or of the giving or receiving of payments or benefits to achieve the consent of a person having control over another person, for the purpose of exploitation. Exploitation shall include, at a minimum, the exploitation of the prostitution of others or other forms of sexual exploitation, forced labour or services, slavery or practices similar to slavery, servitude or the removal of organs . . .. (c) The recruitment, transportation, transfer, harbouring or receipt of a child for the purpose of exploitation shall be considered ‘trafficking in persons’ even if this does not involve any of the means set forth in subparagraph (a) of this article. United Nations Protocol to Prevent, Suppress and Punish Trafficking in Persons, Especially Women and Children (2000)

Located somewhere between ancient fiction and Christian hagiography, the story of Euphemia and the Goth recounts the tale of a young girl from Edessa, Euphemia, whose widowed mother, Sophia, is deceived and manipulated into letting an unnamed Gothic soldier marry her daughter. Later, as the story develops, events take a turn for the worse, and the pregnant Euphemia is taken away to the Goth’s homeland, only to find that he is already married. She is then given as a slave to the Goth’s wife, and suffers terrible abuse before being miraculously rescued. The story is set in ca. 395 ce Edessa, in the context of the invasion of Mesopotamia by the Huns, although possibly composed decades later in the fifth century. As in some other cases from Syriac literature, Euphemia 107

108

Chris L. de Wet

and the Goth reads in many ways like a tale from the genre of the Greek novel. The text exists in Syriac and Greek versions, but Syriac was probably its original language of composition. We have an edition of the Syriac text, with a helpful introduction, by Francis Burkitt.1 The author tells us that he heard the story from the elderly verger of the Shrine of the Confessors Shmona, Guria, and Habbib in Edessa. In this fascinating tale, Sophia and Euphemia are depicted as pious and admirable citizens of Edessa, who are unwilling and unfortunate victims of their circumstances. At the time in which the story is set, the Roman Empire sent its soldiers to Eastern frontier cities, like Edessa, to defend and fortify the city against attacks from the Huns. The billeting of soldiers was not an uncommon occurrence in this period and region.2 When Sophia is compelled to house a Gothic soldier in her home, she decides to hide her young and beautiful daughter from the eyes of this man. The persistence of the Goth is repeatedly noted in the work: And when the Goth had been a long space of time with them it fell out that for an instant he saw the girl [Euphemia] and was greatly inflamed with the desire of her, and with the love of her his soul was taken captive; and her daughter Euphemia in marriage to be his wife. But her believing mother when she heard it was much agitated and was afraid of him, and did not accept his suit. But that bad man did not cease from vexing her with disturbing words in this way, and sometimes it was with fury that he met her, and sometimes with words of gentleness and of flattery, and with mighty oaths he was wheedling her, and displaying gold too and showing it to her.3

Eventually, and after much persuasion, Sophia gives in to the Goth’s request, and allows the Goth to marry Euphemia. Not long after their marriage, Euphemia falls pregnant. Sometime thereafter, when the Roman troops are recalled, the Goth wishes to take his wife with him to his home (we are not sure where this is). But before they leave, Sophia

1

2 3

I thank Kate Cooper, Judith Evans-Grubbs, Marianne Kartzow, Florian Battistella, David Pitz, and Antti Arjava for their comments and feedback on this study. I will use Burkitt’s (1913) Syriac text and his translations, although slightly adapted and updated, as the basis for the text of Euphemia and the Goth in this study. Besides the first analyses of the text by Von Gebhardt and Von Dobschütz (1911) and Burkitt, and a very brief dictionary entry by Brock (2011a, 153), there are few secondary works focusing primarily on Euphemia and the Goth. Wood (2010, 95–100) discusses the text within the context of broader late antique Syrian political thought, while others like Harvey (1996), Horn (2018, 175–8), and Saint-Laurent (2012; 2019) focus on aspects of motherhood, children, and female sanctity (respectively) in the story. To my knowledge, there are no studies examining slavery in Euphemia and the Goth. Fear 2007b, 439. Euphemia and the Goth 6.16–7.9; Burkitt 1913, 131–2.

Slavery in Euphemia and the Goth

109

takes Euphemia and the Goth to the martyr-shrine of the holy Confessors Shmona, Guria, and Habbib, where Sophia asks that the martyrs act as sureties for the marriage and the protection of her daughter. The Goth, too, agrees to this and lays his hand on the coffin and makes an oath: ‘As I deal with her and do unto her, so may God deal with me! Lo, these Holy ones are sureties that I will not grieve her!’4 Unbeknownst to Sophia and Euphemia, though, the Goth already has a wife in a different city in the west. We read: [T]hen he rose up against her like a destroying wolf and stripped off her rich clothing that she was clothed with, and unloosed from her the gold with which she was adorned, and clothed her in the costume of a slave-girl. Then he revealed to her all the treachery he had practised on her, saying to her: ‘I have a wife and I have been married to her, you see, for a long time. But hold your peace, and do not reveal before her or before anybody else what has happened between us: otherwise you will die a terrible death at the hands of her family and tribesmen, for they are well known in our country.’5

After arriving at his home, the Goth gives Euphemia as a slave-girl to his wife. The Syriac term used for slave-girl in this text is ’amtā – Euphemia describes her service explicitly as cabduˉ tā , that is, slavery.6 Because of her jealousy, according to the author, and suspecting that Euphemia was pregnant by her husband, the wife of the Goth treats Euphemia with great cruelty. When Euphemia has a son, the child is poisoned by the Goth’s wife. In retaliation, Euphemia sets up the wife of the Goth to poison herself with the same poison. Euphemia is then taken and imprisoned in a tomb by the victim’s relatives. Like Dorothy in the land of Oz, it is only when Euphemia prays to the martyrs of Edessa that she is miraculously transported back to Edessa. Later, when the Goth returns to Edessa, his deeds catch up with him when acquaintances of Sophia, who know him, see him and take him to her. When he sees that Euphemia has miraculously returned to Edessa, he confesses all of his deeds, and is eventually, after an interesting trial, sentenced to death. Euphemia and the Goth also forms part of a broader cultural and textual tradition, namely that of the martyr cult of Shmona, Guria, and Habbib. These martyrs feature as the spiritual heroes in the story. Burkitt

4 5 6

Euphemia and the Goth 16.3–5; Burkitt 1913, 135. Euphemia and the Goth 15.12–22; Burkitt 1913, 135. See, for instance, Euphemia and the Goth 16.6–7, 18.13–19.15; Burkitt 1913, 136–7. On Syriac terminology related to slavery, see De Wet 2017.

110

Chris L. de Wet

argues, however, that Euphemia and the Goth was written at a time when the cult of these martyrs was already well established in Edessa. So while it shows continuity with Edessene martyrological literature in the general sense, it also stands at somewhat of a distance from traditions like those of the martyrs Shmona, Guria, and Habbib, and even more so from traditions like that of the Doctrina Addai. The text of Euphemia and the Goth is therefore unique in that it provides us with a captivating story that aims to delineate the poignant realities and dangers of war, especially to the vulnerable ones in society, women and children, and calls its readers to a position of solidarity with the Edessene community as being part of the greater Roman Empire. Slavery is one of the major structuring principles in the narrative, and the intersections between slavery, age, gender, and ethnicity are quite pronounced in the text. Euphemia’s tale betrays several harsh realities of being a young woman, not under the authority and protection of a paterfamilias, and a displaced foreign slave in late antiquity. But on a discursive cultural– religious level, slavery is also, at the same time, very closely associated, negatively, with the bonds of marriage and childbirth, a common theme in Syriac and Greek hagiographical literature of the region and era. Thus, my aim in this chapter is to analyse the author’s use of slavery in Euphemia and the Goth, both as a practice and as a discourse. When I speak of slavery as a practice in this regard, I refer to the status of being physically and socially enslaved to another person. In other words, what does Euphemia and the Goth tell us about the practice of slavery in the period, and how does it aid us in understanding the problematics and complexities of late ancient slavery? In my examination of slavery as a discourse in Euphemia and the Goth (which I have elsewhere termed ‘doulology’7), I will ask how slavery as a constitution of knowledge and a language produces specific meanings that reach beyond aspects of institutional enslavement. In this case, I will argue that the discourse of slavery in Euphemia and the Goth shows remarkable similarities to how slavery operates as an ascetical-theological discourse in other monastic and hagiographical texts of late antique Christian Syria and Mesopotamia, such as the Acts of Thomas.

coercion, deception, and cruelty: complexities of late ancient slavery in euphemia and the goth Despite its fantastical character, a hagiographical text like Euphemia and the Goth is, nevertheless, still a valuable source that can aid scholars to 7

See De Wet 2015, 1–44.

Slavery in Euphemia and the Goth

111

better grasp the problematics and complexities of late ancient slavery. Authors of texts like Euphemia and the Goth are required to weave the ordinary and plausible circumstances of daily life in antiquity with the fantastical and miraculous. I will delineate five premises from Euphemia and the Goth that serve to demonstrate the complexity of slavery in late antiquity.

The Means of Enslavement First, the case of Euphemia’s enslavement shows that the means of enslavement in late antiquity was pluriform, and that the enslavement of some individuals often entailed a complex and convoluted set of circumstances that are difficult to quantify and even qualify into one scholarly category. To elaborate on this premise, let us consider the usual categories for the means of enslavement often discussed in scholarship about ancient slavery. Along with home-born slaves, or vernae, at the very top, there are usually four other means of enslavement listed that contributed to the internal supply of slaves in the later Roman Empire, as Kyle Harper states: self-sale, the sale of children (and enslavement of the children pledged for their parents debts8), abduction, and child exposure. External means include captives of war, victims of raiding or piracy, individual kidnapping, and general trans-frontier trading.9 The form of enslavement in Euphemia and the Goth, however, attests to the complexity and opaqueness of these categories. Euphemia’s enslavement does not fall neatly into any of the means of enslavement just listed, and there is nothing ‘legal’ about the eventual enslavement of Euphemia. Technically, and in Roman legal terms, the case of Euphemia is not one that entails bride abduction, and strictly speaking, it is not even kidnapping in the traditional sense of the word. Nor was Edessa raided in the story, and Euphemia taken captive in this manner. As the bride of the Goth, Euphemia is not abducted against her mother’s consent, nor does she elope. Although the consent was coerced, Sophia does give permission for them to marry, and she also consents to Euphemia leaving to go with her new husband, especially after Euphemia falls pregnant, which sometimes even mitigated cases of real bride abduction in antiquity.10 When the Goth reveals his treachery, Euphemia exclaims that the Goth is ‘a robber, 8 9 10

Vuolanto 2003. Harper 2011, 78–91. Evans Grubbs 1989, 62.

112

Chris L. de Wet

stealing away freemen by day’.11 The Goth is here called a robber (Syriac: lest ̣ā yā ) or kidnapper of free persons, simply for the lack of a better word. Euphemia’s might appropriately be called an ancient case of ‘enslavement by deception’. This means of enslavement seems to be quite common in modern instances of enslavement or human trafficking. A study on modern-day bride-trafficking in Southeast Asia has shown that deception was one of the most common recruitment tactics. Some of the quotes by these women include: ‘My lover deceived me to go to meet his parents’; ‘A stranger induced me to go to China to buy clothes’; ‘I was deceived to find a job in Lao Cai province.’ What is more disturbing is that most of the women were conned either by friends or by relatives. Some were also deceived by their (prospective) husbands, as one woman from the study explains: ‘My Vietnamese husband deceived me and trafficked me to China.’12 Thus, in the United Nations’ Trafficking Protocol, quoted at the start of this chapter, there is a clear emphasis on the prevalence of enslavement by deception.13 This complex means of enslavement is one that is neglected in the study of ancient slavery.14 While we may doubt the historicity of the story of Euphemia and the Goth, the events described in relation to Euphemia’s enslavement are nevertheless plausible. It had to be so for the intended audience, and it is especially plausible from the perspective of modern-day victims of human trafficking. The enslavement of Euphemia stands as a testament to the fear of enslavement ancient freeborn folk often harboured, especially the elites. This fear is prevalent in numerous ancient sources. Most famous is one of Augustine’s letters to his friend Alypius, whom he asked to solicit imperial aid against the enslavement of free persons in North Africa.15 Bandits and slave traders were attacking the villages in the surrounding areas in Numidia, and, according to Augustine, it was especially women and children who fell victim to enslavement. Most of the men were killed.

11 12 13

14

15

Euphemia and the Goth 16.5; Burkitt 1913, 136. Stöckl et al. 2017. United Nations Protocol to Prevent, Suppress and Punish Trafficking in Persons Especially Women and Children 2000. For example, in two of the most recent and major reference works on Greek and Roman slavery, namely that of Bradley and Cartledge (2011) and Harper (2011), enslavement by deception is rarely discussed. Augustine, Ep. 10*; CSEL 88, 46–51; see also Harper 2011, 92–4, and the chapters of Kahlos and Evans Grubbs in the present volume.

Slavery in Euphemia and the Goth

113

During times of war, such fear was justified. But Euphemia and the Goth even disturbs the simplistic narrative of captivity and enslavement as a result of war. While it is often the case that war captives, especially women and children, were carried off to foreign lands and enslaved,16 Euphemia and the Goth shows us that the import of foreign soldiers into the Roman army caused numerous difficulties and possible ethnic tensions, including the oppression of local Romans by some in their own army. Euphemia is a different kind of war captive – she is deceived and captured by one of the soldiers who was actually meant to protect her. Thus, while the categories of means of enslavement and slave supply are not wholly unhelpful for us to understand something about the macrodynamics of slavery in antiquity, it should be remembered that these categories are scholarly constructions, and have their limits in the representation of the real experiences of the enslaved of ancient times. There is an urgent need, especially in scholarship of ancient slavery, to tell the unique stories of individual slaves, some who, like Euphemia and many modern victims of human trafficking, might have fallen victim to enslavement because of several events or conditions that cannot be neatly classified into one generic and cold ‘category’ of the means of enslavement.

Radical Asymmetrical and Intersectional Power Dynamics of Late Antique Slavery Second, the practice of slavery in Euphemia and the Goth testifies to the radical asymmetrical and intersectional power dynamics that undergirded late antique slavery. The author of the text masterfully uses these power dynamics as character contrasts in order to make it more shocking to the hearer, and to justify and valorise even more so the heroic miracle of the Confessors at the end. As Bernadette Brooten and Marianne Kartzow remind us, slavery is always shaped by the intersection of various aspects and elements of individual identity, including age, gender, race and ethnicity, culture, religion, and so on.17 This point is vividly clear in Euphemia and the Goth.18 Sophia and Euphemia are sketched as quite vulnerable persons in the Edessene community. On the one hand, they might have been wealthy and part of some stratum of the local elite – we read in the story that the Goth, 16 17 18

On war captivity and slavery, see Gaca 2010; Lee 2007, 123–46. Brooten 2013; Kartzow 2018a; 2010. See esp. Dimambro 2020.

114

Chris L. de Wet

before arriving at his village, ‘stripped off her [Euphemia’s] expensive clothes with which she was attired, and loosened from her the gold with which she was adorned’.19 On the other hand, Sophia and her daughter are devoid of the protection of a superior male. There is no paterfamilias; Sophia is a widow, and Euphemia is her only child. This particularly highlights the vulnerability faced by women, even elite women, who were on their own, so to speak, in such circumstances of war and conflict. Susan Ashbrook Harvey is correct to say that for Sophia and Euphemia, their ‘freeborn status is all they have’.20 Their vulnerability is contrasted with the strength and ruthlessness of the Gothic soldier. The rugged, unrefined, and quite un-Roman manliness of barbarians is often a point of discussion in some late antique sources. While he was a in Constantinople, John Chrysostom delivered a homily (via an interpreter) to a Gothic assembly. In this homily, Chrysostom draws a contrast between the traditional demeanour of the Graeco-Roman philosopher with the ‘wilder’ appearance of the Gothic congregants, ‘who exhibit façades more like lions than men’.21 The statement most likely refers to the seemingly leonine beards and hairstyles and large build of some of the Gothic men. The Goth in our story apparently also had a fierce temper. This is once again contrasted with the beauty and youth of Euphemia.22 Cornelia Horn has convincingly argued that Euphemia was probably a young adolescent, at least by modern standards, with a petite build.23 Horn further remarks: ‘Euphemia and the Goth helps to show us girlhood in the fifth-century Byzantine East as a state of ultimate deprivation of personal freedom, independence, and well-being.’24 The greatest contrast, however, is between the moral virtue of Sophia and Euphemia, and the unbridled vice of the Gothic soldier. Euphemia is sketched as a beacon of ideal Syriac Christian virtue, and all of the terms used to describe the great monks and ascetics of Syria, in Syriac hagiography, are also used to describe Euphemia. First and foremost, the virginity (btuˉ lā ) of Euphemia is stressed at the very beginning of the story.25 Moreover, the author also states that Euphemia was brought up in nakpuˉ tā , that is, modesty.26 This is also why Euphemia is hidden from the eyes of the foreign

19 20 21 22 23 24 25 26

Euphemia and the Goth 15.13–15; Burkitt 1913, 135. Harvey 1996, 41. Hom. Goth. 1; PG 63, 501.11–14. Euphemia and the Goth 5.14–16; Burkitt 1913, 131. Horn 2018, 176–8; see also Harvey 1996, 36–41. Horn 2018, 177. Euphemia and the Goth 5.10; Burkitt 1913, 131. Euphemia and the Goth 5.13; Burkitt 1913, 131; see also 34.14; Burkitt 1913, 146.

Slavery in Euphemia and the Goth

115

male in the house.27 Most interestingly, we also have find the Syriac term iḥidā yā in the text.28 This term is usually translated as single-mindedness, oneness, and being set apart in devotion to God, and is a prime characteristic of Syrian monks. In the context of 5.10, in which it occurs, it first seems to denote the fact that Euphemia is Sophia’s only daughter. Burkitt also translates it in this sense, but there might be a double entendre implied in the Syriac text, namely that Euphemia is not only an only-child, but also singlemindedly devoted to God. Another important word with monastic connotations that we find in the text is the term ’aksnā yā ,29 which literally means ‘stranger’, but which can also refer to a wandering Christian or a monk.30 We will return to these terms later in the chapter when discussing slavery as an ascetical-theological discourse. Although she is not a monk, Euphemia therefore seems to embody almost all of the expected ascetic virtues of early Syrian Christianity.31 The former two terms also imply that Euphemia had sexual honour. Even the wife of the Goth, when first seeing Euphemia, admits that she does not have a slave-girl’s appearance. The Syriac has the phrase lbuˉ šā d’amhuˉ tā , which in this case does not so much refer to the ‘garment of a slave-girl’ (which was removed from her), but her overall appearance or habitus. This reading is supported by the rendering of the phrase in Greek as scheˉ ma doulikon.32 The author of Euphemia and the Goth goes to great lengths to show that she is not of the servile class. It should be remembered that in traditional Roman thought, slave-girls were excluded from embodying these values, especially those of virginity and modesty. While early Christians did believe that slave women could remain virginal and modest, the virtues were usually not associated with the shameful status of being enslaved.33 Not only are Sophia and Euphemia idealised in terms of Syriac Christian piety, but they are also depicted ethnically and culturally as model citizens of Edessa and its civilised Syriac culture and language. The intersections and contrasts of ethnic identities and slavery are

27

28 29 30 31 32 33

Keeping young women hidden for the sake of their modesty was an important value in Christian antiquity. Cynthia Baker refers to this hidden and invisible nature and status of women as anopticism: Baker 2002, 62; see also De Wet 2015, 117–8. Euphemia and the Goth 5.10; Burkitt 1913, 131. Euphemia and the Goth 16.4; Burkitt 1913, 136. Caner 2002, 59. On the virtues of Syrian monasticism, see Brock 1973. See Burkitt 1913, Syriac page 52. De Wet 2015, 222–39.

116

Chris L. de Wet

a particularly prominent theme in the text. Philip Wood’s brief yet erudite discussion of Euphemia and the Goth highlights some important aspects of the uniqueness of the text compared to other Edessene literary works.34 Wood argues that Euphemia and the Goth portrays a vision of Edessene ethnic identity that is rather different from that found in, for instance, the Doctrina Addai. While Euphemia and the Goth is a text that is antimilitary, it does exhibit a more positive stance towards the Roman state, which is different in the context of the Doctrina Addai, a more elite-driven text, which ‘recalls the history of an independent Edessa’.35 The story of Euphemia and the Goth also idealises Edessene identity and solidarity, and the prominence of the Syriac language has a crucial role in the narrative. The figure of the Goth in this tale represents everything that Syrian Christian virtue and Edessene social and cultural solidarity oppose. The author of Euphemia and the Goth makes it very clear that the reason for the young Euphemia’s enslavement was the Goth’s lust. As the story then develops, Euphemia effectively becomes a child sex-slave, and she loses all of her sexual honour.36 The Goth is not named, but the mere mention of his ethnicity and status as a soldier would have called up various connotations to the ancient audience of the text. Von Dobschütz, one of the first scholars to work on the text, remarks that the negative depiction of the Goth should be associated with the ancient perception that all Goths are Arians, which would have been abhorrent for the orthodox of Edessa.37 This view, however, has been rightly challenged by Burkitt,38 and does not stand up to criticism. The religious convictions of the Goth are never mentioned, and the view that all Goths were Arians, or considered to be so, has been refuted.39 Burkitt is correct to rather point to the negative social, cultural, and moral perceptions of foreign soldiers in Christian literature from the East. Burkitt notes the numerous descriptions by Syrian provincials of Goths as being morally reprehensible and uncontrolled in their appetites – slaves to 34 35 36

37

38 39

Wood 2010, 95–100. Wood 2010, 99. Theodoret of Cyrus relays a similar story, in which a local and highly lustful general attempts to abduct a young girl to be his bride. The girl flees to a convent, and after numerous failed attempts, and after being struck with blindness, the general ceases to pursue the girl; Theodoret, Hist. rel. 9.12; SC 234, 426–30. See the introductory remarks of Von Dobschütz in Von Gebhardt and Von Dobschütz 1911, lv. Burkitt 1913, 58–9. See esp. Stanfill 2015.

Slavery in Euphemia and the Goth

117

passion, as many ancients would say. In fact, the text states that the Goth’s soul was ‘taken captive’ by his lust for Euphemia.40 Few are more vocal than the fifth-century writer, Pseudo-Joshua the Stylite, who bemoans the fact that the Gothic soldiers in Edessa were so numerous that they caused social chaos in the true sense of the word: They ejected others from their houses, going in and living in them. Others’ cattle they led away by force as if plundering (an enemy). They stripped some people’s clothing off them and took it away. They used rough treatment on others for the sake of (obtaining) anything whatever. In the streets they and others for the smallest reason. They brazenly plundered the meagre provisions which everyone had, and the stockpile belonging to a few individuals in the villages and cities. They attacked many on the roads. And because there were not enough shelters and inns in the city for them, they stayed with craftsmen in their stalls. In full view of everyone they had their way over the women in the streets and houses. They took oil, wood, salt, and other things for their own needs from the old women, widowed or poor, and they stopped them doing their own work in order to serve them. In a nutshell, they oppressed everyone, nobles and commoners, and no one escaped receiving some (bit) of their wickedness.41

The Gothic soldiers are portrayed as ravaging the Edessene community, and the poor and vulnerable were especially targeted. Pseudo-Joshua continues that the Goths were so numerous that they even had to stay in monasteries, where they disrupted the spiritual discipline and routine of the solitaries.42 Pseudo-Joshua especially vilified the Goths for their excessive eating and drinking habits, their sexual vice, and their unbridled violence.43 Euphemia and the Goth follows in this negative depiction of Gothic identity, and the fact that the Goth remains unnamed in the narrative possibly betrays the stereotype of the Gothic soldier assumed by the author. It was enough, for the author, to simply state that the antagonist was a Gothic soldier.44 As an aside, we should also consider the possibility that the author does not name the Goth in order to erase his memory from history, a damnatio memoriae, as it were. This is a possibility when one considers that, after his 40 41

42 43

44

Euphemia and the Goth 7.20; Burkitt 1913, 131. Pseudo-Joshua the Stylite, Chronicle 86; in Trombley and Watt 2000, 103–5; see also Burkitt 1913, 59–60. Pseudo-Joshua the Stylite, Chronicle 95; in Trombley and Watt 2000, 114. Pseudo-Joshua the Stylite, Chronicle 96; in Trombley and Watt 2000, 114–15; see also Burkitt 1913, 59–60. We should not assume that these negative stereotypes of barbarians were universal. In Gaul, for instance, we have the interesting case of Salvian of Marseilles, who actually reprimands his fellow Christians for their wicked and debauched behaviour with their slaves, and goes to great lengths to argue that not even barbarians abuse their slaves in such a manner; see esp. De Wet 2019.

118

Chris L. de Wet

trial, the Goth is not only punished with the death sentence, but the Edessene citizens also want to burn his body.45 The burning of a criminal’s body had a tradition in the Roman Empire, and in this case it might have been seen as a fitting punishment.46 The burning of the Gothic criminal’s body, firstly, would have signified a retaliation for the shameful violation of enslavement he wrought on Euphemia’s freeborn body: ‘And you were contemptuous of the sacred laws of the Romans . . . bridling a free person with the yoke of slavery.’47 The burning of his body also would have totally erased his existence and memory from the world. The Goth’s body is not burned, but perhaps the story omits his name for the same purpose. The fact that the Goth’s illegitimate male infant with Euphemia is also not named supports the idea that the Goth and his legacy or progeny are not worthy of the honour of memory. While violence against ‘legally’ enslaved persons was not always severely punished,48 illegally enslaving the free was met with the harshest condemnation, as we see here in the case of the Goth. These asymmetrical intersections of power are present throughout the story, and they in fact intensify as the story progresses and as the cruelty against Euphemia increases when she is in the Goth’s village. The restitution of power only comes at the end of the story, when the Confessors rescue Euphemia, and when the legal resolution of the Goth’s crimes is given. The story shows the important juridical role of the Confessors – they step in as guarantors when oaths are made and broken between individuals, and they are especially helpful in bringing justice when Roman law has reached its limits, as in the case of Euphemia’s deception. The Confessors are not set against Roman law by any means; rather, through their miraculous interventions, they help to enforce Roman law, such as freeing a wrongly and illegally enslaved freeborn girl and Edessene citizen and cultivating favourable circumstances that would lead to the prosecution and punishment of the criminal.

Ethnic and Linguistic Alienation This discussion about the intersections between slavery, ethnicity, and culture (and the contrast between the ethnic identity of Euphemia and that

45 46

47 48

Euphemia and the Goth 44.16–18; Burkitt 1913, 152. Haas (1997, 87–8) discusses this practice with reference to a similar social response in Alexandria. Euphemia and the Goth 44.11–15; Burkitt 1913, 152. Harper 2011, 204–9.

Slavery in Euphemia and the Goth

119

of the Goth) brings me to my third premise. Euphemia and the Goth illustrates the intense and traumatic ethnic and linguistic alienation that many enslaved persons of late antiquity experienced. This ethnic and linguistic alienation should be seen as an extension of the natal alienation entailed by slavery, as Orlando Patterson has famously shown.49 At the start of the narrative, Euphemia is utterly silent and invisible, almost commodified. She has no say in her prospective marriage to the Goth; her mother speaks on her behalf in all instances. Euphemia’s silence is, in a word, deafening. Euphemia’s first words are spoken only after the Goth reveals that he is married, and that she is now a slave-girl. Her commodification increases. While Euphemia’s words are directed to the Goth, they are a petition before God and the Confessors: Thank you for your kindness, man, for what you have done for me and for my status as a stranger here! Thank you for your kindness, robber, kidnapping freepersons by day, that you have revealed to me that I am a slave-girl and have bound me with the yoke of slavery, and have not killed me with your sword! These are your promises! This is the covenant of your oaths! So, then, I will call to the sureties that stand between me and you, and that mighty Power which is hidden in the bones of the Confessors.50

In this moment, Euphemia is given a voice, and in this instance she embodies the meaning of her own name. The Greek word, eupheˉmia, carries a sense of religious reverence, praise, worship, or blessing.51 The term can even be used to signify a type of prayer. Thereafter, however, the young Euphemia is cast into a strange and foreign world, one in which she does not understand the language or the customs.52 The text also seems to imply that the Goth’s house was, perhaps, in an outlying rural village, since they were geographically far from a local magistrate or governor.53 Moreover, the prospect of bringing home an exotic and foreign Syrian slave-girl might have been quite appealing to the Goth. When he meets his real wife, he emphasises Euphemia’s ethnicity: ‘She is your slave, and I have brought her for you from Syria.’54 The

49 50 51 52

53 54

Patterson 1982, 13. Euphemia and the Goth 18.2–11; Burkitt 1913, 136. LSJ; Lampe 2010, 578. This was most likely an experience that many slaves, especially children, experienced in late antiquity. For more on this, see the chapter by Pudsey and Vuolanto in the current volume. Euphemia and the Goth 24.20–1; Burkitt 1913, 14. Euphemia and the Goth 18.9–10; Burkitt 1913, 137; see also Dimambro 2020, 321–3.

120

Chris L. de Wet

linguistic alienation from her Syriac mother tongue is made explicitly clear: Now her mistress used to speak with great hatred and great enmity to her, and was using her with indignity and boxing her ears continually. And the girl did not know how to speak in her language or to appeal to her about anything, but only was weeping with sighing and calling the Confessors to her help for she was longing for somebody to speak Syriac with her, and there was no one but that Goth who had taken her away from the Syrian country, as from the length of time he had been in Edessa he had learnt to speak.55

This text highlights how ethnic and linguistic alienation enforced the bonds of domination within domestic slaveholding contexts. When the slave and her mistress are unable to communicate, it means that communication must take place through other means. The communication of tasks and duties might have occurred through simple and impatient (and perhaps physically violent) gestures, and the fact that the slave is unable to talk back or ask questions enforces the domination of the slaveholder. It might also imply that such a slave was only given the most menial, simple, and undignified of tasks, as the text also states. In the case of Euphemia, it is the language of abuse and cruelty that the mistress speaks. The author uses the language of boxing or assault to describe the nature of the communication between the mistress and Euphemia. The fact that the Goth himself is the only one who can speak and understand Syriac also affords him a great measure of power over Euphemia. Everything she might say could be filtered and censored by her master, the Goth. Slavery in Euphemia and the Goth is depicted not only as the denigration and subjugation of a young girl from Edessa, but it is also seen as an assault against the venerable leshā nā suˉ ryā yā , the Syriac language itself. In sum, a young slave-girl like Euphemia, in a foreign land, will always be the outsider, will always be alone, and will always be subject to alienation.

Slave Sexuality and Reproduction Fourth, we also see that slave sexuality and slave reproduction is by no means simple in the late antique context. It is often stated that slaves in Roman society were encouraged to ‘breed’ in order to increase the number of slaves in the household. In this regard, slaves are often referred to as ‘reproductive capital’.56 This was certainly not untrue for the later Roman Empire, in which 55 56

Euphemia and the Goth 19.20–29; Burkitt 1913, 137–8. Kartzow 2015.

Slavery in Euphemia and the Goth

121

we have evidence of high numbers of homeborn slaves being traded.57 But slave sexuality and reproduction was also a complex matter, especially when the master of the slave sexually assaulted enslaved women, which could result in pregnancies. In such cases, the infant remained a slave and, even if adopted, such a person had very limited rights. The rise of Christianity further complicated matters, since slaves were now also expected to show modesty and sexual chastity.58 Many Christian writers strongly opposed the sexual abuse of slaves – which is, of course, proof that it did happen. We also see that slave ‘marriages’, or contubernia, are strictly regulated by Christian authors.59 The case of Euphemia clearly shows the difficulties related to sexuality that could have arisen in domestic slaveholding contexts. The Goth, the master of the house, brings home a pregnant slave-girl, which immediately arouses suspicion with his wife. When a baby boy is born, the wife’s suspicions seem to be confirmed: And when the days had come near for the girl to bear, there was born to her a baby boy, and he was like his father exceedingly; and when the wife of the Goth saw that the baby looked like her husband she was struck with envy, and in great indignation she said to her husband; ‘Look and see how like he resembles you! In this case, therefore, cavilling and lying are of no use to you!’ And when many times in indignation she had said this to him, he said to her: ‘You have authority over her; everything you wish to do, do it, for she is your slave.’60

The problem of slave sexual abuse complicated slave reproductivity in antiquity. In this case, due to her jealousy, as stated by the author, the wife of the Goth is not at all concerned about having a prospective additional slave in the household (Euphemia’s son). The fact that she assumes unfaithfulness on the part of her husband drives her to abuse Euphemia even more. Before the child was born, we read that Euphemia is given hard labour to perform, probably with the aim of causing a miscarriage. The climax of the abuse is the murder of her child through poisoning. Interestingly enough, the Goth, who is the father, shows no regard for his illegitimate child with Euphemia. He also hands Euphemia over to the mistress of the house, and assumes no further responsibility for her. The uncertainty and vulnerability faced by an enslaved woman who bore a master’s child must have been incredible. The pregnant slave-girl is a major anxiety to the domus (and domina) in this instance, and there

57 58 59 60

Harper 2011, 264. De Wet 2015, 225–39. De Wet 2015, 237–8; Brooten 2013. Euphemia and the Goth 20.13–23; Burkitt 1913, 138.

122

Chris L. de Wet

was very little protection for the children of slaves in such circumstances. We therefore have a complex tension between economic and social factors related to slavery, and the case of Euphemia shows that ancient slavery was not always about the numbers.

Stereotypes of Slaves as Structuring Literary Devices Finally, the story of Euphemia and the Goth shows us just how dependent ancient authors, especially those writing narrative texts, were on using stereotypes to speak about slavery. However, there is an interesting tension in the utilisation of slave stereotypes in the story. The author actually seems to use common stereotypes about slaves to prove that Euphemia is not a ‘real’ slave, neither in her legal status nor in her character. The text incorporates the common stereotypes that all slave girls are promiscuous and that mistresses tend to be paranoid, jealous, and physically violent and uncontrolled towards their female slaves. Now while Euphemia, in the eyes of her mistress, appears to be promiscuous, the text intimates from the outset that she is actually the opposite. She was a chaste virgin who was tricked into marrying the Goth, and subsequently became enslaved and lost her sexual honour. The jealousy of the mistress is particularly evident in the narrative. The issue of mistresses beating their slave-girls out of jealousy was not a minor issue in the early Church. The fifth canon of the Synod of Elvira stipulated that a woman who, in a fit of jealousy (furore zeli), beat her slave to death must undergo seven years’ penance if it was inflicted on purpose and five years’ if it was accidental.61 The penalty for the murder of a slave-girl, in this regard, is quite lenient.62 But even so, the matter is more complicated since the woman does not kill Euphemia, but her baby. While scholars like Harvey and Horn have creatively read Euphemia and the Goth in order to show some of the dynamics between mothers and daughters in the Syrian Orient, the story perhaps betrays even more about the fragile relationships between enslaved mothers and their children, especially infants. The murder of Euphemia’s baby is, without a doubt, the most disturbing part of the narrative.63 When the innocent baby shows his affection to the Goth’s wife, it fuels her hatred even more, up to the point that she sends Euphemia to the market and, with the little baby boy at her 61 62 63

Dale 1882, 316; see also Laeuchli 1972, 43, 99–100. Laeuchli 1972, 76. Euphemia and the Goth 21–2; Burkitt 1913, 138–9.

Slavery in Euphemia and the Goth

123

mercy, poisons him by having him ingest poison on a handkerchief. This attests to how vulnerable and under threat slave-girls and enslaved children, especially infants, were in households where the lives of slaves mattered very little. Enslaved mothers constantly lived with the fear of being separated from their children through sale,64 and in Euphemia and the Goth we also witness that the master or mistress of the house might have posed a danger to enslaved children, especially if there was a chance that the master was the father. The story also exhibits one of the most extreme taboos of ancient Roman slaveholding, namely dominicide (that is, when a slave kills a master). Slaves were stereotypically considered to be untrustworthy and domestic enemies.65 Dominicide was usually punished in the harshest way, by not only killing the slave who committed the murder, but also the other domestic slaves.66 However, when Euphemia poisons the wife of the Goth, it is not exactly described as being murder or dominicide proper; after all, the virtue of Euphemia must remain intact. When Euphemia prepares the poison, which was, ironically, the same poison she took from the lips of her baby, which the mistress prepared, we read: So when she was serving them as a maid and was mixing drink for them, the lock of wool with which she had wiped the lips of her son she took out from her handkerchief and dipped it in the cup of wine, saying: ‘I shall see whether she made my son lick anything that he died: and if not, I shall know that it was by a death from God that my son died.’ And when she devised this, she mixed the cup and gave it to the Goth’s wife; and when her mistress took and drank that cup she too slept a deadly slumber, and in the pit that she dug, in it she fell. And the Goth’s wife died and was buried.67

Poison was considered the preferred method of assassination for women, as it was considered cowardly. The fact that the Goth’s wife uses it against an infant further adds to the shame and disgust directed at her character. But Euphemia does not plan to murder the mistress. Rather, when the mistress dies, it is God’s wrath that is exacted upon her for killing the infant. The author seems to show that, although it may appear to be the case, Euphemia is not guilty of the vices of slaves – yet the wicked masters still reap the fruit of their poison tree (no pun intended). Dimambro rightly argues that the women in the story are in fact 64

65 66 67

See the chapter by Pudsey and Vuolanto in the current volume for more details on the sale and separation of children from their parents. Harrill 2005, 147–62. See e.g. McKeown 2007b. Euphemia and the Goth 23.6–14; Burkitt 1913, 140.

124

Chris L. de Wet

depicted as the most violent and unable to control themselves, which again demonstrates the pervasive dynamics of power, gender, and enslavement in the narrative.68 To conclude this section, we see that the author of Euphemia and the Goth did not intend to write this story as a critique of slavery per se. Rather, Euphemia and the Goth is a critique against the illegal enslavement of the freeborn, on the one hand, and unjust domestic slaveholding practices on the other. In Euphemia’s prayer to God, she closes with the following words: ‘You care, O my Lord, for the enslavement of free persons.’69 While it is rare in late antique Christian discourse to find outright critiques of slavery as such, there are many critiques against kidnapping and illegal enslavement of the free, as Augustine’s correspondence to Alypius, for instance, shows. When the Confessors rescue and free Euphemia, they are not really freeing a slave from an unjust master, but rather restoring a freeborn noble-girl to her former status. Woven between the drama of the story’s novelistic character is the tragedy of a beautiful Christian virgin girl being enslaved by a wicked, monstrous, and lustful barbarian, whose rescue is impossible – as it was for many of the enslaved – unless by a miracle. Moreover, Euphemia and the Goth is aimed at demonstrating to its audience the powerful and binding nature of oaths taken before the Confessors, and the dangers of swearing falsely. When Roman law is unable to bring justice and call to account one’s oaths, the Confessors have the power to step in. Euphemia and the Goth could also serve as a critique against unjust domestic slaveholding practices. Late ancient Christian writers lauded slaveholders who were righteous and just, and heavily criticised those who were the opposite.70 The Goth and his wife are portrayed as typical degenerate and impious slaveholders. The fact that the Goth does not care for Euphemia after he brings her home, leaving her to the mercy of a jealous mistress, shows his disdain for the wellbeing of his slave. The actions of the Goth’s wife incorporate all the ancient prejudices against women as being irrational, without emotional control, suspicious, and conniving. I am doubtful whether the author of Euphemia and the Goth had any detailed knowledge about slaveholding practices among the Goths. His descriptions of the slaveholding practices in the Goth’s rural household and village seem so generic and stereotypical that it might as 68 69 70

Dimambro 2020, 326–8. Euphemia and the Goth 17.1–2; Burkitt 1913, 137. See, for instance, my extensive discussions in De Wet 2018a, 22–9.

Slavery in Euphemia and the Goth

125

well have been a conventional Roman household. Euphemia and the Goth, therefore, does not give us any insights into Gothic slaveholding practices, but rather constructs a Gothic slaveholding domus from the stereotypes of late ancient Roman slaveholding.71

slavery as discourse in euphemia and the goth There is, however, another dimension of slavery in Euphemia and the Goth that we have not addressed, namely how it functions, rather pervasively, as a discourse typical of that which we find in other Syrian Christian hagiographical and monastic treatises. In addition to its practices, slavery also operates on another, more discursive, level in Euphemia and the Goth. Slavery was a very common motif in early Syrian-Christian hagiographical literature. The notion of humiliation, or msarrquˉ tā in Syriac, was crucial to Syrian-Christian formulations of asceticism.72 The humiliation of slavery, in particular, was a useful foil to conceptualise the true nature of asceticism. Asceticism was the enslavement and humiliation of the flesh. Thus, monks and ascetics are most commonly depicted as c abdeˉ d’Allā hā , or slaves of God, and slavery was a central motif in the moral-theological discourse of descent and ascent that was so common in the early Christian literature of the East.73 We find numerous narratives that exhibit a pattern where a saint starts in a state of holiness, and then by some wilful action or another shows disobedience to God and descends from their state of holiness into a state of slavery and death. But the enslavement of spiritual descent then acts as a training ground that brings the saint to repentance and a subsequent redivivus takes place, in which the saint moves from spiritual (and sometimes physical) slavery to spiritual (but not always physical) freedom. The use of Philippians 2:6–11, in which Christ is said to descend and take on the form of a slave, was very influential in this regard. The motif of descent and ascent as enslavement and freedom is common in the works of individuals like Tatian, Bardaisan, and Origen. Even the Western tourist in the East, Jerome, makes use of this motif in his Life of Malchus.74 More than any other text from the East, though, the Acts of Thomas structures 71 72 73 74

For more on Visigothic slavery, see the chapter by Lenski in the present volume. Brock 2011b. De Wet 2018a, 22–9. In Jerome’s Life of Malchus, the Palestinian monk, Malchus, leaves his monastery to take care of family business. After he leaves, he is captured by so-called ‘Saracens’ and enslaved along with a woman; see Gray 2015.

126

Chris L. de Wet

its whole narrative around the notion of spiritual and physical enslavement.75 I therefore argue, further, that Euphemia and the Goth should also be read within this tradition that is so common in Syrian Christian hagiography. Euphemia, too, initially finds herself in a safe position of holiness. She is a beautiful and chaste virgin, and her mother intends that she remains as such. As mentioned earlier in this chapter, Euphemia embodies all of the virtues we associate with Syrian monks: virginity (btuˉ lā ), modesty (nakpuˉ tā ), being single-minded or being set apart for God (iḥidā yā ), and, perhaps most importantly, being a stranger or foreigner (’aksnā yā ) in this world. The figure of the foreign and alienated slave finds its pinnacle in Thomas, a foreign slave of God in India. Euphemia is also modelled according to this figure. The difference, however, is that Euphemia is not on a mission to evangelise the Goth and his village. Their wickedness seems to have consumed them already. Euphemia’s mission, as a woman, is to keep her virtue and trust in the Confessors and God. Euphemia therefore starts in a state of holiness and purity in Sophia’s household. But in a moment of weakness and under the force of the Goth’s persistence, Sophia (and not, technically, Euphemia) gives in to the danger of marriage and, thus, sexual reproduction. And almost in the wink of an eye, Euphemia becomes a slave, and before she knows it, she is surrounded by all manner of death. The text makes a point of noting the fact that Euphemia’s clothes and jewellery, which signified that she was freeborn, were ripped off and she was clothed in the garments of a slave-girl. The changing of clothes, from that of the freeborn to that of a slave, is also usually common in these narratives of descent and ascent.76 After enduring much violence and even losing her child, at her lowest point, Euphemia is in a tomb with stinking corpses. The pinnacle of her slavery is death. All this drives Euphemia to recall the sureties given in the presence of the martyrs, and at once she ascends not to heaven, but to Edessa (which is close enough), and she is back in the safety of her mother’s house and free again. Like the Acts of Thomas,77 Euphemia and the Goth affirms the notion that marriage is a pitfall and a form of slavery in itself, as opposed to the 75

76 77

The Acts of Thomas tells the story of the apostle Thomas, who is forcibly enslaved by Jesus himself in order to go and preach the gospel in India. It is a text that is considered to be one of the fundamental works for the development of Syrian ascetical theology; see Klijn 2003; Drijvers 2003. De Wet 2018a, 59–66. Glancy 2012, 3–21.

Slavery in Euphemia and the Goth

127

true slavery to God.78 Christian asceticism, especially that of Syria, was very critical of marriage and sexual reproduction. John Chrysostom writes: ‘For marriage truly is a chain, not only because of the multitude of its anxieties and daily worries, but also because it forces spouses to submit to one another, which is harsher than every other kind of servitude.’79 Sophia and Euphemia, especially, are tragic victims of this, and the text subtly warns against the dangers of married life and sexual reproduction. In this sense, Euphemia and the Goth almost acts as an antiromance. Euphemia and the Goth is therefore very much at home with the rest of Greek and Syriac Christian ascetical literature that uses slavery as a discourse to express the dynamics of servitude to God and the demands of holiness and the ascetic life.

conclusion The practice of slavery in Euphemia and the Goth attests to the numerous complexities of understanding late ancient slavery. While the story gives practically no details about the practices of enslavement among the Goths, the text does illustrate that the means of enslavement in late antiquity were often quite sinister, complicated, and unique to a particular individual. Most of all, Euphemia and the Goth stands as a chilling testament to the suffering and oppression enslaved persons in late antiquity endured, without much recourse to legal or even social protection. The slave, from the point of his or her infancy, was threatened, and even the innocent display of an enslaved baby’s love and affection to a jealous master or mistress could have resulted in abuse and even death. Linguistic and ethnic alienation were useful strategies to keep slaves docile and under the dominion of the master. The text clearly assumes the popular stereotypes of domestic slaves, but also shows that Euphemia is not guilty of these, because she is actually freeborn. As with most Christian texts of the time, the narrative shows no critique of slaveholding per se, although it does criticise the kidnapping and enslavement of free persons, and it also seems to bemoan the unjust and unchristian treatment of domestic slaves as opposed to proper and ‘just’ slaveholding practices. 78 79

On marriage as a burden in late antique ascetic writers, see Vuolanto 2015a, 181–4. Virg. 41.1.11–15; SC 125.236; translation: Shore 1983, 61; see also De Wet 2015, 99– 100.

128

Chris L. de Wet

In sum, if one reads between the lines, the text aims to show just how degenerate the world can be when there are no freeborn Christian-Roman males to take control – the text only has women, barbarians, and slaves as its main characters, excluding, of course, the holy departed martyrs. Therefore, the necessity that the martyrs, as the new patresfamiliae, take up the role of protectors and guardians of the unmarried and the vulnerable. Slavery is also a useful discursive tool for the author of Euphemia and the Goth to describe a life surrendered to marriage and reproduction. It aims to show that the slavery to God is a far better kind of slavery than slavery to the flesh, which involves marriage and procreation. On the one hand, it is used to describe the very heart of asceticism and devotion to God, and on the other, slavery is also likened to lifestyles that stand in contrast to the ascetic life. This type of language no doubt ratified the practice of slavery in early Christian culture. According to this view, it does not matter whether one is a real legally enslaved person or not; what matters is whether one is a slave to God or a slave to the passions of the earthly realm. Such discursive transformations of slavery, unfortunately, severely distracted early Christian thought, which was not able to see the true oppression and savagery of the practice of institutional slavery. What is perhaps even more disturbing is that this story, written almost a millennium and a half ago, witnesses to how easily the vulnerable of society – women and children – can slip into forms of enslavement and servitude, be it in ancient slaving practices or in the deceptive modern traps that lead to human trafficking, as is evident in the UN Human Trafficking Protocol. Euphemia and the Goth testifies further to how difficult, if not impossible, it is for enslaved persons to escape their circumstances – save for a miracle.

6 What Was Jewish about Jewish Slavery in Late Antiquity? Catherine Hezser

Just like other late antique societies, Jewish society in late antique Palestine was a slaveholding society in which slavery was a common phenomenon of daily life.1 Even though the proportional numbers of slaves within the population would not have reached the extent of Roman mass slavery, many similarities existed between Jewish and Roman slave practices and attitudes towards slaves.2 At the same time, we must ask whether Jews, who were subjugated to Roman and Byzantine Christian imperial rule and considered the Torah3 their most authoritative moral guide, developed different perspectives on slavery and treated slaves differently than non-Jewish Romans, whether pagan or Christian, in late antiquity. Was there something specifically Jewish about Jewish slaveholding practices in late antiquity? Did Jews, who commemorated the Exodus from Egyptian slavery in the annual Passover holiday and who were seen as a ‘servile’ people by Roman rulers, develop alternative approaches to slavery? I have to state at the outset that Jewish literary evidence from late antiquity consists of rabbinic literature only. The late antique rabbinic anthologies of the Palestinian and Babylonian Talmuds and Amoraic Midrashim deal with slaves and slavery from a specific rabbinic perspective; that is, they present the views of the self-styled rabbinic religious and intellectual elite. Although the practices of non-rabbinic Jews are also occasionally mentioned in these literary works, whatever is said about 1 2 3

Hezser 2005, 389. Hezser 2016, with examples. I.e. the Pentateuch.

129

130

Catherine Hezser

slavery here represents rabbinic views that were not necessarily shared by other late antique Jews and do not necessarily reflect social reality. Rabbinic views and practices transmitted in the Babylonian Talmud must be examined in the context of Sasanian Persian culture. The focus here will be on late antique Palestinian rabbinic views that developed in the context of Roman-Byzantine society. There are no archaeological and few epigraphic sources available to complement, support, or challenge the image that rabbinic texts create. Therefore, we can only examine and compare slave rules and moral recommendations in rabbinic discourse about slavery rather than gain information about actual slave practices among Palestinian Jews. In the following discussion, the focus will be on aspects in which rabbis developed distinct viewpoints rather than on socio-economic commonalities between late antique Jewish and RomanByzantine society.

poverty, debt slavery, and self-sale Impoverishment through droughts, bad harvests, loss of land, heavy taxation, and large numbers of children seems to have sometimes forced householders to sell themselves or their children into slavery to compensate for debts or to simply allow their children to survive.4 In the Hebrew Bible, such sales of impoverished Israelites to wealthier compatriots are seen as temporary, at least as far as males are concerned.5 Daughters sold into slavery by their fathers were expected to be sexually (ab)used by their masters and therefore believed to be better off if they stayed in the master’s household and were maintained by him.6 The biblical texts admonish the masters to manumit their Hebrew slaves in the seventh year.7 If the debt was not repaid, the situation could become permanent, however, a phenomenon which seems to have occurred quite regularly.8 The Hebrew Bible and the Gospels already contain numerous examples of such arrangements.9

4 5 6 7 8 9

Chirichigno 1993; Hezser 2005, 189, 233–7. Lev. 25:39. Ex. 21:7–11. Ex. 21:2–3; Deut. 15:12. Neh. 5:5; Jer. 34:9. For references to debt slavery in the Hebrew Bible see Bartchy 1988, 541–2. In the Gospels, debtors are mentioned in the ‘Our Father’ prayer in Matt. 6:12 and its parallels (originally from the Sayings Source Q); see also Matt. 18:23–34 for the parable of the debt slave.

Jewish Slavery in Late Antiquity

131

The seizure of a debtor’s children in lieu of payment is mentioned in a rabbinic king-parable transmitted in the Midrash Sifre Deuteronomy.10 A householder is said to have borrowed a large amount of wheat from a king. When ‘he sent [for payment], but he did not pay him anything, the king entered his house and seized his sons and daughters and put them on the auction block’.11 Allegedly, the debtor lacked any documentation of a partial repayment of his debt.12 The use of the parable in a late antique composition suggests that such evocations of the consequences of indebtedness for the family of the debtor were still plausible in that time period. Palestinian Jews were heavily taxed by Romans after the two revolts. By the third century, wealthy Jews were forced to become members of city councils and as such held responsible for ‘producing the tax quotas demanded of their cities’.13 Smallholders who sold their produce to pay taxes could be left without food to feed their families. In the mentioned parable, the debtor’s children are forcefully taken away from their homes and families. They are not meant to repay the debt through their (temporary) labour but lose their freedom permanently. The parable shows how easily freeborn Jews could become enslaved. The biblical recommendations to release (debt) slaves in the seventh year would have been useless in such cases. In Roman society, debt bondage was abolished in the fourth century bce.14 According to Joshel, this measure ‘firmly divided slave from citizen and defined the physical integrity of the citizen against the vulnerability of the slave’.15 In Palestinian Jewish society, on the other hand, freeborn Jews’ slippage into slavery seems to have still been a possibility in late antiquity, at a time when the Roman capture and enslavement of Jews that followed the revolts of the late first and early second centuries was mostly a memory of the past.16 The wealthy creditors were probably mostly

10

11 12 13 14 15 16

Sifre Deuteronomy is a tannaitic Midrash that was probably created in the third century ce; see Fraade 1991, 192n7. Sifre Deut. 26. Fraade 1983, 298–301, on the meaning of apokhi, ‘pledge’. Smallwood 2001, 496. Joshel 2010, 54. Ibid. On the Roman capture and enslavement of Jews during and after the first revolt, see Flavius Josephus. B.J. 3.7.31 (303–4): Titus allegedly sold more than two thousand Jewish women and children as slaves; Joseph. B.J. 3.10.10 (539–41): more than thirty thousand Jews were enslaved under Vespasian.

132

Catherine Hezser

fellow Jews, for poor Jews would be more likely to ask them rather than non-Jews for a ‘loan in the form of subsistence or grain’.17 It seems that rabbis’ awareness of the hardships of some of their fellow Jews caused them to allow the sale of oneself into slavery under certain circumstances. Whereas they generally prohibited self-sale, which would reverse God’s liberation of the Israelites from slavery in Egypt, they exempt the poor person from this prohibition.18 The reason behind this ruling may be that hard work as a slave may be preferable to the death of one’s family members from starvation. The self-sale is recognised as a proper sale: ‘and if he sold [himself], behold, this one is sold’.19 In this case the self-sale is not associated with debt, but with the purchase of work equipment (cattle, utensils, slaves) from the proceeds. The assumption seems to be that a poor householder might sell himself into slavery to a wealthier person to enable his family to maintain their smallholding. The self-sale to purchase slaves is rather odd. Perhaps the tradents imagined that the proceeds from self-sale might enable the family to purchase two younger and/or more able-bodied slaves. In the text’s parallel in Sifra, the purchase of slaves is not mentioned and a biblical proof-text is added: ‘From where [do we know] that a person is not permitted to sell himself and to leave [the proceeds] in his money-bag or to purchase for himself utensils or to purchase for himself a house, unless he is a poor person? Scripture says: “And if he becomes poor and sells himself [Lev. 25:39]”.’20 Probably because of the dangers of sexual abuse, the Mekhilta limits self-sale to men.21 Like Sifre Deuteronomy above, the Tosefta, Sifra, and Mekhilta point to the third century as the time when poverty might have led to debt slavery and self-sale. While debt slavery could easily wreck a family, self-sale was meant to ensure its survival. Rabbis also allowed fathers to sell their daughters into slavery. According to Mishnah Sotah 3:8, ‘a man sells his daughter, but a woman does not sell her daughter’. Poor families may have wanted to get rid of daughters because they contributed less to the family’s income or because they could not afford to give them a dowry. In the text’s parallel in 17

18 19 20 21

Joshel 2010, 54, notes that in earlier Roman society such a loan would be ‘secured on the person of the debtor (or his son) for the very purpose of controlling the labor of the debtor’. T. cArak. 5:8 Ibid. Sifra Behar 7, 80a. Mek. Mishpatim 3.

Jewish Slavery in Late Antiquity

133

the Mekhilta, sons are explicitly excluded from sale.22 With daughters, the moral issue of sexual exploitation had to be considered. In Mishnah Ketubot 3:8 and in the Mekhilta, the father’s right to sell his daughter is limited to the time before puberty, perhaps to avoid associations of a direct sale into prostitution. The Mekhilta also prohibits an adult woman from selling herself.23 Adult women were under the authority of either their fathers or husbands, so decisions about their sale may have been considered these male householders’ prerogatives. In Roman law the paterfamilias determined a child’s life or death and could therefore also decide to abandon, kill, or sell an infant. Evidence of such occurrences appear in Roman legal sources. For example, a six-yearold girl known as Passia was bought by a certain Maximus in Thrace in the first half of the second century.24 The girl’s fate depended on her master’s attitudes towards slaves and treatment of her: she might be exploited and abused, or be ‘lucky’, work hard, and be maintained with food and clothing.25 Whether she was sold or abandoned by her father remains unknown. On the background of Roman fathers’ absolute authority over their children, the rabbinic stipulations that limit the sale of children to girls before puberty may be seen as morally based leniencies. Rabbis may have allowed fathers to sell their young daughters for similar reasons as permitting self-sale: to enable (the rest of) the impoverished family to survive.

the ‘conversion’ of slaves Although rabbis clearly distinguished between slaves and freeborn Jews, they nevertheless feared that slaves owned by Jews could be exposed to idolatry if sold to non-Jews and/or outside of the rabbinically defined Land of Israel. According to Mishnah Gittin 4:6, ‘He who sells his slave to a gentile or to [someone] outside of the Land [of Israel], he goes forth a free man.’ The text has a parallel in the third-century Midrash Sifre Deuteronomy (259). By contrast, no limitations exist with regard to bringing slaves in from outside; that is, purchasing slaves at gentile fairs or in adjacent regions. According to the Tosefta and the Palestinian Talmud, ‘they go to a [gentile] fair and purchase male and female slaves

22 23 24 25

Mek. Mishpatim 3. Mek. Mishpatim 3 and M. Ketub. 3:8. On this example see Sigismund-Nielsen 2013, 286–7. See ibid. 287.

134

Catherine Hezser

there’.26 What could have been the reason behind this distinction?27 What changes were slaves expected to undergo once they became the slaves of Jewish owners? The texts talk about slaves in general rather than distinguishing between formerly Jewish and gentile slaves. Slaves, whether from Jewish or non-Jewish backgrounds, were expected to be involved in Jewish ritual practices associated with food preparation and ritual purity. They would have to prepare food in accordance with the rules of kashrut, help remove leaven from houses before the Passover holiday, participate in the Passover Seder, and adhere to family purity rules (e.g., immersion after menstruation and childbirth; not touching things while in an impure state, at least if they worked in the homes of rabbis and other religiously committed Jews). Although slaves were not considered proper Jews, even if they were born Jewish – being a slave and a Jew was considered incompatible by rabbis, since Jews could have only God as their master – they had to be sufficiently prepared to work in Jewish households. This preparation would have involved practical instructions as well as the immersion and circumcision of non-Jewish slaves. According to the Tosefta, male uncircumcised slaves purchased from non-Jews must be both circumcised and immersed; sons of slave women, even if they are circumcised, must be immersed.28 Otherwise, ‘things which they press against [i.e., touch or sit on] are [considered] unclean’.29 Similarly, the wine they prepare and serve should not be consumed by Jews, unless the slaves who touched it are minors who would not have conscious memories of idolatry. Idolatry is explicitly mentioned in the explanation: ‘And who is [considered] an adult? Anyone who remembers and recognizes idolatry and its service.’30 Even the mere memory of idolatry and its practice, that is, the preparation of libation wine, is said to have rendered a slave unfit for service in a Jewish household, unless he undergoes rituals that are identical to those that rabbis impose on non-Jews who want to convert to Judaism.31 Since

26

27

28 29 30 31

y. cAbod. Zar. 1:1, 39b par. T. cAbod. Zar. 1:8. The negation in the Zuckermandel edition of the Tosefta seems to be an error. See also the 4QDa. 12.10–11: a member of the Qumran sect may not ‘sell his male slave or female slave to gentiles, because they [the slaves] have come with him into the covenant of Abraham’, that is, they have joined the householder’s religion; trans. Cohen 1999, 169. T. cAbod. Zar. 3:11. Ibid. Ibid. On rabbinic conversion rules, see Cohen 1999, 198–238; Segal 2014, 591–3.

Jewish Slavery in Late Antiquity

135

wine preparation could also include agricultural work, the rule is not limited to household slaves but applies to all originally non-Jewish slaves purchased and employed by a Jewish owner. That Jewish slave owners circumcised their slaves in early Byzantine times is also suggested by the Codex Theodosianus, which states explicitly that ‘slaves must not be circumcised by Jews [non debere servos a Judaeis circumcidi]’.32 A statute by the emperor Constantine issued in 336 ce states: It is Our will that if any Jew should purchase a Christian slave or a slave of any other sect whatever and should not greatly fear to circumcise such slave, the person thus circumcised shall be rendered master of his freedom by the measure of this statute and shall obtain possession of the privileges thereof. It shall not be permissible for a Jew who has circumcised a slave of the aforesaid class to retain such a slave in the service of slavery.33

This text almost sounds like the exact opposite of the rabbinic rule that Jews who sell their (circumcised and immersed) slaves to non-Jews would lose their authority over them.34 While rabbis feared that their ‘converted’ slaves could become exposed to idolatry, a category in which they also included Christianity, Christian leaders feared that Jews might convert (potentially) Christian slaves they owned. Both sides created these boundaries to protect their constituencies. At the same time, the rules also link the power of masters over slaves to the power and assumed superiority of one religion over another. The late Roman imperial statute was meant to limit the power of Jews over (potential) Christians now that Christianity had become linked to empire and was eager to express its believed superiority.35 Whether and to what extent Constantine’s statute was enforced in a distant Near-Eastern province remains unknown. Not being able to circumcise their male gentile-born slaves would have prevented Jews from employing such slaves altogether. A later statute by the emperors Gratian, Valentinian II, and Theodosius I, dated to 384 ce, goes one step further and prohibits the ownership of (formerly) Christian slaves by Jews altogether: No Jew whatsoever shall purchase a Christian slave or contaminate an exChristian with Jewish religious rites. But if a public investigation should disclose that this has been done, the slave shall be forcibly taken away, and such master 32 33 34 35

Const. Sirm. 4; par. C.Th. 19.9.1. Ibid.; trans. Pharr 2001, 479. See m. Git. 4:6. See also Linder 2006, 165.

136

Catherine Hezser

shall undergo a punishment suitable and appropriate to the crime. It is further added that if there should be found among the Jews any slaves who are either still Christians or ex-Christian Jews, they shall be redeemed from this unworthy servitude by the Christians upon payment of a suitable price.36

The very necessity of this and later decrees suggests that Jews did in fact continue to own and circumcise Christian slaves in the fourth to sixth centuries ce.37 Interestingly, the late Roman authorities assumed that circumcision and immersion turned the slaves into Jews. The term ‘exChristian Jew’ (ex christianis iudaei) in distinction from ‘Christian slave’ (christiani servi) distinguishes between slaves who have already been circumcised and immersed and others who have not yet undergone these rituals. Both are supposed to be taken away from their Jewish owners, who are threatened with punishment but also promised to be reimbursed ‘by the Christians’ who would then continue to use them as slaves in their own households. Since these measures are to follow ‘a public investigation’, which would have been difficult to carry out in Byzantine Palestine, especially as far as largely Jewish cities and villages in the Galilee are concerned, the efficacy of the decree is questionable. It is possible, though, that slaves who were treated harshly by their Jewish owners approached Christians and told them that they were Christian believers or wanted to convert to Christianity. In such a case they might not have gained their freedom but might have been taken away from their Jewish owners, if Christians were willing to redeem them. A later repetition of the decree in 415 ce 38 rules that ‘slaves of the Christian faith’ in Jewish ownership ‘shall be delivered to the Church’.39 Individual Christians may have tried to get hold of such slaves for their own use and advantage rather than liberating them for good. From the rabbinic point of view, slaves who were circumcised and immersed did not automatically become Jews. In fact, even if they were male, they could not be considered Israelites as long as they were enslaved, since being a slave and a Jew was seen as a contradiction in terms.40 The conversion rituals must rather be understood symbolically, rendering the slaves fit for service in Jewish households. Rather than becoming Jews, 36 37

38 39 40

C.Th. 3.1.5; trans. Pharr 2001, 64. The prohibitions reappeared in the fifth and sixth centuries; see Linder 2006, 165. Linder further writes: ‘Three centuries of continuous efforts to stamp out the enslavement of Christians to Jewish masters proved ultimately unsuccessful’ (at 167). C.Th. 16.8.22. Ibid. 470. See Hezser 2005, 45 with reference to T. B. Qam. 7:5, par. y. Qidd. 1:2, 59d.

Jewish Slavery in Late Antiquity

137

they obtained the status of eved Yisrael, ‘slave of Israel’ or ‘Israelite slave’, which distinguished them from both gentiles and Jews. As members of a Jewish household that performed Jewish religious rituals, the circumcised and immersed slaves were considered to have been brought ‘under the wings of the Shekhinah [that is, the presence of God on earth]’. According to Sifre Numbers, householders who brought slaves ‘under the wings of the Shekhinah’ would ‘increase the glory of God’.41 In Christian households non-Christian slaves were commonly baptised. Acts 16:14–15 refers to the forced Christianisation in the form of baptism of all slaves at the time when their owner, a certain Lydia, became Christian. Like the circumcised and immersed slaves of Jewish owners, the baptised slaves of Christian owners occupied an inferior inbetween status that differentiated them from both slaves and free(born) Christians. While former slaves could become Christian believers, equality existed ‘in Christ’ only, that is, it was ideal and spiritual rather than real.42

upholding jewish family values While rabbis took the Jewish ownership of slave women for granted, they imposed certain rules that limited the consequences of men’s sexual intercourse with slave women for the composition and property of their nuclear families. Whereas biblical texts mention that Israelite men had no reservations against having children with slave women, who would then be recognised as the legitimate children of their Israelite father (see, for example, Abraham and Hagar in Gen. 16; Ishmael is considered Abraham’s son; Gen. 21:12–13), in Roman-Byzantine times rabbis were concerned about guarding the boundaries of the nuclear family and therefore limited the creation of legitimate children and heirs to the householder’s free(born) wife. A possible reason for this development may have been the decimation of land holdings after the two revolts. With the reduction of family-owned plots, the earlier ideal of large families gave way to a limit on the number of heirs who would have to share their father’s property

41 42

Sifre Num. 80. Glancy 2018, 476, writes in connection with Gal. 3:28 and 1 Cor. 12:13: ‘It may be useful to differentiate between the claim that for those in Christ there are no distinctions between slave and free and the claim that slavery presents no obstacles to those who want to join the Christian body. In fact, Paul never clearly makes the latter claim.’ See also Glancy 2002, 49–50.

138

Catherine Hezser

among themselves. Another reason was the rabbinic category of ‘illegitimate unions’ that had an analogy in Roman law. The first-century Jewish historian Flavius Josephus adds slave women and female war captives to the ‘forbidden relationships’ with whom Jewish priests cannot enter legitimate unions.43 Rabbis extended the prohibition on marrying slave women to all Jews and declared the children of slave women slaves, even if their father was a freeborn Jew. According to Mishnah Qiddushin 3:14, the offspring obtains the status of the (freeborn) father in legitimate marriages only. The offspring of sexual intercourse with slave women obtain the status of their mother, that is, they are considered slaves like her.44 These rules would have allowed men to use their female slaves for sexual gratification without having to maintain the children these women might give birth to. The offspring could be kept in the master’s household and be used as slaves or sold off at the slave markets. A similar regulation existed in Roman society, where only the children produced in iustae nuptiae, that is, marriages between Roman citizens, were considered legitimate, whereas children that emerged from intercourse between Roman citizens and slave women were considered slaves.45 In Judaism this matrilineal principle was a rabbinic innovation. It also determined who was considered a Jew, namely, the child of a freeborn Jewish mother.46 The children of slaves, on the other hand, including those of female slaves who were born Jewish or had undergone immersion, could never be considered Jewish as long as they remained enslaved. The Mishnah’s innovative ruling47 indicates the much greater distinction between slaves and free(born) individuals in the Jewish society of Roman-Byzantine times in comparison with earlier periods, a contrast that was prepared in Hellenistic and early Roman times. For rabbis, the Jewish family could employ slaves as members of their wider households, but these slaves and their offspring could not become the householder’s children and heirs. The status of a legitimate heir was reserved for children born to the householder’s free and married Jewish wife only. Yet there is also evidence that the edges of this system were blurry. Although Mishnah Yebamot 2:8 rules that a householder may not marry a slave woman he

43 44 45 46 47

Joseph. A.J. 3.12.2(276). M. Qidd. 3:12. Gai. Inst. 1.81. On the matrilineal principle, see Cohen 2001, 263–307. M. Qidd. 3:12.

Jewish Slavery in Late Antiquity

139

slept with and subsequently manumitted, rabbis concede that if he did marry her, ‘they do not take [her] away from him’, that is, she can retrospectively be considered his wife. In the Tosefta, the same principle is applied to Jewish women and male slaves.48 Besides the socio-economic reasons already mentioned, which rabbis probably shared with Roman jurists, moral concerns also determined their negative views of (unions with) slave women, irrespective of whether these women came from Jewish or gentile backgrounds and whether they were former Roman war captives or bought from fellow Jews. They were always seen as sexually promiscuous, irrespective of whether they had engaged in sexual unions voluntarily or were violated by their owners and/or other men. According to Tosefta Horayot 2:11, no Jewish man is eager to marry a freed slave woman because she ‘is [assumed to be in the status of] one who had been freely available’, that is, she is considered a prostitute, in contrast to a freeborn proselyte, who is expected to have ‘guarded herself’.49 Some rabbis seem to have wished to end sexual intercourse between Jewish males and slave women altogether and tried to instil feelings of guilt in men who engaged in such relationships. The late antique Midrash Leviticus Rabbah comments on Proverbs 14:9: ‘Fools mock at making amends for sin’, with a statement attributed to R. Yudan in the name of R. Levi: ‘This [verse] refers to men who treat [sexual intercourse] with female slaves as permitted in this world. The Holy One Blessed Be He will hang them by the hair of the crown of their heads in the future [world] to come’ (Lev. R. 9:5). Afterwards, Psalm 68:22 is quoted: ‘Surely God will smite through the head of his enemies the hairy scalp of him that goes about in his guiltiness’, followed by: ‘All people shall say: Let this man perish in his guilt! Let this man perish in his guilt!’ Such strong condemnations of sexual intercourse with female slaves are absent from the earlier tannaitic documents. Although the Mishnah already threatens with guilt offerings,50 such punishments could not actually be imposed after the destruction of the Temple. While a fine is imposed on men who had sexual relations with freed slave women who were manumitted as toddlers, no fine is mentioned for the rape of slave women manumitted after the age of three years and one day.51 Could the late antique rabbinic exhortations to

48 49

50 51

t. Yebam. 4:6. Ibid. But see Sifre Emor 1:7 (49b–50a), where manumitted slave women and female proselytes are equated in this regard – both are considered harlots. m. Ketub. 2:2–4. Ibid. 3:1–2.

Catherine Hezser

140

refrain from sexual relations with slave women be meant to prevent stronger boundaries between Jewish men and female slaves of a pagan or Christian background? Or were rabbis more concerned about sexual morality, that is, men’s control over their bodies and desires, than potential enticement to idolatry? In any case, the sexual use of slaves is now seen as detrimental to the well-being of the late antique Jewish family. Perhaps rabbis were familiar with similar Christian views. Glancy notes that in the late third century, Lactantius urged Christian men to have sexual relations with their wives only and not with slaves.52 She dismisses the relevance of earlier Stoic recommendations that appealed to men’s self-control of their urges. Rather, Christian theologians of the third and fourth centuries saw the Christian family as a bulwark of Christian morality in contrast to pagans: ‘Ambrose urged men to avoid sexual relations with slaves in order to avoid giving their wives grounds for divorce. Jerome contrasted Christian with pagan morals.’53 In such a climate, which stressed family values, rabbis may have been similarly inclined to implore their fellow Jewish men to safeguard the sexual ‘purity’ of the Jewish family. Like Christian leaders, they distinguished Jewish values from those of pagans, especially at a time when paganism was increasingly condemned within society. It seems that by outlawing sexual intercourse with slave women, late antique rabbis entered a moral competition with Christian leaders for the upholding of (stoically inspired?) family values.

limits to physical violence against slaves Since slaves were the property of their owners and had no legal rights, masters could treat them however they wanted. Physical violence towards slaves would have been considered a proper means of discipline, if it was executed by the master himself or an appointed supervisor. In Roman law slave owners even had the right to kill their slaves, and this right was allegedly based on the ‘law of nations’.54 In the early fourth century, Constantine assured owners that they would not be held legally responsible if their slaves died as a consequence of physical discipline or restraint.55 In fact, violent physical punishment was justified as a proper

52 53 54 55

Glancy 2002, 58. Ibid. Inst. Iust. 1.8.1. C.Th. 9.12.1.

Jewish Slavery in Late Antiquity

141

reaction to ‘very wicked deeds’, ‘to obtain better behaviour from their slaves’.56 In this rescript, Constantine explicitly states that no investigations should be made: ‘It is our pleasure that masters are not held guilty of murder by reason of the death of a slave, as often as they exercise domestic power by simple punishment.’57 While biblical law permits masters to physically discipline their slaves, it also protects slaves against being killed in this process: ‘When a man strikes his male or female slave with a rod, and he dies under his hand, he must be avenged. But if he survives a day or two, he is not to be avenged, since he is his property.’58 A master who kills his slave, whether intentionally or unintentionally, must be held responsible for his death. If the slave is badly damaged but does not die immediately, the master remains guiltless. The boundary of legitimate behaviour that is set here is the slave’s death. According to Israelite and Jewish belief, only God may take away human life. If a master kills his slave, he illegitimately substitutes for God. The most violent punishment of slaves is legitimised with reference to the master’s ownership of the slave, though. In rabbinic discussions, ‘Hebrew’ slaves are exempted from the permission to chastise them physically: ‘He who wounded a Hebrew slave is liable on all counts.’59 Violence against one’s own ‘Canaanite’ slaves is permitted, though.60 The distinction between ‘Hebrew’ and ‘Canaanite’ slaves is biblical but was hardly maintained in Roman times. As we have already seen above, originally non-Jewish slaves were made to undergo conversion rituals (circumcision and immersion) to work in Jewish households. Jewish-born slaves were not considered proper Jews during the time of their enslavement. Thus, the distinction between formerly Jewish and formerly non-Jewish slaves was blurred. While they were enslaved, their status as eved Yisrael was all that mattered. Does this mean that the Mishnah’s injunction applies to all Jewish-owned slaves then, including those who had undergone the required ‘conversion’ rituals? Or is it specially addressed to the owners of Jewish-born debt slaves?61 In the Tosefta, the owner’s liability is extended beyond the biblical rule62 in two ways. Firstly, no time limit is set to the slave’s survival after 56 57 58 59 60 61 62

Ibid. 9.12.2. Ibid. Ex. 21:20–1. m. B. Qam. 8:3. Ibid. 8:5. See above and Lev. 25:43. In Ex. 21:20–1.

142

Catherine Hezser

the inflicted injury. Secondly, the owner’s responsibility covers all injury leading to the slave’s death within the spatial limits of his property: ‘[he is not punished] unless hitting him and his death [occurred] in his domain’.63 If he hit his slave and then sold him to someone else, however, and the slave died under the new owner’s authority, the former owner is exempt, since the slave did not ‘die under his hand’.64 Recommendations to treat slaves mildly appear repeatedly in late antique rabbinic texts. Some of these texts specifically mention impoverished Jews who had to sell themselves into slavery. Such a slave ‘is not to go behind you carrying you in a sedan chair; and he is not to go before you in the bathhouse carrying your utensils’; he should ‘be with you in food, with you in drink, and with you in clean clothes, so that you are not to eat a fine piece of bread and he eats bread of coarse grain, you drink old wine and he drinks new wine, you sleep on down and he sleeps on straw’.65 The lifestyle projected here is that of a wealthy individual able to afford expensive food and wine, wearing fine clothes, and presenting himself in the public sphere. He is urged to treat his debt slave as he would treat his equal. The text suggests that debt slaves should be seen as impoverished fellow Jews rather than servants. Obviously, no historical conclusions about the actual treatment of debt slaves can be drawn from this proposition. Some of the tasks from which debt slaves are supposed to be exempt here – for example, carrying their master’s utensils to the bathhouse – were also expected of disciples of sages who ‘served’ their masters (shimush chakhamim). Talmudic texts present rabbis as examples of mild and even indulgent treatment of their slaves, without specifying whether the slaves were debt slaves or permanent slaves, formerly Jewish or non-Jewish: ‘Now did not R. Yochanan, from whatever he would eat, give to his slave and recite the following verse in this connection: “Did not He who made me in the womb make him”? [Job 31:15].’66 What is particularly interesting here is the emphasis on the slave’s humanity, his being created by God like any other human being. This notion went against common ancient associations of slaves with animals and their treatment as some kind of nonhuman property (chattel).67 Another example of a rabbi who is said to

63 64 65 66 67

t B. Qam. 9:22. Ex. 21:20. Sifra Behar 7:2, 80a. y. Ketub. 5:5, 30a. Slaves are equated to animals by Aristotle; see Pol. 1254–6 and Isaac 2004, 212.

Jewish Slavery in Late Antiquity

143

have behaved towards his slave as he would towards a free person is the story about R. Gamliel’s mourning and accepting condolences on the occasion of his slave Tabi’s death.68 Unlike R. Yochanan’s anonymous and generic slave, Tabi is said to have been ‘worthy’, though.69 He is presented as an exceptional slave and ‘disciple of a sage’.70 The examples of R. Gamliel and R. Yochanan are presented as extraordinary behaviours. Yet they also indicate that late antique rabbis valued good master– slave relationships and did not share the common ancient objectification of slaves. Some Stoic philosophers also propagated the mild treatment of slaves. Seneca, for example, stressed that slaves were as human as their masters and should be treated with gentleness. At the very beginning of Epistula 47 to Lucilius, he praises his addressee for living ‘on familiar terms with your slaves’, befitting ‘a person of your good sense and education’. He counters those who see slaves as different, stressing that ‘they are human beings’ and ‘housemates’ (contubernales), similar to ‘lowborn friends’.71 Slaves could have free minds, whereas free persons could be enslaved by their desires.72 Later patristic writers sometimes took Paul’s reference to Onesimus as Philemon’s ‘beloved brother’73 as an example of how slave owners should relate to their slaves. Fitzgerald points to the late fourth- to early fifth-century bishop Theodore of Mopsuestia, who, in his commentary on Paul’s letter, stresses the ‘kindness’ and ‘affection’ with which Philemon is meant to treat his slave.74 Although some Stoics, church fathers, and rabbis shared notions about the humanity of slaves that should result in mild treatment, their anthropological and theological views were different. Rabbis believed in the unity of body and soul, which required both to be valued equally. By contrast, the Christian dismissal of the body and emphasis on the spirit could lead to a disregard of slaves’ bodily misery.75

68 69 70 71 72 73 74

75

m. Ber. 2:7. Ibid. m. Sukk. 2:1. Sen. Ep. 47.1. Ibid. 47.17; trans. Graver and Long 2015, 134 and 137. Phlm. 1:1. Fitzgerald 2010, 359, with reference to Thdr. Mops. Comm. Philm. 2.278.15–16 and 2.283.16. Glancy 2002, 10, points out that slaves were commonly seen as bodies that were vulnerable to abuse. Even slaves who became Christian believers could be treated harshly as far as their bodies were concerned.

144

Catherine Hezser

the redemption of jewish slaves As a consequence of Roman imperialism, tens of thousands of Jews were captured and enslaved by Roman soldiers between the first century bce and second century ce.76 In the third century, when the local revolts were quelled, the empire-wide pax Romana seems to have created a rather peaceful situation for Jews in Roman Palestine. At the same time, the Roman military and administrative presence may have provoked the occasional uproar among the Jewish population, and the Roman punishment may well have been the occasional capture and enslavement of individuals. Such measures would have served as manifestations of Roman power and prevented uprisings. Therefore, late antique rabbis felt inclined to continue discussions about the redemption of originally Jewish slaves and war captives. Rabbis were particularly concerned about these captured and enslaved individuals’ forced loss of their Jewishness, through having to serve an earthly master, being exposed to idolatry or converted to Christianity, and being sexually exploited and becoming impure. The Tosefta states: ‘Just as an Israelite is obliged to redeem a [formerly] free [Jew], he is obliged to redeem their slaves’, that is, formerly Jewish-owned slaves that now serve a non-Jewish master.77 The expansion of the rule to include not only slaves who were born Jewish but also those whom their former Jewish owners had ‘converted’ to Judaism stands in direct contrast to the statute of Constantine and later Roman emperors mentioned above, according to which Christians should seize such slaves from their Jewish owners. From the rabbinic point of view, the redemption of Jewish and Jewish-owned slaves from non-Jewish owners was considered a liberation from an idolatrous environment and a restitution into the monotheistic Jewish communal context (‘They take from non-Jews fields, houses, vineyards, cattle, and male and female slaves, because this is like rescue from their hands’78). Since slaves were commonly seen as ‘blank slates’ whom their owners would inscribe, those who had experienced the impact of Judaism were meant to remain in that social, cultural, and religious environment, even if they did not have the status of free adult Israelites as long as they were slaves.

76 77 78

Hezser 2005, 229–30, with references. T. Git. 4:2. T. Moʿed Qat. 1:12.

Jewish Slavery in Late Antiquity

145

A narrative transmitted in the Tosefta presents the example of a handsome young boy, a native of Jerusalem, whom R. Yehoshua discovered at a slave market in Rome.79 There were rumours that the boy, if he remained enslaved, would ‘be put to disgrace’, that is, sexually exploited. Allegedly, the boy’s ethnically Jewish, Jerusalemite origin was not a sufficient motivation for the rabbi to redeem him. He rather examined his Torah knowledge by asking him a biblical question. When the boy is able to answer and proves himself worthy of redemption, ‘he redeemed him with a lot of money and sent him [back] to the Land of Israel’.80 In the Babylonian Talmud, the opposite case is told of an enslaved Jew who was not deemed sufficiently observant to be worthy of redemption.81 A man had sold himself to Lydian masters and asked R. Ammi to redeem him. R. Ammi rejected this request not only because the man had voluntarily submitted himself to slavery rather than being forcefully enslaved by nonJews; there were also rumours that he had eaten meat that was not ritually clean, that is, the rabbi considered him not sufficiently Jewish in a religious sense of the word. In both of these cases an ethnic Jewish background is not considered sufficient reason for religious leaders to redeem a slave for a probably high amount of ransom money. The type of enslavement (forceful seizure by non-Jews) and Jewish religious knowledge and observance of the individual are considered crucial for determining his worthiness of being redeemed and reintegrated into a Jewish context. Rabbis had a reason for their lack of enthusiasm to redeem even forcefully captured Jews from their non-Jewish owners. According to the Mishnah, ‘they do not redeem captives at a price that is higher than their purchase price, because of the good order of the world [tikkun haolam]’.82 Those who formulated this rule seem to have feared that the payment of an exorbitantly high ransom fee might lead non-Jews to capture and enslave even more Jews to make a profit. The situation of the entire Jewish community rather than the individual enslaved person seems to have informed this argument. To maintain good relations between Jews and non-Jews at the turn of the third century ce, that is, to prevent Romans from capturing and enslaving even more Jews, limits had to be set to the redemption of those who were already in non-Jewish ownership. Their benefits had to be weighed against those of the rest of the

79 80 81 82

T. Hor. 2:5–6. Ibid. b. Git. 46b–47a. M. Git. 4:6.

146

Catherine Hezser

Jewish community. Extraordinarily high ransom fees could probably be paid through collective efforts only, that is, money that was meant for internal charitable purposes would have to be given to non-Jews.83 Obviously, relatives had a high interest in liberating their captured family members from non-Jewish ownership. The Babylonian Talmud relates a story about a certain Levi b. Darga, who allegedly paid 13,000 golden denars to redeem his daughter.84 If a wealthy individual could afford such ransom fees, the communal coffers would not be affected, but he might nevertheless set an example that might encourage unscrupulous kidnappers. Therefore, rabbis would not recommend such actions either.85 Imperial statutes transmitted in the Codex Theodosianus urged Christians to redeem their coreligionists from Jewish slave owners.86 The redemption of Christian slaves is presented as a moral imperative in patristic writings.87 Harrill notes that many Christian sources ‘discuss . . . public displays of generosity from ecclesiastical common chests’.88 Yet not all church fathers were in favour of such expenditure. In the early second century, Ignatius of Antioch writes in his letter to Polycarp: Despise not slaves, whether men or women. Yet let not these again be puffed up but let them serve the more faithfully to the glory of God, that they may obtain a better freedom from God. Let them not desire to be set free at the public cost, lest they be found slaves of lust.89

Like the rabbis of the narratives mentioned above, Ignatius reckons with the possibility that the slaves may not be worthy of redemption with communal money. They may not actually adhere to Christian moral values even if nominally Christian. He recommends the maintenance of their status as slaves. As slaves of faithful (and probably wealthy) Christian owners, they are said to be better able to serve God than if they were freed and enabled to live according to their own desires. The ‘better freedom’ promised here would be achieved through bodily servitude. It is reduced to a merely spiritual and moral state. In this way, ‘freedom in Christ’ is used to justify the continuation of bodily slavery.90 83 84 85 86 87 88 89 90

See b. Git. 45a. Ibid. See Abaye’s comment in b. Git. 45a. C.Th. 1.1.5. Grieser 2010 provides a good overview. Harrill 1995, 178. Ign. Pol. 4.1–3; trans. Lightfoot and Harmer 2004 (originally 1891). See also the many references to slavery of desires and freedom in Christ in Eusebius of Caesarea’s Praeparatio evangelica. On the idea of spiritual redemption in ancient

Jewish Slavery in Late Antiquity

147

Due to the belief of the unity between body and soul, rabbis could not recur to such distinctions between bodily enslavement and spiritual freedom. Therefore slaves, whether owned by Jewish or non-Jewish owners, could not obtain the status of an adult (male) Israelite with its requirement of full Torah observance. Accordingly, the redemption of slaves may have been considered justified in exceptional cases of religious piety and observance only.91 The seemingly more intense Christian insistence on redemption, on the other hand, was not meant to give slaves authority over their own bodies and destinies. It was rather meant to bring their ‘souls’ under the custody of wealthy Christian owners or the institution of the Church.

conclusions The discussion has shown that Jewish slave practices in late antiquity can be understood properly only if seen within the Roman and Byzantine Christian context. The question ‘What was Jewish about Jewish slavery in late antiquity?’ can therefore be answered in a relative way only. Aware of the dire poverty in which many of their compatriots found themselves, rabbis permitted the self-sale and sale of children under certain circumstances. Debt slavery and self-sale were also practised elsewhere in the Near East and seem to have occurred among the destitute Roman population, even if the legal and political authorities may have outlawed such practices.92 Like Jewish owners, Christian owners would customarily ‘convert’ their slaves to their own religion and try to prevent the transfer of such slaves into non-Jewish or non-Christian ownership. In late Roman times, Jewish and Christian religious leaders seem to have ‘pulled strings’ from opposite sides, trying to preserve the religious identity of the respective households. Biblical as well as Stoic ideas would have inspired both rabbis and church fathers to admonish slave owners to avoid violence and to regard their slaves as human beings. Masters were urged to keep their sexual urges in check and to uphold Jewish family values. Such recommendations were meant to demarcate the boundaries of the nuclear Jewish family in accordance with its religious identity, succession, and heritage. While the redemption of slaves that shared one’s own ethnic and religious background was held up as a moral ideal, religious leaders feared the

91 92

Christian texts see Glancy 2002, 99. See also the chapters of De Wet and Evans Grubbs in this volume. See also Ramelli’s chapter in this volume. Glancy 2002, 80–5, on self-sale; Chirichigno 1993 on debt slavery in the ancient Near East.

148

Catherine Hezser

impact of a large-scale redemption on communally owned funds and on social cohesion. Therefore, rabbis tried to restrict redemption to religiously knowledgeable and observant Jewish-born slaves seized by Romans, and Christian theologians stressed the greater importance of spiritual redemption over bodily freedom.

7 Divining Slavery in Late Ancient Egypt: Doulology in the Monastic Works of Paul of Tamma and Shenoute Christine Luckritz Marquis

The title of this chapter plays with the problem of understanding the role of slavery in late ancient Egypt. Scholars tend to agree that Egyptians, with a large tenant farmer population, relied far less heavily on enslaved labour than their counterparts elsewhere in the Empire. These smaller numbers mean that enslavement is harder to trace as a phenomenon. Thus, to discern how slavery functioned and who the enslaved were requires a certain amount of ‘divining’ or well-informed guesswork. Luckily, a wealth of extant documentary evidence and literary, largely monastic, texts are available for such an endeavour. To do such imaginative work responsibly requires careful attention to brief references to the enslaved, to the larger world in which they moved, and to how scriptural appropriation of slave imagery among monks masks the presence of actual enslaved individuals within the community. As a backdrop for focusing on the presence of the enslaved among Coptic monks, I first attend to documentary evidence, noting how challenging delineating the enslaved from the free person could be for late ancient Egyptians themselves, and thus by extension for modern historians attempting to study the phenomenon. With such ambiguity in daily life in mind, I build on past scholarship that has recognized how Christian authors exacerbated such ambiguity from the outset through the scripturalizing of enslavement. I do so by carefully examining the extension of such biblical, doulological logics in the works of two monks, Paul of Tamma and Shenoute of Atripe. Having traced the contours of the problem, I suggest a possible way forward by reconsidering a particular argument between Shenoute and a neighbouring political figure known as

149

150

Christine Luckritz Marquis

Gesios. By attending to brief references to enslaved people in our sources (including rhetorical uses of enslavement) and to the spaces we know enslaved people occupied, we can better reimagine enslaved people back into the narrative of late ancient Egyptian Christianity. I argue that mentions of real enslaved persons in a handful of Shenoute’s texts show that Christian monks interacted with enslaved persons, and that at least some of the enslaved may have been Christian. Shenoute’s story suggests that we would be wise to reimagine more Christian communities populated with enslaved individuals, recognizing that the obscuring of their presence does not negate their existence.

linguistic ambiguity A fundamental challenge facing the historian trying to find enslaved people in late antique Egypt is one of terminology, both in Greek and in Coptic.1 In Coptic, the two main words denoting enslaved individuals are hmhal and kauon/kyauon. Both terms may be translated ‘enslaved’, but the latter seems to have been considered more pejorative.2 Each also might mean servant, though this is more common with hmhal. And, following the Greek, the phrase šeˉre šeˉ m might mean either small child or enslaved individual.3 The blurriness of the term highlights the comparable subordination of children and enslaved individuals to the patriarch. In particular, it points to the lack of legal recourse among both children and the enslaved. When we do find distinction, it mainly emerges in the degree to which one ought to beat one’s slave versus one’s child.4 Beyond the ambiguity rooted in generic terms, there is also a lack of clarity in terms meant to identify social statuses. As Sohair Ahmed’s recent work gathering Coptic terms for professions, trades, and occupations indicates, interpretation of social status is often left to the modern historian.5 Terms like kyooure might mean ‘enslaved’ or ‘porter’, leaving 1

2 3

4

5

The focus in this chapter is on the Coptic. For Greek terms related to slavery, see esp. the chapter by Vierros in this volume. All translations are my own. All the Coptic texts, unless otherwise noted, are translated from the Coptic found on the Coptic Scriptorium website (Abraham, Our Father: urn:cts:copticLit:shenoute.abraham; Not Because A Fox Barks: urn:cts:copticLit:shenoute.fox). Layton 2014, 172–3, 206–7. The confusion emerges from the infantilized and emasculated state of the fully grown enslaved individual, especially the male; see Glancy 2002, 24–6. Saller 1996, 151–7, and De Bruyn 1999 suggest infrequent beating of children. Laes 2005 argues children were regularly beaten. Ahmed 2010; 2011.

Divining Slavery in Late Ancient Egypt

151

the free or enslaved status of the referenced individual unmarked. Moreover, specific jobs like manure cleaner or even baker often included tasks that occupied a mixed staff of both enslaved and free throughout the Roman Empire.6 Only those who worked in the given shop or house and their master, if there was one, knew who was free and who was enslaved. Unless a person’s enslaved status was pertinent and therefore recorded, it is often impossible to know whether or not an individual was enslaved. As this initial discussion of linguistic ambiguity underscores, how to delineate the enslaved from the free in our sources is rarely a straightforward task. The issue presents itself on two levels. First, there is the blurring of boundaries in Roman Egypt. Thus, an enslaved person and a free person might frequently work alongside one another at the same job. Or, the enslaved and the household child might often be treated in similar ways. Second, there is the interpretive activity of modern scholars. While historians seek clear distinctions between enslaved and free servant, linguistic ambiguity in Coptic presses us to consider how muddied those boundaries might have been.

status ambiguity and the law Papyrological testimonies support what linguistic attention suggests: the enslaved or free status of individuals was often not readily evident. Roger Bagnall had already noted this in 1993: ‘Roman society . . . contained plenty of people whose exact status may have been unclear to those around them, as well as those who moved from free to slave and slave to free.’7 One aspect of the confusion was that on the street, as people encountered one another, there was not necessarily a physical marker (such as clothing) to distinguish the enslaved from the free, non-elite around them.8 The one physical marker that might have allowed delineation was the tattoo of the enslaved individual, but it is not clear that this practice was common; it was largely performed as a punishment for flight.9 Such physical alterations would have made the sale of the slave far more challenging, as slave sellers were already required to guarantee a slave was not a flight risk. That many sellers attempted to mask the flight of enslaved persons for profit is indicated by the need for legislation

6 7 8 9

Joshel and Petersen 2017, 121. Bagnall 1993b, 227. Joshel and Peterson 2017, 95. Elm 1996; Gustafson 2000; Burrus 2003.

152

Christine Luckritz Marquis

binding sellers to their promises.10 Moreover, as Bagnall highlights in his analysis of P.Grenf. 2 78,11 the ability to demonstrate oneself as not enslaved could prove to be quite challenging. In this papyrus, the petitioner argues that his wife and children have been illegally enslaved. For his wife, he cites the free status both of her parents and her siblings. Proving her freedom was crucial, for his children’s status depended on that of his wife. While it is but one example, the petition underscores how suddenly one might be enslaved and how challenging it would be to prove oneself a free person. That such a situation was a real fear for the general population is reflected in the oracular Sortes Astrampsychi fragment found in Oxyrhynchus. Among the possible fates one might have foretold was: ‘you will be sold as an enslaved person’.12 Given how unclear distinctions might be to the outside observer, the danger of being enslaved was a very real one. One’s status as free required constant vigilance to maintain. More recently, Ari Bryen has revisited the issue of status in late Roman Egyptian papyri vis-à-vis the broader issue of violence. He argues that the act of petitioning was itself ‘about seeking to define one’s status in the first place’ and that the right to claim the experience of violence was meant to mark oneself as of a status capable of being legally violated.13 In this construction, status is perpetually constructed and contested. This matches well the example of the enslaved woman and children above. Her status was altered (illegally, as her husband asserts) and must now be renegotiated. While this testimony indicates that the status of enslaved was largely one to be avoided, there were cases where a clearly enslaved person sought to mark their status as such in order to access the social and legal capital of their master. Bryen offers the example of a certain Ision, concluding that his petition against an assaulter was likely not acted on (as are other examples in the collection) because of his status as enslaved.14 He suggests that in order for the claim to be taken seriously, Ision’s master, Chairemon, would have had to sue the assailant. Perhaps Ision hoped to capitalize on the reading of his enslaved body as an extension of his master’s. While Chairemon was legally authorized to treat Ision as an appendage to his own person, Ision’s failure indicates the legal limits of 10 11 12

13 14

Arzt-Grabner 2010. Bagnall 1993b. P.Oxy 67 4581, col. 3b, line 1; Stewart 2001, 124–5. My thanks to Ville Vuolanto for this reference. Bryen 2013, 45, 74–81. Bryen 2013, 74–6; P.Ryl. 2 144, 219.

Divining Slavery in Late Ancient Egypt

153

appealing to such definitions. The enslaved had no legal recourse even when they acknowledged themselves as extensions of another. This example also highlights the limitations of looking for the enslaved in papyri: to mark oneself as enslaved, even if legally true, was a detriment. That Ision’s appeal was unsuccessful points to why very few individuals are signified as enslaved in papyri. Roman law required that the enslaved, even when referenced, ultimately be swallowed up into their master’s status. The humanity of the enslaved was meant to be erased, the individual subsumed either into the master’s person or, as sales receipts indicate, the master’s property. Beyond petitions, census records and sales indicate that the enslaved were interpreted as one among many things owned by the master. Such evidence tells us nothing about what the experience of the enslaved was. Despite this depressing reality, there may yet be some hints of how slavery functioned. In particular, census records give us a glimpse of the more complex realities of a household (oikos).15 Households might include not only adult siblings, their spouses, and progeny, but also a mixture of nonkin free people and enslaved individuals. Census records vary, some indicating only blood relatives and enslaved family, while others also include non-familial members living on the land. Although census records may seem to be merely lists of individuals, the grouping of non-kin alongside free family members and enslaved persons belies the cultural logic through which communal relations were routed by imperial power. Units of power were households, which included the master, their family, and all others who worked for them. Village households tended to be smaller than their city counterparts and were less likely to include handfuls of non-kin workers, but rather relied upon tenant farmers (and presumably their families), who rented either housing with a small parcel of land or lived in village-like settlements known as epoikia near the arable soil.16 These latter settlements became increasingly common across late antiquity, the owner building housing and communal structures (such as a bakery and bath) for all the individuals who lived and worked on his land. In this way, access to housing and the necessities of life functioned as a form of payment. As noted at the outset, the history of tenant farmers is intimately linked to the story of slavery in Egypt. Bagnall notes that there may have been extensive tenant farmers available in the Egyptian countryside, serving as 15 16

Bagnall and Frier 1994, 12–14. Bagnall and Frier 1994, 66–70; López 2013, 86–90.

154

Christine Luckritz Marquis

a workforce limiting rural slavery in a way not found in other portions of the Roman Empire.17 The possible competition between free farmer and enslaved underscores that slavery existed in a complex web of power dynamics and that the various labour forces might influence the existence and treatment of one another. Thus, to better understand enslavement in Egypt requires likewise exploring the situation of the tenant farmer. If, as already argued above, proving one’s free status could be so difficult, how were tenant farmer and enslaved person delineated? If discerning another’s status and defending one’s own status was so messy, perhaps we ought to think of slavery as residing on a sliding scale with tenant farmers. This would open up the possibility for the movement from free tenant farmer to slave and the reverse, and such movement would likely depend on the whim of the landowner. Not that the free tenant farmers lacked any power compared to the enslaved; rather, the power dynamics between each and the master were by far more similar than they were different.

the shared precarity of enslaved and poor While Egyptians did not exploit large numbers of enslaved people for labour in the ways other regions of the Roman Empire did, they were still intimately aware of and involved in the practices of slavery as those bodies were moved through Egypt towards other portions of the Roman Empire. That is to say, even if we accept that estimated levels of slaveholding in Egypt never paralleled those of other imperial regions, Egyptians witnessed and participated in the enslavement market. Moreover, the very existence of slavery, no matter how minimal, introduced an incredible amount of precarity for all but the wealthiest echelons. The danger posed by violence from the enslaved (the enslaved beating the master) and from rebellion of one’s own slaves (loss of wealth from the flight of a slave) was largely imaginary on the part of slaveholders. The occasional realizations of such violence only heightened the fears of the wealthy minority.18 But it was in fact the enslaved and the poor who were far more prone to suffer. The precarity of the enslaved proceeded from the view of them as

17

18

Bagnall 1993b, 227–30, 237–8; Bagnall and Frier 1994: most enslaved worked in households and thus were more numerous in cities (69–71). Enslaved beating a free person: Plin. Ep. 3.14; enslaved flight: P.Panop. Beatty 1.149. Seneca claimed that requiring distinctive dress would visualize for the enslaved their numbers, inviting revolt; Sen. Clem. 1.24.1.

Divining Slavery in Late Ancient Egypt

155

subhuman property, as bodily extensions of their owners, and the logics that undergirded this stance.19 Even if only a small number of individuals were technically enslaved, the very existence of slavery in Egypt introduced the potential for enslavement of a large swath of the freeborn, especially the poor. What primarily distinguished the enslaved from the free was legal protections against violence. Beyond autonomy to protest violence done to their body, there was little difference between the daily life of tenant farmers, that of the enslaved, and that of local poor villagers. Because of this, we may reasonably speculate beyond the limits of what documentary papyri show to suggest that the challenges of discerning status were used by some of the wealthy to exploit the poorer free. The legal binding of tenant farmers as indebted to the owner of a given epoikia expresses a cultural logic that subordinated the farmer to the landowner akin to the subordination of enslaved to master, a move that undoubtedly exacerbated existing ambiguities of status (and concomitant treatment). In a culture where the enslaved were erased from the record except in cases of violence or taxation, reconstructing something of the enslaved experience requires broadening our scope. Papyrological testimonies will not give a direct window into the enslaved experience, but they illuminate the societal oppression that many suffered. If we take seriously that the risk of enslavement was real in late ancient Egypt, then any study of slavery must contend not only with enslaved people but also with all of the potential enslaved or ambiguously free. Possible legal recourse ought not be confused with actual access to power. Thus, to properly study enslaved people among Coptic monastic authors requires that this status ambiguity and the precarity it introduced into daily life remain part of our picture of the late ancient Egyptian world. But before we turn to that reconstructive work, we must first address how Paul of Tamma and Shenoute obscured actual slavery by deploying ascetic interpretation of enslavement in Scripture.

scriptural enslavement and ascetic comportment The complexity around status multiplies in the gradually Christianized landscape of late antiquity. For while Jesus’ message may have been about liberation, the emerging religion’s adherents quickly appropriated 19

In general, a lack of freedom, with concomitant inability to have legal children and bequeath one’s belongings to them, denied the enslaved identification with Roman (hu)manhood (Saller 1996, 121).

156

Christine Luckritz Marquis

scriptural passages both to defend and obscure actual enslavement. As Jennifer Glancy has shown, the rhetorical move of spiritualizing slavery sits at the root of Christianity and is by no means unique to the Egyptian context.20 But given the prominence of asceticism in Egypt’s landscape, it is unsurprising that ‘doulological logics’ were creatively deployed to articulate an ascetic self through slavery, further obscuring actual enslaved people from view.21 This reading of Scripture through and onto the ascetic body scripturalized slavery, not only spiritualizing the enslaved experience, but also binding the ascetic-body-as-enslaved with scriptural interpretation. Scripture was read through the lens of God as master and ascetic as enslaved in a way that encouraged suffering.22 More generalized spiritualizing of slavery undergirded the construction of the ascetic self as enslaved, in particular in the letters of the Apostle Paul. Paul’s articulation of communal relations maintains the current status quo with regard to slavery (e.g. Phlm., 1 Cor. 12) even as Paul attempts to appropriate the category, describing himself as an ‘enslaved of Christ Jesus’ (Rom. 1:1)23 and others as enslaved to God as well.24 For example, the Coptic abbot Shenoute points to Pauline language in his letter to female monastics under his care in Abraham, Our Father. He describes: ‘Paul is the enslaved of Jesus’ and likewise ‘we are his [God’s] enslaved’.25 He expands, noting that Jesus himself ‘took the form of an enslaved person [Phil. 2:7]’, thus paradoxically embodying both master and enslaved simultaneously.26 Scaffolding the relations of Christians (here sisters in community) onto the cultural praxes of enslavement, Shenoute reprimands the sisters for their unwillingness to properly ‘care’ for one another in a hierarchically appropriate manner. Shenoute spiritualizes slavery, obscuring any actual abuse that may be occurring within the community. We shall return to a more in-depth exploration of this letter and Shenoute’s use of doulological rhetoric below. Another example of ascetic use of ‘enslaved of God’ rhetoric is found in a version of Paul of Tamma’s treatise on living in one’s cell. The text found in MONB.GU I.10, provenanced from Shenoute’s White Monastery, makes use of the phrase only once: ‘For, they are stripped of their 20 21 22 23 24 25 26

Glancy 2002, 29–38. On ‘doulological logics’, see De Wet 2015, esp. 1–44. De Wet 2020. See also Gal. 1:10, Phil. 1:1. 1 Cor. 7:22, Col. 4:12, Phil. 1:1, Eph. 6:6. Shenoute, Abraham, Our Father. Shenoute, Abraham, Our Father.

Divining Slavery in Late Ancient Egypt

157

sackcloth before God. Because of this, there is not mercy in them towards the enslaved of God who sit in their cell.’27 An almost identical version of the passage is found in MU0166, but the phrase ‘enslaved of God’ is replaced by the Coptic word for ‘man’ (roˉ me).28 Tito Orlandi, the editor of these texts, notes that it would be rather odd to delete mention of God, and so the White Monastery version must be a later addition.29 Thus, language of enslavement was intentionally added into Paul of Tamma’s text at some later date by a scribe, doulological logics so permeating the community’s (or at least the copyist’s) worldview that the passage was only made meaningful through appeal to slavery. But if Paul of Tamma did not use language of enslavement in that verse, he did still deploy the logics of slavery in an extant letter. The addressee and beginning of the letter are now lost, but the latter portion and Paul of Tamma’s name are present. While he never uses the term ‘enslaved’ here, he appeals to Phlm. 21 near the end of the letter to demand the recipient’s obedience to his message. In the scriptural passage, the Apostle Paul asserts his authority to expect Philemon to defer to Paul regarding the enslaved Onesimus’ treatment. Mimicking the Apostle, Paul of Tamma redeploys these words to express his own anticipation of his addressee’s obedience. In this new, asceticized context, one in which obedience was a crucial aspect, the virtue of ‘listening’ to the commands of another mark one as a successful ascetic. That Paul of Tamma used this scripture in a new way was meant to bind the ascetic to Paul of Tamma as master, deploying the logics of slavery, even if the unknown ascetic recipient was not explicitly being viewed as Paul’s enslaved. In this way, Paul of Tamma asceticized Phlm. 21 to express the relationship between ascetic master and pupil as akin to that of the Apostle Paul as master and Onesimus as enslaved. Ascetic renunciation was also formulated through intertextual reading of passages alongside one another to craft new meanings.30 In both Paul of Tamma and Shenoute, we find just such creative interpretative work occurring around the language of enslavement. Paul’s own selfidentification as ‘enslaved of Christ’ is productively read alongside Hebrew Bible prophets, themselves popular models for Egyptian ascetics. In particular, Paul as enslaved was intertextually woven with the image of

27 28 29 30

Orlandi 1988, 102. Orlandi 1988, 96. Orlandi 1988, 146. Clark 1999.

158

Christine Luckritz Marquis

various prophets as ‘enslaved of God’ sent to correct a disobedient people. Paul of Tamma places the ‘enslaved’ in parallel with the ‘prophets’, indicating that the prophets were God’s enslaved, too.31 Shenoute is more robust in his linking of the prophets and Paul in order to construct an ‘enslaved’ model that the ascetic ought to emulate. In the passages from Abraham, Our Father noted above, the language of enslavement continues. Having imaged Jesus as taking on ‘the form of an enslaved person’, and Paul as ‘the enslaved of Jesus’, Shenoute continues his list: ‘and like all the apostles and all the prophets, the ones who became the enslaved of the Lord, Christ Jesus’. Then, Shenoute evokes both Jer. 7:25–6 and 44:4–5, eliding the two passages for his purpose. In Shenoute’s hands, God/Scripture speaks directly: ‘I have sent my enslaved, the prophets, from the beginning. I sent them and you did not listen.’32 Here, Shenoute uses the Coptic hmhal, a translation of the Septuagintal Greek, doulos. While the Hebrew people disobeyed the prophets in the original context, the clear audience here is the sisters under Shenoute’s care. Building his argument for their obedience, Shenoute fashions himself as enslaved prophet sent by God to the enslaved sisters under him. Like the Apostle Paul and early bishops such as Ignatius of Antioch, Shenoute’s rhetoric reflects appeals to ‘mutual slavery in Christ’ even as his discourse reasserts his power over the sisters.33 In this way, he wraps his image of appropriate ascetic behaviour in scriptural images of enslavement, even as he maintains and fortifies the hierarchical elements that place those sisters under his mastership.

children, enslaved, and monks in shenoute’s community Like the malleability of language and status, the scripturalizing of the ascetic body challenges our abilities to find the enslaved. And yet, by broadening our scope, as done above with tenant farmers, we find 31 32 33

Orlandi 1988, 96, 104. Shenoute, Abraham, Our Father. Shaner 2018, 50. In particular, chapters three and five explore how power discourses in the writings of Paul and Ignatius sought to resolve troubling power dynamics when enslaved individuals held leadership roles. Shenoute’s deployment of this strategy is somewhat different in that the sisters here seem to be concerned about being demoted to the status of enslaved, rather than being concerned about enslaved individuals holding positions of authority over them. More on this below.

Divining Slavery in Late Ancient Egypt

159

enslaved individuals hiding in plain sight in Shenoute’s community.34 The Shenoutean federation interfaced with actual enslaved people and with those perhaps perceived as enslaved or enslaveable. Among those whose status may have seemed ambiguous are the children who resided with the monks. While the precise circumstances of their arrival remain unclear, there certainly were children being cared for by the Shenoutean monks, and it seems probable that at least some of these children were gifted to the monastery. In his rules for the community, Shenoute mentions proper care of children among them.35 Caretaker monks were to be some of the most advanced in ascetic practice, Shenoute offering warnings to any caretaker who is too playful.36 While some children may have joined the community along with their parents (but were clearly no longer under their direct care), others were plausibly given to the monks as a thanks-offering or donation to God. The practice of child donations later became more common at the Monastery of Phoibammon in the eighth century.37 As Arietta Papaconstantinou has shown, donation contracts make clear that monks benefitted from the labour or wages that such children were able to offer the monastery. Just as the labour of exposed children who became enslaved benefitted their masters, so too the labour of the donated child belonged to the monk-as-master (the language of ‘master’ sometimes appearing in the donation documents). That is, by the eighth century, the Monastery of Phoibammon explicitly housed children regarded through a doulological lens, some donation documents even referring to the children with the terms hmhal and kauon/kyauon.38 As Caroline Schroeder rightly cautions, we must be wary of retrojecting what these later documents describe back into fifth-century monasticism.39 And yet, it is tempting to wonder if some of Shenoute’s fellow monks did, in fact, treat their young counterparts in ways that resonated with contemporary doulological logics. Apparently, some monks who oversaw the children’s

34 35

36 37 38

39

My work is deeply indebted to the approach of Joshel and Petersen 2017. Shenoute, Canons vol. 9 MONB.DF 48 (ed. Crum 1905, 84), MONB.YX 56 (ed. Amélineau 1888, 280–1); Schroeder 2018. Shenoute, Canons vol. 9 MONB.DF 186–7 (ed. Leipoldt 1913, 105–6); Schroeder 2018. Papaconstantinou 2002a; 2002b. P.Kru.97: hmhal and kauon/kyauon are used interchangeably, indicating that in the eighth century they were synonymous. Schroeder 2018, 8.

160

Christine Luckritz Marquis

work had to be rebuked by Shenoute for beating the children.40 As Schroeder notes, Shenoute’s concern in this rule seems to be the hypocrisy of the adult brethren rather than the mistreatment of the children.41 Such penal abuse often was issued against both children and the enslaved. The recent excavation of a first- to fifth-century cemetery at Kellis, in Upper Egypt, presses deeper consideration of what such abuse could look like. The skeletal remains of a two-to three-year-old child’s body from the cemetery show the child suffered repeated intentional beatings.42 While we cannot be certain whether the child was of freeborn or exposed/enslaved status, the injuries sustained indicate that such beatings might bruise and break bones. Julia Hillner has made the compelling case that the extension of corporal punishment to adult monks (male and female), evidence for which is found both in Pachomian and Shenoutean rules, invited the blurring of status between child, slave, and monk.43 Indifference towards child beatings, except inasmuch as it represented hypocrisy, indicates a larger cultural position on physical abuse in Shenoute’s monastery. The ambiguity between the treatment of children, enslaved, and monks suggests doulological logics informed the treatment not only of children within the monastery but also of all its members. So while it is impossible to know whether children within the monastery might have been donated and therefore viewed by some as (quasi-) enslaved, their presence and treatment suggest the presence of doulological reasoning in the construction of the community.

suffering, labour, and ascetics as god’s enslaved Shenoute’s writings also reveal that monastics sometimes worried about being read by others as enslaved individuals. Shenoute’s rebuke of disobedient female monastics evidences anxieties among the sisters about the ambiguity between themselves and the enslaved. Above, I drew on Abraham, Our Father to highlight Shenoute’s interpretation of the Hebrew prophets as enslaved. The letter was originally written to female monastics under Shenoute’s authority. As Rebecca Krawiec has shown, Shenoute deploys and queers familial relations, especially around inheritance, to construct a new spiritual lineage situating him as the paterfamilias

40 41 42 43

Shenoute, Canons vol. 9 MONB.YX 56; Schroeder 2018. Schroeder 2018, 4. Wheeler et al. 2013. Hillner 2009.

Divining Slavery in Late Ancient Egypt

161

of all other monastics in the federation.44 Across the entire text, Shenoute repeatedly returns to his depiction of the patriarch Abraham as bequeathing an inheritance to the obedient, hard-working monastics. Krawiec interprets in this language Shenoute’s desire to construct proper spiritual kin relations: brothers, sisters, father; the father imparting his ‘wealth’ to his children. But Shenoute’s letter makes equally clear that other household members – the enslaved – were also invoked in the argument between Shenoute and the sisters. From the letter’s opening, Shenoute differentiates between Isaac, who receives an inheritance, and Ishmael, who does not. He twice uses hmhal to refer to Hagar, and by implication her son.45 The obvious translation into English is ‘enslaved’. As he shifts to the next generation, he again uses the term hmhal and the verb rhmhal, ‘to serve’, to describe Esau’s relationship to Jacob. Here, how to translate hmhal is more interpretative. Given its coupling with the language of service, perhaps it would be better to translate as ‘servant’ and read Shenoute as intentionally playing on the term’s multivalent meaning. Having established his mixed use of hmhal, Shenoute continues his description of biblical ancestors before turning to rebuke the sisters, who have been refusing to do the work assigned to them. As Krawiec notes, Shenoute is angered over what he sees as misinterpretation of Scripture to defend ‘refusing monastic labor’.46 Shenoute’s argument thus seeks to link proper labour (hice and šphice) to inheritance and kinship. Recall from above that Shenoute refers to the ancestors, the prophets, and Paul all as hmhal, which I have translated as ‘enslaved.’ Elsewhere in the letter, he equates their example with ‘suffering labour with others even until death’.47 To belong to the appropriate lineage requires being the right type of hmhal, which means performing the assigned labours (hice and šphice) – the type of serving that biblical predecessors had done before. As Ellen Muehlberger has noted, the terms for ‘labour’ that Shenoute uses throughout the letter, hice and šphice, also carry the connotation of suffering.48 Thus, the type of labour the heirs comprising Shenoute’s community are called on to perform may produce suffering. That the works assigned are difficult and unpleasant seems to be 44

45 46 47 48

Krawiec 2002, 161–74; Krawiec 2017. Krawiec also generously discussed my interpretation. Shenoute, Abraham, Our Father. Krawiec 2017, 465. Shenoute, Abraham, Our Father. Muehlberger 2017, 181 n7.

162

Christine Luckritz Marquis

something Shenoute and his opponents agree upon. Where the fallout has occurred is whether such labours are appropriate to the status of monastics. Shenoute’s use of hmhal throughout the letter makes clear that some of the sisters had argued that the labours they were assigned and the accompanying treatment by their superiors effectively ‘enslaved’ them. Thus, at the end of the letter, Shenoute rebukes the sisters, quoting them back at themselves: ‘Therefore, let us not blaspheme, saying, “Those who rule us are our masters and we are beneath them in the manner of these enslaved.”’49 The women’s complaint thus becomes clear: they are being treated as if enslaved. While I would not want to overemphasize this grammatical choice, the presence of the demonstrative particle ‘these’ (nei) may hint that there are actual enslaved individuals nearby to whom the complainants could point. Continuing, Shenoute attempts to invert the logics of master–enslaved, asserting that those who seem to be the masters are in fact the enslaved, and that the lowly monastic is actually the master. Useful to Shenoute in his counter argument is the idea that all monastics are enslaved to God/Jesus Christ.50 But equally helpful for his purposes is the ambiguity between ‘servant’ and ‘enslaved’ that he established by paralleling Ishmael and Esau at the outset. These two rhetorical moves, coupled with his rhetoric about the necessity of suffering labours, allow him to make the claim that they have misread the situation. Again and again, Shenoute returns to the language of hmhal to describe the ancestors, until halfway through when he makes clear that the interpretation of his interlocutors, that they are ‘enslaved’, is inaccurate. Citing John 15:15–16, Shenoute asserts that just as Jesus shifted from calling his disciples ‘enslaved’ to ‘brothers’, so too should the sisters understand the nature of their status within the monastery. Jesus tells the disciples his plans, noting that masters do not share their business with their enslaved. Shenoute suggests a parallel between Jesus and himself, for Shenoute is likewise sharing his knowledge with the sisters rather than leaving them ignorant like the enslaved. Weaving Matthew 10:16, 28, into his argument, Shenoute goes on to argue that any shame or abuse they are receiving is because they are extensions of their master, Jesus. Thus, through his interpretation and intertextual reading, Shenoute accepts their view of themselves as akin to the secular enslaved, then converts their status to kin, before finally offering them a new enslavement to Christ. 49 50

Shenoute, Abraham, Our Father. Shenoute, Abraham, Our Father.

Divining Slavery in Late Ancient Egypt

163

The move of making everyone enslaved obscures actual treatment of individuals within the community as suffering enslavement. Shenoute attempts to flatten one hierarchy (master–enslaved) in order to emphasize the hierarchy he prefers (father–children). But given the conflation of enslaved and children, especially around abuse, and the ways in which monks might also be subjected to corporal punishment, it is understandable that these sisters are protesting their assignments and concomitant (mis)treatment. As I have noted, the delineation of one’s status as free and not enslaved was not clear-cut, and we can imagine it was much more so the case for those in a monastery. Those who joined were expected to renounce any material possessions to the monastery. This requirement effectively removed any resources to fight mistreatment as well as any means for bequeathing an inheritance, an ability that was often used to demarcate the boundaries between free and enslaved persons. Moreover, Shenoute’s rules explicitly required obedience. Any disobedience might result in beatings (as for an enslaved individual) or expulsion, leaving an ex-monk without any possessions or resources for survival. Thus, committing to the monastery likely made it difficult for mistreated monastics to delineate their status from that of surrounding enslaved individuals. Shenoute’s evocation of enslavement did not end with the letter’s arrival to its addressees. Incorporated into Shenoute’s Canons, it was read multiple times over the years to all the monks. In this way, year after year, all the monastics within the Shenoutean federation were reminded of the proper ways to imagine themselves enslaved to God/Christ and to ignore any resemblance in their daily lives and treatment to enslaved people.51 By emphasizing the scripturalized, spiritual enslavement of the entire community and the necessity of this for successful inheritance of God’s promise, Shenoute erased actual mistreatment as well as the presence of any enslaved or formerly enslaved individuals within the community.

shenoute’s relations with the labourers of gesios Another place the enslaved emerge in Shenoute’s corpus is around his disagreements with a certain Gesios. Fleshing out the larger late antique context in which Shenoute’s community lived highlights the blurry status distinctions between poor and enslaved in the neighbouring villages, especially around labour practices.52 In previous scholarship, Shenoute’s 51 52

Muehlberger 2017, 178. For full chronology, see Brakke and Crislip 2015, 193–9.

164

Christine Luckritz Marquis

engagements with Gesios have been drawn upon to depict the archimandrite as anti-pagan and violent.53 Gesios, a wealthy landowner from the nearby metropolis of Panopolis, somehow found himself at odds with Shenoute and became the target of Shenoute’s vitriol. As Ariel López has shown, crucial to Shenoute’s argument was the (mis)treatment of the poor.54 In this way, the argument was about contested economic and cultural practices. Such poor lived in neighbouring villages and epoikia, the landowner-established settlements described above. And it is among such poor that we should imagine there were enslaved people sharing in labour, and that such poor may well themselves have held an ambiguous status vis-à-vis enslavement. In the earliest datable text Shenoute wrote against Gesios (his sermon Not Because a Fox Barks), language of enslavement opens his argument. He contrasts Gesios as ‘the enslaved of mammon’ to himself as ‘the enslaved of Christ’.55 He then links this juxtaposition to fatherhood (and thus inheritance) in ways similar to his rhetorical tactic in Abraham, Our Father. Because Gesios’ father/master is the devil, he has mistaken his wealth/inheritance as his own. By contrast, because Shenoute has chosen Jesus as his father and all belongs to Christ, Shenoute’s inheritance is the whole of creation as a gesture towards future heavenly rewards. Since everything belongs to Christ, Shenoute can assert that complaints about Gesios’ people and stewards stealing from Gesios’ home to give to Shenoute stem from Gesios’ bad pedigree. The belongings are not Gesios’ but Christ’s, and Shenoute is not a thief, but merely acting as Christ’s enslaved (o nhmhal), collecting them on the behalf of ‘his brothers’.56 Shenoute makes clear that some portion of these offerings is used for care of the mistreated poor. Shenoute then counters that Gesios is the true robber, accusing him of numerous abuses of the poor: stealing animals, carts, hay, and any other possessions from the poor; overcharging widows, the poor, the elderly, orphans, and strangers for meat; making the poor raise his livestock with no compensation; and extorting grain payments from them for baths they were forced to build but do not use.57 Piling one accusation atop another, Shenoute builds a formidable list of abuses of the poor against Gesios. But

53 54 55 56 57

Bell 1983; Gaddis 2005, 151–207. López 2013. Shenoute, Not Because a Fox Barks. Shenoute, Not Because a Fox Barks. Shenoute, Not Because a Fox Barks.

Divining Slavery in Late Ancient Egypt

165

the disagreement between Shenoute and Gesios also circulated around the ambiguous status Gesios’ treatment created for those working for him. As part of his accusation of ‘forced labour’ (more on which below), Shenoute asks Gesios, ‘Are the villages your epoikia?’ He continues, asking if Gesios built their homes (the way a landowner would for tenants of an epoikia), with the obvious implication that Gesios did not. That is, part of Shenoute’s concern seems to be not the labour practices per se, but upon whom they were being placed. Individuals living on an epoikia might expect different treatment, but when free villagers were subjected to such treatment it became abuse. Equally critical to Shenoute’s argument against Gesios are his repeated assertions that Gesios forces the poor to labour. He conveys this force through the causative verbal formation and three different words for labour in Coptic: kba (‘forced labour’), hice (‘suffering labours’), and hbeˉye (‘works’). Hice, in Abraham, Our Father, had held positive connotations for labour the sisters must suffer as Christ’s enslaved in order to receive their inheritance in Christ. By contrast, Shenoute here leans into the suffering connotations of the term to amplify the horrors of Gesios’ treatment of the poor, the term being coupled not only to ‘forced labour’, but also ‘violence’ (cˇ inkyons).58 And lest there be any doubt that Shenoute invokes enslavement labour conditions, he ends his piece with a prayer comparing the enslavement of the Israelites and the ‘works’ Gesios has made the poor do.59 He asserts that just as the Israelites were not able to speak freely because of their enslavement, so too the poor cannot because of their near enslavement to Gesios, and that thus the poor were not able to fully serve (rhmhal) God.60 While he does not explicitly ever accuse Gesios of enslaving the poor, his accusations of taking their labour without pay coupled to his scriptural allusion to the Israelites leaves little doubt that he sees Gesios as treating the poor as enslaved people. Ariel López has argued extensively that this and other texts regarding Gesios point to Shenoute’s career-long campaign of caring for the poor. But as López notes, the ‘poor’ for Shenoute seem to cover a rather broad swath of the population. Leaning into the ambiguity of the Coptic term heˉ ke, Shenoute thus refers both to individuals with a modest number of possessions (including land) as well as to beggars.61 I would argue that

58 59 60 61

Shenoute, Not Because a Fox Barks. Shenoute, Not Because a Fox Barks. On violence and parrhesia (or to ‘speak freely’), López 2013, 34–41. López 2013, 43–5.

166

Christine Luckritz Marquis

Shenoute’s concept of the ‘poor’ also encompassed the enslaved. While his concern about proper treatment vis-à-vis social status allowed him primarily to address those villagers who were free, his focus on Gesios forced him also to deal with enslavement. In Acephalous 26, Shenoute extensively attacks Gesios, including an assertion that Gesios’ mistreatment of ‘many enslaved and numerous timid poor persons’ has led them away from a focus on God.62 Thus, we explicitly find Shenoute coupling the poor and the enslaved who work for Gesios and a hint that the number of enslaved was fairly substantial. While the mentions are brief, Shenoute on two occasions confirms that Gesios (and others like him, probably) owned enslaved individuals, some of whom interacted with Shenoute and were even Christians.63 In the opening of his God Says through Those Who Are His, whose audience included people tied to Gesios, Shenoute makes clear that the fates of masters and their enslaved are not tied together.64 He speaks to the believing enslaved persons who must suffer under their masters’ lacking faith, assuring them that in the afterlife there will be no enslavement. Given Shenoute’s purpose in addressing those tied to Gesios and this opening crack against masters, it is little stretch to imagine that Shenoute is speaking to enslaved Christians who are owned by Gesios. This supposition becomes certain when taken with Let Our Eyes, a letter in which Shenoute defends having broken into Gesios’ home to steal his pagan idols. Written to the inhabitants of Panopolis, Shenoute attempts to control the narrative: maintaining that Gesios’ doors opened by divine intervention, Shenoute seeks to protect others in Gesios’ house who are obviously being accused of abetting Shenoute’s robbery. He asserts that no one else collaborated with the monks, ‘neither enslaved nor gardener nor anyone else connected’ to Gesios.65 That Shenoute must repeat this claim more than once indicates that his relationships with the poor and enslaved people linked to Gesios had led Gesios and others in the community to suspect that enslaved individuals colluded with Shenoute. Given Shenoute’s tendency to sometimes attract large audiences for his sermons, it should thus not surprise us to find enslaved individuals among the other poor Christian attendees. When Shenoute later refers to his burgling of Gesios’ home, he does so to defend himself against accusations

62 63 64 65

Shenoute, A26; Behlmer 1996, 101–2. De Wet 2018a, 105, is correct that secular slavery was present in Shenoute’s context. Shenoute, God Says; Amélineau 1907, 262. Shenoute, Let Our Eyes; Emmel 2008, 196.

Divining Slavery in Late Ancient Egypt

167

of ‘gathering a mob’.66 Shenoute wrote a short text, referred to by Stephen Emmel as Shenoute’s ‘anti-pagan c.v.’, in response to a different set of anti-pagan activities in which Shenoute later found himself embroiled.67 Mentioning the Gesios event, Shenoute claims he only took ‘seven monks’ with him, not a large crowd. His action is one in a series of activities Shenoute notes he was involved in, though he assures that he has ‘done nothing in a disorderly fashion’. The accusation of gathering a mob seems indeed to have been a heavy charge placed against Shenoute.68 Such bands in late antiquity, like that which Shenoute is accused of having gathered, would have consisted of poor youths and enslaved people.69 Shenoute’s clear self-representation as advocate of the poor and enslaved made such accusations by his opponents seem more plausible; he was already affiliated with those people. Thus, even as he attempts to defend himself from such an accusation, we catch a glimpse of the enslaved Christian in the midst of Shenoute’s community. The presence of enslaved people in such bands or crowds likewise appears elsewhere in late fourth- and early fifth-century Egypt. In particular, references to Ethiopian enslaved people by authors who were either themselves monastics or who were closely tied to an ascetic network indicate that such enslavement was known in the Egyptian context. In the early fifth century, Palladius, an ascetic turned bishop, could accuse Theophilus, the archbishop of Alexandria, of gathering Ethiopian enslaved persons (whom the archbishop himself may have owned) to attack monks in the north-western deserts of Egypt.70 That Theophilus returned the accusation, stating that the monks themselves had gathered a band of poor and enslaved people to threaten him in Alexandria, indicates that there may well also have been enslaved people among the monks of Lower Egypt.71 That is, gathering a band or crowd of people was a common occurrence (and common accusation) between parties competing for power.72 There can be no doubt that the crowds Shenoute gathered as audiences and the crowds he was accused of deploying against Gesios included enslaved individuals. His heated arguments with Gesios make

66

67 68 69 70 71 72

Shenoute, ‘anti-pagan curriculum vitae’; Emmel 2017, 394. I follow Emmel’s translation of meˉ eˉ še as ‘mob’. Emmel 2017. Emmel 2017, 389–90. On such bands in North Africa, see Shaw 2011, 633–7. Pall. V. Chrys. 7; SC 341.144–6. Thphl. Al. Syn. Ep. 2.3, in Hier. Ep. 92.2; Hilberg 1913, 150. On crowd-gathering among Egyptian Christians, see Luckritz Marquis, 2022.

168

Christine Luckritz Marquis

clear to whom enslaved Christians were truly indebted: Shenoute as abbot-master standing in for God as true master.

conclusion For historians reconstructing the story of Shenoute’s monastic federation, one of the most interesting questions remains who built the Shenoutean monastic buildings. Archaeological studies and art historical restorations on the White and Red monasteries underscore the scale and elaborate workmanship that went into their construction and decoration.73 Thus, the buildings would have required and exhibited great wealth on the part of the monastic community. Such seems largely in keeping with what papyri a century later show – that churches and monasteries in Egypt owned about 15 per cent of the arable land in a given district.74 Control of so much land, like the Shenoutean buildings themselves, indicates the need for a large labour force. It is about these crevices in the evidence that historians are left wondering: did the monks alone build complex architectural structures and work the growing land given to them, or were enslaved individuals and neighbouring poor part of the labour force? Given my reading, we should see Shenoute’s acts less as liberation of the poor and enslaved and more as redirection of their labour. As I have shown, the neat delineations we might wish for between poor, enslaved, and even monk are often lacking in our evidence. We are on surer ground when we realize that all those who built the monasteries would have been ‘enslaved of Christ’, but among them were undoubtedly actual enslaved and status-ambiguous poor. The discussion of Shenoute’s use of enslavement language above resonates with this picture. While Shenoute far more frequently appeals to enslavement as a scripturalizing, spiritualizing rhetoric for his own community, or as an evocation of violent mistreatment of the poor, his brief comments about the fates of actual enslaved people indicate that enslaved people did live in his larger social sphere. Enslaved people also would have (however infrequently) moved through Panopolis and therefore passed the monastery on their way to being sold. While the evidence for enslavement in late ancient Egypt can seem meagre and challenging, it is possible to 73 74

Brooks-Hedstrom 2005; Bolman 2016. Hickey 2007, 298; SB XX 14669; P.Lond.Copt. 1075.

Divining Slavery in Late Ancient Egypt

169

responsibly imagine enslaved people back into history. In order to have a fuller picture of enslavement in Egypt (and I would argue elsewhere in the late ancient world), historians need to employ enslaved people in places we know they must have been and look for them in places/texts where they may be hiding in plain sight.

8 Rural Slavery in Late Roman Gaul: Literary Genres, Theoretical Frames, and Narratives Uiran Gebara da Silva

Historians must tread carefully when they deal with both modern and ancient slavery, for every time they brand the practice as an archaic form of domination, it reappears adapted to new ages. Although most of the modern world, since the nineteenth century, has strived to make the ownership of one human being by another illegal, slavery keeps returning in the most insidious ways. At the turn of the century, when modern societies claimed to have finally gotten rid of it, slavery began to resurface, masquerading as free labour all across the globe1 (and not just in the Global South2), bringing the need to discuss and to enforce policies against ‘practices similar to slavery’ back to international forums.3 It is striking that while this notion of an ‘insidious return’ has recently appeared in studies on rural slavery in the ancient world,4 it seems not to have influenced ones focused on late Roman Gaul. The purpose of this chapter is to address the presence of rural slavery in late Roman Gaul (fourth to fifth centuries ce) as described in Gallo-Roman literary evidence. After a brief historiographical survey, I address the methodological and theoretical challenges offered by literary sources, discussing the ways in which ideology and literary genre frame representations of rural

1 2 3 4

The research presented in this chapter was funded by a FAPESP postdoctoral scholarship. I must mention the very helpful comments made by Fabio Joly on the early draft of the paper that I presented at the conference in Tvärminne. I would also like to thank those who provided comments and input during the conference, in particular Antti Arjava, Elizabeth Fentress, and Pieter J. J. Botha. The pioneering work by Kevin Bales was recently revised and updated: Bales 2012. GRETA 2018. For this terminology, see the appendix 2 in Bales 2012, 277. Harper 2011; 2012; Vera 2012; Vlassopoulos 2016; Wickham 2005.

170

Rural Slavery in Late Roman Gaul

171

slaves. Subsequently, I present an analysis of relevant passages taken from Ausonius, Salvian of Marseilles, Paulinus of Pella, and the Querolus, a late Roman comedy, highlighting the permanence of more intensive forms of rural labour exploitation in the countryside. To close, I make some remarks regarding the challenges that new readings and narratives about ancient slavery pose for our understanding of the late Roman Gallic countryside, while indicating the need to stress rural slavery’s qualitative differences from other kinds of rural labour exploitation.

narrative, ideological, and literary frames for rural slaves During the early twentieth century, scholars’ views on ancient slavery were far from homogenous. While for some it was the most distinctive feature of the ancient world, for others it was merely a secondary form of labour exploitation; sometimes the issue disappeared from scholarship, only to return later in the form of heated debates in which the ideological oppositions were made more explicit than usual.5 By the second half of the twentieth century a consensus had been achieved, and slavery was considered the most important form of labour for the classical societies in antiquity, if not in quantitative terms then at least in qualitative terms. With the ancient world being seen as largely founded on agrarian economic relations, rural slavery was then considered a paradigmatic modality of slavery.6 This consensus was then supplemented by a vision of the end of ancient slavery that was an integral part of the paradigm of ‘Decline and Fall of the ancient world’.7 In the traditional modern narratives about antiquity, slavery ended with that age, ceasing to be a relevant form of labour control and disappearing together with (or causing) the end of the Roman Empire.8 As argued in the theories on the crises of ancient slavery, from the third century onwards it was replaced by the colonate, considered a prototypical form of medieval serfdom. Scholarship considered the colonate as a form of labour control that fulfilled an equivalent role in the modern narratives about the medieval societies.9

5 6

7 8 9

Yavetz 1988; McKeown 2007c. The work that most clearly asserts such consensus is that of Finley 1998; 1999; see also Vlassopoulos 2016. On the idea of ‘Decline and Fall’: Ando 2008. Harper 2012. The crisis of slavery was an issue marking major divisions between the Finleyan view and the one exposed by Marxist historians, such as Ste Croix 1998 [1983] and Carandini 1980, although both sides agreed on the rise of the colonate.

172

Uiran Gebara da Silva

In recent decades, however, there have been important changes in how historical studies deal with the subject of rural slavery in the later Roman Empire. The grand narratives about the end of the ancient world gave way to a more loosely defined late antiquity.10 Inside this new framing, the twentieth-century consensus that had claimed ancient slavery waned in that period was replaced by uncertainty about its ending,11 and by a new awareness of the numerous references pertaining to slaves in late antique literary sources,12 as well as in archaeological evidence.13 This new awareness presents two related issues that are important for my argument. The first concerns the limited impact that this new awareness had on studies about rural slavery in late Roman Gaul. On the one hand, slavery in early Roman Gaul has never received much attention, and most scholars have followed the ideas of Fustel De Coulanges, who described the late Roman Gallic countryside as mostly or exclusively occupied by peasants and coloni.14 While some scholars have opposed this perspective, the bulk of their reasoning was mostly based on archaeological evidence15 and, until recently, this opposition had little impact on general narratives about late Roman Gaul.16 The second one concerns how recent analyses of late antique slavery that employ revisionist modernist interpretations of ancient economy tend to underestimate slavery’s particular social features as a form of exploitation and domination. Ancient Economy scholars have replaced the Marxist and Weberian models with New Institutional Economics.17 On the one hand, although there were fundamental differences between the more economically oriented Marxist ‘slave mode of production’ and the more culturally defined Weberian ‘slave society’,18 both Marxist and Weberian models equated rural ancient slavery with modern plantation slavery. On the 10 11 12 13

14 15

16

17 18

Giardina 1999; Ando 2008. Ironically, a movement inspired by Finley’s ideas; see Finley 1998. Harper 2011; 2012. Recent research on medieval slavery moves in a similar direction, arguing in favour of the permanence of slavery in the Early Middle Ages, see Rio 2017. See also the pioneering work of Dockes 1982 and Bonnassie 1991. De Coulanges 1894; Wightman 1978; MacMullen 1987; Whittaker 1993a; Finley 1998. Daubigney & Favory 1974; Samson 1989; Thompson 2003; Ouzoulias 2006; Roymans & Zandstra 2011. This situation has started to change. Wickham 2005 draws attention to the presence of rural slavery in fifth-century late Roman Gaul. See also Harper 2012 and Grey 2011b, who present the literary evidence for slavery in Gaul but do not address it as a particular historical problem. Very representative of this approach is Scheidel, Morris, & Saller 2007. For the paradigmatic use of those ideas, see respectively Carandini 1980 and Finley 1998.

Rural Slavery in Late Roman Gaul

173

other hand, both also stress the differences between ancient and modern economies. Since recent scholarship is strongly challenging both this equivalence of ancient slavery with modern plantation slavery,19 and this differentiation of ancient and modern economies,20 the replacement of the old models with New Institutional Economics is perhaps unsurprising. Moreover, the models derived from New Institutional Economics, instead of ‘exploitation’ and ‘political control’, have established concepts such as ‘diversity of portfolio’, ‘intensification and abatement’, ‘capital’, ‘growth’, and ‘development’ as part of the common language for studies of the economic history of the later Roman Empire (including late Roman Gaul). One of the benefits of this new approach is that it makes it possible to look with fresh eyes at the evidence of rural slavery in the early and late Roman Gallic countryside. The downside is that rural slavery here is considered to be just one form of labour among others, albeit perhaps a somewhat more intensive one.21 I will return to this issue in my final remarks. Besides the framing created by scholars’ narratives and models, there are also challenges presented by the ideological and literary framings of ancient literary sources. What I am hereby referring to as ‘framing’ is the way in which the formal structures – and the social contexts of enunciation and production – of any discourse shape and imprint its referential content. On the one hand, the social contexts of production and enunciation weave class, status, worldviews, and religious or political alliances into the texts, while on the other hand, the framing produced by poetical or prose genres acts upon the referential content, not realistically reflecting it, but recreating it within an expected literary context. In this way, the genre of a given text can be seen as the ‘ideological’ framing that helps to recreate the outside world within a world of texts. The division between formal structures and social contexts is analytically necessary, because, while the reader eventually sees these two dimensions of the text as a whole, the scholar must first analyse them separately, before trying to understand their subsequent articulation or integration into a single thing. The text and its internal ideology, the genre, act as mediators for the social forces being represented. In turn, the texts and the genre are themselves transformed by the social forces.22

19 20 21 22

Roth 2007. Harper 2012. As, again, in Harper 2012. For this methodological perspective articulating social context and language, I was inspired by the concept of ‘mediations’ proposed by Sartre 1963 and by the definition of the literary genre as an ‘ideology of the texts’ formulated by Conte 1994. For

174

Uiran Gebara da Silva

The analysis below follows a chronological sequence: the order from the earlier to the later texts is Ausonius, Querolus, Salvian of Marseilles, and Eucharisticus. I have selected passages from these authors for their significance as sources, having been widely used by scholars since the nineteenth century in historical works about late Roman Gaul. Also, I do not intend to create here the illusion of a historical evolution. In truth, I believe that this chronological sequence helps to deconstruct the narrative of the decline of slavery since it presents slavery as an unchanging aspect of the social relations depicted in the passages.

ausonius The Aquitanian poet was seen, not so long ago, as one of the most impressive cases of upward social mobility in late Antique Gaul,23 but nowadays scholars tend to propose a more nuanced view of his presence at the imperial court.24 However, there is a certain agreement among scholars that, by the end of his life, he was one of the wealthiest men in Gaul.25 Moreover, until recently, classicists tended to depict Ausonius as a scholastic poet, whose style and themes were seen as quite representative of the decadent cultural landscape produced by the elites of an Empire in decline.26 This view has been challenged by a new wave of studies on late antique poetry that have recast not only Ausonius’ work, but the whole of Latin and Greek poetry of late antiquity, on its own terms, instead of judging it against ‘classical’ criteria.27 This new state of affairs has affected the historical and poetical analysis of Ausonius’ work. What was once seen as artificial, including the ubiquitous allusions to classical Latin and Greek writers (even when he writes about everyday matters), the long descriptive passages, the playful and humble elocution, and the loosely structured form of his poetry, are now understood as an expected poetical development from Golden and Silver Age poetry, with Ausonius being one of the best poets of what Michael Roberts has termed ‘the jeweled style’.28

23 24 25

26 27 28

discussions of late antique social representation in literary sources, see Wood 1992 and Miles 1999. Hopkins 1965; Jones 1992. Skinner 2013. Matthews 1990, 81–4; Jones 1992, 352; Sivan 1993; Brown 2012, 185–207; Skinner 2013. Isbell, 1974; Evelyn-White 1988a. Fontaine 1981; Roberts 1988; 1989. Roberts 1989.

Rural Slavery in Late Roman Gaul

175

There are not many references to slaves, and even fewer to rural slaves, in the texts written by Ausonius. When they do occur, they are always very stylized with classicizing allusions, as in the case of those in the Eclogae or the Cento nuptialis.29 Moreover, whenever he mentions rural workers, it is indeed usually impossible to say if they are free or unfree, as is the case in De Herediolo (although in the Mosella he specifically mentions coloni).30 However, matters are different in Epistula 26, a letter to Paulinus of Nola. Here, Ausonius offers a glimpse of the activities of a rural slave in Gaul. His depiction of the rural slave’s working life employs clever Plautian allusions, for the iambic letter is written emulating the early Latin playwright’s style.31 On the one hand, the poet weaves the whole episode within a poetical framing that does not cover or hide the events, but in fact places the man working for him in the proper ideological setting: Philo, who is bailiff (vilicatus) of my estate, or as he himself wishes, the administrator (ἐπίτροπος – for your Greekling thinks that a fine-sounding name which shows the gift of the classic tongue), unites with his complaints my prayers, which reluctantly I myself dispatch. (. . .) This fellow, when light harvest had oft belied his promises, came to hate the name of bailiff; and, after sowing late or much too early through ignorance of the stars, made accusation against the powers above, carping at heaven and shifting the blame from himself. No diligent husbandman, no experienced ploughman, a spender rather than a getter, abusing the land as treacherous and unfruitful, he preferred to do business as a dealer in any salemarket, bartering for ‘Greek credit’ (. . .) and by changing profits into losses and losses into frauds, he makes himself rich and me poor.32

Ausonius here employs the comedy genre as a marker of Philo’s low status, even though what he describes is some sort of upward social mobility. On the one hand, he is depicting the former baillif (vilicatus or vilicus) as a servus callidus, the quintessential character in Plautian comedy. On the other hand, he reveals some of the activities that were still 29

30 31 32

Auson. Cent. nupt. 5.63–5 (Oblatio Munerum), Ec. 23.10–11,15–16 (De feriis romanis), Epigr. 36, 37. Auson. Hered. 24, Mos. 163–7 for coloni. Auson. Ep. 26. Auson. Ep. 26.1–34: Philon, meis qui vilicatus praediis,/ ut ipse vult, ἐπίτροπος,/ (nam gloriosum Graeculus nomen putat./quod sermo fucat Dorius)/ suis querellis adserit nostras preces,/ quas ipse lentus prosequor. (. . .)/ hic saepe falsus messibus vegrandibus/ nomen perosus vilici,/ semente sera sive multum praecoqua/ et siderali inscitia/ caelum lacessens seque culpae subtrahens/ reos peregit caelites./ non cultor instans, non arator gnaruris,/ promusque quam condus magis,/ terram infidelem nec feracem criminans/ negotiari maluit/ mercator quo libet foro venalium,/ mutator ad Graecam fidem,/ (. . .) ac lucra damnis, damna mutans fraudibus/ se ditat et me pauperat. Latin text and translation by Hugh Evelyn-White (1988b, 94–8).

176

Uiran Gebara da Silva

expected from a bailiff in the late Roman Gallic countryside. While Philo now seems to be a freedman still working for Ausonius as some sort of trader (negotiator), when the writer describes his previous activities as a bailiff, this opposition is important, for the poet has to describe Philo’s lack of skill and talent for dealing with agrarian toils. As a former villicatus, he would have acted as a sort of manager, overviewing the work of others in the countryside, who would often have been other slaves. Being incapable of taking care of the sowing, the ploughing, and the harvest, the servus callidus had a knack for bargaining (and that is why Ausonius seems to have promoted him to his new function as a trader). One should bear in mind that there is a contradiction between the poetical framing and the situation for which the letter was written, but this contradiction is productive, in the sense that it helps Ausonius to deal with the inherent contradictions that exist in the relation between masters and slaves. The poetical framing he employs presents many of the literary stereotypes associated with slaves: Philo is incompetent, unreliable, cunning, treacherous, and mischievous. Additionally, it presents Philo as a person with initiative and autonomy, a feature that creates tension in the master–slave relation that lies at the heart of Plautian comedy. For, while the Roman literary representation of slaves tends to present the good slave as someone who always obeys, it also recognizes a slave’s aptitude for autonomous reason.33 On the level of the language used and the poetical allusions, Ausonius creates an intellectual landscape in which he can deal with these contradictions. On the level of the depicted relationship between master and former slave, the mention of the rural activities previously overseen by Philo (including harvesting, sowing, husbandry, ploughing), while rendered in the Plautian style, must not be cast aside as merely a feature of the genre. Firstly, Plautian comedy focuses on domestic and urban slaves’ activities, and secondly, these rural activities are fundamental for the presentation of Philo as a former bailiff. In this sense, it seems to cast rural slavery as something unexceptional in Aquitaine. Regarding this presentation, Ausonius is probably joking about Philo’s incompetence, for he would not have been promoted to the position of freedman if he were quite as bad as Ausonius describes him. Nevertheless, while the poet recognizes Philo’s skills for the new role, he is also reminding the receiver of the letter, Paulinus of Nola, and Philo himself, of the latter’s slave origins. At the same time, Philo’s framing as a servus callidus suggests 33

Fitzgerald 2000.

Rural Slavery in Late Roman Gaul

177

the improvement of his social condition, justifying his role as a negotiator, but also puts him in his proper place, as a cliens and as a former slave.

querolus There is no real description of the activities of a rural slave in the late Latin comedy Querolus. The fifth-century play34 takes place in a rural setting (in contrast to the urban setting of the majority of Plautus’ and Terence’s comedies), in what appears to be the household of a small landowner. While there is no specific description of slaves doing rural work, the play presents an interesting caricature of a poor man’s rural household crafted by the writer of the play, a poet of higher status.35 The plot of the play revolves around a Stoic (though not Christian) moral message,36 for it consists of Lar (the household spirit) convincing Querolus (the small landowner) to accept the things that Fortune (the goddess) brings him. Querolus complains that his father died and left him no inheritance, while Mandrogerus, a clever comedic parasite disguised as a fortune teller, tries to con Querolus. When the con backfires on Mandrogerus, the result is Querolus’ discovery of his inheritance, and him finally being convinced by Lar. Different from earlier Latin comedies, the slave is not important to the plot, and the structural role of servus callidus that moves the play forward37 is shared between the parasite Mandrogerus and the Lar Familiaris.38 The household has two slave characters, Zeta (or Geta), a woman, and Pantomalus, a man, but only one of them, Pantomalus, actually appears and acts in the play. Two passages of the play concerning the representation of rural slaves are worthy of mention. In the first one, Lar makes a pun using Pantomalus’ name:39

querolus:

34

35 36 37 38 39

I have a slave I can’t endure, Pantomalus, and his character fits his name.

There is no consensus concerning the dating of the play. I am following Jacquemard-Le Saos 1994. Grey 2011a, 41. Jacquemard-Le Saos 1994; Vidovic 2009. Segal 1968; Konstan 1983. Vidovic 2009. Which could mean ‘All-the-evils’.

Uiran Gebara da Silva

178

Lucky you are, Querolus, if you have only one; many people have several Pantomaluses. quer. But don’t I hear many men praising their slaves? hous. Their slaves are still worse. quer. Why do they praise them, then? hous. Because they don’t know what they lose by them.40

household god:

Here, the ideology of slavery is presented and reversed, when the spirit downplays the negative effects of the bad behaviour of slaves to Querolus. A further development of this reversal appears as part of a long monologue spoken by the slave: Pantomalus: And this Witness (Arbiter) I’m going to get now – what an accursed villain he is! He gives his slaves less food and more work than is right. Heavens! He’d get a dishonest profit from an empty bushel basket, if he could! If these two men meet by chance or intention, they give each other advice.41

This passage consists of Pantomalus’ complaint against his master, a complaint that is attenuated by his recognition that the neighbour, Arbiter, is much worse with his slaves than Querolus is. As the master complains about the slave, the slave complains about the master, but the reasons for their complaints differ. While the master criticizes the laziness of the slave, the slave accuses the master of cruelty and greed. Such a reversal of the ideology of slavery is a common feature of classical Latin comedy and therefore its presence in Querolus is unsurprising. In keeping with the rules of the genre, it fuels the recreation of the Saturnalian atmosphere in Latin comedy,42 an atmosphere that destabilizes and reverses the social order represented inside the domus – at least for the duration of the play, for the order will be re-established at the end. Besides the depiction of the bad dominus, the second of the two passages quoted above is one of the few instances where heavy manual labour is mentioned in the play. Actually, no specific rural tasks are 40

41

42

Querol. 2.25: Querolus: Servus mihi est quem tolerare nequeo, Pantomalus et mente et nomine./ Lar: Felicem te, Querole, si unus tibi est Pantomalus, multi Pantomalos habent./ Q: Sed plures audio qui suos etiam laudant./ L: Isti peiores habent./ Q: Cur igitur laudant?/ L: Quia quid perdiunt nesciunt. Latin text by Catherine Jacquemard-Le Saos (Querolus 1994, 16) and translation by George Duckworth (1943, 906). Querol. 6.73: Pantom. Ille autem Arbiter ad quem nunc eo, quam sceleratus est homo! Servis alimenta minuit, opus autem plus iusto imperat. Inverso Hercle modio, si liceret, turpe captaret lucrum. Latin edition by Catherine Jacquemard-Le Saos (Querolus 1994, 46–7) and translation by George Duckworth (1943, 930). Segal 1968; Konstan 1983.

Rural Slavery in Late Roman Gaul

179

described in the play in relation to either Querolus’ activities or the slaves’. In this way, although the play depicts a rural setting, the slaves therein imagined seem to be dedicated to domestic, rather than agricultural or pastoral, tasks: cleaning and organizing the house, fixing the roof or the door, bookkeeping, and so on.43 Querolus, as an instance of the representation of Roman slavery within Roman comedy, is undoubtedly quite an interesting case for studying the resilience of the connections between ideology and literary forms. This is because, when taken alone, the play’s highly idealized representations of slavery seem merely to be the ideological reminiscences or topoi required by the genre. The main justification for this impression is that the domestic roles of the slaves depicted in the play and the presence of the Saturnalian reversal obey the generic rules of Latin comedy. However, there is no contradiction between following these rules and depicting an actual late Gallo-Roman rural household. In this sense, the kind of slavery depicted in Querolus can be classified as the typical form of domestic slavery employed by small landowners in the countryside, whereby a small landowner is able to afford a couple of slaves for domestic functions, but not for agrarian work.44 When this depiction is compared with other GalloRoman instances of slavery in the countryside, it begins to appear to be a performative of actual historical social relations in late Roman Gaul.

salvian of marseilles Religious matters substantially inform the descriptions of slaves found in Salvian of Marseilles’s writings. The particular religious angle of Salvian’s treatise, De Gubernatione Dei, became surrounded in considerable controversy in the twentieth century, as scholars debated its use as a source for the study of late Roman social history. Scholarship was divided into two camps: the first, following the paradigm of the ‘Decline and Fall of the Roman Empire’, asserted the usefulness of Salvian’s descriptions of the spoliation of the poor by the elites; the second rejected these descriptions, claiming that Salvian’s religious aims distorted his representations of reality.45 Recent scholarship has recast the problem of Salvian’s

43 44

45

Querol. 6.67. This is something that can also be seen in the passages of the Eucharisticus analysed below. A useful exposition of the scholarship can be found in Lambert 2002; see also Wickham, 2005, 62–4.

180

Uiran Gebara da Silva

ideological and religious bias, by rooting his particular kind of Christianity in the social and ideological context of the Gallo-Roman Christianization, thereby showing how it evolved differently from the ‘Augustinian’ or Italian orthodoxies.46 In that sense, Salvian is no more ‘ideological’ or ‘radical’ than other Christian writers. Additionally, recent scholarship has started to see Salvian’s defence of the rural poor in more nuanced terms, by unveiling the ideal of patronage that structures his critique of late Roman elites.47 This issue is very important for the examination of the presence of slaves in the text of De Gubernatione Dei. Salvian proposes a kind of progressive social distribution of responsibilities, in which the wealthiest, the most powerful, and the wisest are proportionately more accountable for their wrongdoings than are the poor, the weak, and the ignorant. Accordingly, Salvian’s moral theory reserves more severe punishment for those of higher status than for those of low condition: It is generally agreed that slaves are wicked and worthy of our contempt. But, be that as it may, free-born men of noble rank are the more to be reproached if in their more honorable condition they are worse than slaves. Hence the inevitable conclusion is not that bondmen ought to be absolved from responsibility for their wrongdoings; but that the majority of the rich are more to be condemned in comparison with slaves.48

Following the ideological framework delineated in the passage above, slaves appear throughout the treatise, fulfilling an important role as a moral counterpoint to the behaviour of noble and rich Romans. The ideology of slavery and its reversal are key components of Salvian’s treatise, for he accuses the rich and powerful Romans of being more slavish than the slaves themselves are. However, different from the Latin comedy, what he depicts is not a social order in which the rolereversal game resolves the conflict, but a hierarchic cosmic order in which the slaves serve the Romans and the Romans serve God. Since Salvian often represents the Romans themselves as God’s slaves, while

46 47 48

O’Donnell 1983; Lambert 1999; 2000;2002; Grey 2006. Lambert 2002; Wickham 2005, 527–8; Grey 2006; Brown 2012, 433–532. Salv. Gub. 4.29: Malos esse servos ac detestabiles certum satis est, sed hoc utique ingenui ac nobiles magis execrandi si in statu honestiore peiores sunt. Quo fit ut ad illum perveniri exitum rei huius necesse sit non ut servi sint a reatu nequitiae suae absoluendi, sed ut plurimi divites magis sint servorum comparatione damnandi. Latin text by Georges Lagarrigue (Salvien 1975, 254) and translation by Eva M. Sanford (Salvian 1947, 108–9).

Rural Slavery in Late Roman Gaul

181

highlighting the importance of libertas for the traditional Roman order, Salvian’s employment of slavery should be seen as quite polyphonic.49 In this context, the treatise contains two significant passages showcasing rural slaves. The first one is quite notorious, for it is the one in which Salvian equates the rich and the noble with Circe, the sorceress: Recalling the example of that evil sorceress of old who is said to have changed men into beasts, we might say that all who are received on the farms of the rich are transformed as if by Circe’s potions. For the owners begin to count those whom they have received as outsiders and aliens, as their own property; they turn into slaves men known to be free-born.50

This passage was often cited in debates about the Roman colonate, being employed to describe a general assimilation of free coloni into a condition similar to that of the slaves.51 Salvian’s description of the transmutation of the coloni into slaves has two important implications. Firstly, it describes an illegal and immoral behaviour, which he frowns upon, censures, and denounces (that is, it represents an informal subjugation of the coloni by the local elites, which is in fact opposed by the formal process enacted by Roman legislation). One should pay careful attention to the language used by Salvian here: the ones being changed into slaves are received by the landowners as outsiders and aliens (extraneos et alienos). They are so described, according to Salvian,52 because barbarian enemies forced them to flee from their homes and fields and into the supposed protection of a honestior. The second implication is even more important. The excerpt above shows that, for Salvian, there is a category of rural slaves to whom the coloni would be likened through this illegal assimilation. He is employing rhetorical amplification, for the coloni do not become actual slaves. However, for Salvian’s argument to take effect, his audience must

49

50

51

52

For two very useful analyses of Salvian’s writing as part of a Christian discourse on slavery (a doulology), see De Wet 2018b; 2019. For an approach that places the Roman concept of libertas at the heart of Salvian’s treatise, see Lambert 2002. Salv. Gub. 5.45: Et exemplo quondam illius maleficae praepotentis quae transferre homines in bestias dicebatur, ita et isti omnes qui intra fundos divitum recipiuntur, quasi Circaei poculi transfiguratione mutantur. Nam quos suscipiunt ut extraneos et alienos, incipiunt habere quasi proprios; quos esse constat ingenuos, vertuntur in servos. Latin text by Georges Lagarrigue (Salvien 1975, 346) and translation by Eva M. Sanford (Salvian 1947, 151). See Whittaker 1993b. It is not my intention to address the issue of the Roman colonate here. It is sufficient to mention that my position is close to that of Carrié 1997 and Grey 2007, 2011a, for whom the binding of the coloni to the land had no influence on their formal status as free. Salv. Gub. 5.44.

182

Uiran Gebara da Silva

acknowledge both the existence and employment of rural slaves in Gaul and that their situation is considerably worse than that of a free or bounded peasant. The moral conclusion anticipated by Salvian is that they should abhor the very fact that free people are constrained to being assimilated to a condition akin to slavery. The other significant passage, which also has to be understood as part of Salvian’s exposition of his moral theory, describes the duties of a rural slave: Unquestionably those men ought to have been more fervent in service to God whom he had especially enriched by the most abundant evidence of his favor. What is more right and fitting than that those whom their Lord seemed especially to have favored by his gifts should themselves make an earnest effort to please him by their religious worship, particularly since God lays no heavy or burdensome demands on us? For he does not call on us to plow or hoe, to spade up the earth or prepare the ground for vines, nor, to sum up, does he exact from his slaves what we require of ours.53

Salvian here is talking specifically about Aquitaine, and not Gaul in general. This region is often associated with wine production,54 a species of productive activity associated with intensive labour, which might typically mean slave labour. It should not be surprising, therefore, to come upon such a description of tasks demanded from slaves. Again, Salvian’s account of the condition of slaves serves as an attack on the rich Romans. He states that the Romans, as opposed to the slaves, are servi to a good master, a bonus dominus, God. For, while the Roman masters demand from their slaves the execution of several physically strenuous chores, pictured here as typical activities in the fields, such as ploughing, digging, and sowing, God does not demand from his free Roman servi the same toils. Unsurprisingly, according to Salvian, the noble Romans fail to accomplish even the least laborious demands that God makes of them. In these last two passages of De Gubernatione Dei, the situation of both the free poor and the rich in the countryside is contrasted with that of

53

54

Salv. Gub. 7.9: Officisiores absque dubio deo esse debuerant quos peculiariter deus abundantissima beneficiorum suorum dote ditaverat. Quid enim rectius aut quid dignius quam ut quibus per munera sua dominus quasi specialiter videbatur placere voluisse, iidem quoque specialius domino cultu ac religione placuissent, praesertim cum a nobis deus nil onerosum nil grave exigat. Non enim nos ad aratra aut ligones vocat, non ad scindendas terras neque ad vineas pastinandas, non denique illa exigit a servis suis quae exigimus a nostris. Latin text by Georges Lagarrigue (Salvien 1975, 436) and translation by Eva M. Sanford (Salvian 1947, 191). Sivan 1993, 69; Balmelle et al. 2001; Leveau 2007, 659.

Rural Slavery in Late Roman Gaul

183

slaves, the first in social and the second in moral terms. Although one cannot ignore the ideological and religious framing of De Gubernatione Dei, the slaves are omnipresent in Salvian’s treatise, and while the majority of their appearances belong to domestic or urban situations, the passages just examined offer a window onto the existence of rural slaves in his late Roman Gaul.

eucharisticus The Eucharisticus is a short hexametric poem whose autobiographical subject-matter prompted nineteenth- and twentieth-century scholarship on the later Roman Empire to use and abuse it as supposedly one of the clearest pieces of evidence of social decline and barbarian violence in the fifth century.55 The praising of the poem as historical evidence was often paired with its deprecation as a poetical work.56 However, recent studies have acknowledged that, since the most accepted hypothesis regarding the author of the poem identifies him as one of Ausonius’ grandsons, the Eucharisticus should be read along the same literary and historical lines one would now read the Ausonian opera.57 So, while one should resist the temptation of seeing in the Eucharisticus a supposedly genuine image of an individual’s private life (characteristic of the modern autobiography), still, the poetical structures employed by the poet are descriptive in their nature. They amount to an accurate use of poetical tools as descriptio and evidentia, as well as of the hexametric metre.58 As such, they help to create a convincing image of the troubled life of a rich Aquitanian, and that image’s purpose is to support the poet’s thanksgiving to God for His careful guidance all along the newly humbled man’s existence.59 In this context, all the references to property and wealth become fundamental points of the narrative, because the exposition of the narrator’s possessions and their subsequent loss reveal how God taught the Aquitanian aristocrat to become a good Christian, and to value the holy and communal life over any comfortable and mundane existence.

55 56 57 58 59

Evelyn-White 1988a; Matthews 1990; Jones 1992; McLynn 1995. Courcelle 1948; Evelyn-White 1988a; Matthews 1990. Fontaine 1981; Roberts 1988; McLynn 1995. Roberts 1988. Although the effect is very similar to that of Augustine’s Confessiones, the events presented in the two works, as well as the interpretation of God as Providence, are very dissimilar, contra Coşkun 2006.

184

Uiran Gebara da Silva

There are several mentions of slaves in the poem, but two in particular concern rural slavery. The first one presents the aristocrat putting an estate earned by marriage to good use: [A]nd soon forced both myself and my thralls to exchange seductive idleness for unwanted toils – inciting such as I could by the example of my own labour, but compelling some against their will with a master’s sternness. (. . .) forthwith I hastened to bring fallowed lands under tillage, and promptly to lavish pains in renewing the exhausted vineyards in the manner I had learned.60

In the above passage, Paulinus prides himself on being capable of managing the rural underlings, with some of them being inspired by his example, and many of them having to be coerced into work. The language used here undeniably refers to unfree workers, whereby the ‘master’s sternness’ (domini adstringendo rigore) is a fitting means of compelling them out of their rest and into hard work. The second passage presents a clear distinction between the use of slaves for agrarian and for domestic activities: Here no fresh revenues were like to give rise to great hopes – no tilth tended by owned labourers, no vineyards (on which alone that city relies to procure from elsewhere every necessary of life), but, as a refuge for my loneliness, only a house in the city with a garden near, and a small plot, not destitute of vines, indeed, and fruit-trees, but without land without tillage. (. . .) Further, for the outlay which the needs of life demand, I made it my hope to earn them by renting land, so long as my house remained well stocked with slaves, and while my more active years furnished me with undiminished strength.61

Paulinus lost his estates in Burdigala and in Epirus, and all his relatives were dead, with his properties being reduced to a single estate in Massilia. He then claims that the land lacks ‘owned’ labourers, not being cultivated by slaves when he bought it. Again, he prides himself on being capable of making the land productive, clearing the fields and building a house there. 60

61

Paul. Pell. Euch. 191–98: (. . .) cito meque meosque coegi,/ quos potui exemplo proprii invitando laboris,/ quosdam autem invitos domini adstringendo rigore./ (. . .) protinus et culturam agris adhibere refectis,/ et fessis celerem properavi inpendere curam/ vinetis conperta mihi ratione novandis. Latin text and translation by Hugh Evelyn-White (1988b, 301–2). Paul. Pell. Euch. 523–38: nec spes magna novis subitura ex fructibus esset,/ non ager instructus propriis cultoribus ullus,/ non vineta – quibus solis urbs utitur ipsa/ omne ad praesidium vitae aliunde parandum -/ sed tantum domus urbana vicinus et hortus/ atque ad perfugium secreti parvus agellus,/ non sine vite quidem vel pomis, sed sine terra/ digna coli (. . .)/ Porro autem expensas, vitae quas posceret usus,/ conductis studui ex agris sperare paratas,/ donec plena magis servis mansit domus et dum/ maiores melior vires mihi praebuit aetas. Latin text and translation by Hugh Evelyn-White (1988b, 344–5).

Rural Slavery in Late Roman Gaul

185

He claims that he did all this by himself, but he may be implying here the work of his slaves mentioned in a later verse. Paulinus then states that he decided to lease the land to tenants (one must assume these were coloni), and the renting of the land granted him money that allowed him to keep a comfortable house filled with some slaves designated for domestic affairs. This, in turn, safeguarded some of the formerly rich Aquitanian’s dignity and maintained the appearances of his born status. Here, by stating that he only resorted to tenancy when no other option was available, he reveals the importance of slave labour for viticulture in Massilia. Considering the poem as a whole, Paulinus repeatedly presents himself as a servant of God, and the ideology of slavery conveys a similar hierarchical cosmic order to that of De Gubernatione Dei.62 Additionally, the representation of slaves has another purpose: they are employed to depict the lifestyle of a rich Aquitanian while also stressing the importance of slaves for the most intensive forms of labour. The poem, then, reveals that, even as late as the early fifth century, the ownership of slaves was still regarded as an important status symbol, but it also reveals that, at least in Aquitaine and southern Gaul, one would expect slaves to be used in more intense agrarian activities.

conclusions In conclusion, this reassessment of the sources for our knowledge of rural slavery in late Roman Gaul has been guided by two concerns. The first has been the need to bring these sources into their proper literary context, striving to investigate the social history of the Gallo-Roman countryside by incorporating methodological and theoretical insights offered by cultural history. The second has been the expectation that such a reassessment should better equip us to face the challenges that the new readings and narratives about ancient slavery pose for our understanding of the late Roman countryside. In this chapter, I have attempted to bridge the gap between cultural and social history by treating the literary structures as linguistic mediations of ideology and of social relations. Instead of just collating the literary instances of slavery in Gaul, my aim has been to understand these occurrences within their proper literary and social contexts. In this sense, what 62

However, in the Eucharisticus, Paulinus seems to be trying to show he was a better Roman than the ones described by Salvian, almost as if it were a response to the treatise; see McLynn 1995.

186

Uiran Gebara da Silva

the sources might suggest is the existence of slavery as an important social relation, not only for the upper elites, but also for the medium or even small landowners in Gaul. The rural slaves described in Ausonius’ poems, in Salvian’s treatise, in the Querolus, and in the Eucharisticus are shown performing either domestic or agrarian activities, and, suggestively, they seem to play a key ideological role in these texts, especially in the De Gubernatione Dei. From this perspective, rural slavery seems to be deeply integrated in the literary and social reality represented in this sample of texts from late Roman Gaul. It seems that, contrary to expectations sustained within twentieth-century scholarly narratives and theoretical frameworks, the world shown in these texts takes rural slavery to be a normal feature of the Gallo-Roman countryside. As I stated earlier, scholarship on late Roman Gaul is slowly incorporating this perception.63 However, one of the challenges for that incorporation is the resilience of the narrative according to which rural slaves are assimilated to other rural workers in the later Roman Empire.64 Since Gaul is the paradigmatic area of the Empire for such supposed assimilation, the awareness that slavery there maintained its distinction from other forms of free or bonded labour is fundamental.65 In this sense, the existence of slavery in Gaul must be understood as part of a wider examination concerning the permanence of slavery in the later Roman Empire. Here, an additional difficulty is created by models based on New Institutional Economics, as they represent slavery as just one form of workforce among the landowners’ diverse portfolio. This interpretation both recreates, in a new guise, the narrative of assimilation, and tends to underestimate slavery’s particular kind of subjugation and its impact on the whole of society. This also means that the deep ideological impact slavery leaves on culture is both naturalized and disconnected from the relation of exploitation at the core of this form of domination.66 Nevertheless, it is exactly the permanence of these cultural expressions that should be understood as relevant signs of the continuing existence of the specific violence attached to slavery. Although the most recent models are better at revealing the multiple productive options available to the landowners and the coexistence of multiple labour relations in the 63 64 65

66

See notes 13 and 14. Harper 2012; Vera 2012. A distinction that has been highlighted by both the studies on Roman coloni and on early medieval slavery. See Grey 2011a; 2011b; Rio 2017; Wickham 2005. Harper 2012, in an apparent contradiction to the extensive impact of slavery in Roman society presented in Harper 2011.

Rural Slavery in Late Roman Gaul

187

countryside, the older models revealed slavery’s role in structuring several aspects of Roman society. The existence of ideological expressions similar to those of classical slavery in late Antique Gaul shows that slavery was still important in late Roman and early Merovingian Gaul as the most intensive form of labour exploitation. Outside Wickham’s late antique spots of peasant autonomy,67 Gallic elites kept depending on it for their larger rural properties.

67

Wickham 2005, 545–51.

part iii SLAVERY, SOCIAL HISTORY, AND THE PAPYROLOGICAL AND EPIGRAPHICAL SOURCES

9 Slaves in Sixth-Century Palestine in the Light of Papyrological Evidence Marja Vierros

Greek papyrological evidence from the Palestine area is scarce; thus, the main source material for the sixth century ce consists of only two papyrus dossiers: one from Petra and the other from Nessana.1 The recent publication of the Petra papyri provides some long-awaited data on the onomastics of slaves and their existence in households of the elite and in the Christian community of Petra. The more abundant papyrological corpus from Egypt does not give a clear image of slaves this late in time. This article provides the first detailed and contextualized study on the evidence for slaveholding in the Greek documentary papyrological material of the Palestine area in the sixth century. In the first section, I give a brief overview of both the Nessana and Petra dossiers. In the second part, I review the attestations in the papyrus texts from these dossiers as well as the terms used for possible slaves in each case. The context of all attestations is described as far as the surviving fragments allow us. For convenience’s sake I have divided the attestations into three groups: those that certainly refer to slaves, those that have certain reference to slaves and also have some terminology referring to emancipation or former slaves, and those where the reference to slavery is uncertain. In the third section, I discuss the personal names of the slaves, possible slaves and/or servants. I then briefly summarize what we know about the value, prices, and ages 1

Earlier archives from the Near East also survive, like the Babatha archive (Trismegistos Archive ID 41) from the second century, texts from Masada (mostly first century), and Greek and Latin texts from the garrison of Dura-Europos (third century). Some individual papyri written in the Near East closer to the sixth century have been found as well, e.g., P. Monts.Roca IV 95, ‘A Greek Letter from Syria’, and P.Bour. 25, a letter written in Apamea (both dated from the fourth to fifth centuries), but these do not mention slaves.

191

192

Marja Vierros

of the slaves. In the final section, I recapitulate what the Greek papyri have to offer for our knowledge of slavery in sixth-century Palestine.

the sources: papyrus dossiers from petra and nessana Petra, located in southern Transjordan, was the capital of the Nabateans from around the first century bce and was annexed to the Roman Empire in 106 ce. Petra became the metropolis of the Arabian province soon after that. By the sixth century, it had become the metropolis of the third Palestinian province of the eastern Roman Empire, also known as Salutaris.2 A carbonized papyrus dossier was found in 1993 in the Byzantine Church of the Virgin Mary.3 The total number of published texts is 87. Some of the texts are very fragmentary, and their lengths vary; some come from rolls of several metres’ length but may lack half or even three quarters of the text. This dossier forms the only papyrological source known from Petra. The date range of the Petra texts mostly falls within the sixth century – that is, during the reigns of Emperors Justinian, Justin, and Maurice. The documents are, for the most part, related to property: contracts of sales, donations of landed and other property, and documents concerning the taxation of such property. The texts are mostly Greek monolinguals, with only some Latin text surviving (P.Petra V 55). The local Aramaic and Arabic languages, on the other hand, manifest in the place names and in some personal names.4 The people whose property is discussed are among the elite members of the church, a Christian community that belonged to the Roman province of Palestine. The central figure is a person called Theodoros, son of Obodianos (born ca. 512–14), whose path we can follow from a young man to a deacon and eventually to the archdeacon of the church where the papyri were found. The texts were all written in or near Petra, except for one text, which was written in Gaza (P.Petra I 2), since Theodoros resided in Gaza at that time. In eleven documents we find references to slaves (not all of them certain).

2 3

4

Fiema 2002, 1–2. The church was excavated in the early 1990s as part of an American excavation project (Fiema et al. 2001). The papyri were found in the little side room on the left of the apse. The papyri were conserved by a team from the University of Helsinki, led by Professor Jaakko Frösén, and later published by his team in collaboration with a team from the University of Michigan. The first volume of the papyri was published in 2002 (Frösén, Arjava, and Lehtinen 2002), and the fifth and final volume in 2018 (Arjava, Frösén and Kaimio 2018). Al-Jallad 2018.

Slaves in Sixth-Century Palestine

193

The city of Nessana (modern-day Auja Hafir, Israel) was discovered in 1935. It is located about 175 km north-east from Petra and it belonged to the same third Palestinian province. The material found on the site was published in three parts. Volume III is of interest in this chapter, as it includes the documentary papyri – 195 texts in total.5 Almost half of these are labelled ‘minor documents’. The papyri constitute several archives, in Greek and Arabic, from the fifth, sixth, and seventh centuries ce.6 Part of the papyrus collection was found in a storage room of the Church of Mary, Mother of God, and mostly concern property rights and economic and legal matters. The second, and larger, part was found in the monastery Church of S. Sergius and S. Bacchus. The Nessana papyri constitute several different archives – for instance, the soldier’s archive (TM Archive 220) and the archive of the descendants of Patrikios (TM Archive 65). None of the three papyri that mention slaves, and thus are discussed in this article, apparently belong to any specific archive.

attestations of slaves in petra and nessana An overview of the papyri from Petra and Nessana that mention slaves is given in Table 9.1. Fourteen documents mention or deal with slaves or possible slaves. The percentage from the published Nessana papyri is very low, under 3 per cent, but from Petra it is higher, about thirteen percent.7 Thus, we may gather that slaves were not the main topic when property was sold, transferred, or taxed, but not a negligible topic either, especially in Petra. Usually, slaves come up in property division documents, though the type of document is not always clear. In what follows, I will describe the attestations of slaves in the Petra and Nessana papyri and investigate the role of possible slaves in each document on a case-by-case basis in the context in each document. I have divided the attestations into three groups, where the first contains certain references to slaves, the second has terms used for slaves and some terms connected to emancipation (freedom) mentioned, and finally the third group contains the uncertain cases of slaves. Different terms were

5

6

7

P.Ness. III, 1958 (ed. Casper J. Kraemer). The literary texts, including, e.g., fragments of the Aeneid and biblical texts, were published as vol. II of Excavations at Nessana, 1951 (P. Ness. II, 1951). A query in the Papyrological Navigator, with provenance Nessana and time set to sixth century, yields 108 texts that are in Greek and have transcription [accessed August 23, 2019]. Nessana: three from 108 texts (see note 6), but only one of the texts concerns slaves with certainty; Petra: from 87 published texts, 11 documents form almost 13 per cent, but if we take only the nine certain cases it is 10 per cent.

t a b l e 9 . 1 Papyri from Petra and Nessana dealing with slaves. Document

Date

Slaves mentioned

Role of slaves

1) certain references to slaves

P.Petra II 17

505–37? ce

as part of property being divided

P.Petra I 1

537 ce

two slave couples: Salamanios, δοῦλος & Almasia οἰκέτις, Kyriakos, δοῦλος & Ampelion, ?, ἀνδράποδα Gany-, δούλη

P.Petra III 28

559 ce

P.Petra V 58

after 559 ce

P.Petra III 36

6th century

P.Ness. III 89

P.Petra V 72

late 6th–early 7th century after mid-6th century after 565 ce

P.Petra V 57

after 569 ce

P.Petra III 31

582–92 ce

194

Category

2) certain references to slaves with connection to freedom

P.Petra V 52

two slave boys (ages 6 and 4), δουλικὰ σώματα + οἰκέται οἰκέτης, δοῦλος, παιδίον, σώματα ἀνδράποδα, Hadrianos, son of Thekla κοράσιον and παιδίον are traded freed [slaves] the words freedom and ἀνδραπo´δῶν occur Kalemeros, The word παῖς occurs freedman (?) Ouaelos, s. of Mouelos, formulaic phrase [ἀ]νδρα[πόδων]

as part of inheritance (marital property arrangement) as property being divided topic of the contract, nature of the contract unclear as property being divided one object in an account beneficiaries in a will?

main topic of contract, the word freedom occurs often

(continued)

3) uncertain references to slaves 195

P.Petra III 19

539–40 ce

former οἰκέτης Magdios

P.Petra V 73 P.Ness. III 31 P.Ness. III 165

c. 580? ce 6th century 6th–7th century

word παῖς occurs οἰκέτης word δουλάριον occurs (in plural)

property had been registered on his behalf earlier

196

Marja Vierros

used for slaves in these documents, and I will discuss them where they first come up.

Certain References to Slaves a. P.Petra II 17: Three brothers, Bassos, Epiphanios, and Sabinos (the names of the father and the mother are lost) divide a significant amount of property between themselves. The receivers of the three allotments are determined by lot. The property consists of several subsequent divisions, generation after generation, including grain fields, vineyards, land (72 different plots of land, where the individual plots seem to be small), houses in different locations, and also slaves. The slaves, entitled ἀνδράποδα (andrapoda) as a formal category of property, are assigned to two allotments (quoted below), only for Bassos and Epiphanios. It is likely that the first share, which by lot went to Sabinos, had more other forms of property to account for the lack of slaves. Furthermore, it seems that the division really did happen by lot, so the portions were indeed set equal in value. However, it is difficult to estimate what the value of these slaves would have been, since neither the sizes of the plots nor the value of each piece of land or housing is ever mentioned. ll. 64–7 (Bassos’s lot): [ c. 7 ὀμοίως ἔλαχεν ὀ] αὐτος εὐλαβ(έστατος) Βάσσος εἰς τὸ κατ’ αὐτὸν μέρ(ος) [τρίτον ἀπὸ ἀνδραπo´δῶν] α ̣ὐ ̣τῶν· [Σα]λ ̣αμά ̣νιον δοῦλον, υἱὸν Οουε[ δούλης, καὶ ] ̣ Α̣λ ̣μασια οἰκέτ ̣ιν̣ ̣ [. . .] [ . ἔλαχεν καὶ] Ἐπιφάνιος etc. [Likewise] the said most reverend Bassos [received by lot] for his [third] part: [From] their [slaves]: The slave Salamanios, son of [the slave woman] Ooue [ . . .and. . .] the household slave Almasia. . . ll. 136–8 (Epiphanios’ lot): ὀμοίως ἔ ̣λ ̣αχεν ὀ αὐτος αἰδ(εσιμώτατος) Ἐπιφάνιος εἰς τὸ κατ’ αὐτὸν μέρος τρίτον ἀπ ̣ὸ ἀνδραπ[o´δῶ]ν ̣ αὐτῶ[ν·] Κυρι̣ [α]κ ̣ὸ ̣ν ̣ δ ̣ο ̣ῦλον, υ ̣[ἱὸ]ν ̣ Κυριακῆς δούλης, κ(αὶ) ̣ Ἀμπέλι[ον, θ]υ ̣γα̣ ̣τ ̣έρ[α c. 14] Κυριακοῦ σύνβιον. ̣ Likewise, the said most respectable Epiphanios received by lot for his third part: From their slaves: The slave Kyriakos, son of the slave woman Kyriake, and Ampelion, daughter [of. . .] the partner of Kyriakos.

Notably, even within such a wealthy family with landed property, there were only four slaves mentioned in this division of property. Furthermore, it seems that both these pairs of female and male slaves were couples, allowed to remain together, even though separating them could have

Slaves in Sixth-Century Palestine

197

made the distribution of property more even across three allotments. The unity of the servile nuclear family is thus taken into consideration. The practice of not separating slave families was also in accordance with legislation by Constantine, which prohibited the separation of slave families on imperial estates.8 Salamanios and Almasia, both with names of local origin (see below), form one couple (the editors think that Almasia is mentioned as the partner of Salamanios in the lacuna before her name).9 While Salamanios is clearly entitled as a slave (δοῦλος),10 the term for Almasia apparently is οἰκέτις.11 Whether the ambiguous term οἰκέτης/οἰκέτις (oiketeˉ s, m. / oiketis, f.) meant a free or an unfree person has been debated without a clear outcome.12 In this document, Almasia being married or in cohabitation with a slave, and seemingly belonging to the same category of property called ‘slaves’, ἀνδράποδα, seems to strengthen the argument that οἰκέτης/οἰκέτις did not designate free status or at least that the distinction between free and enslaved was not clear at this time (see also below, with P.Petra III 19). 8

9

10

11

12

See comments on P.Petra II 17.65–7 (ed. L. Koenen, J. Kaimio, M. Kaimio, and R. W. Daniel), 113. Cod. Iust. 3.38.11: Possessionum divisiones sic fieri oportet, ut integra apud successorem unumquemque servorum vel colonorum adscripticiae condicionis seu inquilinorum proxima agnatio vel adfinitas permaneret. quis enim ferat liberos a parentibus, a fratribus sorores, a viris coniuges segregari? igitur si qui dissociata in ius diversum mancipia vel colonos traxerint, in unum eadem redigere cogantur. ‘Divisions of property should be made such that close blood relation or a tie by marriage among slaves, or tenants affixed to the soil or resident there (coloni vel inquilini), remains intact upon the succession of a master. For who would tolerate that children should be separated from parents, brothers from sisters, husbands from wives? If therefore anyone has separated slaves or bound tenants and placed them under different ownerships, they are compelled to reunite them as one’ (334 ce; cf. C. Th. 2.25.1). Translation F. H. Blume, as modified in Frier 2016. The suggested restoration for the passage is [Σα]λ ̣αμά ̣νιον δοῦλον, υἱὸν Οουε|[δης δούλης, καὶ τὴν αὐτοῦ (or τούτου) σύνβιο]ν ̣ Α̣λ ̣μασια οἰκέτ ̣ιν̣ ̣ [θυγα|τέρα name of the mother δούλης] (see P.Petra II, 113). The words δοῦλος/δούλη undoubtedly refer to slaves in Petra. These words are also used in the Nessana texts, but with a different meaning: writers referred to themselves with this word expressing humbleness, to be understood as ‘(your/Christ’s) slave’. Admittedly, the editors are not entirely certain that the form is feminine, as the letter iota could be read as an eta (P.Petra II, 114). The name of Almasia is also indeclined or missing the final nu of the accusative case, both quite possible phenomena that need not hinder the interpretation of Almasia as the other half of this pair of slaves. In the Appianus estate in Egypt (the so-called Heroninos archive, ca. 249–68 ce), οἰκέται supposedly referred to free permanent employees who received salary (ὀψώνιον), employed as such because of their poverty (Rathbone 1991, 92, 106–16). Harper (2011, 513–18) argues strongly that οἰκέτης was synonymous with δοῦλος and designated non-free status. He also discusses the problems the Appianus estate texts pose to this view, nevertheless rejecting that οἰκέται could mean free, and mentions that it is sometimes difficult to distinguish freedmen from slaves, because they could still be called δοῦλοι or οἰκέται.

198

Marja Vierros

Kyriakos and Ampelion form the other couple (the word σύνβιος [sunbios], ‘partner’, confirms that they were indeed a couple). Again, Kyriakos is clearly a slave (δοῦλος, doulos), but the servile status of Ampelion is unclear, as there is no space for the word δούλη (douleˉ ) after her name. This may be a scribal omission or might actually mean that Ampelion was not a slave, but in either case the distinction between free and enslaved had become increasingly blurred in late antiquity with the emergence of various categories of semi-free people.13 This Kyriakos on line 137 is probably not identical with the tenant farmer of the same name mentioned on l. 186. We may note that both Salamanios and Kyriakos are designated through their mothers’ names: ‘Salamanios, son of [the slave woman] Ooue-’ and ‘Kyriakos, son of the slave woman Kyriake.’ This designation has slightly different wording than the normal patronymic usage (with plain genitive), since the word ‘son’ is mentioned especially. The same holds true with Ampelion; the word ‘daughter’ can be read with hesitation and the name of the mother (?) would follow in the gap. Almasia’s possible patro-/matronymic is not preserved at all. b. P.Petra I 1: The document is an agreement between Theodoros, son of Obodianos, the future archdeacon, and his cousin and bride Stephanous, at the time of their marriage. A female slave (δούλη), the beginning of whose name (Gany-) is preserved in a fragmentary part of the text (l. 67: τῆς δούλης Γανυ ̣[]), clearly belongs to the property, the possible scenarios of inheritance for which are discussed in this text. The slave woman probably belonged to the property of Palladia, grandmother of both Theodoros and Stephanous. c. P.Petra III 28: This text is a division of slaves between Patrophilos, son of Bassos, and the widow Elaphia, daughter of Sabinos.14 Two slave boys seem to constitute the only property divided in this document (l. 20: δουλικὰ σώματα δύο ἀρρενι[κὰ], ‘two male slaves’; cf. l. 55). The older boy is said to be approximately six years old, and the younger four years (ll. 32–3). Elaphia was the widow of Elias, son of Isakios, who had had some claims, together with Patrophilos, to the two slaves. Elaphia might have been related to Patrophilos herself, but, in this document, her role is to safeguard the rights of her three underaged children to the inheritance of their father, acting possibly as their guardian. The passages where the division is 13 14

See comments on P.Petra II 17 (ed. L. Koenen, J. Kaimio, M. Kaimio, and R. W. Daniel), 131. In l. 45 the document refers to itself as a ‘written division’ and the word ‘servile’ appears in the previous line: διὰ τὸν εἰρημένον δουλι[κὸν – ca.24 -] τόνδ ̣ε ̣ τὸν ἔνγραφον διαμερισμ[ὸν), and on ll. 51–2 the document seems to be mentioned as ‘division of slaves’, although the word ‘division’ itself is in the gap: [οὗτος ὁ διαμερισμὸς] δούλου κύριος καὶ βέβαιος ἔστα[ι], ‘[this division?] of slaves will be valid and firm’.

Slaves in Sixth-Century Palestine

199

stated (ll. 22–4 and 35–40, below) are somewhat fragmentary, but apparently Patrophilos received a half share, including a person called Ioannes; this must have been the older boy’s name. Elaphia’s children receive 3 solidi, clearly as a compensation for getting the slave who was two years younger. ll. 22–4:

ἐξ ἡμισίας μοίρας ἀπο[- ca.27 -] τος ἐ ̣ν ̣ τ ̣ῷ ̣ [συνοι]κησίῳ δύο οἰκ ̣η[τῶν] [- ca.20 - ἑνὸς] ̣ οἰκήτου διαφέροντος τῷ εἰρη[μένῳ εὐδοκιμ(ω)τ(ά)τῳ Πατροφίλῳ - ca.6 -] the half share . . . in the cohabitation of two servants [. . . one] servant belonging to the said [most honourable Patrophilos. . .]

ll.35–40:

ἔλαχεν μὲν Πατρο-] φίλῳ τ ̣ῷ ̣ εὐδοκιμ(ω)τ(άτῳ) εἰς τὸ κατʼ [αὐτὸν μέρος ἥμισυ] [ca.12 -] Ἰωά[νν]η[ς, ̣ τ]ῇ ̣ δὲ κοσμιωτ(άτῃ) Ἐλαφίᾳ [- ca.27 -] τ̣ ̣[ ]σ ̣ ̣ ̣ τ̣ ̣[ ̣ ̣ ̣ ]τ ̣[ ]ν ̣ καὶ παρα[- ca.27 -] ̣ εἰς τὸ κατʼ ̣ αὐτοὺς μέρος ἥμισυ [- ca.27 -] μετὰ νομισμάτων τριῶν παρὰ κ[εράτια - ca.21 -]

to the most honourable Patrophilos [was allotted] in [his half share. . .] Ioannes, and to the most virtuous Elaphia [and . . . was allotted] in their half share. . . with three solidi minus [ x ] carats. . .

It seems likely from the text that both, or at least one of, the slave boys were born from the cohabitation of two οἰκέται (see l. 23 above), one belonging to Patrophilos, the other to Elaphia. The fact that the boys are clearly slaves, property being divided, means that their parents were as well – that is, οἰκέτης indeed means ‘slave’ here. Normally, servile offspring would belong to the mother’s owner, and the separation of families was outlawed in Justinian’s time, but the Syro-Roman Lawbook approved of the division of the offspring of dotal slaves; according to the editors, a contract may have initially been drawn between the owners to divide the yield of their human property.15 d. P.Petra V 58: This text was originally a long, formal notarial contract. One party is Theodoros, son of Obodianos, the other presumed to be Isakios, son of Elias. Isakios is also the son of Elaphia, known to us from P.Petra III 28. Several different terms possibly referring to slaves are used in this text: δοῦλος (l. 32), οἰκέτης (ll. 2 and 59), παιδίον (l. 56), σῶμα (l. 47), but they are usually not preserved with very much immediate context. It is 15

Arjava and Vesterinen 2007, 94; for the Syro-Roman Lawbook, see §95, 108, 119; Selb and Kaufhold 2002, 128–31, 160–3, 172–5.

200

Marja Vierros

important to note that, since slaves seem to be the main topic of this contract, these different terms are most likely used synonymously (including the term οἰκέτης). The words παῖς (pais) and its diminutive form παιδίον (paidion) can both refer to (free) children or slaves.16 Mark Golden explores the term παῖς, noting that it is used both for slaves of any age and for children in Classical Greek (less in poetry, but the usage exists in Aeschylus) and suggests that the underlying reason why this happens is related to the similar legal status of children and slaves. He also suggests that the similarity of the verb ‘to strike’, παίω (paio), to the noun παῖς was a factor in the spread of the noun to denote both these groups of people, since corporal punishment was legally used for both.17 The word τέκνον (teknon, or tekna in plural), on the other hand, seems to unambiguously refer to children in the Petra papyri.18 On two occasions the name Alpheios appears in the genitive case in association with the slaves (ll. 47: [] ϲ ̣ωμάτων Ἀλφε ̣ίο̣ ̣υ τοῦ ̣[ ] ‘Alpheios’ slaves’ or: ‘the slaves, Alpheios’ and 59: [] α ̣ὐτοῦ οἰκέτου Ἀλφε ̣ί[ου] ‘the same oiketeˉ s ̣ of Alpheios’ or: ‘of the same oiketeˉ s Alpheios’). It seems very likely that he is mentioned as a slave-owner connected somehow with the transaction, although it is not impossible to interpret Alpheios as the name of the slave. In line 56 he might also be the ‘said’ person, also in the genitive ([μ]ετὰ τοῦ παιδίου τοῦ εἰ[ρημένου] ‘with the young slave of the said’). e. P.Petra III 36: This text is a draft of a division of property, again very poorly preserved. The property divided is diverse, but slaves are mentioned as part of the estate (l. 40: ἀνδρά̣ ̣ποδά τε καὶ κτήματα ‘. . .slaves and properties. . .’). It is possible to identify Hadrianos, son of Thekla, mentioned later on, as a slave, as he is designated with a matronymic (l. 113: Ἁδριανὸς ̣ Θέ ̣κ ̣λ ̣α ̣ς ̣ Αχζαμος ‘. . .Hadrianos, son of Thekla, Achzamos. . .’). The directly following name, Achzamos, could be the name of another slave, if we consider this as a list. Parental identification through the mother is one possible indication of servile status, as the father of a child born to a female slave was not recognized by law.19 This is the only reason for possibly identifying Hadrianos as a slave; the 16

17 18

19

This was common in the Ptolemaic period; see, e.g., P.Cair.Zen. I 59043, where the same slaves are referred to as σώματα and παῖδες. See the discussion and more examples in Stanton 1988, 469–70, 474–6, also other diminutives like παιδίσκη (paidiskeˉ, fem.) or παιδάριον (paidarion, neut.) can be used in the same way. Golden 1985, 91–104. Cf. Stanton 1988, who discusses the use of τέκνον in papyri as an affectionate word for biological children as well as an unaffectionate term for children as what I would call a legal category. See Westerman 1955, 145 (the old law of inheritance of status by the mother was kept in force by Justinian in 530). However, there is also a group of freeborn but ‘fatherless’ children (ἀπάτορες, apatores) in the census declarations of Roman Egypt, referring to

Slaves in Sixth-Century Palestine

201

document gives no other indication about servile status. However, if we compare the mothers’ names mentioned for slaves in P.Petra II 17 above, we notice that the phrase is not the same here, since the word ‘son’ is not spelled out. Phraseology can vary from document to document, naturally, but we cannot be absolutely sure that Hadrianos is a slave, or possibly someone born out of wedlock. This also affects the status of his mother, Thekla. If Hadrianos is a slave, most likely his mother was as well; but in other scenarios she might be a servant or some other low-status citizen. Nonetheless, slaves are mentioned in this document as property being divided. f. P.Nessana III 89: This interesting papyrus records the activities of a small community (κοινo´της ἡμῶν). It is a sort of a diary and account written by a person called Zunayn, who belongs to a group of people travelling apparently as a caravan. Two slaves, both of unknown age, were acquired in exchange for two camels and a sum of cash. The terms used were κοράσιον (korasion) for the slave girl and παιδίον for the slave boy. The first term is not attested in Petra. The word παιδίον clearly denotes a slave here. ll. 21–2 δ ̣[ο]θὲν τιμῖς το ̣ῦ κορασίου ν[ο(μίσματα)] γ ̣ [κα]ὶ καμαίλον (νομίσματος) ϛ γ´ (ὁμοῦ) θ [γ´] κ ̣αὶ τιμῖς το πεδίου (νομίσματα) ϛ κ ̣α ̣ὶ κ̣ αμελιν (νομίσματα) δ γ´ (ὁμοῦ) (νομίσματα) ι γ´ σὺν τοῦ μισθοῦ τοῦ γράψαντος τὸ ὀνιακὸν αὐτοῦ. ̣ l. τιμῆς, καμήλον, τιμῆς, τοῦ, παιδίου, καμήλιον 22 τῷ, μισθῷ, ὠνιακὸν Paid as price of the slave girl 3 solidi plus camel worth 6 1/3 solidi total 9 1/3 and as price of slave boy 6 solidi plus camel worth 4 1/3 solidi total (including fee of the man who drew up bill of sale) 10 1/3 solidi.

From the prices mentioned, we can see that a slave boy or a girl was more expensive than a camel, though the price of the camels was not stable, ranging from 5 to 15 solidi in Nessana. In the same document, the price for a she-ass is given as 5 1/3 solidi.

Certain References to Slaves with Connection to Freedom The usual term for an emancipated slave, ἀπελεύθερος (apeleutheros), is not attested in the Petra or the Nessana papyri.20 Instead, we find the term

20

children born outside of legitimate marriages. In the census declarations, mothers of slaves are necessarily not mentioned either; Bagnall and Frier 1994, 155–9. The use of this term seems to peak in the papyrological corpus in the second century ce. The query ‘απελευθερ’ (i.e., including both nominal and verbal forms) in the Papyrological Navigator (accessed 23 October 2018) yields 301 results, 265 of which come from documents dated after 100 ce, while only 47 come from after 250 ce and 3 are dated to the sixth century (the term also appears in P.Gen. IV 184.5 (551 ce), though this

202

Marja Vierros

ἐλεύθερος (eleutheros) twice, the verb ἐλευθερόω once, and an incomplete form ἐλευθ[ερ] (in P.Petra III 22 and P.Petra V 57). The adjective does not seem to refer to a freedman in P.Petra V 51.46, and the fragmentary form in P.Petra III 22 is not supplied with enough context to make any convincing claims (hence, these two are not included in Table 9.1 or the discussion). P.Petra III 31.66, on the other hand, possibly does, as it refers to a person (see d.). a. P.Petra V 52: A testator of a will is taking freed slaves into consideration (l. 15: . . .τῶν παρ’ ἐμοῦ ἐλευθηρουμένων. . . ‘. . .and the [slaves] that I have freed. . . ’). In addition to slaves, for instance, holy places and ‘needy travellers’ are mentioned, but the fragmentary state of the papyrus does not reveal any further details about these freedmen. b. P.Petra V 72: This fragmentary papyrus seems to contain three texts, but the relationship between each text is unclear. Text B apparently talks about slaves, ἀνδράποδα, and freedom, ἐλευθερία, but these words do not occur in close proximity (ll. 4 and 9). In text C, the title of a person called Stephanos (or X, son of Stephanos) is abbreviated as οἰκ(), which I have interpretated as οἰκονόμος (oikonomos) in the edition, because there is another person, Sabinos, whose title, οἰκονόμος, appears both written out (l. 36) and abbreviated as οἰκ() (l. 38). It is therefore unlikely that the abbreviation could also stand for οἰκέτης. c. P.Petra V 57: This long roll, with almost four metres of text, is unfortunately poorly preserved. The document is called ‘Agreement on a Slave’ in the edition, but only the word ‘agreement’ (ὁμολογία) is preserved in the text without a direct connection to slaves. It was drawn up by Fl. Paulos from the village Khaph Arrabbon (ἀπὸ Χαφ Α̣ρρ̣ α̣ ̣ββων), and addressed to Theodoros, son of Obodianos. Only Paulos signed it (using a proxy), so the agreement was unilateral. A certain presbyter named Leontios is also mentioned, his name occurring near a few instances of the word ‘freedom’ (ll. 87 and 97). Some words in the text suggest a sale, but there is no vocabulary about houses or land. There is the word παῖς, however, mentioned at the beginning (l. 26: τὸν εἰρημένον παῖδ[α] καὶ κ.[] ‘the said boy/slave and’) as well as in the end, connected to the word ‘written’ (l. 174: π ̣α ̣ῖδ̣ ̣α ̣ ἐγγραφ[]). This could mean that the written agreement at hand concerns a slave or slaves. The person named Kalemeros, who is so often mentioned in the text (ll. 57, 66, 74, 119), is likely the slave and the topic of the agreement. The editor states that Kalemeros might be able to receive freedom under certain circumstances, but this is not does not appear in a PN search due to a typo). This result must also be interpreted against the fact that the second century ce yields more papyri than other centuries.

Slaves in Sixth-Century Palestine

203

a manumission document. Moreover, it appears that something had been given or might be given to Theodoros by Kalemeros (ll. 119 and 179), with Paulos confirming that he would lay no claim to it.21 Unfortunately, the fragmentary nature of the text obscures the exact nature of the agreement. Perhaps it is a sale with some sort of condition by which Kalemeros might be given his freedom and this is somehow connected to the clergyman Leontios. Or it could be some sort of dedication of a slave to the church of Theodoros. We must also notice that a sum of money, 20 solidi, is mentioned in the text (ll. 51, 60) and, at the end of the document, a fine of 18 solidi is set (l. 147), if the conditions of the agreement are not fulfilled. d. P.Petra III 31: This document is a donation from Theodoros to his cousin Hierios. Due to the fragmentary condition of the papyrus, it is unclear what the property comprises. The word ἀνδράποδα, ‘slaves’, occurs in a list of different types of property (l. 73: [ἀ]νδρα[πόδων]) – that is, it refers generically to a category of slaves, not to any individuals. An interesting reference to a person named Ouaelos, son of Mouelos, titles him as ‘free’ (l. 66: . . . καὶ Ουαελου Μουελου ἐλευθέρου ‘. . .and Ouaelos, son of Mouelos, free’). It is possible that he is a freedman who is using a patronymic after gaining his freedom. He does not have an honorific title. However, it is not entirely certain that the word ‘free’ designates freedom from servitude.

Uncertain References to Slaves a. P.Petra III 19: This text is a request for the transfer of taxation by Leontios, son of Valens, and Theodoros, son of Obodianos, concerning a vineyard in the possession of Theodoros. Magdios, son of Euzoios, a former οἰκέτης is mentioned: ll. 5–6

. . . ὑ ̣π ̣ὲρ Μα ̣γδίο(υ) Εὐζοΐο(υ) οἰκέτο(υ) αὐ ̣το(ῦ) γεν ̣ο ̣μέ ̣ν ̣ου. ‘. . .on behalf of his former oiketeˉs, Magdios, son of Euzoios.’

Magdios belongs to an earlier generation than Leontios, as he was the former οἰκέτης of the late Horion, grandfather of Leontios. The vineyard had earlier been registered in Horion’s name on behalf of Magdios (in the deˉmosios coˉ dex). The situation is not elaborated on further. We thus cannot be sure whether the title οἰκέτης designated a slave or free person (the term could also mean a steward or an 21

Arjava and Gagos 2018, 140–2.

204

Marja Vierros

estate manager; see also the discussion under P.Petra II 17). For some reason, Magdios was not registered as the taxpayer for this property. His patronymic reveals that he was most likely free at the moment of writing the document, but he himself does not have an honorific title (cf. Horion, who is ‘most blessed’ – that is, deceased). Since he is said to be the former οἰκέτης, he had been manumitted (if οἰκέτης meant slave), then he took a patronymic into use; or he was a (free) estate manager whose service had ended but whose honorific title is missing for another, unknown reason. In any case, the taxes for the vineyard he probably owned had been paid in another person’s name, and the property had also somehow ended up in Theodoros’ possession from his paternal inheritance. b. P.Petra V 73: This text only has a fragmentary reference either to children or slaves (l. 18: παιδῶν []). Nothing in the document suggests a more precise meaning (see the discussion above with P.Petra V 58, on the ambivalence of the word παῖς). c. P. Nessana III 31: In this division of property from Nessana, we also have an οἰκέτης mentioned. Depending on the interpretation, however, the οἰκέτης has or does not have a patronymic. This is related to the discussion above about the free or unfree status of the οἰκέτης. l. 43

. . . τῶν κληρονόμω̣ν ̣ [Ἀβραα]μίου ̣ Θε ̣μου οἰκέτου ̣ heirs of Abraham, Themos’ oiketeˉ s, or: heirs of oiketeˉ s Abraham, son of Themos

The editor in his translation assumes that the οἰκέτης is a steward of Themos. However, Themos could perhaps be understood as a patronymic (as in a similar phrase on l. 39),22 and the οἰκέτης would be Abraham, son of Themos. This would mean that in Nessana the term might have slightly different meaning than in Petra. d. P. Nessana III 165: This fragmentary private letter (sixth–seventh century ce) contains the word δουλάριον, doularion, a diminutive of the word δούλη, in plural without immediate context (l. 4: τῶν δουλαρίων). It is uncertain whether δουλάριον refers to male or female slaves, although the word derives from the feminine form. In Nessana, the words δοῦλος or δούλη do not necessarily denote slaves per se (see above, note 10), so the possibility exists that the diminutive works in the same way.

22

See comments on P.Petra III 19 (ed. T. Gagos, L. Koenen, R. W. Daniel), 38.

Slaves in Sixth-Century Palestine

205

slave names From the Petra papyri, twelve personal names for (possible) slaves or former slaves are given; no such names have been preserved from Nessana. Since servile names are poorly attested in late antiquity, the evidence from Petra is important. These names are collected in Table 9.2, each of them discussed below. While the name may be an indication of the origin of a slave, ethnic identification based on the name alone must be made carefully. From the papyri, it becomes apparent that names common among free persons, such as Kyriakos, Hadrianos, Thekla, and Ioannes, were also used for slaves. Jean Straus notes that the slave names that appear in the slave sales from Roman Egypt belong to almost all possible categories presented in a typology by Max Lambertz.23 Straus concludes that there is no clear expectation for slaves to carry only certain names by which they could be identified as slaves, since free persons could carry the same names.24 The evidence from Petra seems to corroborate this: the persons certainly identified as slaves carry names that are also common among free persons. The identification of Hadrianos and Thekla as slaves by using the matronymic was discussed above. Magdios, on the other hand, was an οἰκέτης, and is designated with a patronymic; thus, his status is

t a b l e 9 . 2 Names of slaves, possible slaves, and former slaves in the Petra papyri.

23 24

Male

Female

Hadrianos, son of Thekla (?) Ioannes Kalemeros (?) Kyriakos, son of Kyriake Magdios, son of Euzoios (?) Salamanios, son of OoueOuaelos, son of Mouelos, eleutheros (?)

Almasia (?) Ampelion (?) GanyKyriake Thekla (?)

Straus 2004, 249; Lambertz 1907. Straus 2004: 260.

206

Marja Vierros

unclear (see above). Ouaelos is also identified by his father’s name, as son of Mouelos, and he does appear to be ‘free’ – thus, a former slave. Perhaps identification via patronymic was possible for former slaves, if the father was known; it seems less likely that Mouelos was the former master. The lack of an honorific title can also indicate servile status. However, since most of the Petra papyri are very fragmentary, names often survive without much context, and the presence or absence of honorific titles is difficult to determine.

‘Common’ Names The most common of the names in Table 9.2 is undoubtedly Ioannes (Ἰωάννης).25 The name is clearly of Semitic origin, but it was well spread in different parts of the Greek-speaking world as well. The name Kyriakos (Κυριακός) is quite well attested and occurs frequently with patronymics and honorary titles – that is, as the name of freeborn citizens.26 His mother’s name Kyriake (Κυριακή) is not as common as the male version of the same name.27 The name is also attested as a slave name.28 The name Hadrianos also seems to refer to many freeborn men.29 These three names are not tied to any particular region and are attested around the Empire. The name Thekla (Θέκλα), too, is more common in papyri than in inscriptions.30

Names with a Semitic Background The name Salamanios in exactly this form does not seem to appear in other papyri or inscriptions, but it Grecizes the name Salamā n, from the common root š-l-m in Hebrew, Aramaic and Phoenician and s-l-m in Arabic, either of which root can be realized with a number of vowel combinations.31 The 25

26

27

28 29

30

31

LGPN Database query: 319 attestations; TM People: 3271 in total (including different variants and Coptic forms) [accessed 30 October 2018]. LGPN Database query: 80 attestations; TM People: 230 attestations in the form Κυριακός, 589 times in total (including different variants and Coptic forms) [accessed 30 October 2018]. LGPN Database query: 34 attestations; TM People: 7 attestations [accessed 30 October 2018]. For example, Cyriacus and Cyriace in Rome; see Solin 1996, 311; 2003, 445, 447, 1478. LGPN Database query: 19; TM People: 61 attestations in total [accessed 30 October 2018]. LGPN Database query: 26 attestations; TM People: 144 in total [accessed 30 October 2018]. Honigman 2004, 283–5 with attestations; P.Petra II 17.65–7 comm. (p. 114). In an inscription from the Eastern desert of Egypt, SB XXII 15481 (fourth–sixth centuries),

Slaves in Sixth-Century Palestine

207

part of a name that probably indicates the mother of Salamanios – Οουεperhaps comes from Οουεδης, diminutive of Αουεδης, * ̔ uwayd ˍ ā , as in masculine Αουιδος, Αουειδος, and so on, Latinized as Avidas, from Arabic ‘Awid ˍ . The name Almasia is perhaps an Arabic feminine counterpart of alMa‘ṣiyyah.32 Magdios is also Semitic, its counterparts in Nabatean and Arabic meaning ‘creditable, laudable’, and so on.33 Magdios’ father’s name, Euzoios, derives from Greek εὐζωέω, ‘live well’, and is attested once from northern Sinai.34

‘Slave’ Names The names that have elsewhere been attested for slaves include Kalemeros and Ampelion.35 Kalemeros, ‘good day’, is uncommon in papyri, and the male individuals in the papyri are not unambiguously slaves.36 Ampelion is a slave name in the Ptolemaic period,37 and this name derives from the word ἄμπελος, ‘grape-vine’, with multiple variants (such as Ampelia, Ampelis, Ampelinos, and Ampeline). The fragmentary female name Gany- might be, for example, Ganyma (cf. the male name Ganymas);38 moreover, the male name Ganymedes is attested as a slave name.39

The Prices of Slaves Straus collected all the prices appearing in the papyri40 but argues that there are too many moving parts in each case to say anything definite about the prices of slaves. Furthermore, cases where, in the same time and place, one slave is sold double the price as another eludes the drawing of

32 33

34 35

36

37 38

39 40

Salamanis appears as a ‘slave of Christ’. Salamana also appears as a slave name in the city of Rome, for which see Solin 1996, 605. P.Petra II 17.65–7 comm. (p. 114). Nabatean MGDYW or Arabic Maǧ dy, P.Petra III 19, 5 comm. (p. 38); LGPN has one Μάγδις from Thrace, third century bce. A short funerary text SEG XXVIII 1460, 2–3 (ce 350–499?): Εὐζόει. On the slave names Calemerus and Calemera in Rome, see Solin 1996, 212; 2003 101–4; on Ampelion and other variants of the slave name, see Solin 1996, 512; 2003, 1155–6, 1472. TM Name ID 25104 Καλήμερος: P.Prag II 191, Vo 2 (ce 200–399); P.Oxy. LVIII 3932.11 (ce 500–599). A female named Καλημέρα is given as a slave (δούλη) in SB V 8007 (fourth century). P.Dryton I 4.7 Ἀμπέλιον (acc. of Ampelios, variant of Ampelion); TM Name ID 1993. P.Mich. V 326 (‘Division of inherited land and slaves’, 6 April 48 ce), where Ganymas is a child slave. The name appears in the form Γανύμαν (ll. 10, 23), apparently the accusative instead of nominative, which the editors think is Γανυμᾶς. Solin 1996, 327; 2003, 508. Straus 2004.

208

Marja Vierros

price tables. Individual capabilities could always add to or lower the value of a slave. The Emperor Justinian fixed the prices of slaves in 530 ce to 20 solidi for an adult slave (or 30 if s/he was exceptionally qualified) and 10 solidi for a ten-year-old slave.41 Taken at face value, this suggests that every year of a slave child was worth one solidus. Two documents from Egypt from approximately the sixth and seventh centuries mention prices; both come from Hermopolis Magna. In a sales contract of a slave (SB XXII 15969, 491–518 ce), a slave by the name of Nepheros (described as dark-skinned) is sold for 8 solidi. According to the editor, this sum was equal to a soldier’s yearly wage in the sixth century.42 In another sale (SB XVIII 13173, 629 ce (?)), a twelve-year-old Ethiopian slave girl named Maura is sold for 4 solidi. The sale document is very elaborate, signed by many witnesses, but the sale of the one slave is the only topic. Among the Petra papyri, the only monetary value that can be securely connected to slaves occurs in P.Petra III 28: 3 solidi is the difference between the four- and six-year-old slave boys. This sum must have been thought to compensate for the slave-owner’s expenses in keeping up a child that was not yet able to work. Three solidi is one solidus more than we could expect from the Justinian regulation, but the boys’ age difference could in reality be nearer to three years than two (in fact, the older boy is said to be ‘more or less six years’ [l. 32 ἐτῶν ἓξ πλίω ἢ ἐλάσω]). Three solidi is not an insignificant sum but is also not very much either when compared to some other payments attested in the archive.43 The sum mentioned in P.Petra V 57, the agreement concerning a slave, was 20 solidi. This number might refer to the value of Kalemeros as a fully grown slave (either at the moment of drafting the agreement or in the future). Therefore, we can estimate that Justinian’s regulation was followed quite well in Petra in the case of slave prices.

41

42 43

Cod. Iust. 7.7.1.5: sancimus servi pretium sive ancillae – - – viginti solidis taxari, his videlicet, qui usque ad decimum annum suae venerunt aetatis, in decem tantummodo solidis ponendis: sin autem aliqua arte praediti sunt exceptis notariis et medicis, usque ad triginta solidos pretium eorum redigi sive in masculis sive in feminis ‘We ordain that male or female slaves, shall be valued at 20 solidi – - – provided that those who have only reached their tenth year shall only be valued at 10 solidi; if they have learned a trade, they shall, except in case of notaries and physicians, be valued up to 30 solidi, whether they are male or female’. Translation by F. H. Blume, as modified in Frier 2016. Hoogendijk 1996. In comparison, Stephanous’ dowry is three pounds (litra) of gold (P.Petra III 18); the fine (πρόστιμον) for that document was 24 solidi; in P.Petra III 29 and P.Petra III 31, the fines were 36 and over 30 solidi. In P.Petra I 3, a sum which is apparently paid annually is 12 solidi (for taxes?).

Slaves in Sixth-Century Palestine

209

The Nessana account also gives prices, roughly 10 solidi for the boy and the girl – that is, following Justinian’s regulation. Another Palestinian text, albeit earlier, BGU I 316 (359 ce), a sale written in Ascalon (ἐν κολωνίᾳ Ἀσκ[άλωνι]), attests to even higher prices: the slave (δοῦλος), a fourteen-year-old-boy named Argoutis – his other name might have been Gallos – was sold for 18 solidi. The buyer was a soldier who had been transferred to Arsinoitopolis.

conclusions The new find of the Petra papyri offers many new examples that add to our previously meagre knowledge about slaves outside Egypt in the sixth century. Slaves were clearly a part of property belonging to at least some members of the elite of Petra, since they are attested in around 10 per cent of the surviving documentation. That is notably more than what is preserved from the Nessana documents. In Petra, we have no secure evidence of slaves being sold, as we do in Nessana. Slaves in Petra were kept within the family, divided up with the remaining property through inheritance, and there is also indirect evidence of emancipated slaves. Legislation seems to be followed within pricing, although the evidence is thin. With regard to keeping family units together, there is evidence for and against following Justinian’s regulations. Slave couples were not divided in a division agreement, but two young boys who possibly were brothers were divided in another agreement. The latter follows the legal practice in the Roman Near East as evidenced in the Syro-Roman Lawbook. The terminology used for slaves varies, and several different terms were used even within one document to denote slaves. Interestingly, we do not find the word δοῦλος or its derivatives being used in a Christian sense of ‘slave of Christ’ in the Petra documents, which happens in Nessana. Also, some terms denoting slaves in Nessana differ from those used in Petra. The evidence from Petra points to the direction that the οἰκέται held an unfree status; for Nessana this is clearly more uncertain. The origin of the slaves is never spelled out, but the personal names suggest both Semitic origins and Grecized communities; nothing directly suggests slaves imported from elsewhere.

10 Enslaved Children in Roman Egypt: Experiences from the Papyri April Pudsey and Ville Vuolanto

The ancient Mediterranean was home to a youthful population. Demographic dynamics favoured a relatively high proportion of children and adolescents who were visibly present in their multitudes in the cities, towns, and villages; a significant proportion of these young people were of slave status, some having been enslaved or, as is more likely, born into slavery.1 The presence and lives of enslaved persons are documented plentifully in textual, material, and literary evidence from across the Greco-Roman world, though almost all of this material reflects the concerns and attitudes of the slave-owning echelons of society, and we have no extant narrative testimony from any enslaved person from Roman Antiquity. Further, though the existence of children and young people is mentioned in key sources on, and studies of, Roman slavery, a specifically youth-focused perspective is meagre.2 1

2

On the scale of slavery in Roman Empire, see Scheidel 2011, 288–92, with further references. His estimation is that in urban contexts in Italy, one third of the population might have been enslaved, and potentially one out of fifteen (ca. 7 per cent) of the rural population of Roman Egypt, would have been of slave status. See Pudsey 2013 on slave numbers. Research on children and Roman slavery has mostly appeared as a tangential point in discussions of slave supply (see Scheidel 2011, 297–308, with further studies), education (see Keegan 2013), child work (see Bradley 1991, 103–24; Laes 2008; Gamauf 2012), and sexual relations (see Keegan 2013; McKeown 2007a; Laes 2003). In only a handful of studies has the focus been towards experiences of enslaved children (see Wieber 2012; Herrmann-Otto 2012) – significantly, many of these deal with the early Christian world (Kartzow 2018b; Brooten 2015; MacDonald 2014, esp. 36–48, 60–5, 151–3). See also Sigismund-Nielsen 2013; Busch and Binsfeld 2012; Herrmann-Otto 1994. For inspiration and comparative points, we have consulted research on enslaved children in the Antebellum South, which has been more squarely focused on the agency and experience of children: Duane 2017; King 1995; Pargas 2011; Pasierowska 2016; Schwartz 2010;

210

Enslaved Children in Roman Egypt

211

There are several important questions to be asked: how were the experiences of slavery shaped by age? What were the social and cultural factors that determined the experiences and outlooks of children who were enslaved? How can we understand notions of ownership in the context of children, and relations between children of slave and freeborn statuses? Source material from Roman Egypt of first to sixth centuries ce provides us with a rare glimpse into these concerns in antiquity; surviving documentary papyri include details of transactions, private letters, documented travel, movement, and labour, and tensions and concerns around the physical health of children of slave status. This chapter aims to reconstruct some of the backgrounds, experiences, and concerns of children and young people in Roman Egypt who were to enslaved, through a systematic approach to this material. Such a reconstruction would be near impossible from source material outside of Egypt, where epigraphic, philosophical, and other texts exist in varied contexts and with sparsity. The papyri from Oxyrhynchos in Egypt offer a distinct opportunity to contribute two new approaches: first, a detailed study of relevant evidence from one city, and second, an approach that focuses on bringing together perspectives in slave studies with those in childhood studies. Our reconstruction relates to an outcome of our project, PAIDES: Children’s Lives in Roman Oxyrhynchos, for which we have systematically examined some 1,900 literary and over 5,300 documentary papyri, hundreds of material objects, and all of the nearly 100 inscriptions and some literary texts from the city of Oxyrhynchos. Through these, we explore the lives of children within a particular physical, social, and cultural environment, and analyse the roles, relationships, and experiences of children within that environment.3 For our present purposes, those texts which can be connected both to slavery and children, together around seventy-five, cover over one seventh of the documentary cases in our database from this one ancient city. In what follows, we present these cases and use our analysis of them to reconstruct what we can of the lives, experiences, surroundings, and social environment of children of slave status in Oxyrhynchos. As such, this

3

Wieber 2012. Publications specifically on children’s experience of slavery will be forthcoming from a current Leverhulme-funded project, Enslaved Childhoods: Redefining Roman Slavery, Ulrike Roth. A fuller variety of children’s lives and experiences, with enslaved children’s experiences contextualized with other documents on children and youth in Oxyrhynchos and wider Roman Egypt, takes place in Pudsey and Vuolanto 2021. See now also Pudsey 2015; 2017; Pudsey and Vuolanto 2017; Vuolanto 2015b; Vuolanto 2017.

212

April Pudsey and Ville Vuolanto

study would be the first systematic reconstruction of aspects of children’s lives as slaves in an ancient Graeco-Roman city.

enslaved children in households From Oxyrhynchos we have twenty-one preserved census returns that also include minor children; in five of these households there appear children registered as slaves.4 They appear to have lived in relatively large households, in all likelihood with peers of similar age groups. In one household, for instance, a man in his sixties lists as members of his household his son, aged thirty-two, his son’s wife, aged thirty-one, their son aged four, who was born away from home. There is another son mentioned, a daughter of nineteen years, and another daughter who had died. Additionally, we have a enslaved woman with her two children; unfortunately no ages have been preserved for them. Here we apparently observe two children of slave status living with their mother, though without their father or any other relatives, in a household alongside an extended family that also includes a four-year-old boy.5 Census data for children of slave status shows them as more likely to be living with surviving mothers and siblings than either individually or with their fathers. There are some surviving notifications of birth for slave children in Roman Egypt – though only one of these relates Oxyrhynchos6 – and a number of slave children were registered in the city registers as members of metropolitan-status households at the age of thirteen or fourteen.7 The reason for this is evidently that slaves of metropolitan-status owners were subject to a reduction in poll tax, with the consequence that registering enslaved people’s births or their coming of age (at the age of fourteen) was a way of ascertaining their reduced-tax status. As can be observed in the household census more broadly, one feature of the behaviours around completing the household census was the conscious non-recording of those eligible for taxation (boys over fourteen years of age). It is highly 4

5

6 7

See Table 10.1. Of course, this is a small fraction of the total in even the recorded households; many children of slave status may not have been recorded as such for various reasons, or recorded at all, and many census returns are incomplete due to accidents of survival. P. Oxy XXXIII 2671 (216/7), the return relates to Leonidou in the Herakleopolite nome. See Table 10.1 for other cases. See Pudsey 2013 table 24.3. P. CtYBR inv. 1386, 143–4. Chrest. Wilck. 217 (172–3 ce); PSI VII 732, 153–4; P. Oxy. LXVII 4584 (100/101 ce); P. Oxy. IV 714 (122 ce); P. Oxy. XII 1451 (175); PSI XII 1230 (203 ce). See also SB XXII 15210 (77 ce), which may be a enslaved boy with sibling owners.

Enslaved Children in Roman Egypt

213

table 10.1 Recorded enslaved children in Oxyrhynchos census returns. P. Oxy XXXIII 2671 (216/17 ce)

215-He-3

SB XXIV 16011 (11 CE?)

P. Oxy. XII 1548 (202/3 ce)

201-Ox-1

P.Flor. I 4 (245 ce)

243-Ox-1

P. Oxy. VIII 1110 (187–8 ce)

187-An-1

This census return relates to Leonidou in the Herakleopolite nome. A man in forties and his 7-yearold son, a number of slaves (man aged 50, his sons of 14 and 5). A man with two daughters (15 and 20), and four slaves: man aged 18, girl aged 13, woman aged 35 and her 9-year-old son (13-year-old lives with 20-year-old and her husband; 9-year-old boy lives with the declarant and his 15-year-old daughter). Two brothers have slave boys aged 3, 0, 0, and a girl 5, with their mother 24 and one more slave 19. (Not from Oxyrhynchos.) See Pudsey 2013 table 24.3.

likely that similar tax-focused thinking played a role in the recording of slaves.8 From the principate onwards, the slave population across the Roman world was largely maintained through slaves born to slave couples on estates and households, often even by straightforward slave-breeding. Indeed, the perceived value of vernae, home-bred slaves, was much greater than of enslaved people.9 Inferring from the comparative viewpoint outside of Egypt, especially from the patterns of slave reproduction, it is highly likely that a great proportion of them were vernae, that is, born into slave families.10 The registration documents from Oxyrhynchos mention the origin of the slave, and in all but one of these cases these slaves were indeed

8 9 10

Bagnall and Frier 1994, 1–30, and boys over 14. See Scheidel 2011; Herrmann-Otto 1994. Bagnall and Frier 1994, 53–74.

214

April Pudsey and Ville Vuolanto

homeborn – which would reflect a long-term relationship with the other household members. In one of these documents we meet a boy who was to have his own epikrisis (registering of his coming of age) in the same year, and who wants to register a slave owned jointly by himself and his sister: the head of the household and the enslaved boy are purportedly exact age peers. In another registration, made by a certain Trunnia, mother of two and owner of three children, there appears an eleven-year-old daughter, a thirteen-year-old son, and three vernae of slave status aged five, nine, and one of unknown age.11 The importance of social interaction with peers within one’s age group is an influential aspect of child development in terms of broader social and cultural interactions, and we see here one important aspect of slavery often overlooked in scholarship of ancient slave studies: at least in some areas of the Roman world, people who were born into slavery, though subject to ‘social alienation’ in some senses, were raised and subject to peer interactions with those of free status, likely with varying ethnic, religious or cultural backgrounds. The one exception not mentioned to be a verna is thirteen-year-old Melas, who was a foundling.12 Indeed, many of the slave children, both girls and boys, would have shared this background, as foundlings would have been collected and raised as slaves – a practice not specific just to Egypt, but across the Graeco-Roman world. As many of the quite numerous (six from Oxyrhynchos alone) wet-nursing contracts specify, foundlings were found in dung heaps and they were given to wet-nurses (often freeborn themselves) to bring them up usually for two years, and then to return to their finders as slaves. In one of these contracts, it is even pointed out that it was drafted on the very same day as the child was found.13 These nurslings were unquestionably brought up for profit: not only was it clearly stated 11 12

13

Chrest. Wilck. 217 (172–3 ce); P. Oxy. 12 1451 (175 ce). PSI XII 1230. There are all together six certain cases preserved from Oxyrhynchos that mention an epikrisis of enslaved children. Wet-nursing contracts: P. Rein II 103 (26 ce); P. Rein 2 104 (26 ce; suckling infant girl found from a dung heap to be wet-nursed for two years); SB 16.12952 (68 ce: a child found from a dung to be raised as a slave); PSI 3 203 (87 ce: an infant found ‘today’ from a dung heap, to suckle and nurse for two years); SB XVIII 13120 (81/82 ce); P. Oxy. LXXVIII 5168 (18 bce). See Pudsey 2013. For foundlings (and child abandonment as a neighbourhood phenomenon in Graeco-Roman worlds) see Evans Grubbs 2011; 2013. See also P. Oxy. I 37 (49 ce) and P. Oxy I 38 (49–50 ce: a court case – wet-nurse for a boy found from a dung heap. The master slips in to the house and takes the child – but the nurse and her husband claim the master had taken their freeborn boy instead, as the foundling boy had died earlier. Court decided in favour of the parents getting their child back).

Enslaved Children in Roman Egypt

215

they were to become property of their finders, but also they were assets in trade. One document shows, for example, that after an enslaved girl was wet-nursed for two years, and thus survived her infancy, she was immediately sold after weaning. Thus, the decision to nurse the girl was clearly to draw profit from her.14 Rescuing and nourishing a small child often had very little to do with the kindness of strangers.

separation from the family of origin: the mobility of enslaved children There were many individuals who were sold and resold as children. Enslaved children regularly moved from house to house, town to town, and even province to province. Even the local sales were often enough to separate the children from their parents and their peers, friends, and familiar community. In one case there was an eight-year-old girl sold in a deal between two women, in another a girl aged around twelve years was moved to another house, and in a third one it was a fourteen-year-old boy, Sarapous, who was moving to another house due to a partial sale.15 Infants were not only subject to becoming slaves in a new household as foundlings, but there were also infants subject to sale and moving to new households with their slave mothers.16 This issue of sale along with family members is consistent with the family groupings of slaves we see dominating the household census returns, and rather bucks the trend of what we see across the Roman world (and slave societies) more broadly, where enslaved families are typically separated. In one case there appears, for example, a sale of a woman of thirtyfive years of age along with her two children: a daughter and a one-year-old son. It is also worth noting the age of the mother in this example is the oldest known recorded age for a slave woman in the Roman period.17 In some cases, we observe the enslaved moving, after being sold, across longer distances and often between major cities in Egypt as a common experience. A fifteen-year-old girl, originally from Pentapolis, was purchased by a buyer from Oxyrhynchos. In another case it is mentioned that a now twenty-four-year-old enslaved woman had been bought in the Little

14

15 16 17

PSI XII 1228 (188 ce: sale of a half share of a slave girl. See also Col. Inv. 502, a contract for nursing a slave child, first or second century ce. P. Oxy. II 263 (77 ce); SB XX 14285; P. Bingen. 62 (89 ce). SB XX 14395 (170 ce) and P. Oxy. IX 1209 (251–2 ce). P. Oxy. II 375 (c.79 ce).

216

April Pudsey and Ville Vuolanto

Oasis when she was either two or twelve years old. A sick (‘leucoma’) teenager aged nineteen, a hereditary slave from the father, was sold between individuals at Antinoopolis and Oxyrhynchos.18 Such sales could sometimes follow each other in a rapid sequence. In a document written in Oxyrhynchos a seller vows the details of the history of the individual: a seven-year-old boy was sold for a fourth time: the second selling had been in the Small Oasis, the third and fourth in Oxyrhynchos.19 It seems that in all these cases the sale would imply also the separation of the child from all of their former family or household relationships, both kin and non-kin, and thereby further impact the emotional lives of these children and their natal and social alienation. For some children, a sale would imply a move over great distances and a loss of familial culture and even the language. In one case, for example, a girl, Rufina, who was originally a ‘Moor’ from Caesaria Mauretania (present-day Cherchell, near Algiers), was sold in Rhodes – where the document was drafted – when she was ten years old or in her teens. Eventually she ended up in Oxyrhynchos, as did also Balsamea, a seventeen-year-old Oshroenian Mesopotamian, who was part of a contract of purchase made in Tripolis of Phoenicia. A boy, Zoilos, aged nine years, had been sold twice and was recorded as ‘more or less’ Macedonian by origin when he appeared in Oxyrhynchos.20 Another boy, originally named Prokopton but now renamed Aptus, was sold in Oxyrhynchos in 265, seemingly at the age of two. He already had a history of two previous sales behind him – both in Bostra in the province of Arabia.21 These are very large distances and would have presented swift and impactful changes in cultures, languages, and interactions. The sales of children are also discussed in some private letters, for instance when a friend hints to an enslaver based in Oxyrhynchos that an Alexandrine lady was interested in buying three young male

18

19 20

21

P. Ross. Georg. III 27 (234–5); P. Amst. II 46 (217 or 227); P. Ant. III 187a (198 ce). See also P. Oxy. XLIX 3477, 270 ce (A verna girl of sixteen years of age was sold in Cynopolis in the Delta); another verna, Isidora, aged twelve years, was subject to purchase in (Small or Great) Oasis: P. Oxy. LV 3784 (227/8 or 281/2). P. Oxy. LX 4058 (158–9). Rufina: P. Oxy LX 3593 (238–44); Balsamea: P. Oxy. XLII 3053 (252); Zoilos: PSI XII 1245 (237 ce). P. Oxy. XLII 3054 (265 ce) His age is contested by the editors as implausible; still, all three sales mentioned in the document would fit into this time limit.

Enslaved Children in Roman Egypt

217

acrobats,22 or when a son writes to his mother because he needs help in selling his small boy, seemingly unaware of how much he should get for the boy from the market.23 It is often impossible to know which enslaved children were separated from their mothers and other family members when they were young. In our documents those who had been sold, and thus had the highest probability of separation, are quite naturally overrepresented in written transactional records, compared with those children who did not leave any official traces of themselves. Still, it is clear that a constant threat of being sold to a foreign household, or even to a foreign land, was a very real one and, for children in particular, one that must have been felt with fear and with a great deal of intensity – for some, this risk became a reality. This forced mobility of enslaved children indeed reflects not only their economic worth to the enslavers and their social vulnerability, but it also highlights elements of slavery as a system, such as enforcing social death and dishonour; indeed, it is a part of the inner logic of slavery that people do not have access to their communities of origin, their families, traditions, and stories: their heritage, their identity, their continuity are negated. As Orlando Patterson crystallizes the implication of these processes: as slaves cannot anchor their ‘living present in any conscious community of memory’, they encounter ‘natal alienation’.24 This process can hardly be seen more clearly than in these sales of children, and, all too tragically, was likely to have been experienced traumatically by them.

the ‘value’ of enslaved children: work and education Even small children unquestionably had some economic value to the slaveowning echelons of society. This is also confirmed in a marriage contract that mentions mikrois paisi – small slave children – as a part of the dowry. As one would expect, the enslaved – including children – were mentioned regularly as a part of inheritance, for example in a case in which an inheritance was to be distributed among seven children after the death of their parents. There were many slaves, among them Sinthoris, a twenty-five-year-old woman with her ten-year-old daughter, whose ownership was to be shared by the 22

23 24

P. Oxy. XXXVIII 2860 (second century ce). See also P. Oxy I 117 (second or third century ce): Chaereas writes to his brother, wanting him to sell the children (otherwise unspecified). PSI XIII 1359 (second or third century ce). See Patterson 1982, 5–16. We owe thanks to Chris de Wet in pointing out the relevance of Patterson’s concept of natal alienation for our chapter.

218

April Pudsey and Ville Vuolanto

siblings.25 A further asset was the reproductive capacity for girls; thus, for example, in one case an enslaved girl of twelve years old was given as an inheritance, with a standard formula ‘with all her future descendants’. In another case a girl called Nike was given as a security of loan with her future offspring.26 As vernae were a major source of the slave supply (or put otherwise, as ‘breeding’ of slaves on estates made good business sense to most), a girl who had survived childhood mortality, to an age when all her reproductive capacity lay in front of her, would have been viewed by an enslaver as a potentially lucrative opportunity for amassing capital. It was only in their early teens that the prices for enslaved women could have equalled those for enslaved men.27 The great estates had enslaved children as a part of their work force; thus, Apollonius urges another man, a manager of a farm, to take care that ‘little Polydeuces is to work with Melas the carpenter, and not to leave his side’.28 We meet also Totoes, still a minor and named after the father of his owners, who is hired to work for another household – he would have been able to return to his home after the contract’s duration had been completed.29 In a series of documents we frequently find the enslaved as workers in estates alongside free young people. Some of these were seasonal workers, identified as such because of their wages; these include family groups, with children working with their brothers, fathers, and uncles – with half a pay.30 In some of the account lists, however, it is unclear, whether the words used of the workers refer to slaves more generally, or specifically to child slaves. If a document mentions both paides and paidiskoi working without pay; paides helping in grape harvesting, carrying bricks, using donkeys to carry hay and bricks; paidarioi turning over terrain and carrying something; can we infer from these clearly distinguished distinctions that paides might be enslaved adults and paidiskoi enslaved children?31 In another document, there were payments both to a paidarion and to paidia; in still other to 25 26 27

28

29

30

31

P. Ifao 3.5 (second century ce); P. Oxy. XIV 1638 (282 ce). SB XX 14285 (59 ce); P. Oslo II 40a (150 ce). See more generally Scheidel 1996. As Harper (2011, 293) points out, this may also be linked to their sexual potential. P. Oxy. XLII 3066 (third century ce) not clear if free or slave, even if name could refer to slave status. PSI VI 710 (second century ce); unfortunately, we do not know the duration of the agreement. P. Princ. II 96, r. col. I (551–7 ce) with Vuolanto 2015b. See also PSI VIII 953 (567–8 ce) with families of Goths serving in Apion estates in Oxyrhynchos after the Gothic wars in Italy. SB XIV 11960 (ca. 260): a receipt/calculation for the work done.

Enslaved Children in Roman Egypt

219

paidariois and korasiois; and once a paidarion was to be remunerated for the fish he had bought.32 There are no self-evident solutions here, but distinctions between concepts used and work attributed to different people seems to imply that child labour would also have been included here – as does comparison with cases where there are differences in payments for the same work. Enslavers also sought profit by taking care that enslaved young people would acquire new skills. There are five apprenticeship contracts from Oxyrhynchos that relate to slaves. It would appear usual for such individuals to continue living with their enslavers, and to stay with their apprentice master from sunrise until the sunset (this was also typical of many apprenticeship contracts for free young people). Three of these apprentices were apprenticed into the weaving business, one of them a girl, and two other boys were to learn shorthand writing.33 In terms of everyday experience for these young people we can do little more than speculate around the ways in which individual experience and outlook would have differed between slave apprentices and free apprentices – or indeed, whether in a larger workshop the master would have made much distinction between the many people learning and working. Enslaved children and freeborn children also appear together in some simple notifications of death to local officials, which include slave children; in one of these, from the mid third century, there appear a freeborn boy and two enslaved, the older of which was sixteen, who died in the same household. One is left to speculate if this was due to an epidemic or a major accident.34 However, taking these pieces of evidence 32

33

34

P. Mich. XI 619 (after 183): a paidarion and paidia were paid. SB XVI 12764 (first or second century ce): this is an account list for salaries paid to the slaves (paidariois/ korasiois) of a certain Thoth. P. Haun. III 36 (350–450): accounts, with a paidarion Stephanios to be remunerated because of his fish. See also P. Oxy. 68.4683 (426 ce): paidariois supply for two double jars of Theban wine; P. Oxy. LXVIII 4680 (419 ce): Kyriake, a girl pediske, an order to supply oil. BGU IV 1021 (third century ce): a slave boy, wool carding, three years (the length of the contract suggests that the boy in question was unskilled), lives with his owner but stays with his master from sunrise to sunset. P Oxy. XLI 2977 (239 ce) wool carding for five years, wages rise according to training (with first six months no pay – which again refers to an unskilled worker), to work from sunrise to sunset. P. Oxy. IV 724, a slave boy learning shorthand, living with his apprentice master for the period, which is at maximum two years – it is stated that he can return sooner, if he has learned and the master does not want to keep him. P. Oxy. XLI 2988 (second century ce), an apprenticeship contract for shorthand writing, but very fragmentary. P. Oxy. XIV 1647 (late second century ce): an underage female slave, an apprenticeship contract to a weaver for four years. P. Oxy. LXXIV 4998 (253/4). See also P. Oxy. XLIX 3510 (79/80): a slave boy belonging to declarant’s wife, Diogenes, had died (also another slave had died, belonging to declarant’s brother, in turn).

220

April Pudsey and Ville Vuolanto

together, it is clear that the young people would often have had access to peer relationships both with enslaved children and with free children, regardless of their own background. Such children quite frequently grew up alongside free children and adolescents of similar ages, and shared with them their cultures, social practices, and social environments.35 Can we see here any chronological change in slavery, or in the living conditions of enslaved children? Foundlings, picked up to be suckled by nurses and raised as slaves, indeed appear mostly in first-century documents. The young slaves we see referred to in epikrisis and gymnasial registrations, and as apprentices, are more common in the second- and third-century documents, again in step with what we see across the cities of the province. On the other hand, we note that our examples of sales and travel of children as slaves are more frequently found in the third-century documents. However, these trends are perhaps less due to temporal and cultural changes and social practices, and more due to patterns in documentation. The nature of survival of documentary papyri relating to child slaves in Roman Egypt is no different to the patterns of survival of many other types of documentary papyri; that is, more documents have been excavated, worked on, transcribed, and published from the second and third centuries in key sites than elsewhere. Moreover, the documentation of certain deeds and contracts would have changed across time and place. Thus, taken together, the cases we have here from Oxyrhynchos give us little detail of change in the practice of slavery itself, or of change in the everyday lives of enslaved children, between the earlier Roman empire and late antiquity.

conclusions: slavery as lived experience Enslaved children were everywhere in Roman society. They lived, worked, and experienced life in homes, in fields, and in workshops with apprentices in the city. As the census documents and information concerning the work conditions of enslaved children show, in Roman Egypt their lives were often intertwined with those of free children – perhaps even more so than between enslaved adults and adult free persons. It was not unusual for families of slaves to reside together and to continue to be in constant interaction, especially through work and learning. As for leisure and play, the papyri have very little to say, but interaction was expected at least when concerning the younger children. Studies of material cultures of children in Roman and Late Roman Egypt, however, hint at the roles of 35

See also Pudsey 2017 and Vuolanto 2015b.

Enslaved Children in Roman Egypt

221

shared peer interactions and play with toys, in household and courtyard settings – these would invariably have involved slave and free children playing together.36 Also, the comparative material from late antiquity shows that enslaved and freeborn children were in frequent interaction – which was even encouraged as a way for freeborn children to socialize into their adult social roles (as enslavers) at an early age.37 Therefore, any firm distinction between social environments for children of slave and of free status was not as clearly defined as the enslaved/enslaver dichotomy we see firmly entrenched in many Roman elite texts might imply. These lines of distinction were rather more porous, particularly in the sphere of everyday life. Slavery was not necessarily a fate for the rest of a child’s life. Although in our material from Roman Egypt we do not meet any enslaved children manumitted, we do have one seventeen-year-old and two eighteen-yearold individuals being freed – in one of these cases it was a thirty-year-old freeman who had paid for the manumission fee for ‘her mistress’. In one case, a thirty-five year-old verna was manumitted by an owner of over sixty years old, thus the owner had known her for all her life.38 Relationships between enslaved children and their enslavers, once those children had grown up, were necessarily longer-standing than in other cases of individuals purchased or moved as adults. We do not want to suggest, by any stretch of imagination, that the experience of slavery would have been positive or in any way beneficial for those enslaved. Still, the affective aspects of such relationships, in particular between girls and women, can be glimpsed through some documents. In one case, we meet a small enslaved girl with an escort on her way to her singing lessons; her owner, an elderly childless woman, had the idea of bringing the girl up to take care of her in old age; however, the girl was knocked down by a donkey driver while ‘crossing the city’ and was badly injured – 36 37

38

See Swift, Stoner, and Pudsey, forthcoming. See e.g. Joh. Chrys., An Address on Vaingrory, 68–9 (peer relations), 71 (learning to behave properly towards slaves), as well as Vuolanto 2013, 65–6, and De Wet 2015, 164– 5. Manumissions: P. Turner 19 (101 ce), a seventeen-year-old verna (Sinthoonis); manumission fee paid by Ammonios, a thirty-year-old freedman, to the owner; SB 1 5616: a girl of eighteen years (first century ce); P. Oxy. XXXVIII 2843 (86 ce), a woman of thirtyfive years, homeborn, manumitted by her owner, who had known her since her birth; PSI XII 1263 (166–7 ce), a testamentary manumission: Sintheus frees an enslaved woman and her daughter, with a monthly pay of grain and right to live in the house; P. Scholl V (first or second century ce): slave of eighteen years (?) to be manumitted (a testamentary manumission).

222

April Pudsey and Ville Vuolanto

this is why we hear of her.39 In another report, we see a tragic case in which an eight-year-old boy of slave status died after falling from an upper floor window while he was watching dancing girls in the street.40 Another enslaved child, having the worst of luck in her free time, was a twelve-year-old girl who was swimming with her friends when she was caught in a sluice and was drowned.41 We know of these young people because of the various accidents the children encountered. Our information comes through these types of documents: if the everyday life of the ancient population was seldom worthy of comment by the ancient sources, even those preparing documents and letters on papyrus, a child’s normal life was less remarkable, and the fate of an enslaved child even less so. These kinds of cases hint that long, familial and even intimate relationships were also possible for enslaved children, some of whom clearly could live – at times – a life with time for play and some agency in their everyday interactions. However, it is clear that they did not have much say in their lives. Certainly, most of the enslaved children – even these same children with somewhat affectionate relationships and time for play – had to work very hard early on, and many, if not most, would have been subject to extreme physical, sexual, and psychological abuse. This is something we do not see recorded in documentary papyri, but a topic dealt with by scholarship based on other evidence outside Egypt.42 One case to remind of the harder realities in our corpus is a thirdcentury notification of a runaway slave, about fourteen years old, named Philippus, pale skinned, speaking badly, broad nosed, and wearing a woollen tunic and a used shoulder belt.43 As is most often the case, here our sources tell us about the boundary conditions for the children’s experiences, rather than about the lived experiences themselves.44 We know nothing else about this Philippus, but what we can see here is that 39 40

41

42

43 44

P. Oxy. L 3555 (60–130 ce). P. Oxy. III 475 (182 ce). For this and the previous case, as well as references to a number of other accidents of children in the Roman world, see Graumann 2017, 270–2. An unpublished third century ce papyrus, a doctor’s report, linked to the Ancient Lives Project (Dirk Obbink): ‘From Aurelius Philantin . . . son of Neoptolemus Aurelius . . . I . . . certify that . . . a slave girl, by accidental death after being caught in a sluice . . . while swimming with her friends . . . twisted and lifeless body . . . by drowning. . .’. See also e.g. Brooten 2015 on the vulnerability of slave children and the effect of their separation from their parents; and Laes 2011, 153–5, 163–6, 223–34, 259–61 on child slaves and work, violence, and sex. See also Roth (in preparation). P. Oxy. LI 3615 (third century ce). See also Vuolanto 2017, esp. 21–2.

Enslaved Children in Roman Egypt

223

a slave was first and foremost seen through his status as a subjugated person, or, rather, as someone else’s property – regardless of his age. Slave children, even when treated well, were always under their master’s total domination. What we observe is a localized snapshot of how the practice of slavery operated in respect of children and young people’s everyday lives.

11 Late Antique Slavery in Epigraphic Evidence Mariana Bodnaruk

While slavery was entrenched in the high Roman Empire, it continued well into late antiquity. The aim of this article is to evaluate the overall state of the evidence and to discuss the epigraphic and material culture of late Roman slavery, focusing on inscriptional sources for slaves in the later Roman Empire. I contend that the comprehensive overview of this type of evidence enables one to reintegrate the textual, visual, and physical aspects of artefacts of the period and to move the discussion about slavery and material culture forwards, by employing the intersectional approach. To be sure, late Roman slaves were not ‘class for itself’, while ‘class in itself’ they certainly were. They left a visible footprint in archaeological sites, and the epigraphic record that survives from the later Empire holds anthropological information about Roman slave culture: social interactions and a hierarchy of statuses, values, and beliefs. I argue that inscriptional data can be as ambiguous as literary texts, and, more importantly, that the violent subjugation of slaves in the later Roman Empire found its basis, particularly, in the material evidence. Examining the epigraphic representation of late Roman slavery from the late third to the sixth century, I use intersectionality as an analytic tool that is crucial to understanding the complexity of exploitation and oppression in the late Roman slave-owning society. First, I overview specific genres in the late antique ‘epigraphic habit’ recording slavery. I explore inscribed imperial edicts mentioning slaves, census inscriptions in their provincial rural contexts, honorific inscriptions set up by slaves, votive inscriptions giving a glimpse on enslaved war captives, funerary inscriptions with slaves as both dedicators and

224

Late Antique Slavery in Epigraphic Evidence

225

dedicatees, and, lastly, private identification inscriptions, in particular slave collars. Then, I turn to different categories of slaves and of unfree labour in general in the later Roman Empire as recorded in the inscriptions. I examine both ubiquitous domestic slaves in private ownership (servi domestici) and public municipal servitude (servi publici) late into the fifth century. Next, I go through a series of inscriptions, recording various types of slaves in their respective labour contexts: more specifically, urban and rural slavery. Finally, I assess inscriptions mentioning liberti (freedmen) as well as coloni and adscripticii (free and dependent tenant farmers). I end with a brief discussion of the socio-economic realities of coerced labour. I conclude with a delineation of what the epigraphic sources reveal about late antique slavery and the intersections of gender, class, and ethnicity.

late antique legal and census inscriptions recording slaves I begin with inscribed imperial pronouncements such as edicts and tariffs, as well as epistulae (letters) and rescripts regulating slave prices and taxes or imposing and guarding compulsory labour. Numerous but highly formulaic epigraphic sources, in the form of lists, are populated with slaves who appear gendered and classed on the basis of age while being commodified and priced, sold and bought, owned and disowned (manumitted) via market-frames and legislation. The edict issued in 301 by the Emperor Diocletian, comprising a justificatory preamble and a list of ‘prices for the sale of individual items that it is not lawful for anyone to exceed’, extant only in inscribed copies, contains a slave chapter.1 The slave prices fixed in the imperial utterances shed light upon the value of reproduction in the Roman slave system, distinguishing male and female slaves by price. Dated to the fifth or sixth century on palaeographic grounds, a late antique civic tariff recorded in an inscription from Anazarbus in Cilicia taxed slaves brought into the city

1

All translations are mine unless indicated otherwise. The chapter on slave prices has not been preserved complete in any inscriptional copy, either in the original Latin or the Greek translation of the tariff list found throughout Achaea; see Giacchero 1974, ch. 29, 208 (Latin) and 209 (Greek). However, the integration of the fragments from Iasos and Aphrodisias with the preserved text from Aezani allows the reconstruction of almost the entire slave chapter in its original Latin version; see Salway 2010, 19.

226

Mariana Bodnaruk

for sale among other commodities subject to customs, itemizing them above cattle.2 Legal epigraphic documents related to the necessary work of construction document coerced corporate and slave labour. An inscribed imperial epistula from Rome, dated to between 296 and 312, records that a reorganized building guild (collegium fabrum tignariorum) became bound to its duties together with the sons and slaves (filios vel servos) of artifices (craftsmen artisans).3 Exceptionally, the fourth-century marble tabula (tablet), dated on epigraphic and onomastic grounds, recording the circitores (slave guardians of aqueducts) of Tivoli, has a total of at least nineteen active servi publici ascribed to the single aqueduct (forma).4 A bilingual inscription preserves an imperial rescript from 527 issued in response to a petition in order to protect the properties of an oratory of St John from the exactions of soldiers and biocolitae (militia):5 ‘the estates mentioned in (these) complaints and their peasants (γεωργοί) and registered peasants (ἐναπo´γραφοι), and overseers (φροντισταί), and [tenants (μισθωταί)] should be free from [the passage of soldiers or militia (βιοκωλυταί)] and from those units who are known to be permanently garrisoned next to these estates’.6 This late Roman constitution mentions coloni (γεωργοί), free tenant farmers, as well as adscripticii (ἐναπόγραφοι), dependent coloni. The adscripticii or ἐναπόγραφοι (‘registered’) γεωργοί were land-bound cultivators recorded in census registers under the name of the landowner (δεσπόται) on whose estate they lived and whose land they worked. A law of the late fourth century on bound tenants in Thrace proclaims that ‘although they [i.e. registered coloni] seem to be free-born in their condition, even so they are to be considered as slaves of the land to 2

3

4

5

6

Dagron and Feissel 1987, 170–85, no. 108, pl. XLV. It is not clear from the content whether it was a municipal tariff on local trade or an internal imperial one. The editors suggest the latter, noting that the rough methods of assessing the value of goods recorded were more suited to large consignments than to taxing the produce of local peasants. AE 1941, 68, with Fabiano 2019. The absence of slaves from construction inscriptions can be explained by the fact they, for the most part, were highly formulaic documents, adhering to a set pattern. CIL XIV 3649. This inscription preserves a list of public slaves, indicating their children. The prominent predominance of men in this roster of Roman slaves, who looked after the aqueducts of Rome, is related to their labour employment, which would explain the malebiased sex ratio. The aqueduct slaves, whose job was to patrol and prevent damage to an aqueduct, continued to perform their function into the later Roman Empire. CIL III 13640=ILCV 23=AE 1894, 68; see Feissel 2010, 242 n60, 253, 275 n86. A new edition of this text is anticipated shortly. Amelotti and Migliardi Zingale 1985, 97–9 (=CIL III 13640); CSLA, E00869 (trans. Nowakowski).

Late Antique Slavery in Epigraphic Evidence

227

which they have been born’.7 This fourth-century pronouncement is a confirmation that bound conditions were enforced even on the lives of free-in-status adscripticii, within the coercive apparatus of the landlord’s power and sanctioned by the social legitimacy of the slave regime. Late antique inscriptions also offer unique testimony for large, slavebased properties, in the form of census registers. These have been well studied by Kyle Harper, who emphasizes a self-reproducing enslaved agricultural workforce in the late Roman context.8 Although the slave mode of production was not a defining feature of late Roman rural economies,9 slavery was an important component of agriculture in the fourth-century East. This is confirmed by the fragments of census records inscribed on stone, preserved from eleven cities in the Aegean islands and along the western coast of Asia Minor10 dated to the fourth century on epigraphic and onomastic grounds, possibly to the early 370s.11 The Greek census lists show farms that had two, four, eight, sixteen, twenty-one, and twenty-two slaves. However, the Thera inscription, first published in 2005,12 lists 152 slaves, with their names and ages, belonging to a single landowner. The balance (partially identified: sixty-three females and fifty-six males) supports the view that, in late antiquity, the sex ratio of the slave population was not heavily tilted towards males, even in the countryside. The large number of children also proves that natural reproduction through slave families was important among the slaves on Thera. Although Ramsay MacMullen claimed from epigraphic evidence available to him that slavery was not prominent in agricultural production,13 it was in fact instrumental in the intensively managed estates in these regions.

servi and liberti as awarders: setting up honorific and votive dedications I now turn to a genre of the honorific statuary featuring slaves as awarders. Some time between the years 339 and 341, a marble statue of Fabius 7

8 9 10 11

12 13

Cod. Iust. 11.52.1. However, the act of registration did not change the legal status of these individuals as free men. See Grey 2011b, 503 on an early-fourth-century legislation, which kept the positions of a colonus and a slave strictly separated. Harper 2008. Banaji 2010, 181–214, 231–40. Jones 1953, 54 (Magnesia), 55 (Thera), 56–7 (Lesbos and Chios). On the dating see Harper 2008, 85–6, contra Jones 1953, 49, followed by MacMullen 1987, 362. Geroussi-Bendermacher 2005, 340–4 and 354–8 for photographs. MacMullen 1987.

228

Mariana Bodnaruk

Titianus, consul and praefectus urbi (city prefect), was erected in Rome.14 A slave, known as Peregrinus, despite his subjugated status, acted as an awarder of the statue to his master, a Roman senator, demonstrating pride in his connection to a powerful person. Peregrinus must have been a domestic slave (servus domesticus) of the senatorial household of Titianus in Rome. The statue must have been set up within the private properties of Titianus, as the slave could not erect a public monument in one of the prominent sites of the city.15 To be sure, no honorific representation of slaves is known from late antiquity. Moreover, the monument is the only surviving statue dedication by a slave, whose name is preserved as a part of the honorific inscription, and who must have occupied a very important position in Titianus’ household.16 Next, a dedicatory inscription on an altar of the goddess Victoria, set up on the occasion of the victory of a Roman army over the Juthungi near the Rhaetian provincial capital Augusta Vindelicorum, records a raid by the tribe deep into Roman Italy in the winter of 259/60 that carried off ‘many thousands of Italian captives’,17 but was eventually stopped by a Roman general near the Lech in April 260. The epigraphic evidence usually does not allow the identification of slaves in Rome and elsewhere in the Roman world who had their origin among subjugated peoples, such as, for example, a tribe of the Bavares Mesegneitises in Mauretania Caesariensis in Africa. A Roman governor of that province in the last quarter of the third century erected a dedication to the Dei patrii and Dei Mauri conservatores (indigenous and Mauran gods) ‘to commemorate his crushing of the tribe of the Bavares Mesegneitises and his carrying off their families into captivity along with all the booty’.18 Inscriptions equally record actions in the public sphere, such as dedications and benefactions, made mainly by liberti. A Christian mosaic inscription coming from the coemeterium Pamphili (catacomb of Pamphilus) on the Via Salaria Vetus in Rome is dated to 290–324.19 It 14 15

16 17

18

19

CIL VI 1717=ILS 1227. The private character of this dedication thereby indicates that it may have come from a family villa of the honorand. LSA–1422 (Machado). AE 1993, 1231. Late antique source materials yield some important glimpses on slaveimport across the frontiers and enslavement in frontier wars. Romans during the later empire were equally regularly captured and enslaved by barbarians along all the frontiers. CIL VIII 21486=ILS 4495. It could have been a local rebellion, when under P. Aelius Aelianus the struggle against the Bavares was renewed between 270 and 280 (or 278 and 280); see Witschel 2006, 167. AE 1921, 80=ICVR X 26460.

Late Antique Slavery in Epigraphic Evidence

229

probably recalls a work done by freedman Vitalio together with Quodvuldeus in the cubicle of the martyrs Theophilus and Pontianus: ‘In God the Father almighty, Vitalius the freedman made [this] with Quodvultdeus for his lord Theophilus and lord Pontianus, welldeserving in rest.’ The martyrs Theophilus and Pontianus were buried in a double cubiculum (room) in the pars infima (the lowest part) of the cemetery by the beginning of the fourth century.20 According to Maria Costantini, it is so far the latest attestation of the term libertus.21 However, since the inscription yields only a very imprecise dating, other possible candidates for the last mention of the term at later dates are at hand.22

freedmen and slaves as commemorators and commemorands in sepulchral inscriptions Freedmen were much more likely than slaves to be able to afford the erection of tombstones to their masters and patrons. A series of funeral inscriptions set up by former slaves to their patrons reflect the social depth of patron–freedmen relations, if not the institutional dynamics of manumission in late antiquity. For example, three epitaphs of late Roman aristocrats belonging to the ordo equester (equestrian order) were dedicated by freedmen to their patrons. Dalmatius, ex protector (former or honorary imperial bodyguard), patron of Volusius and Sabatia liberti, was commemorated by his freedmen in Savaria in Pannonia Prima.23 Another fragmentary funeral inscription unearthed in Sirmium, in Pannonia Secunda, of equestrian proximus (deputy chief) Postumius Leo is dated some time between the reigns of Diocletian and Constantine.24 The erection of Leo’s monument was entrusted by Achilleus, also a holder of a post in the sacra scrinia (imperial bureaux), to his friends Crescentius and Dyscolius, and to his freedman, whose name is fragmentary. Yet another epitaph from Milan possibly commemorates the vir perfectissimus (‘most perfect man’, 20 21 22

23

24

Borg 2013, 96. Costantini 1997, 181; see ICVR II 6152. E.g. CIL III 9623 (Salona), a sixth-century epitaph, in which Alexandria, the wife of freedman Ursus, expressed confidence that for his merits, kindness, and faith her deceased husband would join the blessed. CIL III 4185 (first half of the fourth century). Although no excavated mausoleum has been published from Savaria, this inscription comes from a fourth-century sepulchral building (titulus). Judging from its physical characteristics, it must have been built as part of the funerary structure, perhaps the mausoleum. AE 1998, 1052.

230

Mariana Bodnaruk

an equestrian rank title), Iulianus, who was a master and the most beloved patron of Quintus, perhaps the freedman, who set up the funerary plaque.25 The tomb inscriptions for aristocratic families record predominantly male freedmen as commemorators. Similarly, funerary inscriptions set up by nutrices (family freedwomen or slaves) in Rome were for nurslings of equestrian or senatorial status. A dedication for L. Septimia Pataviniana Balbilla Tyria Nepotilla Odaenathiana, clarissima puella (girl of senatorial rank) is inscribed on a tombstone erected by a wet-nurse, Aurelia Publiana Elpidia, to her ‘sweetest and most loving patroness’.26 Despite the tria nomina (triple name), she was liberta (freedwoman) and not ingenua (free-born).27 The long-term contact between wet-nurse, nursling, and its family, recorded epigraphically, indicates an affectionate relationship that existed between nursling and nutrix (nursemaid), yet sentiments of dependency and patronage would still have been in the background.28 Further, in the late fourth century, an otherwise unknown Maecilia Rogata erected a tomb for her father, Maecilius Hylas, nutritor (male nurse) of the senatorial children Caeionius Camenius and Caeionia Fusciana.29 The rank of the patron is, however, not always indicated.30 Funerary monuments set up for slaves are rare in late antiquity. Occasionally, tombs were erected for slaves and freedmen by their patron or patroness. Euporia, who is described as the most honest and faithful θρεπτή, received a relief sarcophagus commissioned by her πατρῶνα (patroness), Artoria, in Rome.31 Another Greek inscription, from fourth-century Laodicea, might mention a foster child, but the term

25

26

27

28 29

30

31

AE 1982, 408 (second half of the fourth or the end of the fifth century). AE editors suggest ‘libertus’ for the restoration of the last line. CIL VI 1516=ILS 1202 (late third or early fourth century). Niquet 2000, 91 n36 suggests that it was initially placed in semi-public space. Evans 1991, 214, liberta. A nutrix, the most commonly freed female slave and a person who frequently had significant emotional ties to the freeborn members of the family, set up a commemorative monument for her senatorial patrona. For a discussion of this problem, see De Wet 2015, 128–41. CIL VI 21787=ILS 8533=ILCV 96a. As the epitaph comes from the catacomb of Priscilla on the Via Salaria Nova, the daughter of the freedman was perhaps Christian. A marble tablet from the coemeterium ss. Marcellini et Petri preserves the epitaph, in which the formula liberti fecerunt (‘freedmen made (this)’) appears, probably in a dedication to their patroness Sperantia: ICVR VI 16500 (first half of the fourth century). Another undated sarcophagus’ inscription from the same catacomb carries a probable remains of the libertis libertabusque (‘freedmen and freedwomen’) formula: ICVR VI 16747a. ICVR IX 26042. For the sarcophagus, see Dresken-Weiland 2003, 39. The tomb pertaining to the catacomb of Priscilla on the Via Salaria Nova is dated to 290–324.

Late Antique Slavery in Epigraphic Evidence

231

θρεπτή may refer to a household slave rather than an adopted daughter.32 Domestic slaves were sometimes commemorated together with close relatives of the dedicators since they occupied a special place in the familia (household). A Christian epitaph commemorates ancilla Duiona, a female slave of Valens and a wife of Dexter, buried at Manastirine, the suburban cemetery in Salona.33 Despite the familiar formulas of contemporary early Christian funeral inscriptions, such as ancilla Dei or ancilla Christi, there is, however, no reason to doubt the slave status of Valens’ ancilla, commemorated by a male awarder within the household. Other slaves set up a burial for themselves and paid it off some years prior to their death. In principle, the burial of a sarcophagus was not limited to members of the highest strata, but, as the inscriptions show, in rare cases it was also possible for persons who belong to a rich household in various functions, such as the aforementioned Euporia.34 The eunuch Aedesius, who died aged twenty-five, describes himself as neofitus (newly converted Christian) in the inscription of his relief sarcophagus from the catacomb of San Sebastiano in Rome (Figure 11.1).35 As no patron of Aedesius is mentioned, it may be assumed that he himself took care of his funeral and provided the sarcophagus during his lifetime. A Greek funerary inscription from Nicomedia in Bithynia honours the family of Gaius, οἰκονόμος (steward).36 He commemorated himself with his wife and little son by making an expensive and lavish funerary monument. Yet others were commemorated by their relatives, who were also often of an unfree status. Certain filters may have been applied to select which slaves or freedmen were memorialized. Infants are almost entirely

32 33

34 35

36

MAMA I 163. Hübner 2013, 517 with n9; Destephen 2010, 142–3. AE 1892, 32=CIL III 13124=ILS 8252=ILCV 3870 (426 or 430). The epigraphic formula is reminiscent of the Christian titles famulus dei or Christi (servant of God or Christ), or the less common servus Christi (slave of Christ) in the contemporary epitaphs. The amount of the fine for a burial violation of 216 solidi recorded in the inscription, see Marin et al. 2010, 97, no. 91, seems to be a considerable sum for this simple maid, perhaps a domestic slave. However, several epitaphs from Concordia, dating from the first half of the fifth century, attest heavy fines in gold pounds. Moreover, the dedicator is not explicitly named. It could either be her master, Valens, or her husband, Dexter. The length of the inscription, corresponding to its cost, suggests rather the first solution, while the clumsiness of the expression would indicate a more modest social level, which would be that of the husband. Dresken-Weiland 2003, 40. AE 1982, 82=ICVR V 13443 (late fourth century or early fifth century). On the inscribed decorated sarcophagus, featuring the raising of Lazarus, see Dresken-Weiland 2003, 39– 40. TAM 4.276 (third or fourth century).

232

Mariana Bodnaruk

figure 11.1 Funeral inscription on the Christian sarcophagus of Eunuch Aedesius (late-fourth to first quarter of the fifth century ce). Catacombs of San Sebastiano, Rome. Photograph by Archivio Fotografico della Pontificia Commissione di Archeologia Sacra, with permission of PCAS.

invisible in the columbaria inscriptions. One funeral stela from Rome set up for a slave child Clarus, who died aged three, was dedicated by Ianuarius and Sozusa. The name Sozusa, the feminine form of Sozon (from σώζων, servans (saving)), is frequent among slaves and freedwomen. They had to be wealthy enough to afford to commemorate their brother in a sepulchral inscription. Another fourth-century Greek dedication, pertaining to the cemetery area between the Via Appia and Ardeatina, mentions husband and wife, liberti of the same patron, who put their children to rest.37 The imperial freedman (Augustorum libertus) Aurelius Sozon erected a tombstone for a relative in the catacomb of Priscilla.38 From the same catacomb also comes an inscription in which another Augusti libertus appears with his family.39 In the last two cases, the term libertus is

37 38 39

ICVR IV 12513 (fourth century). ICVR IX 25009=ILCV 763a=AE 1999, 174 (286–324 or Severan (AE)). ICVR IX 25069 (third century). Another funeral inscription recorded in the catacomb of Priscilla records an ἀπελευθέρα (freedwoman) as a tomb’s dedicator: ICVR IX 26085 (third century).

Late Antique Slavery in Epigraphic Evidence

233

a part of the formula Aug(usti) lib(ertus), which, until about the middle of the third century, was still a sign of prestige. In a few cases funeral inscriptions include a status-specific title, such as collibertus, fellow freed slave, but this habit is not quite as rare as it is often considered to be.40 A mosaic inscription, pertaining to the Sepulchrum Aureliorum (sepulchre of Aurelii) on the Via Labicana in Rome, was placed by Aurelius Felicissimus for three Aurelii, two of whom are called fratres (brothers) and colliberti, while the woman is designated as virg(inia).41 An early fourth-century fragmentary inscription placed for a certain Crescens by the colliberti and, according to the integration by Antonio Ferrua, by the collibertae, comes from the catacomb of Praetextatus.42 Another inscription comes from the cemetery ad Decimum in Rome, where the colliberti laid to rest a certain Sperantius.43 The term colliberti seems to have been added at a later stage. The epitaph of Pusinna, who was buried in the locus of Severus in Rome, mentions collibertus de tit(ulo) Marci, a sepulchral building from 336 or shortly before.44 Pusinna died aged nine months, that is, too young to be considered a liberta, and therefore the term collibertus, used by the dedicator, does not indicate a relationship of reciprocity with respect to the deceased, but perhaps with respect to the titulum Marci. If there were slaves owned by religious institutions, it is likely that there were also slaves who were freed from such institutions. Severus, named in the inscription, was probably originally a slave of the parish of St Mark, then freed but remained bound, by legal obligations related to his condition, to the titulus where he had been servus. In the absence of comparisons, it is conceivable that he would define himself collibertus to emphasize the spirit of brotherhood that permeated the religious environment in which he was working.45 Coming from a burial over the cemetery of Callixtus, the epitaph is thus associated with early Christian manumission practices.46

40 41

42 43 44 45 46

Solin 2008, 129. ICVR VI 15931 (225–74). A Christian funeral inscription of colliberti, ICVR I 1032a, last recorded in the SS. Quattro Coronati in Rome, is dated by the editor to the fourth century or later; however, it also belongs instead to the third century. ICVR V 14146. ICVR VI 15731 (third or mid-fourth century). ICVR VI 17340. Costantini 1997, 181. Also, in the Catacomb of Callixtus, Petronia Auxentia, clarissima femina (woman of senatorial rank), received a tomb from the slaves who had been emancipated by her: ICVR IV 10085 (perhaps from the fourth century).

234

Mariana Bodnaruk

With regard to the manumissio in ecclesia (emancipation in the church),47 Christian inscriptions of liberti and slaves are not representative of demographical reality, as they are numerically much lower than the bulk of men of servile status and freedmen converted to the new religion. Another epitaph from the cemetery of Priscilla attests to the manumission of seven slaves by the Christians Secundus and Rufina on the occasion of their daughter’s demise.48 This inscription is a further proof of the persistence in Christian circles of the juridical relationship that linked a libertus to his patron in the Roman world.49 Freed slaves were not on an equal footing with the ingenui, being bound by lifelong obedience (obsequium) to their former master, as well as the daily labour services (operae libertorum). The obsequium had its root in the authority that the patron derived from the patria potestas (‘power of a father’), that is, from the original power of life and death that the pater familias (‘father of the household’) had over the members of his family and therefore also over the former slaves. Obsequium is, however, not only a moral attitude that implies respect and devotion towards a patron in recognition of his authority, but it is also inexorably linked to the economic reality of forced labour.

inscribed objects and the materiality of slavery: slave collars Ultimately, slave collars (collares servorum) are part of the epigraphic genre of inscriptions on instrumentum domesticum, domestic or everyday material. Most collared slaves appear to have come from Rome and its environs from Christian slave-owners and senatorial slaveholders.50 Over forty collars and pendants survive from antiquity,51 and they are virtually all from the fourth century. Many include Christian symbols such as crosses or the chi-rho. One reads: ‘I am the slave of Felix the archdeacon. Retain me lest I flee.’52 Even if these collars are not explained by a Christian uneasiness towards the use of facial tattoos, they are still powerful testaments to the enduring risks of flight and the strict preventive 47 48 49 50

51 52

Cod. Iust. 1.13.1. ICVR VIII 23272 (second half of the fourth century). Costantini 1997, 181. The concentration of slave collars in the milieus of the imperial elite at Rome is complemented by the survival of slave collars from Sardinia or Africa. Thurmond 1994. Sotgiu 1973–4.

Late Antique Slavery in Epigraphic Evidence

235

measures against it.53 Two of the collar inscriptions from Rome are commonly interpreted in Italian scholarship as the collars of dogs54 belonging to Q. Clodius Hermogenianus Olybrius, praetorian and city prefect.55 However, Yann Riviere’s study interprets the use of such objects as a strategy to deter theft, as exemplified by the phrase noli me tenere, non tibi experet (‘hold me not, it is not expedient for you’) engraved on both bronze collar pendants, where the powerful Roman senator appears rather as dominus (master) of his slaves.56 The basic slave identification inscription, tene me ne fugia (‘hold me, lest I flee’), testifies to a domestic use of slaves in the urban milieu of not only traditional aristocracy but also new ‘aristocracy of office’: ‘Retain me and return me to Apronianus palatinus at the Mappa Aurea on the Aventine because I have fled.’57 This inscribed collar was, however, found in Tusculum in Latium around the neck of a skeleton, presumably a slave of the western court official (palatinus) Apronianus in the late fourth century. Another slave collar with a bronze disc found in Rome announces the reward for returning the runaway slave to his or her master, Zoninus (Figure 11.2).58 Dated to the period between the fourth and the end of the fifth century, it reads: ‘I have run away! Keep hold of me! When you bring me to my master Zoninus, you receive a solidus.’59 The solidus (gold coin), 1/72 of a pound of pure gold, would be a reward too high for the return of a dog by any calculation. Far from being dematerialized texts, inscribed collars embody the lived experience of subjugation of urban slaves.

categories and the use of late roman slaves I now turn to examining the economy of slavery, that is, the enmeshment of slaves as labour force in the late Roman state. Slaves were employed in a wide range of workplaces that can be roughly divided into four categories: household or domestic, public or municipal, urban crafts and services, and agriculture. In turn, inscriptions equally provide information on a range of

53 54 55 56 57 58 59

Harper 2011, 258. LTUR III, 76, s.v. horti Q. Clodi Hermogeniani Olybri (E. Papi). Orlandi 2011, 428. CIL XV 7199; 7199a. Riviere 2002, 162. CIL XV 7182. CIL XV 7194. Trimble 2016.

236

Mariana Bodnaruk

figure 11.2 Zoninus’ slave collar (fourth to fifth century ce). From Rome. Photograph by Museo Nazionale Romano – Terme di Diocleziano, with permission of Ministero per i beni e le attività culturali e per il turismo – Museo Nazionale Romano.

occupations of late Roman slave-owners, including administration, military, and clergy.60 In the later Roman Empire, domestic work was still mainly performed by slaves (servi domestici). Female slave labour was a crucial part of the economy of slavery in the household, even if they were producing services rather than a marketable product. The exceedingly complex relationship between master and slave is exemplified in the vocabulary of Roman slavery. Verna, alumnus, delicatus designate privileged links between the servus and the dominus, outlining the whole lexicon of emotional situations. The selection of vocabulary in late Roman epigraphy, including servi, famuli, ancillae, mancipia, familiarii, and pueri, confirms that their number was considerable. However, late antique alumni, delicati, and pueri were not always slaves.61 60 61

Hillner 2001. Also, θρεπτός/θρεπτή (alumnus/alumna) may mean the same as Latin expositus/exposita, an abandoned child.

Late Antique Slavery in Epigraphic Evidence

237

Vernae, slaves born from slave mothers in the master’s household, were often the outcome of male owners’ sexual use of their domestic slavewomen, although the father could also have been another slave. However, vernae could be held in great affection by various members of the household.62 A fragmentary funeral inscription from Aquileia dated to the fourth century commemorates a dulcissima (‘the sweetest’) verna, whose name is not preserved on the stone.63 Other forms of familial relationship were also possible, such as those among slave families: the stela for verna Clarus was commissioned by his brother Ianuarius and sister-in-law Sozusa.64 The sepulchral monument erected by this slave married couple displays a portrait of a nude boy, who holds a circular object in his left hand, while touching a large rooster with his right hand. The status of alumnus, an abandoned child taken into a household, was also that of a slave and such a person often bore a simple servile name. Extended informal ties between nutritor and alumnus are attested epigraphically.65 For example, a funeral inscription from Rome on a sarcophagus with a representation of a man in the toga contabulata records alumnus Cointus (Κοΐντος).66 In the epitaph, he appears as a recipient of the sarcophagus acquired for him by libertus Flavianus Augustus when both were still alive. Another Christian honorific (or possibly sepulchral monument) from Rome documents the alumnus Marcus Servilius Servilianus acting as awarder of a memorial to a Roman senator (Figure 11.3).67 Alumni continued to appear in the inscriptions in the period from the second quarter of the fourth to the first quarter of the sixth century.68 Moreover, the natural child of one free or freed parent and one slave is equally termed alumnus. It can, however, be difficult to distinguish in epigraphy. Grave inscriptions set up not by immediate family members or spouses are rare, yet dedications by alumni show comparatively close personal connections in the familia. A fragmentary funeral inscription from 62 63

64 65

66 67 68

Mander 2013, 124. CIL V 8601. Another inscribed fragment from Aquilea recording a verna is undated: InscrAqu-2, 1452. Zaccaria 2016, 204, table 8, nos 1–2. AE 1984, 127. Boulvert 1974, 326 n330. For a discussion of alumni in inscriptions, see SigismundNielsen 1987; Bellemore and Rawson 1990; Smodlaka Kotur 1994; Rawson 2003. CIL VI 17956. For the sarcophagus, see Dresken-Weiland 2003, 253. CIL VI 31810=41322. Brancato 2015, nos. 145 (518), 241 (after 415), 833 (340), 954 (501–600), 970 (351–99), 978 (390–425), 980 (350–450), 985 (326–75), 997 (326–75), 999 (326–75), 1001 (400– 499), 1086 (315–50).

238

Mariana Bodnaruk

figure 11.3 Inscription on the Christian monument recording its awarder, alumnus Marcus Servilius Servilianus (312–37 ce). From Rome. Photograph by the author, © Roma, Sovrintendenza Capitolina ai Beni Culturali.

Aquileia was dedicated by both filii and alumni to their pater familias, whose name is not preserved.69 Here alumni, who appear as cocommemorators of their dominus, are undoubtedly servi, as they are unambiguously differentiated from the biological sons of the deceased. However, a foster child would also be called alumnus. An early fourthcentury epitaph from Lycaonia, which was engraved on the tomb of the θρεπτός Tyrannos, reads: ‘Aurelius Gourdos, a presbyter, erected [the tomb of] Tyrannos his foundling son in remembrance.’70 The omission of the praenomen (personal name) on the tombstone is no proof of a slave status, although it contrasts with the duo nomina (double name) of the dedicator. In Latin tradition, the delicia/deliciae/delicati were slaves of freedmen and free Romans,71 generally of young age, considered as favourite objects of their masters and often elected to keep them company throughout the day. Yet, delicati were also, more generally, small children of any possible legal status. In the late antique inscriptions, slaves feature as part of the emotional life of the family to which they belong. A funeral inscription from Onaeum in Dalmatia was set up by Valeria Severina and Messor, slave of Firmio, for their innocentissimus (‘the most innocent’) delicatus Victor, who died aged nine.72 Delicati commemorated in 69 70 71 72

CIL V 1741. Sterrett 1888, 197. Laes 2003. CIL III 1905.

Late Antique Slavery in Epigraphic Evidence

239

epitaphs are not only young; they are also commemorated with pronounced emotional responses of grief and mourning. A sepulchral monument was installed in Rome by Turrania Polybia for Turrania Prepusa, who lived only seven years and three months: ‘for delicata, the sweetest soul suddenly stolen by the injustice of fate, so that she could not savor the benefits of her foster mother intended for her’.73 A third-century funeral stela from Fulginiae or Hispellum was erected by soldier (evocatus) Suetus Paullinus for delicatus Apolaustus, son of Caius Lamavus.74 Another funeral inscription was set up by M. Allius Firminus to his son Ursinus, delicatus of C. Septimius Carpophorus, who, aged eleven, drowned in the sea following a shipwreck.75 A neutral term, famulus, denoted a slave member of the household. A verse epitaph of the influential senator, urban prefect Junius Bassus, who died while in office in Rome in 359, mentions such family slaves:76 While governing the people of his city and the house of the senate, he died, to the everlasting tears of the city. Nor were his own servants (famulis) allowed to carry his bier, but it was the burden of the Roman people, vying [for the honour]. . ..77

The funerary inscription claims that family retainers were ‘not allowed’ to carry the coffin of their master (dominus), and that ‘the people’ vied to bear the burden instead. At first sight it seems to be ‘no more than a bit of flattery, implying spontaneous enthusiasm on the part of a grieving public to honor a popular figure’78 as the aristocratic funerals were traditionally family affairs. However, Alan Cameron concludes that Bassus was awarded the honour of a public funeral, and that is why members of the elite at large rather than family retainers carried the coffin. The epitaph is influenced by funerary and consolatory poetry, but with the mention of famuli the social context rather than literary genre was the prime consideration. Eunuchs were also slaves in private ownership. However, they were equally imperial slaves attached to the emperor’s household, the familia Caesaris. Thus, Aedesius, who made a brief confession of faith in his 73 74 75 76

77 78

CIL VI 27827. Xenia, 8 (1984): 42, no. 7 (Ambrogi). CIL III 1899=ILS 8516. CIL VI 32004=41341a-b=ILCV 90. Multiple identification tags make it clear that archaeologically invisible slaves were ubiquitous in the aristocratic domus (household), with its hierarchy of human personnel, but had no assigned spaces, often sleeping outside in the portico of the peristyle. Trans. Cameron 2002, 290. Ibid.

240

Mariana Bodnaruk

funeral inscription (see Figure 11.1), was either employed in a rich private household79 or in the imperial household as eunuch chamberlain. By the late fourth century court eunuchs could rise through palatine service and become senators themselves. Two praepositi sacri cubiculi (superintendents of the sacred bedchamber) Antiochus and Parthenius are styled viri clarissimi (‘most distinguished men’, a senatorial rank title) on the only surviving inscription mentioning this office engraved on a bronze tablet from Rome.80 Palatine eunuchs were almost always initially imported ‘barbarian’ slaves, usually from outside the Empire.81 Slaves as weapon-bearers are continually found accompanying their masters among the ranks of the soldiers, and in the case of an emergency they could be drafted into the late Roman army. The funeral inscription of an eighteen-year-old Taurinus was found in Fano on the Adriatic coast near the Via Flaminia: ‘Here lies puer Taurinus, eighteen years old, son of Aurora, of the Invicti seniores.’82 Dietrich Hoffmann rightly indicates that puer here rather has a meaning of ‘slave’, listing contemporary inscriptions of similar usage.83 This is confirmed by the fact that, of the deceased’s parents, only the mother is mentioned on the inscription, whose simple name Aurora identifies her as a member of the lower strata and perhaps as an unfree foreigner.

79 80

81

82

83

Dresken-Weiland 2003, 222 n116. CIL VI 31946=XV 7131 (late fourth or early fifth century). Hopkins 1963, 64 wrongly dates the inscription (perhaps tabula immunitatis (tablet with payment immunity act)) to the first half of the fourth century. Antiochus may be identical with chamberlain Antiochus (in office between 414 and 419/20). The inclusion into the ordo senatorius (senatorial order) was of a great importance for praepositus in the first place in social terms. Nevertheless, Claudian still describes Eutropius as a slave and a eunuch who rose from an ignominious station; see Claud. In Eutr. 1.181–4. Hunt 1996, 569 states that the imperial eunuchs at the late Roman court were ‘usually freed slaves from Armenia or Persia’. CIL XI 6289=ILCV 531 (late fourth or early fifth century); Hoffmann 1970, 376. While Hoffmann totally ignores the excessive Christian decoration of the monument, Binazzi 1989, 194 no. 125 suggests that puer has perhaps, in the present case, the meaning of ‘recently baptized’, in consideration of the age of the deceased. Although Taurinus has only one onomastic element, in the fourth century onomastics was definitely moving towards an extreme simplicity that led to the use of a simplex nomen (simple name). However, puer in the sense of boy – seldom found on grave inscriptions – usually applies to smaller children, and puer, in place of filius, is an unusual formula. Hoffmann 1970, 377 with n572. ILCV 289, 637, 693, 769, 1358, 2968, 3077, 4607, 4721. However, not all of these inscriptions give an age indication, notably, the famous ‘Turpillian stone’ (ILCV 3077). ILCV 1751, referred to by Hoffmann, does not record any puer.

Late Antique Slavery in Epigraphic Evidence

241

However, as Hoffmann himself points out, the problem with this interpretation is that slaves were legally excluded from military service.84 So, Taurinus would not have been a real soldier, but only a slave and a man-at-arms of a soldier or an officer. The decree of 406 by Honorius explicitly mentions such weapon-bearers, who mostly come from the enslaved and who accompany their masters in war, when, in view of the emergency situation at that time, slaves were also called upon to join the troops as recruits with assurances of their future release.85 However, it is not possible to apply the military service of Taurinus to the edict itself, even if it is assumed that it was in force for a long time. For then, as is expressly stated in the decree, he would have been given the freedom to serve as a regular soldier and no longer as a servant, and would therefore no longer be called a puer on the tomb’s text either. If one finally asks why, as it is customary, the master did not set up the tombstone of his deceased puer himself, it could be explained by the fact that he had also fallen or otherwise perished, and so perhaps – as Taurinus’ filiation is indicated through the maternal line – his mother Aurora, who could have accompanied the army unit as a sutler, arranged the burial herself.86 Municipal and state slaves (servi publici) were the slaves of the Roman people. The public urban servitude, as an institution, is well attested at the beginning of the fourth century and continues to appear with regular frequency throughout the century, though the range of activities gets increasingly restricted. Originally a public slave, the οἰκονόμος Gaius was set free by the philanthropy of the citizens.87 Slaves who filled the post of city steward might be sufficiently affluent. Gaius, former δούλος δημοσίους (public slave), as the inscription states, went into the service of a private man named Tryphon after his release as an οἰκονόμος. Οἰκονόμοι τῆς πόλεως (treasurers of the city) (usually termed arcarii in the West) dealing with public finance are attested in numerous inscriptions from the high Empire, although public finance, like public record-keeping, was never the exclusive preserve of slaves.88 Public slaves continued to be acquired and employed in both eastern and western cities deep into the fifth century. Yet there is no securely datable inscription later than 250 and the number roughly datable to the

84 85 86 87 88

C. Th. 7.13.8 (from 380). C. Th. 7.13.16. Hoffmann 1970, 377. TAM 4.276. Weiss 2004, 59–69, cat. no. L86.

242

Mariana Bodnaruk

fourth century on epigraphic grounds is meagre, with none from the later centuries. In the fourth century their activities were still quite diverse, including the upkeep of aqueducts. Provincial cities continued to manage their public water systems with servi publici in late antiquity. Circitores were public slaves, a specialized crew of the civic water administration, belonging to urban cura (care), to oversee and prevent damage to the water system. The inscription from Tivoli lists nineteen circitores, three decani (‘chiefs of ten’), and their twenty-eight children.89 The number proves that Rome’s aqueducts may not have been suffering from any lack of servi publici when this inscription was carved. According to Noel Lenski, the aquarii (public slaves who attended to the aqueducts) were among the few servi publici whose ranks were maintained well into late antiquity.90 Some public slaves were kept as skilled artisans. The funeral monument from Rome dated to the late third- or early fourth-century records the officium (employment) of the freedman or slave Crescens, who was vermiculator.91 The vermiculator was perhaps an officialis (servant), a floor-layer, who specialized in the polychrome mosaic-work for pavements (opus vermiculatum). The term officialis, however, well attested for slaves in the early fourth century, shifts its meaning almost entirely by the century’s end, when it came to designate a member of the officium (office) of the imperial bureaucrat. The most suggestive piece of evidence about the status of prostitutes in late antiquity is a slave collar from North Africa: an inscribed lead band found during the excavations of the Temple of Apollo at Bulla Regia, found still on the skeleton of a prostitute with the familiar formula ‘retain me if I flee’. Not only is this fourth-century relic a testimony that the life of prostitution was carried out ‘within the entire apparatus of the master’s power’,92 as prostitutes usually did not have an individual owner (yet were rather in public ownership), but it also offers physical evidence of her slave status. Furthermore, the collar powerfully embodies practices of enslavement that shaped the material culture. MacMullen opened debate over the extent to which slavery was a rural or an urban phenomenon in the period, and the importance of the economic role(s) played by slaves.93 He endorsed the proposition that slaves

89 90 91 92 93

CIL XIV 3649. Lenski 2006, 348. AE 1988, 48; see Buonocore 1983, with drawing. Harper 2011, 310. MacMullen 1987.

Late Antique Slavery in Epigraphic Evidence

243

had become largely a parasitic element in the economic activity of the period, using the epigraphic sources to demonstrate that slaves were largely absent from rural contexts, and present only in very limited circumstances in domestic and urban settings. MacMullen’s theory triggered a strong reaction from Ross Samson, who dismissed the epigraphical evidence as unsatisfactory to quantify the balance of urban and rural slave labour force, and argued in favour of a continuing slave presence in Roman villas in rural contexts.94 Cam Grey cautioned not to overemphasize the difference between rural and urban slavery in late antiquity, as countryside slaves did not need to be replaced as a workforce, for their exploitation had never been the dominant mode of production.95 Subsequently, Harper used new epigraphic evidence unavailable to MacMullen and Samson as proof of the exploited labour of slaves that was fundamental to agricultural production on estates in the fourth century both in the East and in the West.

other types of compulsory labour: freedmen, coloni, and adscripticii Late antiquity yields less epigraphic evidence than a substantial series of manumission inscriptions from southern Macedonia, mostly of the third century, confirming the importance of household slavery among the upper tier of village families. However, the inscription from Thera pictures a rural slave population affected by adult male manumission. The manumission of adult male slaves is presumed on the grounds of the age structure. It has led Harper to believe that manumission was an incentive used to motivate slaves and to create internal hierarchies on the estate.96 Manumission inscriptions themselves were commemorative and not legal documents, even though they record the act of a slave-owner freeing slaves. An undoubtedly Christian inscription from Moesia Prima commemorates manumission performed on account of his good will (pro bona voluptate) by Ingenuus, vir devotus (‘devout man’), perhaps palatinus, of his slave Toma.97

94 95 96 97

Samson 1989; 1992. Grey 2011b, 507. Harper 2011, 74–6. IMS 4.177=IMS 4.118 (fifth or sixth century). The text of the inscription was scrawled in the brick while still wet. However, unlike graffiti, it bears more official character, despite the material used.

244

Mariana Bodnaruk

The epigraphic evidence of slavery in the western provinces of the Empire is scarce, perhaps due to regional differences in the epigraphic habit. However, a corpus of epigraphic references to coloni reveals them to be even less commonly recorded in stone than slaves.98 The casarius (‘cottager’) Santulus Flavius Aurelius, colonus or a servant that belonged to an estate, died and was buried in Rome in 347.99 His employers are mentioned as such, but it is uncertain if they were responsible for the commission of his undecorated sarcophagus. Another funeral inscription recording a casarius as commemorator is a Christian epitaph of uncertain provenance in Rome.100 Criminal or undesirable behaviour on the part of free men could result in subjection to the colonatus perpetuus (‘perpetual colonate’) as a judicial sentence,101 similarly to enslavement for crimes, that is, becoming slaves of the state (servi poenae) as a result of a legal penalty. Diether Einbach believes that adscripticius as a term denoting a dependent colonus was not in use before the end of the fourth century.102 Although being dependent on the landowner, as technically free coloni, adscripticii were not slaves. The term adscripticius is documented for the first time in the law of 224,103 which concerns the children of ancillae or adscripticiae, but it is considered to be a later interpolation. The same explanation is given for the phrase adscripticia conditio in the law dated a century later.104 The stone from the borderlands of Pisidia and Kibyratis contains the partially preserved rescriptum Iustini et Iustiniani de possessionibus oratorii S. Iohannis apostoli (Rescript concerning the possession of the oratory of St John).105 The bilingual inscription confirms that the Latin term adscripticius corresponds to the Greek ἐναπόγραφος. The imperial letter places the estates owned by an oratory and the peasants cultivating them under the care of the ruling emperors. The Latin version reads: [Since it is proper that our serfs be kept immune, it is absolutely necessary that the es]tates [belonging] to the rever[end oratory of Sa]int John the Apostle [should benefit from] the same provision. [For this reason] we state, by (the power) of the

98 99 100 101 102 103 104 105

Samson 1992, 219. CIL VI 9237=ILCV 589=ICVR IV 10851. CIL VI 9238=ILCV 588=ICVR I 3467. LTUR II 361 (Papi), s.v. a Furca. C. Th. 14.18.1 (from 382). Einbach 1977, 142, 204. Cod. Iust. 8.51.1. Cod. Iust. 3.38.11. CIL III 13640=ILCV 23=AE 1894, 68.

Late Antique Slavery in Epigraphic Evidence

245

present divine decision that, [the estates] mentioned in complaints and their peasants (coloni) and ascribed peasants (adscriptici), and overseers (curatores) and free tenants (conductores) should be [free from ev]ery burden, from both [the passage of sol]diers or mili[tia (violentiae prohibitores)] and from those soldiers who are kno[wn] to be permanently garrisoned [next to] these estates. May no one of them dare to inflict [any] damage upon [them] on any occasion, if the [complai] nts are true. . ..106

reproducing intersectional inequalities in the later roman empire Overall, the perspective of intersectionality, a theoretical framework that can account for relations of domination organized around gender, race/ ethnicity, and class, conceptualizes the material foundations of social relations as an integrated and unified process. The analysis of the relations of production (narrowly construed) as the dominant force shaping social relations had seldom addressed gender-based forms of oppression and domination. To recognize it means to emphasize the interdependence between relations of production and reproduction, or to capture the role of gender in forms and structures of oppression that shape social matrices in terms of material conditions. The maximum prices in the slave chapter of Diocletian’s edict show the value of (reproductive) labour of an adolescent female slave equivalent to her male counterpart, while otherwise it differs significantly with the prices tilted towards men. Furthermore, the sex ratio of the Thera inscription leans more towards females among slaves over the age of thirty. This could imply that male slaves experienced higher mortality. However, if manumission is the underlying cause of this ratio, it shows that female slaves were rarely manumitted when they were physically capable of reproduction. The fertility of slaves remained valued, since children born to slave mothers within the Empire were one of the two most important sources of slaves. Hardly any statistics are available on the late antique Roman slave supply. However, the second of the two most important sources of slaves in late antiquity was surely trade. Epigraphic source material underrepresents slave-import across the frontiers and enslavement in frontier wars, 106

CSLA, E00869 (trans. Nowakowski). The rescript is thus aimed at protecting the estates and various categories of peasants cultivating them that were owned by an oratory (εὐκτήριον), meaning probably a local shrine. The different kinds of peasants mentioned in the text were troubled by both transient and locally garrisoned soldiers and militia, who were apparently looting villages or violently exacting due supplies.

246

Mariana Bodnaruk

but it yields some important glimpses.107 Different tribes of the semidesert between the Mediterranean and the Sahara were a giant reservoir of human bodies, trafficked by the Roman slave trade. There is visual evidence for black slaves in Roman antiquity, including the fourth-century mosaic of the Maison d’Isguntus at Hippo Regius in Africa Proconsularis as well as mosaics at Piazza Armerina in Sicily. In turn, the ideology with which Roman foreign conquests were interpreted opposes servitus (servitude) and libertas (freedom), which the Romans lose when, as happens in the wars on the Empire’s borders, they become captivi, prisoners, and, in short, slaves of the barbarians. Religious attestations, primarily in funeral practices, of the Christian ‘epitaphic culture’ bring to light a burgeoning Christian slave population, particularly in Rome, conscious of its social status and religious affiliation. In the inscription on his explicitly Christian sarcophagus from Rome (see Figure 11.1), Aedesius indicates his sexual status (eunuchus) in combination with his origin from Armenia (natione Armenius), which taken together would imply his initial status as a slave in the Empire. In his indication of a sexual status, a non-Roman origin, and a new adherence to the Christian religion (neofitus), he must have wanted to highlight his social advancement. It was commonly among the courtiers that eunuchs played their role. Usually, although not exclusively, inner-court domestics underwent a triple discrimination as slaves of barbarian origin,108 eunuchs,109 and (allegedly) persons of same-sex sexuality.110 Nonetheless, court eunuchs in the emperor’s service benefited from the power generated by their lofty position in the palatine hierarchy and their participation in imperial ceremonies. Late antique monuments are already intimately connected to the realities of their social and economic environment. Within the inscribed traces of late Roman slavery, embedded in the material cultural remains, shaped by the same lived experiences,111 different identities are recognized. A theoretical framework that intends to account adequately for relations of domination and exploitation organized around gender, race, and class (that is, processes for identities) has to recognize the simultaneity of 107 108

109 110

111

Scheidel 2011. Scholten 1995, 33, estimates that for the fourth and fifth centuries the great majority of chamberlains were drawn from Armenia or Persia. Claud. In Eutr. 1.171 and 2.22, semivir (half-man). Sideris 2002, 161–2. On the same-sex sexuality of eunuchs, see Claud. In Eutr. 1.65–77; On other prejudices against eunuchs in the fourth century, see Guyot 1980, 164–76, and Sideris 2002, 161–2. See George 2013.

Late Antique Slavery in Epigraphic Evidence

247

figure 11.4 Slave victimarii in the suovetaurilia relief of the Decennalia base (303 ce). Forum Romanum, Rome. Photograph by Procopius. Wikimedia Commons, https://bit.ly/2SpmuzC, with CC BY-SA 3.0 licence: https://creative commons.org/licenses/by-sa/3.0.

oppressions. However, my contention is that slaves’ experiences of gender and race in the late Roman state were conditioned by a class-relationship with class as a fundamental axis of oppression.112 To delve into the specificities of structural class relations is vitally necessary to an analysis that explains experiences of enslavement. Only a wholesale analysis of a structure of material relations of production and reproduction, accumulation and dispossession, which affects the multifaceted realms of culture 112

On the use of the concept of ‘class’ in ancient history, see Ste Croix 1998 [1983]. In the scene of the sacrificial procession from the only surviving column base of the tetrarchic five-column monument, erected in Rome in 303, with two naked slaves carrying axes, the visual connotation of class is underscored. The social status distinction of the victimarii (sacrifice attendants responsible for slaughtering animals) is emphasized visually by their nudity and their attributes (Figure 11.4), which refer to their job of cutting up the sacrificial carcasses, in contrast a togate official who supervised the traditional suovetaurilia (sacrifice of a pig, a sheep, and a bull). While nudity on sacrificial reliefs was a pictorial convention defining the social class, it is unknown if such sacrificial attendants were in fact always naked; see Elsner 2018, 87–8, with fig. 60.

248

Mariana Bodnaruk

and ideology, can explain the multiple structures of domination that pattern the late Roman world. In order to do justice to the complexities of oppression in the later Roman Empire, analyses of ‘intersecting oppressions’ effectively render visible the notorious triad of patriarchy, imperialism, and slavery, which were historically solidified in and through one another.

part iv SOCIAL AND RELIGIOUS HISTORIES OF SLAVERY ON THE BORDERS OF THE EMPIRE AND BEYOND

12 Slavery among the Visigoths Noel Lenski

The question of Visigothic slavery has generally been treated in the context of larger studies on the Visigoths1 or of the Mediterranean economy rather than studies devoted to slavery. Even when it has been examined as a thing unto itself, it has tended to get folded into larger narratives of the development of social relations in the early Middle Ages.2 Marxist historians, such as Abilio Barbero, Marcelo Vigil, and Chris Wickham, have attempted to fit the evidence into a historical materialist narrative that posits a decline of slavery and slaveholding in the post-Roman world as states contracted and serfs or peasants replaced slaves as the primary agricultural producers.3 This narrative has been countered by French social historians, especially Georges Duby and Pierre Bonnassie, who argued that slavery continued to predominate in the early medieval West, giving way to serfdom only in the tenth century.4 More recently, American historians Michael McCormick and Kyle Harper have viewed the classical, Late Roman, and Carolingian economic flourishes as heavily dependent on slavery while positing a marked drop in economic performance and the practice of slavery in the fifth through eighth centuries.5 This study shows that Visigothic slavery fits none of these models well. The Visigoths offer clear evidence for a robust slaveholding culture in the sixth and seventh centuries that corresponded with a period of relatively weak 1 2

3 4 5

Thompson 1969, 267–74; King 1972, 159–83. Verlinden 1955, 58–102; Nehlsen 1972, 153–250; Bondue 2011, 128–99; Rio 2017, 62–6, 89–92, 145–52; cf. Lenski 2008. Barbero and Vigil 1978; Wickham 2005; cf. Bloch 1947; 1975. Duby 1974; Bonnassie 1991. McCormick 2001; Harper 2011.

251

252

Noel Lenski

central state authority but relatively strong overall economic performance.6 Assembled from an admixture of Roman and Germanic customs and norms, Visigothic society hosted a unique but highly intensified slaveholding culture worthy of study unto itself.

visigothic history – in brief It is worth examining in brief the history of the Visigothic kingdom and the sources left to us to reconstruct it.7 The Visigoths were the descendants of the multi-ethnic group that migrated into the Roman Empire in the fourth century, defeated the emperor Valens at Adrianople in 378, went on to sack Rome in 410, and were settled with Roman permission in southwestern Gaul (Aquitaine) in 418. The group expanded across southern Gaul south and eastward to the Mediterranean coast by the late fifth century such that they controlled most of what had been the late Roman diocese of Viennensis (in modern France, south of the Loire), as well as a significant sector in northern Spain, from their capital in Toulouse. After their king Alaric II was defeated by the Franks and killed at the Battle of Vouillé in 507, the Goths were forced to abandon Aquitaine and, with the help of their allies the Ostrogoths, to re-establish themselves in Spain, where they came to rule from a new capital in Toledo. By the mid sixth century they reached a peak of power under Leovigild (r. 568–586 or 587) and his son Reccared (r. 587–601), who engineered the reconstruction of infrastructure and urban renewal. This prosperity was facilitated by the conquest of northern and western Iberia from the Suebi, who were fully suppressed by 585, and the recapture of southeastern Iberia from Byzantine forces that had occupied Cartagena and its territory from the 530s until 625. This continual stream of campaigns resulted in regular hauls of captives that served in part to supply the Visigoths with slaves. The Visigoths maintained a scaled-down version of Roman administrative and fiscal apparatus. They eliminated the upper tiers of the Roman administrative hierarchy but maintained Judges (Iudices), Dukes (Duces), and Counts (Comites), who presided over territorial units the size of cities, regions, and provinces. These persons administered justice and assisted in tax collection, which was ultimately conducted through a servile bureaucracy (servi fiscales). The economy was monetized, as attested by literary 6

7

For debates about the relative strength of the Visigothic state, see Díaz and Valverde Castro 2000 and Fernández 2017a; 2017b. For histories of the Visigoths, see Thompson 1969; Collins 2004.

Slavery among the Visigoths

253

and documentary sources and above all by abundant numismatic evidence.8 Throughout the Visigothic period, local leadership – on the part of landholding lords and bishops – remained an ever-present counterweight to central, royal authority. This was compounded by the fact that Visigothic kings were frequently overthrown and thus had trouble establishing enduring dynasties or royal traditions that might have secured their power against local challengers. Nevertheless, royal authority pushed hard to counter these centrifugal forces and managed to maintain measurable unity for the 204 years that the Goths ruled Iberia. It also worked gradually to erase distinctions between Goths and Romans that traced back to the early fifth century. This process culminated in the conversion of King Reccared from Arian to Nicene Christianity in 587, a move that helped unify the kingdom and its normative structures. In the final decades of the kingdom, civil war plagued the monarchy and weakened the forces of unity. The problems were compounded by demographic decline in the wake of ongoing outbreaks of the plague. They were also worsened by efforts to convert the region’s sizable Jewish population forcibly, conducted with especial vigour under Kings Erwig (680–7) and Egica (687–702). These centrifugal trends left the kingdom vulnerable to invasion by Muslim forces from al-Ifriqiya, which arrived in 711 and secured control of the region in just two short years. Although most of Iberia south of the Pyrenees was now ruled under an emirate based in Cordoba, the population remained largely Latinspeaking and continued to follow the basic legal framework established by the Visigoths down to the Reconquista.

sources for visigothic slavery The maintenance of a Latin-speaking populace that largely followed Visigothic social praxis through the Arab centuries was accomplished in no small part thanks to a vigorous tradition of lawgiving that has left us with a sizable source pool for the kingdom. Visigothic law developed over time, and we are fortunate to have significant traces of this process in four normative codes and a modest number of documents.9 The first, the Code of Euric (CE) from c.475, is preserved in a highly fragmentary state, but gives us a relatively clear sense of a code designed to govern the Gothic population of the kingdom following legal principles recombined from 8 9

Pliego Vázquez 2009. On the sources, see Nehlsen 1972, 153–60; Liebs 2002, 157–76, 196–9.

254

Noel Lenski

Roman and traditional ‘Germanic’ practices. Euric’s code operated alongside Roman laws, which were synthesized for the kingdom under the Lex Romana Visigothorum (LRV), published in 506 by his successor Alaric II. This curious source culled laws from the Theodosian Code and other Roman legal texts and reorganized them into a handbook valid for the Roman subjects of the Goths – the majority of the kingdom’s inhabitants. The LRV not only assembled the original laws but also furnished most with an interpretatio – an explanation of the manner in which they were to be applied in a Visigothic context. The third normative source, the Leges Visigothorum (LV; originally known as the Liber Iudiciorum), survives in two recensions, the first completed in 654 under Recceswinth (r. 653–72) and the second in 681 under Erwig (r. 680–7). Both assemble laws dating as far back in the Visigothic tradition as Euric, but update these to reflect a new social order in which the distinction between Goth and Roman had faded to a near nullity. The fourth significant legal source is the series of Formulae Visigothicae (FV), templates for common use in the recording of legal transactions. Forty-five formulae survive in a collection of the early seventh century.10 Finally, we possess the acts of some thirty-six dated church councils held in the Visigothic period, some of which also have relevance to the practice of slavery. These normative sources are supplemented by a few documents, above all a small clutch of wills and over 160 texts inscribed onto slates (pizarras). These can be combined with literary texts, including hagiographies and the sizable corpus of the seventhcentury polymath Isidore of Seville, as well as a handful of remaining literary and ecclesiastical authors. Archaeology also provides abundant material, but little of direct use for the study of slavery. All told, the haul of evidence is thus more than adequate to create a detailed picture of Visigothic slavery. Even a superficial glance at the evidence demonstrates how important slavery was in the overall construction of society. This can be seen most clearly in the three codes most likely to address the problems of slavery. The palimpsest of the CE, fragmentary though it is, preserves forty-four laws with enough content to judge their meaning, of which fourteen (32 per cent) involve slaves; the fragments of the same text reconstructed from the Lex Baiuvariorum contains fifteen laws, of which four (27 per cent) involve slaves. The FV, which offer a good impression of the everyday priorities of Visigothic users of law, consist of forty-five 10

For an English translation of all the formulae that touch on slavery see https://bit.ly/3lK UJhj/.

Slavery among the Visigoths

255

separate texts, of which sixteen (36 per cent) deal in some way with dependents: fourteen with slaves (servi/mancipia) and two with coloni. The LV includes some fifty-four titles with 539 chapters, of which forty-five titles and 215 chapters involve slaves (83 per cent / 40 per cent).11 These are extremely high numbers and percentages, especially when compared with contemporary Roman codes. They cannot, of course, be taken as indicators of the percentages of slaves in the general populace; yet they certainly do indicate, at least in an approximate way, the scale of legal business involving unfree persons and the relative importance of slaves in the political economy.

visigothic slaves as chattel property Some have argued that the servi, ancillae, and mancipia of these codes and other sources were not slaves in any strict sense of the word. Hermann Nehlsen had refuted this position long ago, but it is worth reconsidering the evidence in brief.12 If we take article 1(1) of the League of Nations 1926 Slavery Convention13 as a working definition, ‘Slavery is the status or condition of a person over whom any or all the powers attaching to the right of ownership are exercised’, there can be no room for doubt in the Visigothic instance. There is abundant evidence from all the codes that the servi, ancillae, and mancipia in question were saleable as chattels, as were their offspring, and their sale could occur independently of the land that they might have farmed. This is perhaps clearest in the formulaic slave sale contract of FV 11, which reads very much like a Roman slave sale contract and is drawn up for sales effected in gold solidi.14 But the saleability of Visigothic slaves is also regularly attested in the LV, which regulates price formation, the rescission of slave sales, self-purchase, sale to foreigners, etc.15 For that matter, the LV offers ample attestations to the giving of slaves as gifts, the temporary lending of slaves, the joint ownership of slaves, and one of the formulae indicates they could even be mortgaged.16 11

12 13 14 15

16

García Moreno 2001, 206, arrives at somewhat different numbers but an overall percentage of forty-six. Nehlsen 1972, passim. See https://bit.ly/3gIepio. FV 11. See also the failed sale described at V Patrum Emeretensium 4.3.1–12. LV 5.4.7 (cf. CE 294); 5.4.8 (cf. CE 289); 5.4.11 (cf. CE 290); 5.4.13; 5.4.14 (cf. CE 288); 5.4.15 (cf. CE 291); 5.4.16 (cf. CE 292); 5.4.17–18; 5.7.16; 7.6.1; 9.1.21; 12.2.13–14, 18; 12.3.1. See also Conc. Toletanum IV (a. 633) can. 43. Gifts: LV 4.5.3; 5.4.14, 18; 5.7.2, 16; Conc. Toletanum III (a. 589) can. 8; mortgage: FV 44.

256

Noel Lenski

To be sure, slaves are at times listed as a component of landed property: ‘a field, vineyard or slave’ is a common trilogy associated with landed wealth,17 but there is never any question that individual slaves themselves could be separated from the estates on which they lived and sold as freestanding chattels. This is especially clear in a law that permits the sale of slaves outside the Visigothic kingdom, but forbids a master from selling the same slave abroad more than once before he loses the right to reclaim ownership if a slave has escaped back to Visigothic territory.18 Following the same logic, church canons order the sale of slavewomen of clergymen who have been cohabiting with them and forbid the sale of Christian slaves to Jews.19 Another property connotation of Visigothic slavery was the fact that ownership of chattel and real property could accrue to masters through their servi and ancillae. This was true above all of their labour product, as several pizarras indicate.20 Of course, the same was also true of some tenants, but in contrast with tenants, slaves could also amass property for the master as a result of acquisitions that fell to a slave without her or his labour input – gifts, inheritances, tort settlements, etc.21 The inverse of this meant that the master could also incur liability through the actions of a slave: if a slave committed a crime or delict and the master knew or encouraged it, the master was obligated to pay the requisite composition.22 In many instances, however, the master was also given the choice to refuse to pay and instead to turn over the slave to the plaintiff in tort actions (‘noxal surrender’) or, in the case of crimes, to allow the slave to suffer physical torture, whipping, and even execution.23

gradations of unfreedom For all that slaves could be treated as property, they also retained some aspects of personhood – a contradiction inherent in all slave systems. This 17

18 19

20 21 22

23

FV 9: posessionem cui vocabulum est ill. cum mancipiis, terris et vineis omnique iure eius atque adiunctionibus; cf. FV 21; LV 5.4.13, 19; 5.7.16. See similar catalogues of estate property in Roman law, e.g. CTh 10.8.1 (a. 313): cum adiacentibus et mancipiis et pecoribus et fructibus et omni iure suo. LV 9.1.10; cf. LV 7.3.3, on the illegal sale of kidnapped children outside the kingdom. Sale of women: Conc. Toletanum IV (a. 633) can. 43; sales to Jews: Conc. Toletanum X (a. 656) can. 7. Pizarra 5, 75, 103. LV 5.4.15; 7.1.4; 9.1.17. LV 3.3.9, 12; 5.5.6–7; 6.1.1; 6.4.2, 4, 10; 6.5.10, 20; 7.2.3–6, 9, 14, 22–3; 7.3.4–6; 8.1.1, 6, 8; 8.2.1–2; 8.3.5; 8.4.21; 8.6.3; 9.1.2; 10.1.10. Cf. Nehlsen 1972, 191–220. See below notes 128–31.

Slavery among the Visigoths

257

was especially true of criminality, for, as discussed in the preceding section, slaves were held accountable for criminal acts they committed with the punishment of their own body. Furthermore, any slave who committed theft and was then freed before being tried carried the liability for that theft into their free status.24 Visigothic slaves could also, under specific circumstances, make positive legal claims. In early Visigothic law, slave women were afforded some legal protection against rape or sexual abuse by anyone but their master – who is nowhere legally prohibited from sexually abusing female slaves.25 Then in the mid seventh century the rights of slaves were considerably broadened by a series of laws issued by king Chindaswinth (r. 642–53). He allowed slaves to act as plaintiffs in the interests of a master who could not pursue his own suit because he was too far absent – Visigothic slaveowners regularly owned multiple estates in widely scattered locales, some of which they only occasionally visited.26 He also gave them greater authority over their peculium.27 Furthermore, Chindaswinth began allowing slaves to testify in court, a right forbidden them earlier.28 He granted royal slaves broad rights of testimony while extending more limited powers to other slaves,29 and his son Recceswinth allowed slave testimony in default of freeborn witnesses in cases involving homicide, small property claims, and the flight of fellow slaves.30 Slaves thus acquired limited claims to personhood underpinned above all by legal and economic necessity but also by a growing relaxation of the slave condition in the seventh century. The problem of legal personhood was further complicated by a series of status grades into which unfree persons were divided. At the top of this hierarchy were the royal slaves (servi dominici / regis) just alluded to. The crown controlled an extensive – surely the most extensive – body of slaves in the kingdom, but only some of these counted as ‘royal’ or ‘palatine’. This subset was apparently extremely close to the person of the monarch and held ennobling titles associated with their sphere of

24 25

26

27 28 29 30

LV 7.2.2. LV 3.4.16. See more on sexual violence in Germanic cultures at Wemple 1993. More on Visigothic sexual morality below at notes 135–7. LV 2.2.9; cf. 2.2.7; 2.3.3 [Erwig recension]. LV 7.4.1 (a law of Recceswinth) assumes slaves could appeal cases to a judge. LV 5.4.13. LV 5.7.12. LV 2.2.11; 2.4.4. LV 2.4.10.

258

Noel Lenski

activity: supervisors of the stablemen (stabularii), cupbearers (gillonarii), moneyers (argentarii), and cooks (coqui).31 Servi dominici with the title compulsores exercitus were charged with enforcing conscription into the royal army.32 Such royal slaves were an elite subset of the much larger body of servi fiscales, so-named because they were owned by the royal fisc.33 Their enslavement could result from the commission of acts considered crimes against the state, such as incest or divining on the fate of the monarch, although most servi fiscales were probably born into their status, which was heritable, or appropriated from the slaveholdings of private slaveholders.34 Indeed, being a servus fiscalis was nearly a permanent condition given that, from Chindaswinth onward, one could only be manumitted with the signature of the king, and any property they had amassed continued to belong to the king.35 This did not, however, stop some royal slaves from using their considerable wealth for their own ends, as for example making endowments to churches.36 Much like the imperial slaves and freedmen of the JulioClaudian period, Visigothic royal slaves apparently represented a special category of elite slave endowed with power and some degree of honour despite the enduring impediment of servitude. Privately owned slaves were also broken into subcategories. The Visigoths recognized the elevated status of some slaves whom they designated idonei – an adjective best translated as ‘reputable’, which could be applied to freedmen as well. The title was more than honorific, for the servi idonei were entitled to more favourable treatment in court and less stringent punishments than ordinary slaves (servi inferiores / viliores / rusticani).37 Ecclesiastical slaves formed yet another gradient of unfreedom within the broader status category of slaves.38 Freedmen (liberti) 31 32 33

34

35

36 37

38

LV 2.4.4. LV 9.2.2, 5. LV 7.5.9 also attests to servile notarii of the king. In 683, Erwig forbade any but servi fiscales from being promoted to palatine slaves, see LV 12.1.3 = Conc. Toletanum XIII (a. 683) can. 6 and Erwig’s accompanying Tome (Martínez Díez and Rodríguez 1984–2002, 6.222). Incest: CE Rest. 2; divining: LV 6.2.1; appropriation of private slaves: Conc. Tolet. XIII (a. 683) can. 6. LV 5.7.15–16; cf. LV 10.2.4–5. That some fiscal slaves attained freedom is proven by LV 5.7.19, requiring liberti regis to serve in the army. Conc. Toletanum III (a. 589) can. 15; LV 5.7.16. FV 3; LV 3.3.9 [Erv.]; 3.4.15; 6.1.5; 6.4.3, 7; 11.1.1; Conc. Spalense I (a. 590) can. 1; cf. Conc. Emeritense (a. 666) can. 17. Note that the distinction is most important in cases of delict, where Roman law also recognized a distinction between the relative statuses of slaves, Dig. 47.10.15.44. See Sommar 2010 and below notes 158–60.

Slavery among the Visigoths

259

should also be mentioned as members of a less than fully free class that remained separate from slaves or freeborn individuals. Like Roman liberti, Visigothic freedmen gained citizenship but owed obedience (obsequium) and rights of succession to their former masters (patroni).39 Moreover, in keeping with principles first laid out explicitly by Constantine, liberti who acted ungratefully or dishonourably towards their patroni could be reenslaved.40 Finally, the Visigoths also retained the status grade of bound colonus, a category of quasi-servile personhood first attested under Constantine. The mention of bound coloni is relatively common in the early-sixth-century LRV41 but then largely disappears under the kingdom of Toledo, when the ongoing existence of coloni is confirmed in only a scant handful of attestations.42 Among these is a recently published document from the Monastery of San Martín de Asán recording the donation of ten farms to the monastery in 522, all of which were staffed by manicipia or servi but two of which also had coloni or colonici.43 The relative paucity of coloni in sixth- and seventh-century Visigothic sources has led Luis García Moreno to make the compelling argument that the Visigothic kingdom came largely to dispense with the semi-dependent status of colonus in favour of slaves for its bound labour force.44 This array of unfree statuses is characteristic of post-Roman society, particularly in areas with influence from Germanic cultures. The cultures attested in northwest European law codes of the fifth through seventh centuries displayed considerable status gradation along a spectrum stretching from nobility to full servitude.45 This is certainly true of the Burgundian, Lombardic, and Frankish codes, all of which catalogue several levels of unfreedom and which display relatively well developed 39 40 41

42

43 44 45

LV 5.7.13, 14, 17. LRV CTh 4.10.1–2; LV 5.7.9–11, 17. LRV CTh 4.8.7; 5.9.1–2; 5.10.1; 5.11.1; 11.5.1; Nov. Val. 8–9, 12; cf. Bondue 2011, 141–3. Isid. Etym. 9.4.36; Conc. Spalense II (a. 619) can. 3. FV 36 indicates that colonus can designate a tenant who owed a tenth of his produce plus xenia (supplementary ‘gifts’) to a landlord on a five-year lease agreement; cf. FV 37. This would imply that the Visigothic colonus was not bound to the land in the manner of late Roman coloni adscripticii. The same would appear to be true of the accola mentioned at LV 10.1.15. See also LV 10.1.19, which mentions 10 per cent rental agreements, and should be read in conjunction with LV 10.1.11, which indicates these leases were often terminated by eviction for non-payment of rent. But see LV 5.4.19 for plebei who do appear to be bound to their land (gleba). S. Martín de Asán Document 1 (23 December 522); cf. DTV fol. 1 ll. 33–6. Garcia Moreno 2001; cf. Nehlsen 1972, 166–7; King 1972, 161; Claude 1980, 187–8. Olberg 1991.

260

Noel Lenski

systems of Wergeld – the money value associated with an individual’s personal status, which was graded on the basis of gender, ethnicity, age, and social standing. The functioning of the Wergeld system relied on a legal anthropology based primarily on the law of delict (tort), which expected individuals to prosecute wrongs to their person or property on their own behalf while minimizing the number of offences the state treated as crimes. Because of the history of its development inside a Roman provincial context, Visigothic law assigns less importance to Wergeld, but a closer look reveals that the whole Visigothic system is undergirded by the notion of values assigned to individuals based on their age, gender, and status, as well as a default assumption that disputes should be resolved through the payment of delictal settlements, referred to generally as compositions (compositiones).46 As such, the Visigothic normative sources reflect a society in which traces of status gradation are abundantly on view, among both free and unfree individuals. Out of this tradition of legally prescribed status hierarchy arose a complex fabric of slaveholding that allowed for semi-servile statuses, even if full chattel servitude remained far and away the most common form of dependency.

sources of slaves The supply stream for Visigothic slaves was wide indeed, involving many more avenues into unfreedom than the Roman system that preceded it. The first was captivity, a major source in all periods of Roman history as well. Although our narrative sources for Gothic history are thin, they point to a number of instances of mass captivity events in the fifth century as the Visigoths overran rivals in the Iberian peninsula such as the Suebi and Asturii.47 Warfare with the Franks and Byzantines in the sixth century also yielded captives.48 The LV makes it clear that the trade in captives flowed both in and out of the kingdom, and that provisions were necessary to ensure that those slaves recaptured from the enemy would be returned to their original Gothic owners, who then owed a third of their value to the rescuer.49

46

47

48 49

The clearest evidence comes at LV 6.4.3 (a fine-grained catalogue of compensatory penalties for personal assault), but compositions are literally everywhere in Visigothic law. For the gradation of penalties based on status, in addition to LV 6.4.3, see CE Rest. 2; LV 2.1.7; 2.2.8; 2.4.2–3; 3.1.6; 6.4.1; 7.2.20; 8.3.6, 10, 12, 14; 8.4.16, 24, 29. Hydatius 165[172]–167[174] (s.a. 456–7); 179 (s.a. 457); 237 (s.a. 467); Isid. Hist. Goth. 61; Ioh. Bicl. Chron. 35 (s.a. 575). Isid. Hist. Goth. 54. LV 5.4.21; cf. 9.2.7.

Slavery among the Visigoths

261

Far more common, however, was the production of slaves by birth. Visigothic law, like Roman law before it, recognized the principle that children born of slave women took the status of their mother.50 Moreover, for the Visigoths, the children of male slaves could also take the status of their father even when born of a free mother.51 Isidore, who openly asserts that ‘anyone born always takes the lower parental status’, also attempts to etymologize the distinction between captive and born slaves by claiming that the former (servi) were so called because of having been saved (servati) from death in battle while the latter were born into a familia (famuli).52 A logical extension of slavery by birth was the principle that exposed infants could be enslaved by their rescuers. Classical Roman law had permitted exposure even while forbidding the enslavement of freeborn children – a situation that ultimately afforded a major supply source given that infants were scarcely able to defend their own freeborn status.53 Christian concerns about the dangers of exposure – which could lead to consumption by wild animals or enslavement into a life of prostitution – led Constantine and subsequent emperors to forbid it, even while granting those who rescued infants that were exposed in spite of the prohibition the right to raise them as either slave or free.54 The Visigoths carried Constantine’s law further: the LV clearly prohibits exposure, but, like Constantine’s law, permits recovered infants to be raised as slaves; should the birth parents later attempt to reclaim their child, they could prevail only if they replaced the child with another slave; if they could not, they were themselves to be enslaved to the rescuer in exchange for the child’s freedom.55 Constantine had granted open sanction to the longstanding practice of selling one’s newborn children into a form of servitude that could best be classed as long-term indenture.56 The fifth-century LRV initially upheld Constantine’s new policy for Roman subjects57 even while Goths appear to have been forbidden to sell their children, but by the seventh century the prohibition had become universal.58 Visigothic law also penalized those 50 51 52

53 54 55 56

57 58

LV 9.1.15, 16; 10.1.17; Conc. Toletanum IX (a. 655) can. 10; cf. Isid. Etym. 9.4.45; 9.5.18. See below notes 169–70. Isid. Etym. 9.5.18; 9.4.43; 10.54; Diff. Verb. 525 (PL 83.63). Both etymologies derive from earlier Roman sources. Harris 1999. CTh 5.9.1 (a. 331) with Evans Grubbs 2009; cf. Tuor-Kurth 2010. LV 4.4.1–3. CTh 4.8.6 = CJ 7.18.3; CJ 4.43.2 = CTh 5.10.1 with Vuolanto 2003; cf. Bondue 2011, 132–5. LRV CTh 3.3.1; 4.8.2; 5.8.1 interp. CE 299; LV 5.4.12.

262

Noel Lenski

who kidnapped children to sell into slavery with the same composition required of homicides.59 Self-sale was also forbidden by the Visigothic codes of Euric and Recceswinth, which abide by the classical Roman principle that those who sold themselves to defraud the buyer were to be enslaved!60 Interestingly, however, FV 32 is composed in such way that it acknowledges the statutory prohibition on self-sale but then provides a formula with which to work around it.61 The apparent contradiction arises from the fact that, in contrast with the Romans, who forbade debt bondage and debt slavery, the Visigoths essentially allowed both. While Visigothic law appears to have upheld the Roman principle that one should not create a debt using one’s body as collateral, they did allow the accumulation of debts to result in the enslavement of debtors who could not meet their obligations.62 This too grew from the notion of Wergeld, which provided abundant opportunities for enslavement on the basis of debt obligations incurred not through lending and borrowing but through the commission of a crime or delict.63 In contrast with Roman delictal law, which calculates liability on a sliding scale based on estimated damages and limits credit settlements to the maximum value of the person’s assets, Germanic law tends to set elaborate schedules of fixed penalties for the composition of torts and permits the permanent enslavement of those incapable of meeting the resultant debt obligations. Moreover, where Roman law permits penal enslavement only for crimes (as opposed to delicts) and restricts the possession of penal slaves to the emperor, Germanic law, including Visigothic law, permits penal and delictal enslavement to both the king and private individuals. Enslavement could, in other words, be used by Visigothic plaintiffs as a last resort for the settlement of a tort claim against their defendants. The range of delicts that could result in enslavement to a private plaintiff was incredibly broad. It included the following offences, which usually ended in enslavement only if the defendant was unable to pay the 59 60 61

62

63

LV 7.3.3; cf. LRV CTh 9.14. CE 300; LV 5.4.10. See Duncan Jones 2016, chapter 14, on this seeming paradox. FV 32; cf. Pizarra 40 verso, which effectuates a self-enslavement, without mentioning sale. Contrast LV 2.5.8 and 5.6.5. For the Roman principle, see especially Herz 2015. For an interesting case of enslavement for fiscal debt in the city of Saintes under King Theodorid (r. 453–66), see Vita Viviani 4 (MGH SRM 3.96). Liebs 2001; cf. Nehlsen 1972, 185–250. Rio 2017, 62–6, attributes the emphasis on what she terms ‘penal enslavement’ in Visigothic law to the influence of the Theodosian Code, in marked contrast with what legal scholars argue.

Slavery among the Visigoths

263

requisite composition: giving fraudulent judgements;64 bearing false witness;65 marrying a girl betrothed to another man;66 a female slave owner having sexual relations with her own slave;67 remarrying while a living husband’s whereabouts were unknown;68 kidnapping a freeborn female;69 various forms of ‘adulterous’ sex;70 bigamy;71 child exposure;72 selling free men or women into slavery;73 bringing a false accusation that results in the death of a slave tortured in court;74 administering abortifacients;75 burglary;76 conspiring with a homicide;77 bringing false accusations;78 poisoning, magic, theft, and other crimes;79 kidnapping a slave;80 kidnapping freeborn children;81 forging signatures or documents;82 arson;83 releasing a fugitive slave;84 doctors killing patients during a phlebotomy;85 women who persist in sleeping with clergymen.86 Nor did Visigothic law shy from the Roman principle of permitting enslavement to the monarch, who generally reserved the option of retaining penal slaves or gifting the newly created bondsman to an individual of his choosing. The following crimes could result in enslavement to the king: treason;87 incestuous marriage;88 prostitution;89 fraudulent

64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86

87 88 89

LV 2.1.21. LV 2.4.6, 9. LV 3.1.2. LV 3.2.2. LV 3.2.6. LV 3.3.1–3, 5. LV 3.4.1, 2 (Erv.), 3, 12, 13 (Erv.), 14; 3.6.2. LV 3.4.9; 3.6.1, 2. LV 4.4.1, 3. LV 5.4.11. LV 6.1.5 = CE Rest. 11. LV 6.3.1 = CE Rest. 3. LV 6.4.2. LV 6.5.12. LV 7.1.1. LV 6.2.3; 7.1.5; 7.2.13–14. LV 7.3.2. LV 7.3.3. LV 7.5.2. LV 8.2.1. LV 9.2.1. LV 11.1.6 (Erv.). Conc. Toletanum III (a. 589) can. 5; Conc. Spalense I (a. 590) can. 3; Conc. Toletanum VIII (a. 653) can. 5. Conc. Toletanum XIII (a. 683) can. 1. CE Rest. 2. LV 3.4.17.

264

Noel Lenski

divorce;90 violating marriage restrictions for ecclesiastical freedmen;91 forging money;92 dodging conscription or creating fake draft notices;93 Jews refusing to be baptized and to renounce their religious customs, or apostatizing back to Jewish practices;94 Jews circumcising a Christian slave;95 unconverted Jews continuing to trade with Christians;96 and Christians helping Jews avoid the sale or manumission of their Christian slaves.97 Most disturbing of all, in 694, Egica decreed the enslavement of all Jews in the kingdom for redistribution as chattels to his subjects for the alleged crime of supporting the usurper Suniefred and for continuing to resist conversion.98 The sheer number of offences shows the impact of the Wergeld system on the possibilities for enslavement as well as the connection between ‘debt’ and slavery. In this environment, the chances for impressment into slavery must have been rife and must have turned the judicial system into something of a slave mill. That this was the case is clear from a series of laws attempting to ensure that the unjust enslavement of free individuals could be rescinded.99 Overall then, the system generated a tremendous number of chattel slaves, who were then traded and exchanged through sale, barter, gift, and inheritance. Unfortunately, we have only one documentary source for the conveyance of a slave, a pizarra preserving a barter contract for the exchange of a horse in return for a female slave.100 This paucity of data is a testament more to the exiguous number of surviving documents than to the relative strength or weakness of the slave market. Indeed, the value of the slave – worth one horse – matches well what we know about Roman slave prices, which are often equal to or equated with horse prices.101 And although we lack even a single sale contract with which to establish market pricing, we know from the LV that slave sales were generally 90 91 92 93 94 95 96 97 98 99 100 101

LV 3.6.2. LV 4.5.7. LV 7.6.2. LV 9.2.8–9. LV 12.2.11, 14, 17. LV 12.2.12. LV 12.2.18. LV 12.2.14. Conc. Tol. XVII can. 8 (Vives 534–6). LV 5.7.3–8; cf. 3.2.7; FV 1. Pizarra 42. E.g. CTh 11.17.1 (23 solidi); cf. CIL VIII 4508 = 18643, where the harbour customs tax at Zarai for slaves and horses are equivalent. See also LV 3.1.5, which establishes a rough equivalency between horses and slaves as markers of value.

Slavery among the Visigoths

265

denominated in gold solidi, and the same source attests a number of penalty prices for damage to slaves that indicate slave values could reach as high as 100 solidi but generally averaged around 20 solidi – precisely in line with late Roman prices.102

slave ownership Given the massive scope for entry into servile status, we might now ask who owned all these slaves. Because of the paucity of literary and documentary evidence, our knowledge of the names of individual owners of slaves and landed estates is minimal. Where we do have indications, they point to broad-based slave ownership. As we might expect, elite control of slaves is readily apparent in figures like the ‘Senator Sicorius’ or the ‘Count Eugenius’, both said to have been slave owners in Braulio’s Life of S. Aemilian.103 High-level clergymen such as bishops also owned slaves, some of whom were permanently attached to their church and others whom they inherited from their families.104 The former were unable ever to be fully freed, while the latter could. Jews owned slaves and estates as well, although – in keeping with prohibitions going back to Constantine – they were forbidden to buy Christian slaves or to receive them as gifts, to circumcise or manumit them, to work them on Sundays, or even to hold Christian slaves whatsoever. Indeed, Visigothic legislation against Jews became ever harsher until Egica essentially banned Judaism and ordered the forced conversion of Jews: those who complied could keep their slaves and land, but those who refused were themselves to be enslaved to the fisc.105 At the lower end of the economic ladder, four pizarras lay out account records for smaller estates, all of which included at least some slaves or freedmen on their roles.106 A law forbidding freeborn women from prostituting themselves orders their enslavement and transfer to poor persons in the countryside, an indication that it was not unusual even for the

102

103 104 105

106

LV 3.3.9 (100 solidi); 3.4.16 (20 solidi); 6.3.4 (20 solidi); 6.5.9 (36 solidi). On late Roman prices, see Harper 2010. Braulio VEmiliani 11[18]; 14[21] (Vazquez de Parga, ed.). V Patrum Emeretensium 4.7.2–10; 5.6.28. LRV 16.4.1; LV 12.2.12–14, 18; 12.3.1, 3, 6, 12, 16, 18. See also LV 12.3.19, forbidding Jews to supervise Christian slaves while acting as a vilicus. See also the regulations at Conc. Toletanum III (a. 589) can. 14; Conc. Toletanum IV (a. 633) can. 66; Conc. Toletanum X (a. 656) can. 7. See also King 1972, 130–45; Bondue 2011, 182–9. Pizarra 5; 11; 75; 111; cf. 103.

266

Noel Lenski

impoverished to own slaves.107 It is also confirmed that slaves could own slaves, just as they could in Roman law.108 Perhaps the best documented estate with slaves is described for us in the will of Vincent, a deacon of the church of Huesca, who inherited the estates of his parents and determined to donate them to the monastery of Asanus (Asán).109 In the document, which preserves both his donation and his will, Vincent disposes of at least twentynine properties, twenty-eight of which he gave to the monastery in 551, retaining for himself a usufruct and the right to manumit slaves for their good behaviour. When he died in 576, he confirmed the donation in his will, which is much more fragmentary but mentions one other property and enumerates a series of manumissions at its end. Here he disposes of at least nineteen slaves by name (in two instances named males are freed with an unspecified number of ‘children’): four of these remained in servitude but were transferred to heirs; of the remaining fifteen, most were first freed in the will, but two had been freed inter vivos so that the will only reconfirms their manumissions. Nonetheless, these nineteen or more slaves did not represent the entirety of Vincent’s familia, for in conjunction with all the properties he had donated ‘their coloni and slaves and their peculia’, and, moreover, in keeping with Visigothic inheritance law, he had withheld one quarter of his estates from the donation for transfer to his heirs at law.110 It is difficult to determine the relative size of Vincent’s properties and familia, although it is likely that they were on the small side given that he was a deacon. A provision of the XVIth Council of Toledo in 693 indicates that the smallest church properties had at least ten slaves.111 A controversy at the tenth Council Toledo in 656 reveals that Ricchimer, bishop of Dumio, had given some 500 of his own slaves as gifts to a group of ecclesiastical freedmen he had illegally manumitted.112 The estates of elite Visigoths must have had many more, a fact we can judge from the notice discussed below that elite weddings uniformly 107 108 109 110

111 112

LV 3.4.17; cf. LV 5.4.19. LV 5.4.13, 18; 5.7.16. See the translation of Corcoran 2003; cf. Díaz 1998; Roth 2016. DTV fol. 1 ll. 33–6 (coloni et servi); fol. 2 ll. 1–17 (manumissions); DTV fol. 1 col. 1 ll. 42–4 (one quarter). On the requirement to preserve one quarter for heirs, see LRV CTh 2.20.1; cf. LV 4.2.19. Ariño Gil and Díaz 2003 have shown that Vincent’s family properties were geographically diversified, allowing for the cultivation of the classic Mediterranean triad of grain, olives, and grapes, plus pastoralism – all of which are mentioned in the documents. Conc. Toletanum XVI (a. 693) can. 5 (MGH LNG 1.481). Conc. Toletanum X (a. 656) causae Potamii et Ricchimiri (Martínez Díez and Rodrigues 1984–2002, 5.546).

Slavery among the Visigoths

267

involved dowries of ten male and ten female slaves, making it likely that the largest estates must have had hundreds. Largest of all would have been the slaveholdings of the king. In his History of the Conquest of Spain, Ibn al-Quˉ tıˉ̣ ya claimed that, after the Muslim conquest, the sons of King Witiza had been granted ongoing possession of the royal estates, which consisted of some 3000 properties.113 The figure seems exaggerated, and it is impossible to say how many slaves each property might have contained, but, given the scale of royal holdings and the number of tasks assigned to servi dominici and servi fiscales, the king’s slaves may have numbered in the tens of thousands.

the use of visigothic slaves Most of these slaves were used in farming. FV 8 provides a template for donating land to a church ‘together with its slaves of the following names’, and FV 9 lays out the donation of an estate with its ‘slaves, lands and vineyards’. The same picture emerges from Pizarra 103, a letter from a certain Faustinus ordering Paulus, perhaps an estate manager, to conduct the olive harvest using slaves (mancipios [sic]). A law issued by Erwig ordering landowners to muster their slaves for military expeditions complains that these are otherwise too inclined to pursue private profit by constantly keeping their slaves labouring in their fields.114 Vincent of Huesca’s slaves appear to have been employed exclusively in agriculture when he donated them to the monastery of S. Martin de Asán. So too three of the deeds recently published by Tomas-Faci and Martin-Iglesias from the same monastery record the transfer of more than a dozen properties gifted or granted in 522, 576, and 586, all outfitted ‘with slaves’ (cum servis / cum mancipiis).115 Slaves are also attested in the codes and hagiography working as shepherds and swineherds.116 That said, slaves were clearly not the only form of agricultural labourer employed by Visigothic landowners. As noted above, coloni are attested in documents from San Martin de Asán, and a law of the LV also attests to the use of

113

114 115

116

Ibn al-Kutiya 1926, 2 (Spanish) / 3 (Arabic); please note this name is now more commonly transcribed as ‘Ibn al-Quˉ tˉ̣ıya’. LV 9.2.9; cf. 5.4.13; 12.2.14; 12.3.6, all implying agricultural work. S. Martín de Asán Document 1 (23 December 522); Document 3 (22 February 576); Document 4 (13 December 586). See also Isid. Reg. Mon. 5.7. Shepherds: LV 5.5.6; cf. V Fruct. 2 (Diaz y Diaz 82). Swineherds: LV 8.5.1, 3. See also LV 5.4.13, on slave sales of livestock. Fruct. Brac. Reg. Comm. 9 (PL 87.1117–18) clearly indicates that free young boys (parvuli) were also used as shepherds.

268

Noel Lenski

hired agricultural labourers (mercenarii).117 For that matter, many of the labourers catalogued in the estate records of the pizarras appear to have been neither servi nor bound coloni but free, rent-paying tenants.118 The Visigoths also continued the Roman practice of using slaves as estate managers (procuratores, vilici) and accountants (actores) as well as business managers and secretaries in urban environments, although all these jobs could also be held by free individuals.119 Slaves often served their masters as grooms, attendants, messengers, and factotums,120 and we also have evidence that they could be used as skilled workers – builders, blacksmiths, swordsmiths, bronze casters, even doctors.121 At times slaves were clearly rented to other producers for temporary use, with their wage reverting to their master.122 A single law attests to the master’s use of slave women as prostitutes, but only to forbid this on pain of a public whipping.123 In general, Visigothic society’s draconian penalties against sexual promiscuity make it unlikely that sex trafficking was a major source of income for slaveowners, but it cannot be denied that Isidore of Seville’s etymological explorations betray a rich knowledge of the vocabulary of prostitution and of the details of its practice.124 Thus, even if agriculture was far and away the most common arena of employment for slaves, as also for the free, unfree labour permeated all layers of the Visigothic economy.

117 118

119

120

121

122 123

124

LV 9.1.12. See Pizarra 11, which seems to record payments in cheese to free labourers; cf. Pizarra 111. On estate managers, see LV 2.2.9; 6.1.1; 8.1.5; 9.1.8, 9, 18, 21; 12.1.2; 12.3.19; Cf. Isid. Etym. 9.4.33–35; Diff. Verb. 123 (PL 83.23). On business and clerical work by slaves, see CE 287; LV 2.2.7; 2.5.6, 13; 7.5.9; V Patrum Emeretensium 5.3.10–12; 5.7.1–10. E.g. V Patrum Emeretensium 5.11.19–21; LRV Pauli Sententiae 3.9.56; LV 3.4.6; cf. Isid. Diff. Verb. 94 (PL 83.21); V Patrum Emeretensium 4.7.2–10; 5.13.1–15. Clerics were forbidden from having female slaves as attendants to minimize the chances of sexual impropriety, Conc. Toletanum II (a. 527) can. 3; Conc. Spalense I (a. 590) can. 3; Conc. Toletanum IV (a. 633) can. 42–3. LV 6.1.5; 11.1.7; Conc. Toletanum III (a. 589) can. 15; Conc. Toletanum X (a. 656) at Vives 1963, 322; Isid. Reg. Mon. 5.7. ILCV 1815 = AE 1995, 844a offers epigraphic confirmation of the use of slave labourers (operarii vernuli) in the construction of three churches at Nativola (Acci, near Granada) by the Illustrious Duke Gundiliuva in 607, cf. Velázquez 2007. LV 11.3.4. LV 3.4.17; cf. CE rest. 3, punishing slave women who provided abortions, implying employment as healers or midwives. Isid. Diff. Verb. 1.263; Etym. 10.163, 182, 228–29, 231; 18.42.2; 19.25.5; 29.16.24. See also Conc. Toletanum IV (a. 633) can. 44.

Slavery among the Visigoths

269

the treatment of slaves Visigothic slaves could expect to be treated with generalized dishonour and violence.125 This applied to all aspects of their lives. Even the bread they ate and wine they drank was inferior to that of their freeborn masters.126 They were regularly enchained and beaten, especially with whips.127 The whipping of slaves was always permitted to masters but could also be enacted by the state in instances where testimony was needed against a master: such court interrogations were permitted in cases involving adultery by the master or mistress (of which the slave often had exclusive knowledge), treason, counterfeiting, homicide, sorcery, or collusion.128 Furthermore, most of the crimes and delicts of slaves were to be punished with whipping, sometimes also execution. In these criminal and delictal situations we again find strong traces of Wergeld, a reality that led to significant differences with contemporary Roman law in the Byzantine East or even Ostrogothic Italy. Under Visigothic law, not only slaves but also freeborn people (ingenui) could be subject to public torture, usually by whipping. Nevertheless, ingenui were often (though not always) given the option to settle liabilities with a money composition and thereby avoid torture. In general, this meant that public whipping affected only those free persons too poor to pay a composition – sometimes classed juridically as ‘little people’ (minores). This harsh reality shrank the distance between slave and free in Visigothic society, although even here a difference was marked by the fact that, for any given offence, minores were often prescribed only half the number of lashes compared with slaves.129 Also, minores could still receive money compositions for wrongs done to them, whereas the master received a composition owed for wrongs done to a slave. By the same token, slaves had the advantage 125

126

127

128 129

One could say that the violent treatment by the Visigoths of their slaves shows marked influence – but not complete overlap – with Roman precedents; cf. Lenski 2016. Isid. Etym. 20.2.15. II De Escis. Cibarius est qui ad cibum seruis datur, nec delicatus [‘Concerning foods. “Cibarius” [bread] is what is given to slaves as food, and it is unrefined’]. Isid. Etym. 10.3.9. III. De potu. Crucium vinum est insuave quod servi potant [‘Concerning drinks: “Crucium” is a wine that is harsh which is given to slaves’]. Greek and Roman slaves were also fed inferior bread and wine (Matthews 2006, 164–8). E.g. LV 5.6.1; 6.5.12; 9.1.2; cf. Isid. Etym. 5.27.15–32; Pizarra 22. More on slave punishment at Nehlsen 1972, 226–8; Bondue 2011, 150–70. LV 3.4.10–11; 6.1.4; 7.6.1. E.g. LV 2.2.7; 3.3.8–10, 12; 5.6.1; 6.4.1–3; 7.2.2–6, 13–14, 22; 7.3.1, 4; 7.4.1, 11; 8.3.2, 5, 6, 10, 14–15; 8.4.8, 11, 15, 26, 29; 9.1.1; 10.3.2, 5; 11.2.1–2. Compositions for the killing of a slave by a third party were also valued at half that for a free person, LV 6.5.7, 9.

270

Noel Lenski

that, if their masters chose, they could pay a composition on behalf of their slaves and thereby relieve them of torture.130 Indeed, the consequences of such tortures – frequently 100 or even 200 lashes – often resulted in death or permanent disability.131 Serious offences were also punished with ‘decalvatio’, a legal form distinctive of Germanic cultures. This was a sort of colossally bad haircut; whether the result of scalping or less drastic capillary abuse is debated. Decalvatio was prescribed for slaves who cheated someone of a judicial pledge, kidnapped a free woman or even a female slave, prostituted themselves, sold their own master into slavery, killed a fellow slave on orders from their master, or (beginning in 693) converted permanently to Judaism.132 Here too, however, free people could also face decalvatio for serious crimes, such as conspiracy to murder, treason, bearing false witness, or deserting the army.133 Harshest of all was burning alive, a punishment reserved for slaves who robbed tombs or raped freeborn virgins or widows, but which was also applied to the male slave who had sexual relations with his female owner – who was also to be burned alive.134 In all of these instances, typically servile punishments could also be inflicted on freeborn individuals, but usually for crimes significantly worse than those provoking the same punishments for slaves. Sexual abuse has also been part and parcel of slave societies across history. This was surely true of Visigothic society as well, although we have no direct testimony to confirm it. Male masters are nowhere forbidden in normative sources from sexually exploiting their slave women, and they certainly did so. Indeed, LV 3.5.5 explicitly attests to relations of concubinage between male slaveowners and their female slaves, albeit to forbid any other family member from initiating a sexual relationship with that same woman, as part of the broader process of regulating incest. We just saw that female slave owners were forbidden from sexual relations with their male slaves on pain of public immolation, an inheritance from 130

131

132 133

134

Compensation through fines and torture is everywhere in the LV. See representative examples at LV 7.2.14; 8.1.3–4, 10–12; 8.2.1–2, 5–6, 10, 14–15; 8.3.2, 5–6; 8.4.8, 11, 15, 24, 26, 29, 30–1; 8.5.3; 8.6.3; 9.1.1–2; 10.3.2, 5; 11.2.1–2. Cf. Nehlsen 1972, 187–8. If an innocent slave was injured or killed as a result of torture for a false accusation, the accuser owed replacement slaves to the master, CE 11; LV 2.3.4; 6.1.5. LV 2.2.7; 3.3.8–10; 3.4.17; 5.4.11; 6.5.12; 12.2.14; 12.3.6. See Nehlsen 1972, 224–6. LV 2.1.8; 2.4.6; 3.6.2; 6.2.4; 6.4.5; 6.5.12; 9.2.9; 12.3.2–9, 11–13, 17, 19, 21. A great number of these penalties were issued by Erwig in his persecution against Jews. LV 3.2.2; 3.4.14; 11.2.1. The gruesome execution of a female master and her male slave who engaged in sexual relations traces back to a law of Constantine, CTh 9.9.1 = CJ 9.11.1, which was transmitted into Visigothic law through LRV CTh 9.6.1.

Slavery among the Visigoths

271

Constantinian law. From the reign of Chindaswinth onwards, homoerotic intercourse between males was also strictly banned on pain of castration for both parties, regardless of their status.135 This surely hampered – without entirely foreclosing – the sexual abuse of male slaves by their male masters. We have also seen that the sexuality of female slaves was protected against third party attackers by law, and Visigothic law’s strict censures against sexual promiscuity outside of marriage – together with its recognition of forms of marriage between slaves – may further have hampered unfettered access by male slave owners to their female slaves.136 Gothic discomfort with the sexual abuse of slaves also appears to be evidenced in On the Governance of God, written by the Gallo-Roman Salvian of Marseilles c.440. Salvian, who lived on the fringes of Visigothic territory in Narbonese Gaul, pins the collapse of Roman power in this region on his fellow Romans’ sexual exploitation of their slave women: the Goths, he charges, exercised sexual restraint with their slave women, and the Vandals had even outlawed prostitution, while their Roman contemporaries invited God’s wrath upon themselves by treating their slaves like so many whores and breeding mares.137 The moralizing tone makes it difficult to weigh the value of this testimony, but its coincidence with other indications of sexual restraint among Gothic masters lends it at least an air of credibility. In the seventh century, the Visigothic kings also introduced legislation that shielded slaves from the worst abuses of their masters. Mitigations like this had begun to be enacted in Roman law already in the second century ce and picked up speed under Constantine and Justinian. Constantine had, for example, forbidden slave families to be split apart in property transfers. The Visigoths absorbed this into the LRV, and a law issued by Sisebut (r. 612–21) confirms that the principle continued to be enforced in the seventh century.138 In a law of 316, Constantine forbade the tattooing of a slave on the face, ‘which is formed in the image of heavenly beauty’ – a clear reference to Genesis 1:27.139 This law was not absorbed

135 136

137

138

139

LV 3.5.4, 7; cf. LV 3.6.2; Conc. Tolet. XVI (a. 693) can. 3. On protections for female slaves, see note 25. On Visigothic sex control, see LV 3.2–5, passim. Salv. Gub. 7.6.23–7 (CSEL 8.162–3); cf. 4.6.27–30; 7.3.15–4.20 (CSEL 8.72–3; 159– 61). On the Vandals, see Salv. Gub. 7.22.96–7 (CSEL 8.187). On these passages, see De Wet 2018b. Constantine’s law, CTh 2.25.1 = CJ 3.38.11 (a. 325), is absorbed as LRV CTh 2.25.1. The principle is enforced at LV 12.2.13. CTh 9.40.2 = CJ 9.47.17.

272

Noel Lenski

into the LRV by Alaric II in 506, but one of Recceswinth’s laws punishes with exile any master who disfigured his slave’s face or hands, which were ‘formed in the image of God’, without first undertaking a full judicial inquiry.140 Similarly, Constantine had forbidden the intentional killing of slaves – again on Judeo-Christian principles – even if he also issued a follow-up rescript exempting those masters who killed their slaves in the course of corrective punishment.141 Again, the LRV walks this back, omitting the first law and preserving only the second.142 Chindaswinth, whose favourable legislation towards slaves we have seen above, acted more decisively by enacting a blanket prohibition on private slave killing in the absence of a public judicial decision and punished offending masters with perpetual exile and the loss of their fortunes.143 These laws clearly occasioned controversy, for in his recension of the LV, Erwig (r. 680–7) significantly weakened the law on slave killing and entirely omitted that on mutilation; Egica (r. 687–702) then restored the former and reinstated the latter.144 Thus only in the seventh century did Visigothic law make halting advances towards the sorts of mitigations that Roman law had deployed in the early fourth to limit the abuse of slave bodies. None of this forestalled what appears to have been a massive problem with flight. Much of the ninth book of the LV is devoted to staunching this massive haemorrhage of fugitive slaves. Rules are established to limit the number of days one can host an unknown visitor on one’s property, to require that inquiry be made into their identity, to arrest and interrogate them if they appear to be fugitives, to protect those who failed to hold fugitives despite using chains and ropes, and to reward those who returned fugitives (one tremissis for every thirty miles they 140

141

142

143

144

LV 6.5.13; cf. Conc. Emerit. (a. 666) can. 15; Conc. Toletanum XI (a. 675) can. 6. The Visigoths never came close to abolitionist sentiments, which do faintly surface in fourthand fifth-century Roman sources. Isid. Sent. 3.47.1–3 (CCSL 111.295–6) assumes slavery was a necessary part of the post-lapsarian world. CTh 9.12.1 = CJ 9.14.1 (a. 319); CTh 9.12.2 (a. 329). The religious principle is derived from Exodus 21.20–1 and is followed at Conc. Eliberitanum can. 5 (Mansi 2.6); Bas. Ep. 188.8; cf. Collatio 3.1.1. LRV CTh 9.9.1; cf. LRV Gai Epitome 1.3.1. Leovigild had reaffirmed the master’s right to do whatever he wished against a slave caught thieving, without interference from a judge, LV 7.2.21. LV 6.5.12 [Recc.]. But see LV 6.5.8, again confirming the innocence of masters (and teachers?) who killed a slave (or pupil) accidentally while administering corporal punishment. See also LRV CTh 9.1.8, which threatens with death those who bring a capital charge against a slave falsely. On the killing of slaves by third parties, see Nehlsen 1972, 172–5. LV 6.5.12 [Erv.]; 6.5.13*.

Slavery among the Visigoths

273

needed to travel – a substantial premium, adding further confirmation to the high monetary value assigned to slaves).145 By the same token, expensive penalties were prescribed for those who consciously harboured or assisted fugitives or brazenly attempted to appropriate them as their own.146 This last appears to have been a major problem, for powerful landowners were clearly able to hold fugitive slaves for extended periods, some long enough to establish conjugal unions and bear offspring.147 A general paranoia about the omnipresence of fugitive slaves emerges from a generic letter of request for the return of a fugitive slave preserved in the FV, and from a scene in the Life of Fructuosus of Braga, a bishop whose humble travel attire led to a caning by a stranger who assumed Fructuosus must have been a fugitive slave.148 The problem was almost certainly exacerbated by the decentralized nature of Visigothic power, for the relative autonomy of local landholders gave wide scope for shirking of the law and necessitated self-help and personal policing: the most powerful could impress poorer peasants into slavery and could appropriate the fugitive slaves of less powerful owners with little fear of the crown.

manumission from slavery and the rise of iberian serfdom Slaves could also escape the control of their masters legitimately, through manumission. This could occur either while the master was still alive (inter vivos) or in a will.149 The former occurred either before three to five witnesses or before clergymen, a form called ‘manumission in church’ (manumissio in ecclesia) that was first introduced into Roman law by Constantine.150 Visigothic manumission also followed Constantine’s lead in allowing freedmen (liberti) who harmed or offended their former master (patronus) to be re-enslaved.151 Like their Roman predecessors, the Visigoths appear to have been relatively free with 145 146 147 148 149 150

151

LV 9.1.3, 4, 6, 8, 9, 11, 13, 14, 21; 9.2.7. See also King 1972, 167–70. LV 9.1.1, 2, 5, 7. LV 9.1.12, 15, 16, 18, 19, 21; 10.2.2; cf. CE 277. FV 43; V Fruct. 11 (Diaz y Diaz 98–100). LV 5.7.1–15. On Visigothic freedmen, see Claude 1980. LRV CTh 4.7.1; LV 5.7.2, 9. For Roman manumissio in ecclesia, see Fabbrini 1965. Claude 1980, 163–5 posits a form of inter vivos manumission akin to Roman vindicta (before a magistrate), but I find no evidence. LV 5.7.9–11, 18–20; Conc. Spalense II (a. 619) can. 8. For Roman laws on the reenslavement of liberti ingrati, see CTh 4.10.1–2.

274

Noel Lenski

manumission, leaving us with multiple witnesses to the practice. Out of a total of forty-five extant Visigothic formulae, six lay out charters for inter vivos manumissions (cartula / scriptura libertatis), and one other offers a standard will that includes provisions for testamentary manumissions.152 We have also seen that Vincent of Huesca freed thirteen of the seventeen slaves named in his will. Most of the manumission formulae spell out that granting a slave liberty was thought to be of spiritual benefit to the master.153 This provided masters with a motivation to free their slaves above and beyond feelings of sympathy or bargains to reward good behaviour. This same belief no doubt also fed into the common practice of freeing slaves for immediate ordination in the clergy or entry into a monastery.154 Further reward could accrue to a freedman when the master granted him the gift of his peculium or a donation of land on which to support himself, which appears to have been a common practice, often with the express intent of ‘confirming their liberty’.155 Indeed, care needed to be taken to ensure that properly freed slaves were not unjustly reclaimed back into slavery – yet the number of laws against this abuse indicates it occurred relatively often.156 By the same token, precautions were also set in place against fugitives falsely claiming to be freedmen. When, for example, fugitives sought work as paid labourers or admission to a monastery, they were expected to prove their freedom by producing signed manumission charters or taking oaths.157 We have seen that manumission was possible for royal slaves, but only with the explicit permission of the king, reducing the chances greatly that royal bondsmen saw freedom. The chances were also slim for ecclesiastical slaves. These could be divided into those owned by individual churches and those in the private estates of bishops or clergymen.158 In order to minimize any diminution in the wealth of a given church, the former could be freed only in limited numbers and under increasingly restrictive conditions: the Third and Fourth Councils of Toledo in 589 and 633 required all ecclesiastical freedmen 152

153 154 155

156 157

158

FV 1–6, 21. The formulae recorded at FV 1–3, 6, are followed in several extant charters, see Gil 1991, 71–7. Córcoles Olaitz 2006 should be used with caution. FV 2–5. FV 6; LV 5.7.18; cf. Isid. Reg. Mon. 4.3 (PL 83.872). FV 2, 5–6, 21; LV 5.7.1, 14; V Patrum Emeretensium 5.13.4; V Fruct. 8 (Diaz y Diaz 92); DTV, fol. 2.1–5, 11–14; cf. Conc. Toletanum IV (a. 633) can. 69. LV 5.7.3–8. LV 9.1.11, 14–15; Fruct. Brac. Reg. Comm. 4 (PL 87.1114); cf. Conc. Toletanum VI (a. 638) can. 9; Conc. Caesaraugustana (a. 691) can. 4. Monastics were forbidden to hold slaves, Isid. Reg. Mon. 20.

Slavery among the Visigoths

275

to remain in the permanent patronage (patrocinium) of the church, which insisted on ongoing and heritable bonds of ‘obedience’ (obsequium) that had to be acknowledged in annually renewable vows; and the Ninth Council of Toledo in 655 forbade a freedman’s intermarriage with fully free partners or even the freedmen of other patrons; full freedom was only possible if the manumitting bishop paid double the slave’s value to the church from his private wealth.159 Moreover, while the privately owned slaves of clergy, acquired before ordination or through inheritance, could still be fully freed, the fuzzy distinction between ecclesiastical slaves and the private chattels of clergy was often difficult to draw, leading episcopal successors regularly to seek the revocation of manumissions granted by their predecessors.160 Over time, Dietrich Claude has argued, private manumissions came ever closer to the ecclesiastical rigidity in restricting the liberties of freedmen. Visigothic manumission could entail serious impediments, for some charters insisted that a freedman continue to serve a former master until his or her death.161 The oldest version of LV 5.7.13 indicates that Visigothic freedmen who moved away from their patron’s land lost any rights of testation, and Erwig’s recension of the same law forbids any freedmen to leave the obsequium (‘obedient service’) of their patrons before their death. This habit of locking freedmen into ongoing service to their manumitter appears to explain why Vincent of Huesca needed to reconfirm the freedom of two slaves he had already manumitted while still alive.162 In the mid sixth century, Recceswinth forbade freedmen or their children to marry or insult their patronus,163 and by the late seventh century, Egica extended the right of the patronus to maintain control over his freedmen’s families to the third generation and beyond, essentially 159

160

161 162

163

By 506, the Visigothic church had begun placing severe restrictions on the manumission of ecclesiastical slaves, cf. Conc. Agathense (a. 506) can. 7 (CCSL 148.195–6). For the ever-growing limitations on ecclesiastical manumission, see Conc. Toletanum III (a. 589) can. 6; Conc. Spalense I (a. 590) can. 1–2; Conc. Toletanum IV (a. 633) can. 67–72; Conc. Toletanum VI (a. 638) can. 9–10; Conc. Toletanum IX (a. 655) can. 13– 16; Conc. Emeritense (a. 666) can. 20; Conc. Caesaraugustana III (a. 691) can. 4; cf. LV 4.5.7. See also Esders 2010, 38–9. V Patrum Emeretensium 5.13.1–15; Conc. Spalense I (a. 590) can. 1; Conc. Toletanum IV (a. 633) can. 67; Conc. Toletanum X (a. 656); cf. Castellanos 1998. Claude 1980; FV 3, 5; Conc. Toletanum IV (a. 633) can. 73. DTV fol. 2 ll. 1–2, 15–17. DTV (fol. 1 col. 1 ll. 44–5) indicates that Vincent took advantage of the broad right to re-enslave his freedmen in order to guarantee their ongoing service and obedience. See also LV 5.7.13 on the master’s rights to the inheritance of property from freedmen. LV 5.7.17.

276

Noel Lenski

locking ex-slaves into a permanent state of servility.164 Thus, even as the slave condition was becoming milder, complete liberation from servitude was growing more difficult – rendering late Visigothic slavery ever closer to serfdom. Claude posits that this enserfment of freedmen was triggered by demographic collapse in the wake of repeated outbreaks of the plague, which is certainly a possibility, although the situation in seventh-century Iberia mirrors closely that of Italy, Africa, and the Greek East in the fifth, which raises the possibility that we are witnessing a trans-regional pattern in the evolution of complex slave systems.

the particularity of visigothic slavery As discussed at the beginning of this study, the scholarship on Visigothic slavery has often diverged: some have ignored the intensity of Visigothic slaveholding, others regard it as a straightforward extension of Roman practice, and others still as an early manifestation of the serfdom characteristic of the high medieval West. A fairer assessment is surely that the Visigoths developed their own slaveholding practices, which represented a hybridization of Roman and Germanic customs and relied heavily on the intensive use of slaves in ways that would impact patterns of slaveholding in the Iberian peninsula – setting them apart from other parts of Western Europe – for many centuries to come. We have already seen that Visigothic law imported a considerable amount of Roman practice into a system that nevertheless remained fundamentally Germanic. Above all, the complex gradation of statuses and the ground principle of Wergeld resulted in the regular enslavement of members of the Visigoths’ own society (what Orlando Patterson terms ‘extrusive slavery’), who could be added to those slaves taken from outside (‘intrusive slavery’).165 To demonstrate the singularity of this system and its hybrid nature, it is worth examining three specific examples of Visigothic slaveholding practices that demonstrate at the microscopic level what we assert to be true macroscopically as well. First, like other forms of Germanic law, Visigothic law recognized some form of marital union between slaves.166 This will strike students 164

165 166

LV 5.7.20. This represents a sharp break with late Roman custom, which had limited the ability of patroni to re-enslave ungrateful freedmen to the manumitting master, debarring subsequent generations from this right, cf. Nov. Val. 25 = LRV Nov. Val. 6. Patterson 1982, 38–45. LV 3.2.5: qui servo suo ancillam alienan coniuncxerit (sic); 12.2.14: mancipia vero, que ex christianorum et Hebreorum conubiis nata vel genita esse noscuntur; cf. 3.2.7; 9.1.15–16 and Bondue 2011, 190–4.

Slavery among the Visigoths

277

of Roman slavery as anomalous and perhaps also a concession to the humanity of the slave, yet the Visigothic approach was perhaps even more beneficial to the slaveowners themselves. Above all, slave marriage allowed masters to lock those free individuals – males as well as females – who established lasting sexual unions with their slaves into permanent servitude unto themselves. By 52 ce, Roman law allowed the owners of male slaves to enslave any free woman who refused to heed three warnings to keep away from their male slaves.167 Visigothic law, by contrast, appears initially to have followed Germanic custom in turning offending women over to their own families to receive (capital) punishment.168 Eventually, however, king Leovigild (r. 568–86) combined Roman and Germanic principles by allowing the master of the male slave to gain ownership of the free female if her parents refused to punish her – and he also punished free males who slept with another master’s slave women by having judges publicly whip them.169 In the seventh century, Chindaswinth – whose favourable legislation we have already encountered – gave more positive force to these unions by allowing masters to separate them only if they acted within one year of learning the slaves were living together.170 The recognition of some form of slave unions thus allowed Visigothic law simultaneously to mitigate the slave condition and to employ quasi-marital bonds as yet another avenue into servitude. A second arena in which Visigothic practice stands apart regards the dowries of members of the Visigothic elite – palace grandees and the ‘lords of the Gothic Race’ (seniores gentis Gotorum).171 Gothic dowries were starkly different from their Roman counterparts, primarily because they represented a permanent gift from the bridegroom to the bride rather than a sum of wealth provided by the bride’s paterfamilias for her maintenance, as had been the Roman custom. Because they were, in essence, a bride price, Visigothic dowries were used for status competition between members of the elite until a law issued by Chindaswinth in 645 limited them to property ‘worth one thousand solidi, counting all assets, and in addition we grant the freedom of cataloging and giving 10 male and 10 female

167 168

169

170 171

Pauli Sententiae 2.21A.1; CTh 4.12.1–7 (LRV CTh 4.11.1–6); Nov. Val. 3.30. On this practice, see Lex Burgundionum 35.2–3; Edictum Rothari c.221; Lex Salica 25.5–6 with Zeumer 1899, 592–4 on the earlier Visigothic custom. LV 3.2.3–4; cf. 3.2.7; 3.6.2; 4.5.7. Following Roman law (CTh 9.9.1 = LRV CTh 9.6.1; cf. Nov. Anth. 1), Visigothic law (LV 3.2.2) forbade marriage between female masters and their male slaves or liberti on pain of death. LV 10.1.17. For ongoing concern with the assertion of Gothic ethnicity see Liebeschuetz 2015.

278

Noel Lenski

slaves and 20 horses’.172 This same figure shows up in a Visigothic formula, likely composed before the issuance of Chindaswinth’s law, which lays out a verse version of a dowry charter, including the line: ‘Behold, we hand over ten boys and the same number of girls, together with ten stallions; and we give an equal number of mules among other gifts and armaments, in keeping with the Morgengabe (‘morning gift’) of the ancient Gothic order.’173 This same standardized gift reappears in extant dowry documents of 887, 926, and 962 ce.174 Slaves were, in other words, a very ancient and enduring part of the Gothic bridal trousseau, used to measure and transfer wealth among the elite in a fashion peculiar to – and in part constitutive of – this Germanic society. Finally, in stark contrast with the Romans, for whom slaves were banned from armed service on pain of death, Visigothic kings required landholders to bring a percentage of their slaves with them on military expeditions. To be sure, Roman soldiers regularly owned slaves whom they used as camp followers to perform menial services, but these were never armed, nor did they see combat. Visigothic slaves, by contrast, were used as active combatants, a fact confirmed in a law issued by Erwig in 681 that demanded that 10 per cent of a proprietor’s servi accompany them on expeditions, and that they be armed for combat by the proprietor.175 Granted, this law was issued late in the history of the kingdom, when military recruitment was in crisis, But already in the late fifth century a law of Euric attests to the use of slaves in combat when it treats the question of who owns the plunder seized in combat by the slaves of a proprietor’s wife who were brought with him on expeditions.176 This practice of arming slaves for combat, so antithetical to high Roman norms, became increasingly common as the Empire wore on but reached full flower only under the Visigoths. Ultimately, it grew out of a characteristically Germanic practice of using military service among captives and slaves as a pathway to integration into full societal membership.177 172

173 174

175 176

177

LV 3.1.5: atque insuper decem pueros decemque puellas et caballos viginti, seu in ornamentis, quantum mille solidorum summa esse constiterit, dare debebit. FV 20. Loscertales de G. de Valdeavellano 1976, I no. 119 (a. 887); Andrade Cernadas et al. 1995, II no. 577 (a. 926); Mínguez Fernández 1976, no. 207 (a. 962). On Germanic dowry practices see Zeumer 1901, 138–46; cf. Rio 2017, 146–7. LV 9.2.9; cf. LV 2.5.13; 5.7.19. For fuller references on armed slaves, see Lenski 2009. CE 323; cf. LV 4.2.15; 8.1.8. See also CIL II 714 = ILCV 274 (a. 642), which appears to record the recovery of a Visigothic noble’s body after his death in combat by his cli(e)ntes (freed slaves). Goldmann 1904.

Slavery among the Visigoths

279

Visigothic slaveholding was, in other words, constituted of a hybridization of Germanic and Roman practices. Further influence from economic and political shifts in the early Middle Ages, and from the growing power of the Christian church, also had an impact.178 The result was a highly intensive slave system embedded in its societal (Germanic and Roman), historical (post-Roman), and geographical (Southern Gaul and Iberia) situation.

conclusion Slavery thus played a marked and important role throughout the history of the Visigoths. Even as slavery was giving way to milder forms of dependency in the Roman Empire of the fifth and sixth centuries, it was reaching its zenith in Visigothic Gaul and Iberia.179 In sixth-century Iberia, Visigothic slaves were clearly the chattels of their owners, who appear to have been spread across the social and economic spectrum. Slaves were a key component of the labour force, particularly in agriculture, but also in industry and service. They were supplied through both intrusive and extrusive sources (birth, trade, self-sale, exposure, enslavement for crimes and delicts, captivity), allowing their demographic impact to be every bit as robust as that of the high Roman Empire. They were treated with marked severity, even if the spectrum of statuses attested in Visigothic society subjected other subaltern groups to related abuses. Over time, and particularly under the reign of Chindaswinth and his son Recceswinth in the mid seventh century, a series of mitigations were introduced that softened the hard edges of the slave condition: we know of stronger recognition for slave marriages, greater control of peculium, the right to sue and to offer testimony in court, and the prohibition on slave killing and mutilation. It is also in this period that the reliance on slaves as military combatants increased. Around the same time, slaves and their descendants came increasingly to be tied to the land on which they were born, particularly insofar as manumission began to represent not

178

179

Church influence, not discussed in detail here, included an emphasis on the personhood of the slave (see notes 139–44 on slave killing and mutilation), as well as the ransoming of captives from slavery (Isid. Eccl. Off. 2.5.18; V Patrum Emeretensium 4.10.2; Fruct. Brac. Reg. Comm. 9, 11 [PL 87.1117–20]) and the use of the church as a place of asylum (LV 5.4.17; Conc. Ilerdense a. 546 can. 8; Conc. Bracarense II a. 572 can. 47; Conc. Toletantum XII a. 681 can. 10). On the rise of bound tenancy in late Roman North Africa see Lenski 2017. And in Byzantium, Lenski 2021.

280

Noel Lenski

a passage to full freedom but rather an introduction to a form of semiservility with ongoing obligations to the patronus and his family. The trends attested in the written sources also seem to coincide with patterns we see in settlement archaeology. The sixth century appears to have represented a watershed in Iberian agricultural settlements beyond which more concentrated habitation, which had formerly clustered in and around late Roman villae, gave way to dispersed, small scale villages that show few signs of central planning and appear to have shared capital resources such as bread ovens and storage silos as community resources. The growth of this peasant society, focused on subsistence rather than surplus generation, probably influenced the legislators as they blurred firm status distinctions between slave and free and increased the rights of servi from the sixth century onwards.180 But their new impediments on free labourers combined with the Visigothic tendency to enslave individuals as compensation for delicts appears to have introduced shifts in status that had a long afterlife in the history of peasant dependency in the Luso-Hispanic world. It was, for example, the attribution of heritable guilt (mals usos) that constituted the legal basis for permanent enserfment in high medieval Catalonia.181 Moreover, the robust legal structures enforcing slaveholding that were set in place in Visigothic law also had a long tail, for the ongoing recourse to Visigothic law – often through its thirteenth-century vernacular adaptation, the fuero juzgo – left ample space for the ongoing practice of slavery on the Iberian peninsula after the Reconquista.182 Ultimately, it could be argued, it was the Visigoths who endowed the Luso-Hispanic world with the fundaments of slave culture it would import to the New World, a different story, to be sure, but one that also deserves telling.

180

181 182

The archaeology is consistent, even if the chronology varies regionally from transformations beginning in the fifth century to the seventh. See above all the essays in Quirós Castillo 2009; cf. Quirós Castillo 2007; Tejerizo García 2013; 2016. Freedman 1991. Verlinden 1934; Blumenthal 2009; Phillips 2013; Rio 2017, 64–6, 145–52.

13 Sinner, Slave, Bishop, Saint: The Social and Religious Vicissitudes of St Patrick Judith Evans Grubbs

I, Patrick, a most uneducated sinner and the least of all the faithful and most contemptible among many, had as father a certain Calpurnius, a deacon, the son of the priest Potitus, who was from the village of Bannauem Taburniae; for he had a small villa near there, where I was captured. I was then about sixteen years old.1

Thus begins one of the most unusual and intriguing documents of late antiquity, the Confession of St Patrick, describing the circumstances in which a Romano-British teenager was taken from his home and sold into slavery in barbarian Ireland. The Confessio is Patrick’s defence against criticisms by certain members of the British church hierarchy who were attacking his past and his missionary efforts in Ireland.2 It was written sometime in the second half of the fifth century, but is plagued by chronological vagueness, references to unknown places, and a Latin that has suggested to some it was not his native tongue.3 Nevertheless, the Confessio is still the closest thing we have to a slave narrative from antiquity.4 It is one of 1

2 3

4

Patr. Conf. 1: Ego Patricius peccator rusticissimus et minimus omnium fidelium et contemptibilissimus apud plurimos patrem habui Calpornium diaconum filium quendam Potiti presbyteri, qui fuit vico +bannavem taburniae+; villulam enim prope habuit, ubi ego capturam dedi. Annorum eram tunc fere sedecim. Latin text is from Bieler 1952, vol. 1, 56. All translations are my own unless otherwise noted. For the attacks on Patrick, see Nerney 1949, 501–7; O’Loughlin 2005, 79–95. See Mohrmann 1961, 9–12 and 45–7; and Herren 2013, 15. Patrick’s original language would have been British (Brittonic/Brythonic), but after he came to Ireland, he learned Old Irish. Cf. Confessio 9, cited by Koch 2003, 103 to mean he was a Brittonic speaker. However, Adams 2016, 483–4, finds very little evidence for bilingualism in Patrick’s writings and shows that all of his language can be attributed to biblical uses, with some characteristics of ‘living Latin’. Noel Lenski has suggested that Jerome’s Vita Malchi (which purports to be the life story of the former slave turned holy man, Malchus, as told to Jerome) is a slave narrative. Malchus’ story, however, is mediated through Jerome, whereas Patrick addresses the

281

282

Judith Evans Grubbs

only two authentic writings of St Patrick, the other being the earlier Epistola ad milites Corotici (Letter to the Soldiers of Coroticus), addressed to the soldiers of a British warlord who had attacked and enslaved newly baptized Irish converts.5 Everything else we think we know about Patrick is myth, the product of early medieval Irish promoters of the saint and his cult places in Ireland.6 In the almost two centuries between Patrick’s death (probably around the mid-to-late fifth century) and the first extant reference to him in an Irish hymn, we hear nothing about him.7 Patrick appears frequently in scholarship on late Roman Britain and receives brief mentions in recent work on early medieval slavery, but his testimony has been ignored by scholars of late Roman slavery.8 Indeed, Patrick is conspicuously absent from most recent work in the evergrowing field of late antiquity. This is probably because of his geographical marginality, and because it was assumed that Roman forces pulled out of Britain in the first decade of the fifth century, and therefore post-Roman Britain was not really part of the late antique world that evolved out of the late Empire. But recent archaeological work has changed our knowledge of Britain in the fifth and sixth centuries, and refuted this picture of a ‘Dark Age Britain’ bereft of any vestiges of its Roman past and quickly overcome by Anglo-Saxon invaders.9 Later medieval accretions to

5

6

7

8

9

reader directly. Hayes-Healy 2005 compares Patrick’s depiction of his conversion experience as a shepherd to Malchus’ and sees it as an ‘image and a symbolic understanding of the shepherd’s life that is found frequently in ascetic discourse’ (243). On the Epistola ad milites Corotici see Fear 2007a, who thinks it was also meant to be read by the British ecclesiastical hierarchy, who had failed to penalize Coroticus for his actions. The Latin text is in Bieler 1952, vol. 1, 91–102. See especially Binchy 1962; also Stevenson 1990. On the medieval Patrick, see O’Loughlin 2005, 98–130, and the sources collected in De Paor 1993. Most notable is the Life of St. Patrick by Muirchú moccu Machthéni (Latin text in Hood 1978) from the end of the seventh century. The earliest known reference is the hymn Audite Omnes Amantes, found in a manuscript of the late seventh century and probably composed in the early seventh century; see Orchard 1993. Patrick is not mentioned in Harper 2011 or Garnsey 1996, or in the work of MacMullen 1987, Whittaker 1993b, or Ste. Croix 1998 [1983] on slavery in the late Roman world. Medievalist mentions: Rio 2017, 30–2; McCormick 2017, 249–50. Blackburn 1997, 37–8, puts Patrick in the context of ‘The Old World Background to New World Slavery’. The only extended discussion of what slavery meant to Patrick that I know of is McLuhan 2001. Thompson 1985, 17–18, dismisses the possibility that Patrick’s personal experience affected his view of slavery in general. See Crabtree 2018, 18–49; Haarer 2014; Collins 2012; 2013; synthesis in Fleming 2010, 31–8; also Dark 1999; 2000; 2004. Note especially the continued use (though in a muchreduced state) of Roman settlements at Birdoswald and Wroxeter. Dark is a bit overoptimistic: see Higham 1994 and 2004. Faulkner 2000 has a more pessimistic take.

Social and Religious Vicissitudes of St Patrick

283

Patrick’s life and teachings have also obscured the historical Patrick and cast doubt on his relevance for understanding the late Roman world. But he was a product of that world, and his first-person writings offer a window on how a British provincial experienced the chaotic world of the fifth-century West. Moreover, as I shall argue, Patrick’s experience as a slave shaped his idea of himself as sinner and Christian, and motivated his famous mission to the very people who had enslaved him. At the same time, his story of capture, enslavement, escape from slavery, and return home shares similarities with accounts of enslavement elsewhere in the early-fifth-century Empire, and adds to our understanding of slavery in the late Roman West.10

patrick’s background Scholars have argued over almost every aspect of Patrick’s biography: the date and place of his birth, his educational background, the circumstances behind his innovative mission to the Irish, and his feelings about the Roman Empire and Romanitas.11 Estimates of his birthdate range from the last quarter of the fourth to the mid fifth century, but most scholars now agree that his floruit came in the middle decades of the fifth century.12 The following is my reconstruction, based on a close reading of Patrick’s writings and of much (though by no means all) of the massive Patrician scholarship that has accrued in the past century. Like all of the dozens of attempts to place Patrick in time and place, it is open to challenge on almost every point.13

10

11 12

13

Similarities with other accounts: cf. Aug. Ep. 10* (see below and the chapter by Kahlos in this volume); Thdt. Ep. 70. In Evans Grubbs 2021 I place Patrick in the context of late antique enslavement of children. On this last question, see especially Jones 1996, 112–20; Fisher 2006. See Dumville 1993a. Koch 2003 argues for an ‘early’ St Patrick in the fourth century, pointing out that Patrick does not mention the pull-out of Roman troops from Britain, Pelagianism, or the Anglo-Saxons. Koch identifies Patrick, quite implausibly, with Patricius, the fisci patronus of the usurper Magnus Maximus, who oversaw the execution of Priscillian. Flechner 2019, 33–7, also prefers an ‘early’ Patrick, to fit his argument that Patrick went to Ireland voluntarily, in order to escape curial duties required of him under Roman law (which would no longer have been operative after 410). See the following note and notes 27–8 on Patrick’s status. Out of the vast reservoir of Patrician studies, I have drawn most on Thomas 1979; 1981, 295–346; Hanson 1983; Thompson 1985; Dumville et al. 1993; and O’Loughlin 2005. Jones 1996, 112–20 also has interesting observations. A good review of the scholarship and controversies can be found in Etchingham’s introduction to the 1999 edition of Thompson 1985. Flechner 2019 is an engaging and well-researched account of both

284

Judith Evans Grubbs

Patricius, the man known today as St Patrick, was probably born around the year 400, give or take a few years, in what was then still a Roman province. But by the time he was taken captive sixteen years later, this official Roman presence was gone, after a revolt of the Britons against Rome and the cessation of imperial military and administrative support, both of which occurred in the first decade of the fifth century.14 Patrick makes no mention of these events, and may have known nothing about them. However, the Britain to which he returned after he escaped from slavery in Ireland (and an indeterminate period of wandering, in either Gaul or Britain, about which he is exceptionally vague15) was no longer part of the Empire. Rather, it was in a state of limbo between ‘Roman’ and ‘English’ – what used to be called ‘Dark Age Britain’ and is now known as ‘post-Roman’ or ‘sub-Roman’ Britain. There was still, however, a viable ecclesiastical presence in post-Roman Britain, as we know not only from Patrick’s own writings – the Confessio was clearly meant as a response to criticism from the British clergy – but also from a late-fifth-century account of a visit by St Germanus of Auxerre to southern Britain in 429. Germanus had come to combat the teachings of the British-born Pelagius, who had recently been condemned as a heretic by the Pope and whose followers had caused such turmoil in the British church that the Catholic leadership had asked Germanus to intervene. While in Britain, Germanus supposedly engineered the miraculous defeat of an attack by Pictish and Saxon marauders.16 Patrick’s writings make no mention of Pelagianism or St Germanus, or indeed of the Saxons, although

14

15 16

Patrick’s own writings and the fictional but enduring ‘Patrick’ of medieval and early modern sources. However, his novel (in both senses of the word) hypothesis that Patrick invented a story of captivity in Ireland to disguise a deliberate self-exile goes well beyond any evidence (see note above). Our main source for this is the confused version in Zos. Hist. nov. 6.2.1–2 (ed. Paschoud 1989, 5–6): army in Britain revolts; usurper Constantine (III) crosses over to Gaul (presumably with army); 6.5.2–6.6.1 (Paschoud 1989, 8–9): revolt of Britons; and 6.10.2 (Paschoud 1989, 13): Honorius tells Britons to fend for themselves. Interpretations differ widely, depending on whether literary or archaeological evidence is privileged. For a range of views on the events in Britain in the early fifth century, see Thomas 1981; Wood 1987; 2004; Millett 1990; Esmonde Cleary 1990; Jones 1996; Snyder 1998; Dark 1999; 2000; James 2001; and the essays in Haarer 2014. Esmonde Cleary 2013 compares southern Britain in the fifth century to a ‘collapsed state’, such as Somalia in the early twenty-first century. Patr. Conf. 19–23; see Thompson 1985, 22–7. Constantius, Vita sancti Germani 12–18. Constantius also tells of a second trip by Germanus to Britain, evidently in the 440s (ibid. 25–7), but the historicity of this visit has been doubted; see Barrett 2009. Constantius makes no mention of Patrick; presumably he was not aware of him.

Social and Religious Vicissitudes of St Patrick

285

he does refer contemptuously to ‘apostate Picts’ in his letter to the soldiers of Coroticus.17 This must be because, even when he was not in Ireland, he lived far from the eastern and south-eastern areas of Britain visited by Germanus and most affected by the Anglo-Saxon migrations. The villula of Patrick’s family was in a location accessible to Irish pirates; scholars have variously argued for Northumbria or the West Country or Wales.18 More than four decades ago Charles Thomas suggested that Patrick’s home was not far inland from Carlisle in the vicinity of the Hadrian’s Wall fort of Birdoswald, whose Roman name was ‘Banna’, thus explaining the mysterious (and ungrammatical) village (vicus) of ‘Bannavem Taburniae’ recalled by Patrick at the beginning of his Confessio.19 His hypothesis receives support from recent archaeological excavations showing continued occupation in several of the forts along Hadrian’s Wall in the ‘sub-Roman’ period of the fifth century. Birdoswald is particularly interesting, for there archaeologists have found that one of the fort granaries was converted into a timber ‘hall’ after the building had begun to collapse in the later fourth century.20 Other sites along the Wall indicate post-Roman refortification, presumably by local Romano-Britons against British kingdoms in what is now southern Scotland.21 The material evidence suggested to Kenneth Dark that this area remained the locus of ‘high-status occupation’ by British inhabitants.22

17

18 19

20

21

22

Patr. Ep. 15; also Patr. Ep. 2 and 12, where Picts are referred to (negatively) along with Scotti (= Irish). Nerney 1950, 21–6, argues that Patrick knew of and was refuting Pelagian views espoused by the British clergy who opposed him. But in fact, his evidence really shows how influential Paul’s writings were on both the anti-Pelagians and (quite independently of them) Patrick. Dark 1993 suggests the villula was in the West Country; cf. also Higham 1994. Patr. Conf. 1 (epigraph above); Thomas 1981, 310–14. Thompson 1985, 10–11, disagreed ‘because slave-raiders are unlikely to have launched a major raid under the very noses of the garrison troops’. But an attack on an isolated villula, probably near the coast, is not a ‘major raid’ and Augustine’s report of slave raiders along the coast of north Africa in the same period (see below) shows that brazen slaving activity occurred elsewhere in the western provinces. Wilmott 2000. However, Dark 2004, 196–9, proposed a somewhat different scenario: abandonment of the fort at Birdoswald in the fourth century, with consequent ruin of the granary, followed by non-military occupation in the early fifth century, then a break and later reoccupation by different people. Collins 2012, 32–5 and 101–6; Birley 2014. Collins 2013 interprets continued occupation along the Wall as evidence of former imperial frontier soldiers (limitanei) continuing to man the forts as local post-Roman ‘warriors’. Dark 1992; cf. Dark 1993, 23: ‘what has often been dismissed as “squatter occupation” represents a final phase of “high-status” occupation of villas’. But cf. Higham 1994; 2004.

286

Judith Evans Grubbs

Patrick’s villula was inhabited not only by his family but also a number of slaves, who were seized by Irish raiders along with Patrick.23 However, we should not visualize an elegant villa with elaborate mosaics like those found in southeast Britain in the fourth century.24 Rather, Patrick’s home was probably an agricultural estate, prosperous enough to attract raiders. It may have been reduced to the stage of habitation that archaeologists have called ‘squatter-occupation’, but that does not mean that Patrick’s family were ‘squatters’ – on the contrary, his father was not only the possessor of human chattel, but also held both religious and secular office. At the opening of the Confessio, Patrick declares his Christian pedigree: his father was a deacon and his grandfather was a priest. He adds, however, that he himself was ‘ignorant’ of God and so was enslaved for his sins.25 This may be a later rationalization of his misfortune (found in contemporary Christian writers, as we shall see), but despite his family’s religious offices, Patrick himself may not have had a devout childhood. If he had been baptized, it is unlikely that he would say, ‘I did not believe in the living God, not from my infancy, but I remained in death and unbelief’.26 He reveals more about his background in his other extant work, when he pointedly reminds the soldiers of the British leader Coroticus of his own social position, claiming ‘I was freeborn [ingenuus] according to the flesh; I am born of a decurion father’. Patrick’s family belonged to the Romano-British social and ecclesiastical elite – not by any means comparable in wealth or status to the elite of Italy or Gaul, but nevertheless people of consequence in their community. Moreover, his father owned male and female slaves (servos et ancillas), who fell victim to the Irish raiders at the same time Patrick himself was taken. It is this social position that Patrick later renounced in order to minister to his erstwhile enslavers: ‘I sold my nobility – I am not embarrassed nor do I regret it – for the advantage of others’.27 23 24

25 26

27

Patr. Ep. 10: me aliquando ceperunt et devasterunt servos et ancillas domus patris mei. On which see Scott 2000. I would class Patrick’s villula among the ‘middle range’ of Faulkner’s (2000, 135–7) late Romano-British landowners. Patr. Conf. 1; see below on Patrick’s association of sin and enslavement. Patr. Conf. 27; see Dumville 1993a, 14–15, which suggests that ‘the Christianity of Patrick’s household may have been comfortable and conventional’. Patr. Ep. 10: . . .Numquid a me piam misericordiam quod ago erga gentem illam qui me aliquando ceperunt et devastaverunt servos et ancillas domus patris mei? Ingenuus fui secundum carnem; decorione patre nascor. Vendidi enim nobilitatem meam – non erubesco neque me paenitet – pro utilitate aliorum. . . (‘Did [anyone] ever [ask] from me the pious mercy that I show towards that people who once took me captive and ravaged the male and female slaves of my father’s house? I was freeborn according to the flesh;

Social and Religious Vicissitudes of St Patrick

287

What would curial status have meant in the post-Roman Britain of Patrick’s adulthood? The dozens of fourth- and fifth-century laws enjoining decurions to fulfil their onerous and now hereditary duties would no longer have had any force and the tax collection for which decurions were responsible would have ceased.28 But the prestige and authority held by those of curial status in their communities would have remained, and so Patrick invoked his former secular position in his confrontation with Coroticus, even as he ostentatiously renounced it. Patrick’s writings evoke a society that still identified itself as Roman, with civic and ecclesiastical offices and the continued existence of late Roman education, but that no longer had an imperial military or administrative presence. The difference in his self-definition between his two works is revealing. In the Confessio, when defending himself to the Romano-British church hierarchy who had questioned his fitness to be missionary bishop to the Irish, Patrick gives only his ecclesiastical pedigree, and stresses his alleged lack of education and rhetorical skill compared to his accusers. He presents himself as humble, rusticissimus, in contrast to the dominicati rethorici (sic) opposing him, who nevertheless should listen to a stultus who has received God’s word.29 But when writing to Coroticus, who claimed high rank in the British world, Patrick stresses his status in secular society, not his religious background.30 This difference in self-identification in two works aimed

28

29

30

I am born of a decurion father. I sold my nobility – I am not embarrassed nor do I regret it – for the advantage of others. . .’) I have supplied a subject and verb for the main clause of the first sentence, since Patrick does not. The use of numquid indicates that the answer to his question is ‘no’. Thus Flechner’s suggestion (Flechner 2011; further developed in Flechner 2019) that Patrick returned to Ireland to escape from curial munera, possibly even financing his travels by the sale of family slaves (!), is untenable; see note 12. Patr. Conf. 13: . . .et vos dominicati rethorici audite et scrutamini. Quis me stultum excitavit de medio eorum qui videntur esse sapientes et legis periti et potentes in sermone et in omni re, et me quidem, detestabilis huius mundi, prae ceteris inspiravit si talis essem. . . (‘. . .and you, lordly rhetors, listen and examine yourselves. Who raised me up, a fool, from the midst of those who appear to be wise and learned in the law and powerful in speech and in everything, and even inspired me, hated of this world, above others, if I am such. . .’). Who exactly these dominicati rethorici were, if that is indeed what Patrick wrote, is unclear. Rethorici clearly means rhetorici. Mohrmann 1961, 29–31 thought they were learned clergy, the ones who had been criticizing Patrick; cf. Nerney 1949, 502; McDermott 2004, 26–7. Note that Hanson 1978, 82–3 reads domini cati rethorici (‘lords, clever rhetors’). To judge from certain passages in the Epistola (e.g. Ep. 2), Coroticus had a claim to being a ‘Roman’, even a Christian. Many suggestions have been made as to his home base (e.g. Thomas 1979, 90–1, posits Strathclyde), using evidence from the seventh century and

288

Judith Evans Grubbs

at different audiences suggests that Patrick was more rhetorically sophisticated than his claims of linguistic poverty imply.31 Claims of oratorical inadequacy were regularly made by even the most accomplished speakers in antiquity, including that most skilled of Christian rhetoricians, the apostle Paul, who was a clear literary influence on Patrick.32 Patrick was not the first preacher of Christianity to come to Ireland; he was preceded by the bishop Palladius, sent by Pope Celestine in 431. Palladius’ mission is attested by the Chronicle of Prosper of Aquitaine, which makes no mention of Patrick.33 Scholars generally assume that Patrick’s missionary activities post-date those of Palladius. But he also had a different target, for Patrick was apparently the first Roman to see his mission as aimed at pagan barbarians outside the Empire (as opposed to other missionaries like Ulfilas, who ministered to Christian Romans in Gothic territory).34 By the late seventh century, Palladius had been conflated with Patrick by the church of Armagh, a promoter of the cult of Patrick. But the Patrician documents emanating from early medieval Armagh can tell us nothing about the ‘real’ Patrick.35 For that we must look solely to his own, often enigmatic, writings.

31

32

33

34

35

later. Thompson 1985, 125–43, suggested that Coroticus was a Briton (perhaps even a former Romano-British military commander) based in Ireland, who raided and took captives in Ireland and sold them to the Picts in Scotland and to Scotti in other parts of Ireland. This would explain how Patrick learned so quickly of what Coroticus did with the people and property he took, and why he could claim to have some ecclesiastical jurisdiction over Coroticus. Dumville 1993b finds Thompson’s idea plausible but not certain. Now, however, Fraser 2013 argues (like Thomas) for the western coast of Scotland as Coroticus’ base, specifically Clyde Rock (Dumbarton Rock). Contra Hanson 1983, 36, who says, ‘Patrick’s writing is completely devoid of rhetoric . . . [he] was incapable of writing for effect’. On the other hand, Dumville 1993c, 127, calls the Epistola ‘a powerful, ornate, rhetorical piece of writing’, and Higham 1994, 229, notes Patrick’s ‘self-deprecating’ strategy in referring to his villula. Conybeare 1994 stresses the oral/aural nature of Patrick’s writing: he was meant to be heard by his audience. Now see Adams 2016, 445–84 on Patrick’s Latin, esp. 475–9 on his ‘literary pretensions’ (479). See Fear 2007a. On Paul’s influence on Patrick, see also Nerney 1949; Mohrmann 1961, 6; Shanzer 1993, esp. 196–97; and below. Prosp. Chron. s.a. ad 431: Ad Scottos in Christo credentes ordinatus a papa Caelestino Palladius primus episcopus mittitur (‘Palladius is ordained and sent by Pope Celestine as first bishop to the Irish who believe in Christ.’); cf. Prosp. Contr. coll. 21 (both cited in Charles-Edwards 1993, 1); see O’Loughlin 2005, 37–41. Thompson 1985, 55–65 and 79–94; Fletcher 1997, 82–8. But see Etchingham 1999, xxvii–xxviii. Johnston 2017, 120–3, places Palladius’ mission ‘within [Roman] strategies of frontier management’ (quote at 121). As Binchy 1962 demonstrated. Thomas 2004 thinks that, on the contrary, the Armagh ‘Patrick’ can tell us about the activities of Palladius. On the early medieval Patrick preMuirchú, see O’Loughlin 2005, 96–111.

Social and Religious Vicissitudes of St Patrick

289

enslavement The circumstances of Patrick’s enslavement were far from unique in late antiquity. Ammianus Marcellinus speaks of Irish raiders in Britain in the mid fourth century, about fifty years before Patrick’s capture, and the sixthcentury British writer Gildas describes a crisis in the mid fifth century, several decades after Patrick was taken to Ireland, when Roman Britons faced Irish, Picts, and invading Anglo-Saxons.36 Patrick says he was taken to Ireland along with ‘thousands of others’.37 This vague number must refer not to the particular shipment of slaves in which he was included, but to the number of Romano-Britons who ended up in servitude in Ireland over several decades. That former Romans did comprise a substantial community (or communities) is suggested by the papally sponsored mission to Ireland in 431 of bishop Palladius which preceded Patrick’s missionary activities and was aimed, unlike Patrick’s, at those who were already Christians, rather than at pagan Irish.38 While some of those to whom Palladius ministered may have been Roman traders settled in Ireland or refugees from upheavals in the Empire, others were no doubt descended from captives brought to Ireland in circumstances similar to Patrick’s.39 Archaeological work has shown that there was more Roman interaction with Ireland than previously thought; one scholar has gone so far as it describe it as ‘soft Romanization’.40 Pirates and brigands were a fact of life in the increasingly insecure world of the fourth- and fifth-century West, as central Roman authority waned. Nor was it only in the far-flung province of Britain that children were prey to capture by raiders and enslavement far from home. In 315, the emperor Constantine had fulminated against kidnappers (plagiarii) ‘who inflict on parents the pitiable loss of their (still) living children’ and had imposed harsh penalties on free men and slaves who engaged in 36

37

38

39

40

Amm. Marc. 27.8.5–9, on Irish raids on Britain in mid-fourth century; Gildas, De excidio Britonum (The Ruin of Britain) 21. Patr. Conf. 1: Deum enim verum ignorabam et Hiberione in captivitate adductus sum cum tot milia hominum. . . (‘For I was ignorant of the true God and I was led in captivity to Ireland along with so many thousands of people. . .’). See note 33 above. Thompson 1985, 56, assumes that the Christians in Ireland to whom Palladius was sent were Irish, but they are as likely to have been Romans and others who came as slaves or refugees from Britain or Armorica, or the descendants of traders. See De Paor 1993, 23–37, esp. 35, for the probable existence of communities of enslaved British Christians by the early fifth century. Di Martino 2003, 47. Di Martino also argues that the Romans actually ‘invaded’ Ireland (in the first century under Agricola or in the fourth century), but this is not likely. See Johnston 2017 and Cahill Wilson 2017 for evidence of contacts between the Roman world (especially Britain) and Ireland.

290

Judith Evans Grubbs

kidnapping, the first such law specifically against the kidnapping of children.41 In an incident close to the time of Patrick’s own capture, Augustine of Hippo complained to his fellow African bishop Alypius about the illegal activities of slavers (mangones) in the western Mediterranean. According to Augustine, raiders would swoop down ‘in a herd’ into remote rural areas of the North African coast, yelling and dressed as barbarians or soldiers, and carry off their victims for sale in other provinces. Almost all those so caught, Augustine says, were children (liberi). A few had been sold by their parents, ‘whom they [the mangones] buy not as Roman laws allow, for their services of 25 years, but they buy them straightout as slaves and sell them as slaves across the sea’.42 Augustine relates that ‘very recently’ he had heard of one attack on a certain villula (the same word that Patrick uses for his home), in which the adult male inhabitants were killed and the women and children seized and sold. He had the opportunity to talk with a small girl who had been rescued by his congregation; she said she had been ‘snatched from the home of her own parents’ and when he asked if she had been alone at the time, ‘she replied that it had happened in the presence of her parents and brothers’. One brother, who had come to retrieve her, said that bandits had broken into the house at night, and that when this sort of thing happened, people preferred to hide rather than resist, ‘believing them [their attackers] to be barbarians’.43 Augustine’s congregation in Hippo, in his absence, had learned of a shipment of such enslaved persons and had rushed out and freed almost 120 people, including children.44 Unfortunately for Patrick, however, there seems to have been no one willing or able to rescue him and his fellow captives when they were snatched by the Irish raiders. Looking back, Patrick blamed his enslavement on the godless life he had been leading: For I was ignorant of the true God and I was led in captivity to Ireland along with so many thousands of people – according to our deserts, since we had withdrawn

41

42

43 44

C. Th. 9.18.1 to Celsus, the vicar of Africa. On kidnapping of children, see Harper 2011, 80–1. Kidnapping (plagium) had been a crime since the first century bce under the lex Fabia, which originally dealt with the stealing of another’s slave, but came to be applied to those who kidnapped and forced into slavery a free person; see Robinson 1995, 32–5. Aug. Ep. 10* (ed. Divjak 1987); see also the chapter by Kahlos in this volume. The letter is now dated 428. I translate liberi as ‘children’ rather than ‘free people’ (as others have translated it) because that meaning makes more sense in this context, especially as Augustine immediately goes on to say that only a few were actually sold by their parents. Aug. Ep. 10*.2–3. Aug. Ep. 10*.7.

Social and Religious Vicissitudes of St Patrick

291

from God and did not guard his precepts and were not obedient to our priests, who were warning us for our own safety; and the Lord brought the wrath of his mind upon us and scattered us among many peoples even to the end of the earth. . .45

The belief that widespread misfortune was a result of God’s displeasure at sinful behaviour (not only of individuals but of entire peoples) recurs frequently in the writings of fifth-century Christians who sought an explanation for repeated disasters. The justness of divine chastisement was backed up not only by Scripture but also by the traditional Roman religious rationale that the gods rewarded those who honoured them and punished those who did not. Thus, Patrick’s contemporary, Salvian, thundered against the Gallic Romans whose sinful behaviour and disobedience to God had led deservedly to their defeat and enslavement by the (allegedly) more virtuous Visigoths: Why therefore do you wonder if we are punished, if we are handed over to the law of the enemy, if we are weaker than all? Whether it be our miseries, or our weaknesses, or our upheavals, or our captivities, and our utterly wicked enslavements, they are testimonies of a bad slave and a good master. In what way a ‘bad slave’? Since to be sure, I suffer at least in part what I deserve. In what way a ‘good master’? Since he shows what we deserve, although he does not impose what we deserve. For he prefers to correct us with most clement and most kind castigation rather than that we perish.46

The same cause and effect relationship between sin and slavery appears in the writings of the great preacher and bishop of Antioch, John Chrysostom.47 But the most notable exponent of the idea that enslavement was God’s punishment for sin was Augustine. In Book 19 of the City of God (probably written during the time that Patrick was in slavery), Augustine brilliantly connects original sin, slavery, and the justice of God: The condition of slavery is justly imposed on the sinner. Wherefore we do not read of a slave anywhere in the Scriptures until the just man Noah branded his son’s sin with this word; so he earned this name by his fault, not by nature48 . . . Witness the man of God, Daniel, who in captivity confesses to God his own sins and those of 45

46 47 48

Patr. Conf. 1: Deum enim verum ignorabam et Hiberione in captivitate adductus sum cum tot milia hominum – secunda merita nostra, quia a Deo recessimus et praecepta eius non custodivimus et sacerdotibus nostris non oboedientes fuimus, qui nostram salutem admonebant: et Dominus induxit super nos iram animationis suae et dispersit nos in gentibus multis etiam usque ad ultimum terrae. . . (Bieler 1952, vol. 1, 56–7). Salv. Gub. 4.2. De Wet 2015, 1 (quoting Chrys. Hom. 22 in Eph. 2), 51–64, 103–4, 176. The curse of Noah on the descendants of his son Ham, who had uncovered the drunken Noah’s nakedness, provided a rationale for the much later enslavement of black Africans (assumed to be the descendants of Ham) in the early modern period; see Blackburn 1997, 64–7.

292

Judith Evans Grubbs

his people, and in pious sorrow recognizes in them the cause of his captivity. The prime cause of slavery, then, is sin, so that man was put under man in a state of bondage; and this can be only by a judgement of God, in whom there is no unrighteousness, and who knows how to assign divers punishments according to the deserts of the sinners.49

Patrick would not have read Salvian or De civitate Dei; whether he had read Augustine’s Confessiones is a matter of debate.50 But he had read and thoroughly absorbed the Scriptures, both Old and New Testament, particularly the Gospels of Luke and John, the Psalms, and most of all the Pauline Epistles.51 Quoting the Gospel of John, Patrick tells the soldiers of the British king Coroticus, who had just cut down a group of Patrick’s newly baptized converts, that they would serve in hell along with the devil in eternal punishment because ‘he who does sin is a slave’.52 Some Christians, following a Stoic line of thought, would have taken this metaphorically to mean that the sinner is a slave to sin. But as one who had experienced the consequences of sin for himself, Patrick interpreted Jesus’ words literally.

enslavement as empowerment During his six years in captivity, Patrick’s duties were those of a shepherd. His hours of solitude in the fields gave him ample time to reflect on the 49

50

51

52

Aug.Civ. 19.15, quoted (from LCL) in Garnsey 1996, 216–17. Garnsey 1996, 213–19, discusses Augustine’s innovative and important rationale for slavery. See now De Wet 2018a, 116–19, where Augustine’s discussion of slavery and sin is expounded. Bieler 1952, II, 86, says he ‘might well have known’ Augustine’s Confessiones; Mohrmann 1961, 4–7, and Thompson 1985, 54, dispute this. Dronke 1981 (summarizing earlier views) strongly argues for knowledge of the Confessiones, but is countered by Shanzer 1993, whose arguments against are convincing. In any case, he would not have known De civitate Dei. Reid 2010, 37, sees Patrick and Augustine as sharing ‘similar perceptions of communication between God and men’ that are rooted in the age in which they lived. This is part of a larger, ongoing debate over what non-biblical sources, if any, Patrick used; see Cain 2010, who argues (not completely convincingly) for echoes of Jerome’s Ep. 52 in the Confessio. He seems to have used a text that was ‘partly Old Latin, partly Vulgate, and partly a transitional version’. Bieler 1952, I, 34–5, includes a complete apparatus criticus of all biblical passages quoted. Dumville 1993a, 15–16, sums up Bieler’s conclusions: the mixture of texts used by Patrick indicates that his religious education was gained in the fifth century, not the fourth. But this was presumably as an adult, after his conversion; cf. Howlett 1989. Patr. Ep. 4, quoting John 8:34. Note that Patrick’s Latin is particularly problematic here; he actually says perenni poena gehennam pariter cum ipso [the devil] mancipabunt, quia utique qui facit peccatum servus est et filius zabuli nuncupatur, but surely he means gehennae mancipabuntur or in gehenna mancipabuntur. Therefore I would translate, ‘They will serve in hell with eternal punishment along with [the devil] himself.’

Social and Religious Vicissitudes of St Patrick

293

causes of his enslavement and to develop a deeply personal relationship with God: ‘so every day I grazed the sheep and frequently prayed during the day – more and more my love of God grew and my fear of him and my faith increased, so that in one day (I prayed) up to one hundred prayers and almost the same at night. . .’.53 Like Daniel, who fasted and prayed continuously that God would forgive his people their sins, Patrick saw visions and heard voices in his dreams. One night he was told, ‘You do well to fast, as you will soon return to your own homeland’. And indeed, the next day he escaped and walked some 200 miles across Ireland and found a ship.54 Ultimately, he returned to his parents in Britain after ‘a few years’, including another enslavement of sixty days, to which he only vaguely refers.55 His family asked him never to depart again, but he was drawn back to the place of his servitude. He returned to Ireland in response to another dream in which he saw a man named Victoricus carrying ‘innumerable letters’ containing ‘the Voice of the Irish’ (vox Hiberionacum) begging him to return to them: ‘We ask you, holy boy, to come and walk among us again.’56 This is reminiscent of the apostle Paul, who had a vision in the night of a Macedonian man standing over him and asking him to ‘come to Macedonia and help us’.57 Enslavement was not only a punishment for Patrick; it was an opportunity. He had been enslaved for a reason: so that even after his escape, he 53 54

55

56

57

Patr. Conf. 16. Patr. Conf. 17: deinde postmodum conversus sum in fugam et intermisi (lit., ‘I neglected’, an odd choice of words to describe his flight from his master) hominem quo fueram sex annis et veni in virtute Dei (‘then later I turned to flight and I neglected the man with whom I had been for six years and I came into the virtue of God’). Patrick’s escape from his master evidently bothered his seventh-century hagiographer Muirchú, who invented an episode where Patrick returns to compensate his master for his loss and convert him to Christianity. But his former master cannot bear the thought of obedience to his old slave, and immolates himself in front of Patrick (Muirchú, Vita sancti Patricii 1.11–12, in O’Loughlin 2005, 201–3). Patr. Conf. 21. McDermott 2004, 28–9, suggests this refers to the time Patrick spent with the sailors who gave him passage from Ireland. Patr. Conf. 23, in Bieler vol. 1, 71: Rogamus te, sancte puer, ut venias et adhuc ambulas inter nos. Note that here Patrick calls the Irish Hiberionaces rather than Scotti, although later in the Confessio he refers to filii Scottorum et filiae (‘sons and daughters of the Irish’; Patr. Conf. 41; cf. Patr. Ep. 12) and a benedicta Scotta genetiva nobilis pulcherrima adulta (‘blessed, Irish born, noble, very beautiful young woman’) whom he had converted (Patr. Conf. 42). He uses Scotti to refer to the wicked allies of Coroticus: socii Scottorum atque Pictorum apostatarumque (‘allies of the Irish and the apostate Picts’) (Patr. Ep. 12). In Ep. 16 Patrick says of himself along with his converts: Hiberionaci sumus (‘We are Irish’). See McDermott 2004, 23–4: Scotti are grouped with the apostate Picts, whereas Hiberionaces/Hiberionaci are the Irish whom Patrick himself converted. Acts 16:9; see Nerney 1950, 16.

294

Judith Evans Grubbs

might ‘walk again’ among the pagan Irish, supposedly at their own request. Indeed, Patrick saw himself as having benefited from his chastisement at the hands of God, since through slavery he came to an understanding of God that his ostensibly Christian upbringing had not given him. And he believed the Irish also benefited, for without him they would not have received the word of God and the liberating salvation of baptism. Patrick’s belief that he benefited from slavery recalls the Aristotelian idea that slavery can be beneficial for the natural slave. The Hellenistic Jewish philosopher Philo of Alexandria developed this line of thought further specifically in regard to the biblical enslavements of Canaan (son of Ham) and Esau.58 Patrick obviously would not have read Philo (let alone Aristotle), although Philo’s rationalization of the enslavements of Canaan and Esau did apparently influence the fourth-century Christian thinkers Basil of Caesarea and Ambrose of Milan.59 Augustine himself agreed, in his discussion of Cicero’s De republica, that ‘slavery is useful for some people’.60 Patrick’s views on slavery and sin, however, clearly derive from his spiritual model, Paul, particularly the apostle’s letters to the Corinthians.61

conclusion By his own account, Patrick’s mission was a great success; he baptized ‘many peoples’, including a beautiful Irish noblewoman who decided to dedicate herself to holy virginity.62 Patrick encouraged Irish Christians to become holy virgins and monastics, a decision that in the case of young women (he says) happened against their parents’ will, so that they faced ‘persecution’ from their families.63 He travelled with the sons of Irish kings who offered him gifts according to their custom; he refused to accept the gifts, but nevertheless came under criticism from British church leaders.64 58 59 60

61

62 63 64

See Garnsey 1994; 1996, 157–72. Garnsey 1996, 45–7 and 195–9. On the ‘curse of Ham’, see De Wet 2018a, 104–45. Aug. Civ. 19.21: . . .quibusdam esse utilem servitutem (‘as it is useful for all to be enslaved to God’). Nerney 1949, 498. See Shanzer 1993 and Mohrmann 1961 on Patrick and Paul. McLuhan 2001 talks about Paul’s view of slavery as an influence: ‘a powerful paradigm for the imitation of Christ’ (63). See Avdokhin in this volume for the impact of Paul’s metaphor of slavery in a very different Christian culture. Patr. Conf. 38 and 42. Patr. Conf. 42. Patr. Conf. 37. Patrick’s indignant defence of his conduct as a missionary (esp. Patr. Conf. 48–54) has many similarities to Paul’s defence of himself in his letters to the Corinthians: see esp. 1 Cor. 9:3–19 and 2 Cor. 10–12. I thank John Boyles for this observation. See also Flechner 2019, 168–9.

Social and Religious Vicissitudes of St Patrick

295

But he remained well aware of the reality of sudden and unexpected capture, and he particularly empathized with Christian women who had been enslaved, as he was well aware that they would be subject to sexual abuse and special suffering: ‘continuously they endure terrors and threats without end’.65 He also worked for the redemption of those who had been taken captive by barbarians, as did other Christian leaders in the troubled times of late antiquity. Ransoming of captives, particularly women whose chastity was at risk, had been considered a duty of bishops from the time of Cyprian of Carthage in the mid third century. The sixth-century bishop Caesarius of Arles, a few decades after Patrick, even redeemed nonChristians (barbarians and Jews), probably at least in part with a view to their conversion.66 Patrick in fact knew of the redemptive activities of bishops on the continent, for he tells Coroticus: (This is) the custom of the Roman and Gallic Christians: they send suitable holy men to the Franks and other peoples with many thousands of solidi for the purpose of redeeming baptized captives. But you kill them and you sell them to a foreign people ignorant of God, as if you are handing over the limbs of Christ to a brothel . . . Where sin clearly, seriously, and shamelessly abounds, there freeborn men have been sold, (and) Christians have been led into servitude, especially servitude to the most unworthy and apostate Picts.67

This passage explains how Patrick could both accept servitude as the just recompense for sin and yet protest vehemently against the slaving practices of Coroticus. For those killed or enslaved by Coroticus’ men were definitely not sinners; they were newly baptized Christians, the oil of the chrism still fresh on their faces.68 Enslavement of sinners like the teenage Patrick or (presumably) all pagans was not only just punishment but could be beneficial, but enslavement and abuse of those who obviously were without sin was an abomination – especially when they were enslaved not to other Christians but to unbelievers like the ‘apostate’ Picts.69 Disapproval of Christian slaves serving non-Christian masters appears in imperial laws of 65

66

67 68 69

Patr. Conf. 42. This should be distinguished from the concern of late antique bishops that a master’s sexual exploitation of his slaves damages the master’s soul. Herren 1989 notes Patrick’s sympathy for female converts dedicated to virginity but fails to see that it is explicitly enslaved women for whom Patrick has the most concern. That enslaved women were especially subject to sexual abuse is also a theme of American slave narratives. See Klingshirn 1985. Redemption of those of one’s own faith who were enslaved by nonbelievers was a practice in later periods also; see especially Bensch 1994. Patr. Ep. 14–15. Patr. Ep. 3. Although ‘apostate’ would imply that the Picts were lapsed or heretical Christians, it is likely that it is here merely a term of abuse; see Dumville 1993d. In Ep. 12 (see above)

296

Judith Evans Grubbs

the fourth and fifth centuries that forbid Jews to own Christian slaves.70 In fact, the idea that those outside the religious community could rightly be enslaved goes back to Jewish law, and is essentially an application of the idea found in many slave-owning societies that the slave is (or should be) an outsider.71 Ultimately, in the medieval and early modern periods, this would lead both Muslim and Christian slaveholders to turn to the (mostly) ‘infidel’ (i.e. ‘pagan’) populations north of the Black Sea and in sub-Saharan Africa for their slave supply.72 Not that this was Patrick’s perspective: his concern was with these new Christians that he knew were sinless and thus had been undeservedly enslaved, and for whom he felt personal responsibility. Patrick’s years as a slave shaped not only his perspective on slavery and his feelings for others who were held in bondage, but also his interpretation of the Christian message and his mission to deliver it to his former enslavers. Yes, slavery was the consequence of sin, as the teenaged Patrick had discovered to his cost. But on the other hand, the Christian God was not a respecter of persons, and a former slave of limited literacy had as much right to speak for this God as did the comfortable, educated British ecclesiastics (dominicati rethorici) whom he saw as his critics.73 With his intimate knowledge of the letters of Paul, Patrick would have recalled how Jesus had ‘emptied himself, taking the form of a slave, being born in the likeness of men; and being found in human form he humbled himself and became obedient even unto death, even death upon the cross’.74 Patrick himself longed to suffer martyrdom as the ultimate ‘imitation of Christ’. But barring that, he could at least say that he had imitated his God in his service to the Irish.75 The apostle Paul, who styled himself ‘the slave of Jesus Christ’, had declared that ‘being free from all, I have enslaved myself to all, in order to win over the more’.76 Paul’s metaphor of self-imposed status reversal

70

71

72

73 74 75 76

Patrick despairs because Coroticus is a ‘betrayer [traditor] of Christians into the hands of the Irish [Scotti] and Picts’ – clearly these Scotti and Picts are not Christian. See C. Th. 16.9.1–5. C. Th. 16.9.3–4 does allow Jews to have Christian slaves on condition that they do not try to convert them to Judaism. Leviticus 25:44–6 and Philo Spec. 2.123, both quoted in Garnsey 1996, 26–7. See Patterson 1982 on the slave as outsider; see also Hezser’s chapter in this volume. Blackburn 1997, esp. 31–93; Origo 1955; Epstein 2001; cf. Fynn-Paul 2009 (who is in my view too optimistic and positivist about the actual impact of Christianity on slave-taking). See note 29. Philippians 2:7–8, as quoted in Garnsey 1996, 185 with discussion. Patr. Conf. 59. 1 Cor. 9:19; Romans 1:1. See Garnsey 1996, 173–88. As Garnsey 1996, 18–19, notes, Christian theology (with Paul as its starting point), while agreeing with the Stoic idea of enslavement to sin, differed in that Christians also posited a ‘good’ type of slavery –

Social and Religious Vicissitudes of St Patrick

297

would have had a much different significance for Patrick than for other church leaders. For him, enslavement was a reality, not a metaphor: he had gone from a literal enslavement to willing self-subordination to his former enslavers and become ‘a slave in Christ to a foreign people’.77 Paradoxically, his slave experience had empowered him and made him uniquely suited to preach to both slaves and enslavers, to comfort and to chastise. It was an enslavement that was to have tremendous consequences for the later history of Christianity.

77

slavery to God – which they took from the Old Testament. Cf. Romans 1.1 and see McLuhan 2001. For Chrysostom’s use of the slavery to God/Christ metaphor, see De Wet 2015, esp. 56–74. Patr. Ep. 10.

14 Slave Boys in Paradise? The Text of the Quran and Its Later Exegetes Ilkka Lindstedt

The issues of slavery in classical Islamic law and thought as well as the social reality of slaves and ex-slaves (clients) in the medieval Near East have been gaining increased attention in recent decades.1 Much remains to be done, however. For instance, I am not aware of any specific study (book or article) focusing on slavery and slaves in the Quran or Quranic environment. I will begin this chapter by briefly discussing the Quranic passages that have a bearing on slavery in the late antique Arabian context before moving on to deal in detail with one specific verse, Quran 52:24, which seems to suggest that, in fact, there might be slaves in the afterlife serving the believers. The crux of the matter is the interpretation of the expression ghilmā n lahum, which could be understood as ‘slave boys belonging to them’, ‘young servants of theirs’, or simply ‘children of theirs’. This study endeavours to situate and analyse the Quran’s discourse on slavery, and later Islamic understandings of it, in the late antique Arabian context. The Quran’s depiction of heavenly servants in verse 52:24, it will be argued, is highly original. Moreover, the article will trace the reception of this verse in exegetical works. The later Muslim commentators of Quran 52:24 put forward differing interpretations of the verse but, in particular in the late medieval era, understood it to refer to slaves that

1

I thank Katja von Schöneman and the editors of this volume for important comments on an earlier draft of this study. See, e.g., Ayalon 1971; Crone 1980; Mottahedeh 1980; Pipes 1980; Bacharach 1981; Pipes 1981; Murray Gordon 1989; Lewis 1992; Ayalon 1994; Brockopp 2000; Matthew Gordon 2001; Segal 2001; Bray 2004; Meouak 2004; Bernards and Nawas 2005; Athamina 2007; Richardson 2009; Rotman 2009; Ali 2010; Caswell 2011; Culbertson 2011; Urban 2012; J. Brown 2017; Franz 2017; Matthew Gordon and Hain 2017; Brown 2019.

298

Slave Boys in Paradise? The Text of the Quran

299

serve the believers in the hereafter. Intriguingly, some Quranic commentators, moreover, appear to suggest that these paradisal slaves might be eunuchs.

the late antique arabian context Jonathan Brown notes: ‘Slavery was an integral part of economic and social life in the Late Antique world, and the Quranic revelation accepted it as a reality’.2 But what was this late antique context, especially as regards Arabia? Although we lack any details in contemporary sources of slavery in the pre-Islamic Hijaz (where the Prophet Muhammad lived and died around 632 ce), we have quite a considerable number of texts dealing with capturing slaves in late antique Syria by North Arabians (often called ‘Saracens’ in Greek and Latin texts and Ṭayyā yeˉ in Syriac). These sources have been surveyed and analysed by Noel Lenski.3 Lenski’s study offers a lucid overview of cases reported in the sources in which North Arabians captured slaves, particularly in Syria. Such instances are relatively common in the late antique sources: for example, Saracens reportedly captured Christians near Fayyum in 291 ce; Jerome (writing around 390 ce) mentions meeting one Malchus, who had been a slave among the Saracens but who managed to escape; and in 509 Saracens invaded the provinces of Arabia and Syria, taking captives.4 North Arabian groups also warred against and took captives from each other.5 As elsewhere, the uses to which these slaves were put varied: ‘interbreeding or adoption into the tribal unit, targets for violence or human sacrifice, exchange for ransom, sale to neighboring sedentarists’,6 and so on. Let us turn our sights to the south. Although we have surviving (and in some cases dated) epigraphic data from the late antique Himyarite 2 3

4 5 6

J. Brown 2017, 108. Lenski 2011c. I have to note, however, that in my opinion Lenski’s exposition is inadequate when discussing ethnicity and social history: he treats ‘Saracens’ as constituting a subgroup among Arabs characterized by nomadism. It has recently been argued by Peter Webb (2016) that there is scant evidence of Arab ethnogenesis in the pre-Islamic era and, in any case, most Arabians (a geographical term) were settled rather than nomadic. Moreover, ‘Saracens’ was an exonym not used by the community itself, so the actual existence of such a group and the characteristics assigned to it have to be approached critically. See Lenski 2011c, 237–9, 243, 245–6. Lenski 2011c, 247. Lenski 2011c, 239.

300

Ilkka Lindstedt

kingdom of Yemen (the third to sixth centuries ce), it unfortunately does not provide many details about slaves, slavery, or the slave trade there.7 It appears that the corpus of lapidary inscriptions is not illuminating in this regard. Fortunately, South Arabian texts written on wooden sticks offer some information about some aspects of slavery in Yemen. For instance, one such text is a legal document in which the handing over of twin slave girls is detailed.8 According to this and other documents, it appears that, in pre-Islamic Yemen, children born of a slave mother were treated as slaves. Later, Islamic law takes a different stance (although this is not yet detailed in the Quran) on the basis of the idea that status and religious identity descent patrilineally: if a male Muslim master has sexual relations with his female slave and a child is born, the child is a free Muslim.9 The later Arabic historiographical sources contain many narratives where slaves and slavery are present during the life of the Prophet.10 Given that they are non-contemporary texts, we cannot use them as definite evidence of what took place when Muhammad lived. However, they do not clash with Quranic and non-Arabic evidence. Hence, we can say that the picture that they present of an Arabia where slavery was commonplace is credible. Many early Muslims owned slaves according to the Arabic historical texts. Indeed, the Prophet Muhammad is portrayed as owning (and in many cases manumitting) slaves.11 Gilli-Elewy notes that the texts depict different sources of slaves: Apart from the main two sources of enslavement – slave trade and warfare – other causes of enslavement are mentioned in pre-Islamic Arabia, such as debt slavery, sacrificial enslavement, selling oneself or one’s children, kidnap, and enslavement as punishment. Muhammad prohibited debt as a source of enslavement, just as he banned selling one’s own children and sacrificial enslavement to deities and shrines.12

After this short survey of contemporary non-Arabic and noncontemporary Arabic sources, let us turn to the Quran and analyse its terminology and notions of slavery.

7

8 9 10

11 12

This is according to my digital searches in Corpus of South Arabian Inscriptions (http:// dasi.cnr.it/) with the word ʿbd, meaning servant or slave in Sabaic. Last accessed 7 August 2020. Stein 2015. Puente 2013, 39. For a summary of these texts, see Gilli-Elewy 2017; for some prophetical traditions, see also J. Brown 2017, 108–13. Ibn Saʿd, Ṭabaqā t, I/2, 179–80. Gilli-Elewy 2017, 166.

Slave Boys in Paradise? The Text of the Quran

301

the quranic terminology used for slaves The vocabulary employed in Quranic Arabic to refer to slaves is rather different from other surviving Arabic literature. The word jā riya, which is the most common term in later stages of Arabic to refer to a female slave, does not appear in this meaning in the Quran. The word ghulā m (plural ghilmā n), which is the most usual expression to denote male slaves in classical Arabic, does not in most cases designate slave status in the Quran, though this might be the intended meaning in verses 12:19 and 52:24, as will be seen in the next section. The usual expression used for a female slave in Quranic Arabic is ama, plural imā ʾ (Quran 24:32). A male slave is ʿabd (for instance, 2:178), plural ʿibā d. However, both can also designate a human being in relation to God: a ‘servant’ of God. Quran 16:75 contains a longer expression: ʿabd mamluˉ k, ‘an owned slave’. Incidentally, a common word for a slave in classical Arabic is mamluˉ k, which does not appear in the Quran outside 16:75. This shows the differences between the Quran and other Arabic literature when it comes to slaves and related vocabulary, though it must be noted that, more generally, the semantic fields of the words used in premodern Arabic literature in reference to slaves and slavery are often quite wide and the exact meaning ambiguous. Interestingly, the expression used in 2:177, 4:92, 5:89, 9:60, 58:3, and 90:3 is riqā b (sing. raqaba), literary meaning ‘necks’. For example, 9:60 states (as adapted from the translation of Abdel Haleem): ‘Alms are meant only for the poor, the needy, those who administer them, those whose hearts need winning over, for al-riqā b, and help those in debt, for God’s cause, and for travellers in need.’ The word riqā b is often understood, by both medieval and modern commentators, to mean ‘slaves’. However, it could also refer to people captured in battle, so the idea in 9:60 and elsewhere would be that it is a suitable to use alms and other funds to ransom believers who are prisoners of war.13 If, on the other hand, the original audience of the Quran understood the word riqā b as meaning ‘slaves’, then this choice of word might indicate that slaves were often made to wear slave collars around their necks. Moreover, if the reference is to slaves, these Quranic passages could signify a number of things, for example, buying slaves for service in the early Muslim community, buying pious slaves from unbelievers and releasing them, and so on.

13

Brockopp 2006, 58.

302

Ilkka Lindstedt

Another peculiar Quranic expression is mā malakat aymā nukum, ‘what your right hand possesses’, which is a euphemism used in the Quran (4:3, 4:24, 4:25, 4:36, 24:33, 24:58, 30:28) to refer collectively to one’s slaves. Many of the Quranic contexts (in particular the verses in surah 4) seem to suggest that the expression refers to female slaves. The roots of this expression are unclear. Neither it nor anything similar seems to be attested in pre-Islamic Arabian epigraphic materials. However, the Muslim exegetes are unambivalent about understanding it as denoting slaves.14

general notions of slavery Let us now survey some of the more general notions of slavery in the Quran. Some examples cited here are actually grammatically masculine, but their reference is in most cases not gender specific since a group that includes both men and women would be referred to in Arabic with a masculine plural. The Quranic discourse does not problematize slaveholding as such, but approaches it as a common matter that needs regulation but is not forbidden. Verse 16:75 notes that a master and a slave are not equal: God has given the master good provision from which he or she gives alms. Societal hierarchy and slavery are ingrained in this notion. However, a slave owner has to be respectful towards her or his slaves: ‘Be good to your parents, to relatives, to orphans, to the needy, to neighbours near and far, to travellers in need, and to your slaves. God does not like arrogant, boastful people’ (Quran 4:36).15 Freeing a slave is ordained in the Quran for various offences by the master, for example for breaking an oath (5:89) or killing a believer by mistake (4:92). The idea seems to be that by freeing a slave, the master can atone for these sins in the eyes of God. There are also some Quranic 14

15

See, e.g., al-Ṭabarıˉ, Jā miʿ, XVII, 352, commenting on verse 24:58. He quotes opinions that state that, in this verse, the expression means both male and female slaves. The whole verse reads, in Abdel Haleem’s translation: ‘Believers, your slaves [alladhıˉ na malakat aymā nukum] and any who have not yet reached puberty should ask your permission to come in at three times of day: before the dawn prayer; when you lay your garments aside in the midday heat; and after the evening prayer. These are your three times for privacy; at other times, there is no blame on you or them if you move around each other freely. In this way God makes messages clear: God is all knowing, all wise.’ Chris de Wet commented to me (private communication) that this sounds similar to the Christian and Stoic rhetoric about treating slaves justly. This discourse of slavery (about the just treatment of slaves) is, then, common in early Christianity and early Islam.

Slave Boys in Paradise? The Text of the Quran

303

passages that suggest that freeing slaves is beneficial in general: ‘What will explain to you what the steep path is? It is to free a slave, to feed at a time of hunger an orphaned relative or a poor person in distress, and to be one of those who believe and urge one another to steadfastness and compassion’ (90:12–17). This short survey of the Quran has sketched some of the regulations and notions of slavery in the text. The Quranic discourse fits well the general late antique Near Eastern – and more particularly Arabian – context, where slaves as a group were taken as part of a traditional society. As noted above, we have contemporary (or somewhat earlier) evidence from northern and southern Arabia about the existence of slavery.

female slaves and the construction of gender hierarchy In many systems of slavery, one of their aspects has been the use of female slaves as sex partners, without marrying them, by their male masters.16 Some passages of the Quran (23:6, 70:30) seem to suggest that sexual relations between a male master and female slaves are permitted. Moreover, Quran 33:50 addresses the Prophet Muhammad (quoted in part below): Prophet, We have made lawful for you [that is, as wives or sexual partners] the wives whose bride gift you have paid, and any slaves God has assigned to you through war, and the daughters of your uncles and aunts on your father’s and mother’s sides, who migrated with you.

Other passages actually encourage marrying the slaves: ‘If any of you does not have the means to marry a believing free woman, then marry a believing slave’ (Quran 4:25). Verse 2:221 notes that believing slave women are to be preferred as wives to disbelieving free women: Do not marry idolatresses until they believe: a believing slave woman is certainly better than an idolatress, even though she may please you. And do not give your women in marriage to idolaters until they believe: a believing slave is certainly better than an idolater, even though he may please you. Such people call [you] to the Fire, while God calls [you] to the Garden and forgiveness by His leave. He makes His messages clear to people, so that they may bear them in mind. 16

See Ali 2017, who argues that the Quran is ambiguous on this issue but that the later jurists agreed that the master has sexual access to his female slaves with or without their consent. Islamic law does not have a concept of concubinage similar to Roman law.

304

Ilkka Lindstedt

To summarize, these examples show that the Quranic discourse of concubinage between male master and female slaves is ambivalent. It might be understood as permitted – at least for the Prophet – but it was better to properly marry the slave. This was a debated and changing issue at the time, and later Islamic jurisprudence would offer more details and comments based on the Quranic verses and the traditions of the Prophet.17

quran 52:24 and the possibility of paradisal slaves The discussion above illustrated some aspects of slavery and its discourse in late antique Arabia and the Quran. In the remainder of this chapter, I will deal with one specific Quranic verse, 52:24, and its exegetes. This is an interesting passage, since many later Muslim exegetes understood it to contain a reference to slaves that will serve the believers in the afterlife. Quranic surah 52, called al-Ṭuˉ r, ‘The Mountain’, is an early surah. Traditional Islamic exegesis calls it a Meccan surah,18 and Theodor Nöldeke’s categorization places it among the surahs of the first Meccan period (the earliest layer of Quranic revelations).19 The surah emphasizes eschatological urgency as well as heavenly reward and punishment. After describing the last judgement and the fate of the disbelievers, verses 52:17–28 move on to describe the joys of paradise that the believers will receive. I quote them in the translation of Abdel Haleem, leaving untranslated the phrase ghilmā n lahum that is of importance for this chapter: 17: Those who were mindful of God are in Gardens and in bliss, 18: rejoicing in their Lord’s gifts: He has saved them from the torment of the Blaze, 19: ‘Eat and drink with healthy enjoyment as a reward for what you have done.’ 20: They are comfortably seated on couches arranged in rows; We [scil. God] pair them with beautiful-eyed maidens; 21: We unite the believers with their offspring who followed them in faith – We do not deny them any of the rewards for their deeds: each person is in pledge for his own deeds – 22: We provide them with any fruit or meat they desire. 23: They pass around a cup which does not lead to any idle talk or sin. 24: Ghilmā n lahum like hidden pearls wait on them (yat ˉ̣ufu ʿalayhim). 25: They turn to one another and say,

17 18 19

On this, see J. Brown 2017. E.g., al-Bayḍā wıˉ, Tafsıˉ r, II, 433. Nöldeke 2013, 77 [1909–19, I, 94].

Slave Boys in Paradise? The Text of the Quran

305

26: ‘When we were still with our families [on earth] we used to live in fear 27: but God has been gracious to us and saved us from the torment of intense heat. 28: We used to pray to Him: He is the Good, the Merciful One.’

The joys of paradise that those ‘mindful of God’ (al-muttaqıˉ n) will receive are revealed in detail: there will be couches, beautiful-eyed maidens (ḥuˉ r ʿayn), food, and drink. But are there also slaves in paradise in their service? The question hinges on verse 52:24, which in Arabic reads yat ˉ̣ufu ʿalayhim ghilmā n lahum ka-annahum luʾluʾ maknuˉ n, in particular the expression ghilmā n lahum. For instance, Abdel Haleem translates this as ‘devoted youths’, rejecting clear implications of slave status. The word ghilmā n is the plural of ghulā m, which means ‘youth, boy’, but also ‘(male) slave’. It is one of the most usual Arabic words to refer to a male slave,20 although it is also used to denote a young free male.21 The prepositional expression lahum can be interpreted as ‘belonging to them’, ‘for them’, or similar. The word ghulā m (including its declensions) occurs thirteen times in the Quran, verse 52:24 being one of them. In most instances, it means ‘boy’, ‘youth’, or ‘son’, with no indication of slave status. For example, Quran 19:19–20 uses it to refer to Jesus, and 3:40 and 19:7–8 to refer to John the Baptist. The intended meaning is ‘son’ or ‘boy’. Similar usages can be found in verses 15:53, 37:101, and 51:28 (Abraham’s son) as well as 18:74 and 80 (an anonymous boy). However, the word is also used, in verse 12:19, in reference to Joseph. There the usage is more ambiguous and could be understood as both ‘boy’ and ‘slave’. In the passage, Joseph’s brothers have thrown him into the well. Then, a group of travellers comes by: ‘They sent someone to draw water and he let down his bucket. “Good news!” he exclaimed. “Here is a ghulā m!” They hid him like a piece of merchandise – God was well aware of what they did – and then sold him for a small price, for a few pieces of silver: so little did they value him’ (Quran 12:19–20). Since the travellers sell Joseph after finding him, the meaning of the word ghulā m in this passage could indeed refer to Joseph being treated as a slave.

20

21

For it and similar expressions, see Matthew Gordon 2001, 1–8. The usage of the word ghulā m, literally ‘youth’, to refer to a male slave is derogatory. The implication is that, like a child, a slave is under the authority of his master. A similar problem is found in Greek with the term pais. For the usages in late antique Greek sources, see the contribution of Vierros in the present volume.

306

Ilkka Lindstedt

The previous section outlined the words and expressions that are used in the Quranic discourse to denote slave status. The word ghulā m and its plural ghilmā n are not among the common vocabulary for referring to slaves. The contextual reading of the Quran, then, suggests that it is perhaps not probable that the earliest audience of the Quranic revelation understood the phrase ghilmā n lahum in 52:24 as primarily signifying slaves. However, it is not impossible. To begin with, ghulā m is a very common word in (later) classical Arabic for referring to a male slave. Moreover, although the basic meaning of ghulā m is ‘a youth’, the meaning ‘slave(-boy)’ appears to have developed already in the pre-Islamic era. The word occurs (both in its masculine and feminine forms) in some preIslamic Old Arabic inscriptions written in the Safaitic script. In inscription SIJ 750, which I quote here in Al-Jallad’s reading and translation,22 it appears in its masculine form: l ʿm bn z ̣nʾl w ts²wq ʾl- ḥm -h z ̣nʾl f h lt {ʾ}db l -h ġ lm w ʾ{h ˍ }t -h rh ˍ lt w {ṣ}wn f q[y][t] By ʿm son of Ẓnʾl and he longed for his father-in-law Ẓnʾl so, O Lt [a divine name], grant him a servant boy and his sister an ewe-lamb, and preserve and protect.

In this inscription, Al-Jallad translates the word ġ lm (corresponding to the classical Arabic ghulā m) as ‘servant boy’, although later, in the dictionary section of his book, he translates it as ‘slave boy’.23 Both meanings are possible here and the inscription itself is ambiguous in its usage. What is more, ghulā m appears in some jā hiliyya (pre-Islamic) poems to refer to a slave. The authenticity of the jā hiliyya poetic corpus as a whole is in doubt, but the scholarly consensus is that, while some of it might be spurious, it also contains authentic early material. Some of these occurrences of the word ghulā m are mentioned by Nathaniel Miller.24 For example, in one poem attributed to the sixth-century poet Imruʾ alQays, the figure of a ghulā m appears as a ‘high status slave or servant doing all of the work of hunting, while the speaker merely commands him’.25 If these verses ascribed to Imruʾ al-Qays and other pre-Islamic poets are authentic, it means that the semantic broadening of the Arabic word ghulā m to mean not only ‘youth’ but also ‘slave’ had taken place before the life of the Prophet Muhammad. 22

23 24 25

Al-Jallad 2015, 282. In the edition, [x] refers to a restored letter and {x} to a damaged letter. Al-Jallad 2015, 316. Miller 2016, 156–9. Miller 2016, 156–7.

Slave Boys in Paradise? The Text of the Quran

307

We have also seen that Quran 12:19 could be understood to mean that the travellers that found Joseph in the well deemed him a slave. Thus, it is possible that ghilmā n lahum in 52:24 might have been understood by some among the earliest Muslims to denote slaves. Nevertheless, we must leave the case somewhat open for Quranic Arabic in general and verse 52:24 in particular. The next section outlines what the medieval Arabic Quranic exegetes understood the phrase ghilmā n lahum to mean. According to the sources that I have investigated, many Quranic commentaries interpreted it as signifying slave status. Verse 52:24 has failed to receive much attention in modern scholarship. A notable exception is Nerina Rustomji’s recent significant article dealing with, among other things, servants, slaves, and heavenly partners in the Quranic afterlife.26 She interprets the Quranic ghilmā n in 52:24 to refer to enslaved servants and understands the word as more or less synonymous with the wildā n mukhalladuˉ n, ‘eternal youths’, mentioned elsewhere in the Quran (56:17, 76:19) as the servants of the believers in the afterlife.27 She describes the Quranic aesthetics of the afterlife in the following fashion: As the believers recline and enjoy each other’s company, the youth serve (Q 56:19, 52:23–5, 76:13–9) during the banquet-like setting. The youth then are not only instrumental in serving the non-intoxicating drink (Q 56:18–9) or a drink from the fountain Salsabıˉl (Q 76:17), but they act as features of the purified landscape. While the verses pair the youth with the cups and vessels from which believers drink, their presence in the text provides a vision of paradise as opposed to a functional explanation of how life in paradise is administered. Instead, the role of the wildā n mukhalladuˉ n and ghilmā n is to imbue the landscape with the sheer spectacle of shimmering beauty.28

Apart from Rustomji’s valuable study, verse 52:24 and its exegesis (tafsıˉ r) nonetheless remain understudied. ‘Heavenly slaves’ is a topic that is not treated in the scholarship of the Quran or Islam.

later medieval exegetes on verse 52:24 In this section, I will present what the medieval commentators made of Quran 52:24. The section surveys many esteemed classical commentators, although it cannot claim to be all-encompassing with regard to the

26 27 28

Rustomji 2017. Rustomji 2017, 298–9. Rustomji 2017, 300–301.

308

Ilkka Lindstedt

extensive medieval exegetical discourse. To limit the material under consideration, I have dealt primarily with (non-Sufi) Sunni exegetes. Their treatment of verse 52:24 might be peculiar to them, so the findings of this study cannot be expanded to all Muslim commentators of the Quran. For example, in the commentary of the Sufi scholar Ibn ʿArabıˉ the ghilmā n lahum are understood allegorically to refer to spiritual beings (ruˉ ḥā niyyā t).29 The approach that I am using to interpret the Muslim exegetical texts can be compared with genealogical discourse analysis, which has been employed in Quranic studies by Omaima Abou-Bakr30 and Katja von Schöneman.31 With this method, we can approach ‘the diachronic development of the interpretations. The potential changes over time can be traced using a genealogical approach assuming that they represent a historical lineage’.32 The method is warranted since the exegetes often knew, quoted, reacted to, or rejected the readings of the earlier authorities. Although my approach is, strictly speaking, more chronological than genealogical, I aim to demonstrate in this section the different strands of interpreting verse 52:24. The earliest, eighth- to ninth-century ce Quranic commentators do not offer much on verse 52:24. I remark here that it is unclear whether we can speak of these early figures as authoring books proper. This applies especially to the Meccan Mujā hid’s (d. ca. 722) work, which seems to include scattered reports disseminated by him and compiled by his students.33 Mujā hid’s published Tafsıˉ r does not contain a comment on Quran 52:24. The Tafsıˉ r of Muqā til (d. ca. 767) can also be understood as a group effort to an extent: the work was probably revised by his students, although, in contrast to Mujā hid, Muqā til endeavoured to offer a treatment of the whole Quran.34 Muqā til lived in Iran and Iraq. Although Muqā til is an early exegete whose work survives and who is thus very important from the point of view of modern scholarship, he was not deemed trustworthy by medieval Muslim scholars. Muqā til does present a brief comment on the ghilmā n lahum of 52:24, making it probably the earliest surviving exegetical gloss on the passage. 29 30 31 32 33 34

Ibn ʿArabıˉ, Tafsıˉ r, II, 552. Abou-Bakr 2015. Von Schöneman 2018. Von Schöneman 2018, 17. On the work and its transmission, see Gilliot 2014. Sinai 2014.

Slave Boys in Paradise? The Text of the Quran

309

According to Muqā til, the ghilmā n ‘never become adults’,35 thus equating them more or less with the wildā n mukhalladuˉ n, ‘eternal youths’, occurring in Quran 56:17 and 76:19 (see Nerina Rustomji’s comment above). The same equation will be made by al-Rā zıˉ, as we will see. As for another early authority, al-Farrā ʾ (d. 822 ce), he does not proffer a comment on the phrase ghilmā n lahum. One of the most famous medieval exegetes and historians, al-Ṭabarıˉ (d. 923 ce), is rather disappointing as to verse 52:24. He concentrates on explaining what the comparison of the ghilmā n to ‘hidden pearls’ means but does not elucidate or quote reports on how to understand the word ghilmā n. He does note, however, that the ghilmā n serve ‘cups of drink’ to the believers.36 All in all, then, the commentators who lived between 700 and 1000 ce are rather muted about the interpretation of ghilmā n lahum. Fortunately, the exegetes from the eleventh to the fifteenth centuries offer more information as to how they read the verse. Al-Thaʿlabıˉ (Abuˉ Isḥā q Aḥmad ibn Muḥammad) was from Nishapur. He was known as an exegete and collector of stories of the prophets. He died in 1035 ce. In his Quranic commentary, he explains that the ghilmā n of Quran 52:24 are there to serve (bi-l-khidma).37 He then quotes two dicta ascribed to the Prophet Muhammad according to which every denizen of paradise has at their service a thousand servants, who are referred to with two words in the narratives: khā dim and ghulā m. Neither word takes a clear stance on the slave/free status of those servants. The terms khā dim and bi-l-khidma should be kept in mind, since they might refer, in fact, to eunuchs. We will come back to this question in the next section. Al-Ṭabarsıˉ (also known as al-Ṭabrisıˉ, ca. 1077–1153 ce), who is the only Shiʿi scholar that I include here, grew up in Khurasan and died probably in Mashhad. He wrote three Quranic commentaries (of which the Majmaʿ is the longest) and numerous other works. Al-Ṭabarsıˉ does not explicitly state whether or not he thinks that these ghilmā n are slaves, simply noting that they are there to serve (li-l-khidma).38 In his commentary on the verse under consideration, he also offers an interesting, although brief, ethical reflection on what it means that there are

35 36 37 38

Muqā til, Tafsıˉ r, IV, 146. Al-Ṭabarıˉ, Jā miʿ, XXI, 589–90. Al-Thaʿlabıˉ, Kashf, IX, 129. Al-Ṭabarsıˉ, Majmaʿ, IX, 166.

310

Ilkka Lindstedt

servants/slaves in paradise. What or who are they? He remarks that ‘it is said that the ghilmā n will not have any hardship in serving the people of paradise. Rather, they have pleasure and happiness, since that place [that is, paradise] is not a place of affliction’.39 The ghilmā n are, in al-Ṭabarsıˉ’s opinion, basically automatons. Al-Rā zıˉ (Fakhr al-Dıˉn) was born in circa 1149 in Rayy and died in 1209 near Herā t. He is chronologically the first exegete among the ones that I have surveyed to make the explicit statement that the ghilmā n lahum of verse 52:24 are slaves. After him, this becomes a standard interpretation. Al-Rā zıˉ was a voluminous scholar, composing works in, for example, theology, philosophy, and exegesis. His Quranic commentary,40 interestingly, makes the same association as Muqā til did implicitly and the modern scholar Nerina Rustomji41 explicitly: that the ghilmā n are the wildā n mukhalladuˉ n, ‘eternal youths’ mentioned elsewhere in the Quran. According to al-Rā zıˉ, the ghilmā n go round the believers serving them with cups. However, that these are slaves is unambiguous: lahum is explained as milkuhum, ‘their property’, by al-Rā zıˉ, adding that the believers can freely command, forbid, and require services from these ghilmā n. He then remarks that just as the wine of the afterlife is different from the wine of this life (since the former ‘does not lead to any idle talk or sin’; Quran 52:23), so also the ghilmā n of the afterlife are different from those of this life. The paradisal ghilmā n do not themselves try to attain anything but simply serve the believers, so they are preferable to terrestrial ghilmā n, who have goals and needs of their own. The ghilmā n of the afterlife are constructed as ideal slaves by al-Rā zıˉ. This reflection on the characteristics of the ghilmā n can be compared to al-Ṭabarsıˉ’s musings cited above. However, whereas al-Ṭabarsıˉ takes the viewpoint of the ghilmā n, al-Rā zıˉ is not interested in their wellbeing but notes that the paradisal slave boys are trustworthy servants to their masters. Al-Rasʿanıˉ (ʿAbd al-Razzā q ibn Rizqallā h al-Hanbalıˉ) lived from 1193 to 1262. He was born in Raʾs al-ʿAyn in northeastern Syria and he died in Sinjā r in today’s Iraqi Kurdistan. In his Quranic commentary, he does not spend much ink on the expression ghilmā n lahum but notes that it signifies ‘slaves allocated to them’ (mamluˉ kuˉ n makhṣuˉ ṣuˉ n bihim).42 Thus,

39 40 41 42

Al-Ṭabarsıˉ, Majmaʿ, IX, 166. Al-Rā zıˉ, Mafā tıˉ ḥ, VI, 68. Rustomji 2017, 298–9. Al-Rasʿanıˉ, Rumuˉ z, VII, 449.

Slave Boys in Paradise? The Text of the Quran

311

ghilmā n is glossed with mamluˉ kuˉ n, an unambiguous word meaning ‘slaves’, and lahum with makhṣuˉ ṣuˉ n bihim, ‘allocated to them’. Al-Qurtubı ̣ ˉ (Abuˉ ʿAbdallā h Muḥammad ibn Aḥmad, d. 1276 ce) was a scholar originally from Cordoba. He left his hometown, however, when it was conquered by Christians in 1236 ce, moving to Egypt, where he eventually died. In his Quranic commentary, he is generally interested in the legal aspects that the Quranic discourse entails. In his exegetical note on Quran 52:24,43 al-Qurt ̣ubıˉ begins by noting that the ghilmā n serve the believers fruits, food, and drink. He then discusses the phrase ghilmā n lahum. His preferred meaning is that the ghilmā n are the believers’ own children. Perhaps, he has in mind verse 52:21: ‘We [scil. God] unite the believers with their offspring (dhurriyyatuhum) who followed them in faith.’ He notes other interpretative possibilities, prefaced with the caveat ‘it is said’: that the ghilmā n are someone else’s children or that they are ghilmā n (here in the meaning ‘servant’ or ‘slave’?) that are made specifically for paradise. Al-Bayḍā wıˉ, who died around 1286 ce, worked as the judge of the town of Shıˉrā z. His Tafsıˉ r, Quranic commentary, is based on the earlier alZamakhsharıˉ’s exegetical work. However, the latter had Muʿtazilıˉ (a branch of rationalistic theology) leanings that al-Bayḍā wıˉ omitted from his work. Al-Bayḍā wıˉ deems that the reference in 52:24 is to slaves (mamā lıˉ k).44 Furthermore, he connects the expression ‘wait on them’ (yat ˉ̣ufu ʿalayhim) with the cup mentioned in verse 23: the slaves serve the cup to the believers. This interpretation was made by al-Ṭabarıˉ, quoted above, and al-Bayḍā wıˉ might be using him as the source here. AlBayḍā wıˉ furthermore mentions that ‘it is said (qıˉ la)’ that the ghilmā n actually signifies the believers’ children. However, the expression used by al-Bayḍā wıˉ (‘it is said’) probably means that he did not find this reading very likely. Al-Andalusıˉ (Abuˉ Ḥ ayyā n, d. 1344) was, as his name indicates, from al-Andalus, Muslim Spain, more specifically Granada, although he died in Cairo. In his commentary, he notes quite tersely that ghilmā n lahum refers to slaves (mamā lıˉ k),45 echoing al-Rasʿanıˉ and al-Bayḍā wıˉ. I will end this survey of Quranic commentaries with the one that became perhaps the most popular of the classical commentaries because of its conciseness and clearness, still used by many Muslims today. This is 43 44 45

Al-Qurt ̣ubıˉ, Jā miʿ, XIX, 528. Al-Bayḍā wıˉ, Tafsıˉ r, II, 435. Al-Andalusıˉ, Baḥr, VII, 137.

312

Ilkka Lindstedt

the so-called Tafsıˉ r al-Jalā layn, composed by Jalā l al-Dıˉn al-Maḥallıˉ (d. 1459) and his student Jalā l al-Dıˉn al-Suyuˉ tˉ̣ı (d. 1505). The commentary is very unambiguous when it comes to Quran 52:24, explaining that ghilmā n means slaves (ariqqā ʾ) who are there to serve (li-l-khidma) the believers.46 The Tafsıˉ r al-Jalā layn, then, cements the late medieval Sunni understanding of ghilmā n lahum as being of servile status. According to the evidence cited in this section, while the different exegetes understood Quran 52:24 in varying ways, it became common in the late medieval era to follow al-Rā zıˉ in understanding the ghilmā n lahum as denoting slaves. Before him, the commentators do not proffer explicit statements regarding whether or not they saw these ghilmā n as being of slave status properly speaking. The suggestion that there might be enslaved servants in the afterlife to tend to the pious is quite intriguing, but the exegetes do not unfortunately elaborate how they see this fitting their overall paradisal imagery.

what about eunuchs? As noted above, some exegetes, such as al-Thaʿlabıˉ and al-Ṭabarsıˉ, call the ghilmā n lahum ‘servants’ (sing. khā dim) or say that they are there ‘to serve’ (bi-/li-l-khidma). The choices of words are interesting, since David Ayalon has argued forcefully and at length that in the historiographical literature and belles-lettres the word khā dim (lit. ‘servant’) was used from the ninth century ce onwards euphemistically in the sense of ‘eunuch’.47 Another word, khaṣˉı, has the basic signification ‘eunuch’, but, in addition to it, the word khā dim started to be used with the same meaning. Ayalon has amassed much evidence of this usage in medieval Arabic literature. He argues, in my opinion convincingly, that this usage is already present in the epistles of al-Jā ḥiz ̣, who died in 868 ce.48 Al-Jā ḥiz ̣ also offers an interesting remark stating that, in his opinion, a khaṣˉı has an aptitude for service (khidma).49 Thus, the concept of service, khidma, and being a eunuch are linked: a eunuch is the best servant. Ayalon summarizes his view: In the second half of the 3rd/9th century the term khā dim (‘eunuch’) becomes not only very frequent, but is the predominant one. As for the beginning of the 4th/ 46 47 48 49

Jalā l al-Dıˉn al-Maḥallıˉ and Jalā l al-Dıˉn al-Suyuˉ tıˉ̣ , Tafsıˉ r al-Jalā layn, 525. Ayalon 1979 and 1985. Ayalon 1985, 289–96. Ayalon 1985, 296.

Slave Boys in Paradise? The Text of the Quran

313

10th century and later on, there are no uncertainties whatsoever about the use of khā dim [in the sense of ‘eunuch’] by the historians of that period . . . One should not infer from the above series of arguments that every single khā dim encountered in the historical and related sources is a eunuch beyond any shadow of doubt, especially if he is not a prominent person. Such an absolute certainty does not exist with regard to any term which underwent a similar process of development . . . My sole aim was to demonstrate how overwhelming are the chances that that term means ‘eunuchs’ in this kind of sources.50

The Arabic historiographical and geographical literature of the medieval period is illuminating as regards these slave eunuchs (khadam, the plural of khā dim). Sources state that these slave eunuchs belonged to the ethnic categorizations of, for example, Slavs (ṣaqā liba), Byzantines (ruˉ m), Chinese (ṣˉın), and Africans (suˉ dā n).51 Since Islamic law prohibited castration, even of slaves, a legal loophole was found in the practice of castrating them before they were brought to the Islamic area. Now, Ayalon discussed ‘historical and related sources’ and stated the general rule that if one encounters the word khā dim there, it is probable that the word refers to a slave eunuch rather than simply a servant.52 He did not look into the usage of the word in more religious texts, such as Quranic commentaries. The question that arises is: did the exegetes that used the word khā dim to describe the heavenly ghilmā n or that stated that the ghilmā n are there ‘to serve’ (bi-/li-l-khidma) indicate that these heavenly slaves/servants are eunuchs? That is possible, but direct evidence to that effect is lacking. It is certain, however, that the writers of the Quranic commentaries must have been familiar with this usage of khā dim in the sense of ‘eunuch’, which is very common in the types of literature, in particular historiography, surveyed by Ayalon. The commentators were well-read scholars who were conversant with many genres of Arabic literature. The most explicit identification of ghilmā n in Quran 52:24 as khadam is found in al-Thaʿlabıˉ.53 In addition to his Quranic commentary, he wrote a much-read book on the stories of the prophets called ʿArā ʾis almajā lis fıˉ qiṣaṣ al-anbiyā ʾ and hence took part in the historiographical discourse. Given Ayalon’s well documented studies about the usage of the word khā dim in medieval Arabic literature, it can be suggested with some caution that the commentators surveyed in this chapter who used

50 51 52 53

Ayalon 1985, 298, 307. Ayalon 1985, 303–4. Ayalon 1985, 307. Al-Thaʿlabıˉ, Kashf, IX, 129.

Ilkka Lindstedt

314

expressions from the root kh-d-m to describe the ghilmā n lahum had eunuch slaves in mind.54

conclusions In the beginning of this article, I surveyed the evidence as regards slavery in late antique Arabia. I also introduced the Quranic verses dealing with slaves and slavery. Slavery was common, and the Quran’s approach to the phenomenon is, for the most part, matter-of-fact. Despite this, it is unclear how the first generation of followers of the Prophet Muhammad understood the Quranic verse 52:24: yat ˉ̣ufu ʿalayhim ghilmā n lahum kaannahum luʾluʾ maknuˉ n. Nonetheless, I have shown here that while some later commentators (al-Qurtubı ̣ ˉ and, with a caveat, al-Bayḍā wıˉ) understood the words ghilmā n lahum in Quran 52:24 to refer to ‘their children’, possibly connected to ‘their offspring’ (dhurriyyatuhum) mentioned earlier in the surah, the majority of medieval Arabic exegetes took verse 52:24 to mean that believers will be served in paradise by slaves. I am not aware of parallels to this idea in any Judeo-Christian traditions. Qian

ghilman lahum (Quran 52:24) signifies:

servants (eunuchs?)

slaves Al-Razī (d. 1209) Al-Rasʿanī (d. 1262)

Al-Thaʿlabī (d. 1035)

Al-Baydawī (d. 1286)

Al-Ṭabarsī (d. 1153)

eternal youths Muqatil (d. ca. 767) Al-Razī (d. 1209)

children of the believers Al-Qurṭubī (d. 1276)

Al-Andalusī (d. 1344) Jalal al-Dīn al-Maḥallī (d. 1459) and Jalal al-Dīn al-Suyuṭī (d. 1505)

figure 14.1 Figure illustrating how the commentators of the Quran discussed in this chapter understood the phrase ghilmā n lahum in Quran 52:24

54

As Chris de Wet noted to me (private communication), the focus on the purity and sinlessness in the Quranic discourse might (euphemistically) support the fact that the slaves are eunuchs.

Slave Boys in Paradise? The Text of the Quran

315

compares the ‘heavenly banquet described in the Quran’ to the symposia of Greek antiquity, where the all-male participants leaned on couches and drank wine served by slave boys.55 This is an interesting analogue, but it is left for further research to inquire if there are any direct or indirect influences and what they might be. The minority position of the exegetes – that ghilmā n lahum refers to the children of the believers who serve the latter in paradise – is in no way less interesting. I have not encountered such a suggestion in Judeo-Christian texts either. Although many Quranic commentaries still remain unpublished, according to the sources surveyed here the interpretation of ghilmā n lahum as slaves is somewhat late and does not appear in exegetical works before al-Rā zıˉ (d. 1209). To connect the exegetical discourse to social and political historical phenomena is somewhat problematic, but it might be noted that the period when the simple identification of ghilmā n lahum as slaves becomes paramount (the thirteenth century ce) corresponds with the political power of the Ayyubids and Mamelukes, who made extensive use of slaves in both military and other spheres of life. The Mamelukes (mamā lıˉ k) were elite slave soldiers who eventually became the ruling class of Egypt. Their elite status notwithstanding, it is clear that this is a period of increased demand for slaves in the Near East.56 Perhaps the resurgence of the slave trade during the era and prevalence of slaves in the society is one of the reasons why the commentators of the Quran started to prefer the interpretation of heavenly slaves in verse 52:24 over other readings. It is rather astonishing that so few exegetes offered their insights as to the metaphysical or ethical aspects of the ghilmā n lahum. A notable exception to this among the commentators treated in this article is alṬabarsıˉ, who proposed that the ghilmā n are a sort of automaton, serving the believers with joy since there is no affliction in paradise, even for those of servile status. Al-Rā zıˉ, too, notes that they do not have demands or aspirations. For him, the paradisal ghilmā n are the ideal slaves.

55

56

Qian 2017, 261, n. 58. It is unclear whether this comment by Qian is meant to present a parallel case or direct or indirect influence from Greek to Arabic literature. For the Mamelukes, see Ayalon 1971; Pipes 1981; Ayalon 1994; Franz 2017, 110–20.

Bibliography

Abdel Haleem, Muhammad A. S. (trans.), The Quran. Oxford: Oxford University Press 2004. Abou-Bakr, Omaima, ‘The Interpretive Legacy of Qiwamah as Exegetical Construct’, in Z. Mir-Hosseini, M. Al-Shamarni and J. Rumminger (eds.), Men in Charge? Rethinking Authority in Muslim Legal Tradition. Oxford: OneWorld 2015, 44–64. Adams, James N, An Anthology of Informal Latin 200 BC–AD 900: Fifty Texts with Translations and Linguistic Commentary. Cambridge: Cambridge University Press 2016. Ahmed, Sohair S., ‘Professions, Trades, Occupations, and Titles in Coptic (Alphabetically), Part II’, Journal of Coptic Studies 13 (2011): 183–212. ‘Professions, Trades, Occupations, and Titles in Coptic (Alphabetically), Part I’, Journal of Coptic Studies 12 (2010): 115–48. Ali, Kecia, ‘Concubinage and Consent’, International Journal of Middle East Studies 49 (2017): 148–52. Marriage and Slavery in Early Islam. Cambridge, MA: Harvard University Press 2010. Al-Jallad, Ahmad M., ‘Introduction: The Arabic of Petra’, in A. Arjava, J. Frösén, and J. Kaimio (eds.), The Petra Papyri V. Amman: American Center of Oriental Research 2018, 35–55. An Outline of the Grammar of the Safaitic Inscriptions. Leiden: Brill 2015. Allen, Pauline and Bronwen Neil, Crisis Management in Late Antiquity (410–590 CE): A Survey of the Evidence from Episcopal Letters, Supplements to Vigiliae Christianae 121. Leiden: Brill 2013. Allen, Pauline, Bronwen Neil, and Wendy Mayer (eds.), Preaching Poverty in Late Antiquity: Perceptions and Realities. Leipzig: Evangelische Verlagsanstalt 2009. Amato, Eugenio, Procope de Gaze, Discours et fragments. Paris: Les Belles Lettres 2014.

316

Bibliography

317

Amélineau, Emile (ed. and trans.), Oeuvres de Schenoudi: Texte copte et traduction française, Vol. 1. Paris: E. Leroux 1907. Contes et romans de l’Égypte chrétienne. Paris: E. Leroux 1888. Amelotti, Mario and Livia Migliardi Zingale, Le costituzioni giustinianee nei papiri e nelle epigrafi, 2nd ed. Milan: Giuffráe 1985. Anderson, Gary A., Sin: A History. New Haven, CT: Yale University Press 2009. Ando, Clifford, ‘Decline, Fall, and Transformation’, Journal of Late Antiquity 1.1 (2008): 31–60. Imperial Ideology and Provincial Loyalty in the Roman Empire. Berkeley: University of California Press 2000. Andrade Cernadas, José Miguel, Marta Díaz Tie, and Francisco Javier Pérez Rodríguez, O tombo de Celanova: Estudio introductorio, edicion e indices (SS. IX–XII). Santiago de Compostela: Consello da Cultura Galega 1995. Andrews, James A., Hermeneutics and the Church: In Dialogue with Augustine, Reading the Scriptures. Notre Dame, IN: University of Notre Dame Press 2012. Arbesmann Rudolph, Emily Joseph Daly, and Edwin A. Quain, Tertullian: Disciplinary, Moral, and Ascetical Works, The Fathers of the Church 40. Washington, DC: Catholic University of America Press 1959. Arentzen, Thomas, ‘Voices Interwoven: Refrains and Vocal Participation in the Kontakia’, Jahrbuch der Österreichischen Byzantinistik 66 (2017): 1–10. Ariño Gil, Enrique and Pablo C. Díaz, ‘Poblamiento y organización del espcacio. La Tarraconense pirenaica en el siglo VI’, Antiquitè Tardive 11 (2003): 223– 37. Arjava, Antti, Jaakko Frösén, and Jorma Kaimio (eds.), The Petra Papyri V. Amman: American Center of Oriental Research 2018. Arjava, Antti and Traianos Gagos, ‘57. Agreement on a Slave’, in A. Arjava, J. Frösén, and J. Kaimio (eds.), The Petra Papyri V. Amman: American Center of Oriental Research 2018, 140–53. Arjava, Antti and Marjaana Vesterinen, ‘28. Division of Property’, in A. Arjava, M. Buchholz, and T. Gagos (eds.), The Petra Papyri III. Amman: American Center of Oriental Research 2007, 91–102. Arzt-Grabner, Peter, ‘“Neither a Truant nor a Fugitive”: Some Remarks on the Sale of Slaves in Roman Egypt and Other Provinces’, in T. Gagos and A. Hyatt (eds.), Proceedings of the 25th International Congress of Papyrology, Ann Arbor, July 29–August 4, 2007. Ann Arbor: Scholarly Publishing Office, University of Michigan 2010, 21–32. Athamina, Khalil, ‘How Did Islam Contribute to Change the Legal Status of Women: The Case of the Jawā rıˉ or the Female Slaves’, Al-Qantara 28 (2007): 383–408. Atkins, Margeret, and Robin Osborne (eds.), Poverty in the Roman World. Cambridge: Cambridge University Press 2006. Axel, Brian, The Nation’s Tortured Body: Violence, Representation, and the Formation of the Sikh Diaspora. Durham, NC: Duke University Press 2001. Ayalon, David, Islam and the Abode of War: Military Slaves and Islamic Adversaries. Aldershot: Ashgate 1994.

318

Bibliography

‘On the Term Khā dim in the Sense of “Eunuch” in the Early Muslim Sources’, Arabica 32 (1985): 289–308. ‘On the Eunuchs in Islam’, Jerusalem Studies in Arabic and Islam 1 (1979): 67–124. Outsiders in the Lands of Islam: Mamluks, Mongols, and Eunuchs. Aldershot: Ashgate 1971. Babcock, William S., ‘Sin and Punishment: The Early Augustine on Evil’, in J. T. Lienhard, E. C. Müller, and R. J. Teske (eds.), Augustine: Presbyter Factus Sum. New York: Peter Lang 1993, 235–48. Bacharach, Jere L., ‘African Military Slaves in the Medieval Middle East: The Cases of Iraq (869–955) and Egypt (868–1171)’, International Journal of Middle East Studies 13 (1981): 471–95. Bagnall, Roger S. Egypt in Late Antiquity. Princeton, NJ: Princeton University Press 1993a. ‘Slavery and Society in Late Roman Egypt’, in B. Halpern and D. W. Hobson (eds.), Law, Politics and Society in the Ancient Mediterranean World. Sheffield: Sheffield Academic 1993b, 220–40. ‘Official and Private Violence in Roman Egypt’, Bulletin of the American Society of Papyrologists 26 (1989): 201–16. Bagnall, Roger S. and Bruce Frier, The Demography of Roman Egypt. Cambridge: Cambridge University Press 1994. Baker, Cynthia M., Rebuilding the House of Israel: Architectures of Gender in Jewish Antiquity, Divinations: Rereading Late Ancient Religion. Stanford, CA: Stanford University Press 2002. Bales, Kevin, Disposable People. New Slavery in the Global Economy (3rd ed.). Berkeley: University of California Press 2012. Balmelle, Catherine, Dany Barraud, Jean-Pierre Brun, Béatrice Duprat, Hervé Gaillard, Philippe Jacques, Louis Maurin, et al. ‘La viticulture antique en Aquitaine’, Gallia 58 (2001): 129–64. Banaji, Jairus, Theory as History: Essays on Modes of Production and Exploitation. Leiden: Brill 2010. Barbero, Abilio and Marcelo Vigil, La formación del feudalismo en la penísula ibérica. Barcelona: Editorial crítica 1978. Barclay, John M. G., ‘Ordinary but Different: Colossians and Hidden Moral Identity’, Australian Biblical Review 49 (2001): 34–52. Barnes, Timothy D., ‘Legislation against the Christians’, Journal of Roman Studies 58 (1968): 32–50. Barrett, Anthony A., ‘Saint Germanus and the British Missions’, Britannia 40 (2009): 197–217. Bartchy, S. Scott, ‘Slavery’, in G. W. Bromiley (ed.), The International Standard Bible Encyclopedia, Vol. 4. Grand Rapids, MI: Eerdmans 1988, 539–46. Bazzana, Giovanni B., ‘“Basileia” and Debt Relief: The Forgiveness of Debts in the Lord’s Prayer in the Light of Documentary Papyri’, Catholic Biblical Quarterly 73.3 (2011): 511–25. Behlmer, Heike (ed. and trans.), Schenute von Atripe: ‘De iudicio’ (Torino, Museo Egizio, Cat. 63000, Cod. IV). Turin: Ministero per i beni culturali e ambientali, Soprintendenza al museo delle antichità egizie 1996.

Bibliography

319

Bell, David, The Life of Shenoute. Kalamazoo: Cistercian Publications 1983. Bellemore, Jane and Beryl Rawson, ‘Alumni: The Italian Evidence’, Zeitschrift für Papyrologie und Epigraphik 83 (1990): 1–19. Benko, Stephen, ‘Pagan Criticims of Christianity during the First Two Centuries A.D.’, Aufstieg und Niedergang der römischen Welt 2.23.2 (1980): 1055– 119. Bensch, Stephen P., ‘From Prizes of War to Domestic Merchandise: The Changing Face of Slavery in Catalonia and Aragon, 1000–1300’, Viator 25 (1994): 63–94. Bernard Donna, Rose (trans.), St. Cyprian: Letters 1–81, The Fathers of the Church 51. Washington, DC: Catholic University of America Press 1964. Bernards, Monique and John Nawas (eds.), Patronate and Patronage in Early and Classical Islam. Leiden: Brill 2005. Bernhardt, Johannes C., Die jüdische Revolution: Untersuchungen zu Ursachen, Verlauf und Folgen der hasmonäischen Erhebung. Berlin: De Gruyter 2017. Best, Eugene C., An Historical Study of the Exegesis of Colossians 2,14. Rome: Pontificia Universitas Gregoriana 1956. Bettenson, Henry (trans.), Augustine: City of God. London: Penguin 2003. Bieler, Ludwig (ed.), Libri Epistolarum Sancti Patricii Episcopi: Introduction, Text, and Commentary, 2 Vols. Dublin: Royal Irish Academy 1952 [repr. in one vol., 1993]. Binchy, Daniel A., ‘Patrick and His Biographers, Ancient and Modern’, Studia Hibernica 2 (1962): 71–3. Birley, Andrew, ‘Brigomaglos and Riacus: A Brave New World? The Balance of Power at Post-Roman Vindolanda’, in F. K. Haarer (ed.), with R. Collins, K. J. Fitzpatrick-Matthews, S. Moorhead, D. Petts, and P. J. Walton (eds.), AD 410: The History and Archaeology of Late and Post-Roman Britain. London: Society for the Promotion of Roman Studies 2014, 195–205. Blackburn, Robin, The Making of New World Slavery from the Baroque to the Modern Era 1492–1800. New York: Verso 1997. Blanchette, Oliva A., ‘Does the “Cheirographon” of Colossians 2,14 Represent Christ Himself?’, Catholic Biblical Quarterly 23.3 (1961): 306–12. Bloch, Marc L. B., ‘Comment et pourquoi finit l’esclavage antique’, Annales d’Histoire Economique et Sociale 2 (1947): 30–44, 161–70. Slavery and Serfdom in the Middle Ages: Selected Essays. Berkeley: University of California Press 1975. Blumenthal, Debra, Enemies and Familiars: Slavery and Mastery in FifteenthCentury Valencia. Ithaca, NY: Cornell University Press 2009. Bock, Johann P., Die Brotbitte des Vaterunsers. Paderborn: Bonifacius 1911. Bodel, John, ‘Death and Social Death in Ancient Rome’, in J. Bodel and W. Scheidel (eds.), On Human Bondage: After Slavery and Social Death. Oxford: Wiley-Blackwell 2017, 81–108. Boer, Roland and Christina Petterson, Time of Troubles: A New Economic Framework for Early Christianity. Minneapolis: Fortress 2017. Boggs, Carl, Gramsci’s Marxism. London: Pluto 1976. Bolman, Elizabeth S., The Red Monastery Church: Beauty and Asceticism in Upper Egypt. New Haven, CT: Yale University Press 2016.

320

Bibliography

Bondue, Didier, De servus à sclavus: la fin de l’esclavage antique, 371–918. Paris: Presses de l’université Paris-Sorbonne 2011. Bonnassie, Pierre, From Slavery to Feudalism in South-Western Europe. Cambridge: Cambridge University Press 1991. Borg, Barbara, Crisis and Ambition. Tombs and Burial Customs in Third-century CE Rome. Oxford: Oxford University Press 2013. Boulvert, Gérard, Domestique et fonctionnaire sous le Haut-Empire romain: La condition de l’affranchi et de l’esclavage du prince. Besançon: Université de Franche-Comté 1974. Bourdieu, Pierre, Pascalian Meditations, R. Nice (trans.). Stanford: Stanford University Press 2000. Masculine Domination, R. Nice (trans.). Stanford, CA: Stanford University Press 2001. Sociology in Question. R. Nice (trans.). London: Routledge 1993a. The Field of Cultural Production: Essays on Art and Literature, R. Johnson (ed.). Cambridge: Polity 1993b. In Other Words: Essays toward a Reflexive Sociology, M. Adamson (trans.). Stanford, CA: Stanford University Press 1990a. The Logic of Practice, R. Nice (trans.). Stanford, CA: Stanford University Press 1990b. ‘The Forms of Capital’, in J. Richardson (ed.), Handbook of Theory and Research for the Sociology of Education. New York: Greenwood 1986, 241–58. Distinction: A Social Critique of the Judgment of Taste, R. Nice (trans.). London: Routledge 1984. Outline of a Theory of Practice, R. Nice (trans.). Cambridge: Cambridge University Press 1977. Bourdieu, Pierre, and Loïc J. D. Wacquant, An Invitation to Reflexive Sociology, R. Johnson (ed.). Cambridge: Polity 1992. Bradley, Keith, ‘Introduction’, in K. Bradley and P. Cartledge (eds.), The Cambridge World History of Slavery, Vol. 1: The Ancient Mediterranean World. Cambridge: Cambridge University Press 2011, 1–3. ‘On Captives under the Principate’, Phoenix 58 (2004): 298–318. ‘The Problem of Slavery in Classical Culture’, Classical Philology 92.3 (1997): 273–82. Discovering the Roman Family. Studies in Roman Social History. Oxford: Oxford University Press 1991. Bradley, Keith, and Paul Cartledge (eds.), The Cambridge World History of Slavery Volume 1: The Ancient Mediterranean World. Cambridge: Cambridge University Press 2011. Brakke, David, and Andrew Crislip (trans.), Selected Discourses of Shenoute the Great: Community, Theology, and Social Conflict in Late Antique Egypt. Cambridge: Cambridge University Press 2015. Brancato, Nicolò Giuseppe, Una componente trasversale della società romana. Gli alumni: inscriptiones latinae ad alumnos pertinentes commentariumque. Rome: Artecom 2015.

Bibliography

321

Brass, Tom, ‘Late Antiquity as Early Capitalism?’, Journal of Peasant Studies 32.1 (2005): 118–50. Bray, Julia, ‘Men, Women and Slaves in Abbasid Society’, in L. Brubaker and J. M. H. Smith (eds.), Gender in the Early Medieval World: East and West, 300–900. Cambridge: Cambridge University Press 2004, 121–46. Brock, Sebastian P., ‘Euphemia and the Goth’, in S. P. Brock, A. M. Butts, G. A. Kiraz, and L. Van Rompay (eds.), Gorgias Encyclopedic Dictionary of the Syriac Heritage. Piscataway, NJ: Gorgias 2011a, 153. ‘Radical Renunciation: The Ideal of Msarrqûtâ’, in R. D. Young and M. J. Blanchard (eds.), To Train His Soul in Books: Syrian Asceticism in Early Christianity, CUA Studies in Early Christianity. Washington, DC: Catholic University of America Press 2011b, 122–33. From Ephrem to Romanos: Interactions Between Syriac and Greek in Late Antiquity. Aldershot: Ashgate, 1999. ‘Early Syrian Asceticism’, Numen 20.1 (1973): 1–19. Brockopp, Jonathan, ‘Slaves and Slavery’, in J. McAuliffe (ed.), Encyclopaedia of the Qurʾā n. Brill: Leiden 2006, V, 56–60. Early Mā likıˉ Law: Ibn ʿAbd al-Ḥ akam and His Major Compendium of Jurisprudence. Leiden: Brill 2000. Brooks-Hedstrom, Darlene L., ‘An Archaeological Mission for the White Monastery’, Coptica 4 (2005): 1–26. Brooten, Bernadette J., ‘Early Christian Enslaved Families’, in C. Laes, K. Mustakallio, and V. Vuolanto (eds.), Children and Family in Late Antiquity. Life, Death and Interaction. Leuven & Walpole: Peeters 2015, 111–34. ‘Enslaved Women in Basil of Caesarea’s Canonical Letters: An Intersectional Analysis’, in U. Eisen, C. Gerber, and A. Standhartinger (eds.), Doing Gender – Doing Religion: Case Studies on Intersectionality in Early Judaism, Christianity and Islam, Wissenschaftliche Untersuchungen zum Neuen Testament 302. Tübingen: Mohr Siebeck 2013, 325–55. Brown, Jonathan, Slavery and Islam. Oxford: Oneworld 2019. ‘Scripture, Legal Interpretation and Social Praxis in the Islamic Tradition: The Cases of Polygamy and Slavery’, in N. Berend, Y. Hameau-Masset, C. NemoPekelman, and J. Tolan (eds.), Religious Minorities in Christian, Jewish and Muslim Law (5th–15th Centuries). Turnhout: Brepols 2017, 99–113. Brown, Peter, ‘Dialogue with God’, New York Review of Books, 26 October 2017, online: https://bit.ly/3jTeyR6. Through the Eye of a Needle: Wealth, the Fall of Rome, and the Making of Christianity in the West, 350–550 AD. Princeton, NJ: Princeton University Press 2012. The Body and Society: Men, Women, and Sexual Renunciation in Early Christianity, 20th Anniversary Edition with a New Introduction. New York: Columbia University Press 2008 [1988]. Augustine of Hippo: A Biography, 2nd ed. Berkeley: University of California Press 2000. The World of Late Antiquity: AD 150–750. London: Thames & Hudson 1971.

322

Bibliography

Bryen, Ari Z., Violence in Roman Egypt: A Study in Legal Interpretation. Philadelphia: University of Pennsylvania Press 2013. Buonocore, Marco, ‘Vermic(u)lator’, Zeitschrift für Papyrologie und Epigraphik 52 (1983): 199–201. Burkert, Walter, Greek Religion: Archaic and Classical. Oxford: Blackwell 1985. Burns, J. Patout, ‘Augustine on the Origin and Progress of Evil’, Journal of Religious Ethics 16.1 (1988): 9–27. Burrus, Virginia, ‘Macrina’s Tattoo’, Journal of Medieval and Early Modern Studies 33.3 (2003): 403–17. Busch, Stephan and Andrea Binsfeld, ‘rosa simul florivit et statim periit – Sklavenkinder in römischen Grabepigrammen. Ein Neufund: Die Stele der Iucunda aus Segobriga’, in H. Heinen (ed.), Kindersklaven – Sklavenkinder. Schicksale zwischen Zuneigung und Ausbeutung in der Antike und im interkulturellen Vergleich. Stuttgart: Franz Steiner 2012, 203–29. Butler, Judith, Bodies That Matter: On the Discursive Limits of ‘Sex’. New York: Routledge 1993. Byron, John, Slavery Metaphors in Early Judaism and Pauline Christianity: A Traditio-Historical and Exegetical Examination, Wissenschaftliche Untersuchungen zum Neuen Testament 162. Tübingen: Mohr Siebeck 2003. Cahill Wilson, J., ‘Et tu, Hibernia? Frontier Zones and Culture Contact – Ireland in a Roman World’, in S. Gonzalez Sanchez and A. Guglielmi (eds.), Romans and Barbarians beyond the Frontiers: Archaeology, Ideology and Identities in the North, TRAC 1. Oxford: Oxbow 2017, 48–69. Cain, Andrew, ‘Patrick’s Confessio and Jerome’s Epistula 52 to Nepotian’, Journal of Medieval Latin 20 (2010): 1–15. Cameron, Alan, ‘The Funeral of Junius Bassus’, Zeitschrift für Papyrologie und Epigraphik 139 (2002): 288–92. Cameron, Alan and Jacqueline Long, with a contribution by Lee Sherry, Barbarians and Politics at the Court of Arcadius. Berkeley: University of California Press 1993. Cameron, Averil, The Mediterranean World in Late Antiquity: AD 395–700. London: Routledge 1993. Campos Ruiz, Julio and Ismael Roca Melia, Reglas monásticas de la España visigoda. Madrid: Biblioteca de Autores Cristianos 1971. Caner, Daniel F., Wandering, Begging Monks: Spiritual Authority and the Promotion of Monasticism in Late Antiquity, Transformation of the Classical Heritage 33. Berkeley: University of California Press 2002. Carandini, Andrea. ‘Prefazione: Quando la dimora dello strumento è l’uomo’, in J. Kolendo (ed.), L’agricoltura nell’Italia romana: tecniche agrarie e progresso economico dalla tarda repubblica al principato. Roma: Riuniti 1980, ix–lx. Carletti, Carlo, Epigrafia dei cristiani in Occidente dal III al VII secolo. Ideologia e prassi. Bari: Edipuglia 2008. Caron, Bruce, ‘Orthoposture’, lightblueblog (2010), online: https://bit.ly /3t1r44V. Carr, Wesley, ‘Two Notes on Colossians’, Journal of Theological Studies 24.3 (1973): 492–500.

Bibliography

323

Carrié, Jean-Michel, ‘“Colonato del Basso Impero” la resistenza del mito’, in E. Lo Cascio (ed.), Terre, proprietari e contadidni dello’impero romano. Roma: La nuova Italia Scientifica 1997, 75–150. Casias, Cassandra M. M. , ‘Sinful Slave-Owners in Augustine’s Sermons’, Journal of Late Antiquity 11.1 (2018): 111–28. Castellanos, Santiago, ‘El testamento de Ricimiro de Dumio en el contexto de la consolidación episcopal en la historia tardoantigua’, Hispania Antiqua 22 (1998): 427–37. Caswell, F. Matthew, The Slave Girls of Baghdad: The Qiyan in the Early Abbasid Era. London: IB Tauris, 2011. Chambers, Katherine, ‘Slavery and Domination as Political Ideas in Augustine’s City of God’, Heythrop Journal 54 (2013): 13–28. Charles-Edwards, Thomas M., ‘Palladius, Prosper, and Leo the Great: Mission and Primatial Authority’, in D. N. Dumville and L. Abrams (eds.), Saint Patrick, AD 493–1993, Studies in Celtic History 13. Woodbridge, UK: Boydell 1993, 1–12. Chin, Catherine and Caroline Schroeder (eds.), Melania: Early Christianity through the Life of One Family. Oakland: University of California Press 2017. Chirichigno, Gregory C., Debt-Slavery in Israel and the Ancient Near East. Sheffield: Sheffield Academic Press 1993. Christie, Neil (ed.), Landscapes of Change: Rural Evolutions in Late Antiquity and the Early Middle Ages. Aldershot: Ashgate 2004. Christol, Michel and Jean-Pierre Laporte, ‘Teniet-el-Meksen: un nouveau gouverneur de Maurétanie Césarienne dans les dernières décennies du IIIe siècle et les luttes entre le pouvoir romain et les Bavares (Æ, 1907, 157; ILS, 8959)’, Antiquités africaines 38–39 (2002): 109–24. Christol, Michel and Αndreina Magioncalda, ‘Note su un’iscrizione di epoca tetrarchica: CIL VIII 20836 da Rapidum’, in A. Mastino (ed.), L’Africa romana. Atti del VII convengo di studio, Sassari, 15–17 dicembre 1989. Sassari: Gallizzi 1990, 907–32. Clark, Elizabeth A., Reading Renunciation: Asceticism and Scripture in Early Christianity. Princeton, NJ: Princeton University Press 1999. Clark, Gillian, ‘Bodies and Blood: Late Antique Debate on Martyrdom, Virginity and Resurrection’, in D. Montserrat (ed.), Changing Bodies, Changing Meanings: Studies on the Human Body in Antiquity. London: Routledge 1998, 99–115. Clark, Patricia, ‘Women, Slaves and the Hierarchies of Domestic Violence: The Family of St Augustine’, in S. R. Joshel and S. Murnaghan (eds.), Women and Slaves in Greco-Roman Culture: Differential Equations. London: Routledge 1998, 112–33. Claude, Dietrich, ‘Freedmen in the Visigothic Kingdom’, in E. James (ed.), Visigothic Spain: New Approaches. Oxford: Oxford University Press 1980, 159–88. Cohen, Shaye J. D., The Beginnings of Jewishness: Boundaries, Varieties, Uncertainties. Berkeley: University of California Press 1999.

324

Bibliography

Collins, Rob, ‘Soldiers to Warriors: Renegotiating the Roman Frontier in the Fifth Century’, in F. Hunter and K. Painter (eds.), Late Roman Silver: The Traprain Treasure in Context. Edinburgh: Society of Antiquaries of Scotland 2013, 29–43. Hadrian’s Wall and the End of Empire: The Roman Frontier in the 4th and 5th Centuries. New York: Routledge 2012. Collins, Roger, Visigothic Spain, 409–711. Malden, MA: Blackwell 2004. Combes, Isobel A. H., The Metaphor of Slavery in the Writings of the Early Church: From the New Testament to the Beginning of the Fifth Century, Journal for the Study of the New Testament Supplement Series 156. Sheffield: Sheffield Academic Press 1998. Conant, Jonathan, Staying Roman: Conquest and Identity in Africa and the Mediterranean, 439–700. Cambridge: Cambridge University Press 2012. Cone, James H., A Black Theology of Liberation. Maryknoll, NY: Orbis 2010. Conley, Aaron D., ‘Augustine and Slavery: Freedom for the Free’, in T. Delgado, J. Doody, and K. Paffenroth (eds.), Augustine and Social Justice. Lanham: Lexington 2015, 131–44. Conte, Gian Biagio, Genres and Readers: Lucretius, Love Elegy, Pliny’s Encyclopedia. Baltimore, MD: Johns Hopkins University Press 1994. Conybeare, Catherine, ‘Re-Reading St. Patrick’, Journal of Medieval Latin 4 (1994): 39–50. Córcoles Olaitz, Edorta, ‘The Manumission of Slaves in the View of the Visigothic Formulae’, Veleia 23 (2006): 339–49. Corcoran, Gervase, Saint Augustine on Slavery. Rome: Institutum Patristicum Augustinianum 1985. Corcoran, Simon, ‘The Donation and Will of Vincent of Huesca: Latin Text and English Translation’, Antiquitè Tardive 11 (2003): 215–21. Corke-Webster, James, ‘The Early Reception of Pliny the Younger in Tertullian of Carthage and Eusebius of Caesarea’, Classical Quarterly 67 (2017): 247–62. Coşkun, Altay, ‘The Eucharisticos of Paulinus Pellaeus: Towards a Reappraisal of the Worldly Convert’s Life and Autobiography’, Vigiliae Christianae 60 (2006): 285–315. Costantini, Maria Luisa, ‘La menzione di servus e libertus nelle iscrizioni tardoimperiali di Roma’, in I. Di Stefano Manzella (ed.), Le iscrizioni dei Cristiani in Vaticano: materiali e contributi scientifici per una mostra epigrafica, Vol. 2. Vatican City: Edizioni Musei Vaticani 1997, 181–3. Courcelle, Pierre, Histoire Littéraire Des Grandes Invasions Germaniques. Paris: Hachette 1948. Crabtree, Pam J., Early Medieval Britain: The Rebirth of Towns in the PostRoman West. Cambridge: Cambridge University Press 2018. Crone, Patricia, Slaves on Horses: The Evolution of the Islamic Polity. Cambridge: Cambridge University Press 1980. Crum, Walter E. (ed. and trans.), Catalogue of the Coptic Manuscripts in the British Museum. London: British Museum 1905. Culbertson, Laura (ed.), Slaves and Households in the Near East. Chicago, IL: Oriental Institute 2011.

Bibliography

325

Cunningham, Mary C. and Pauline Allen (eds.), Preacher and Audience: Studies in Early Christian and Byzantine Homiletics. Leiden; Boston, MA: Brill 1998. Daley, Brian E. Leontius of Byzantium: Complete Works. Oxford: Oxford University Press 2017. Dark, Kenneth R., ‘The Late Antique Landscape of Britain, AD 300–700’, in N. Christie (ed.), Landscapes of Change: Rural Evolutions in Late Antiquity and the Early Middle Ages. New York: Routledge 2004, 279–300. Britain and the End of the Roman Empire. Stroud: Tempus 2000. Civitas to Kingdom: British Political Continuity 300–800. London: Leicester University Press 1999. ‘St. Patrick’s Villula and the Fifth-Century Occupation of Romano British Villas’, in D. N. Dumville and L. Abrams, Saint Patrick, A.D. 493–1993. Woodbridge, UK: Boydell 1993, 19–33. ‘A Sub-Roman Re-defence of Hadrian’s Wall?’ Britannia 23 (1992): 111–20. Daubigney, Alain and François Favory, ‘L’esclavage en Nairbonaise et Lyonaise’ in Actes du Colloque 1972 Sur l’Esclavage. Paris: Belles Lettres 1974, 315– 88. Davies, Bronwyn, and Rom Harré, ‘Positioning: The Discursive Production of Selves’, Journal for the Theory of Social Behaviour 20 (1990): 43–63. Davis, Stephen J., Monasticism: A Very Short Introduction. Oxford: Oxford University Press 2018. De Bruyn, Theodore, ‘Appeals to the Intercessions of Mary in Greek Liturgical and Paraliturgical Texts from Egypt’, in L. M. Peltomaa, A. Külzer, and P. Allen (eds.), Presbeia Theotokou: The Intercessory Role of Mary across Times and Places in Byzantium (4th–9th Century). Wien: Verlag der Österreichischen Akademie der Wissenschaften 2015, 115–30. ‘Flogging a Son: The Emergence of the pater flagellans in Latin Christian Discourse’, Journal of Early Christian Studies 7.2 (1999): 249–90. De Coulanges, Fustel, ‘Le colonat romain’ in Recherches sur quelques problèmes d’histoire. Paris: Hachette 1894, 3–186. De Ligt, Luuk, ‘Roman Law, Markets and Market Prices’, in P. J. du Plessis, C. Ando, and K. Tuori (eds.), The Oxford Handbook of Roman Law and Society. Oxford: Oxford University Press 2016, 660–70. De Matons, Grosdidier J., ‘Liturgie et Hymnographie: Kontakion et Canon’, Dumbarton Oaks Papers 34/35 (1980–81): 31–43. Romanos le Mélode et les origines de la poésie religieuse à Byzance. Paris: Beauchesne 1977. De Paor, Liam, Saint Patrick’s World: The Christian Culture of Ireland’s Apostolic Age. Dublin: Four Courts 1993. De Wet, Chris L., ‘Modelling Msarrquˉ tā : Humiliation, Christian Monasticism, and the Ascetic Life of Slavery in Late Antique Syria and Mesopotamia’, in K. Cooper and J. Wood (eds.), Social Control in Late Antiquity: The Violence of Small Worlds. Cambridge: Cambridge University Press 2020, 105–30. ‘“The Barbarians Themselves Are Offended by Our Vices”: Slavery, Sexual Vice and Shame in Salvian of Marseilles’ De Gubernatione Dei’, HTS Theological Studies 75.3 (2019): 1–8, online: https://hts.org.za/index.php/hts/article/vie w/5302/12576.

326

Bibliography

The Unbound God: Slavery and the Formation of Early Christian Thought. London: Routledge 2018a. ‘The Great Christian Failure of Mastery: Slavery and Romanness in Salvian of Marseilles’s De gubernatione Dei’, Religion & Theology 25 (2018b): 394–417. ‘Slavery and Asceticism in John of Ephesus’ Lives of the Eastern Saints’, Scrinium 13 (2017): 84–113. ‘The Punishment of Slaves in Early Christianity: The Views of Some Selected Church Fathers’, Acta Theologica Suppl. 23 (2016): 263–82. Preaching Bondage: John Chrysostom and the Discourse of Slavery in Early Christianity. Oakland: University of California Press 2015. ‘Sin as Slavery and/or Slavery as Sin? On the Relationship between Slavery and Christian Hamartiology in Late Ancient Christianity’, Religion & Theology 17 (2010): 26–39. De Wet, Chris L. and Wendy Mayer (eds.), Revisioning John Chrysostom: New Approaches, New Perspectives. Leiden; Boston, MA: Brill 2019. Destephen, Sylvain, ‘Familles d’Anatolie au miroir des MAMA. 043.14’, Epigraphica Anatolica 43 (2010): 135–48. Di Martino, Vittorio, Roman Ireland. Cork: Collins 2003. Díaz, Pablo, ‘El testamento de Vicente: Propietarios y dependientes en la Hispania del s. VI’, in M. J. Hidalgo de la Vega, D. Pérez y Pérez, and M. J. Rodríguez Gervás (eds.), ‘Romanización’ y ‘Reconquista’ en la península Ibérica: Nuevas Perspectivas. Salamanca: Ediciones Universidad de Salamanca 1998, 257–70. Díaz, Pablo and Maria del Rosario Valverde Castro, ‘The Theoretical Strength and the Practical Weakness of the Visigothic Monarchy of Toledo’, in F. Theuws and J. Nelson (eds.), Rituals of Power: From Late Antiquity to the Early Middle Ages. Leiden: Brill 2000, 59–93. Dimambro, Thomas, ‘Women on the Edge: Violence, ‘Othering’, and the Limits of Imperial Power in Euphemia and the Goth’, in K. Cooper and J. Wood (eds.), Social Control in Late Antiquity: The Violence of Small Worlds. Cambridge: Cambridge University Press 2020, 318–36. Dockes, Pierre, Medieval Slavery and Liberation. Chicago: University of Chicago Press 1982. Downing, F. Gerald, ‘Pliny’s Prosecutions of Christians: Revelation and 1 Peter’, Journal for the Study of the New Testament 34 (1988): 105–23. Downs, David, Alms, Charity, Reward, and Atonement in Early Christianity. Waco, TX: Baylor University Press 2016. Dräger, Paul, ‘Bissula – Eliza – Lolita: Priap als „Sprachlehrer“’, Göttinger Forum für Altertumswissenschaft 4 (2001): 187–219. Dresken-Weiland, Jutta, Sarkophagbestattungen des 4.–6. Jahrhunderts im Westen des römischen Reiches. Freiburg: Herder 2003. Dronke, Peter, ‘St. Patrick’s Reading’, Cambridge Medieval Celtic Studies 1 (1981): 21–38. Duane, Anna Mae (ed.), Child Slavery before and after Emancipation: An Argument for Child-Centred Slavery Studies. Cambridge, MA; London: Cambridge University Press 2017.

Bibliography

327

Duby, Georges, The Early Growth of the European Economy: Warriors and Peasants from the Seventh to the Twelfth Century. Ithaca, NY: Cornell University Press 1974. Duckworth, George (ed. and trans.), Complete Roman Drama. New York: Random House 1943. Dumville, David, ‘The Floruit of St. Patrick: Common and Less Common Ground’, in D. Dumville et al. (eds.), Saint Patrick, A.D. 493–1993. Woodbridge, UK: Boydell 1993a, 13–18. ‘Coroticus’, in D. Dumville et al. (eds.), Saint Patrick, A.D. 493–1993. Woodbridge, UK: Boydell 1993b, 107–15. ‘Verba militibus mittenda Corotici. An Analysis of St. Patrick’s Tract on the Crimes of Coroticus’, in D. Dumville et al. (eds.), Saint Patrick, A.D. 493– 1993. Woodbridge, UK: Boydell 1993c, 117–27. ‘Picti apostatae(que)’, in D. Dumville et al. (eds.), Saint Patrick, A.D. 493–1993. Woodbridge, UK: Boydell 1993d, 129–31. Dumville, David et al. (eds.), Saint Patrick, A.D. 493–1993. Woodbridge, UK: Boydell 1993. Duncan Jones, Richard, Power and Privilege in Roman Society. Cambridge: Cambridge University Press 2016. Dunn, Geoffrey, ‘The Poverty of Melania the Younger and Pinianus’, Augustinianum 54.1 (2014): 93–115. Dupont, Anthony, Shari Boodts, Gert Partiens, and Johan Leemans (eds.), Preaching in the Patristic Era: Sermons, Preachers, and Audiences in the Latin West. Leiden; Boston, MA: Brill 2018. Einbach, Diether, Untersuchungen zum spätantiken Kolonat in der Kaiserlichen Gesetzgebung unter Besonderer Berücksichtigung der Terminologie. Bonn: Habelt 1977. Elm, Susanna, ‘“Pierced by Bronze Needles”: Anti-Montanist Charges of Ritual Stigmatization in Their Fourth-Century Context’, Journal of Early Christian Studies 4.4 (1996): 409–39. Elsner, Jaś , The Art of the Roman Empire AD 100–450. Oxford: Oxford University Press 2018. Emmel, Stephen, ‘Shenoute for Historians: The Pneueit Incident (A Monastic Leader and Anti-Pagan Violence in Late Antique Southern Egypt)’, in D. Brakke and S. Davis (eds.), From Gnostics to Monastics: Studies in Coptic and Early Christianity in Honor of Bentley Layton. Leuven: Peeters 2017, 369–407. ‘Shenoute of Atripe and the Christian Destruction of Temples in Egypt: Rhetoric and Reality’, in J. Hahn, S. Emmel, and U. Gotter (eds.), From Temple to Church: Destruction and Renewal of Local Cultic Topography in Late Antiquity. Leiden: Brill 2008, 190–9. Epstein, Steven A., Speaking of Slavery: Color, Ethnicity, and Human Bondage in Italy. Ithaca, NY: Cornell University Press 2001. Esders, Stefan, Die Formierung der Zensualität: Zur kirchlichen Transformation des spätrömischen Patronatswesens im früheren Mittelalter. Ostfildern: Jan Thorbecke Verlag 2010.

328

Bibliography

Esmonde Cleary, A. Simon, ‘Southern Britain in the Fifth Century: A “Collapsed State”?’ in F. Hunter and K. Painter (eds.), Late Roman Silver: The Traprain Treasure in Context. Edinburgh: Society of Antiquaries of Scotland 2013, 45–53. The Ending of Roman Britain. Savage, MD: Barnes & Noble 1990. Etchingham, Colmán, ‘Preface to Paperback Edition’, in E. A. Thompson, Who Was St. Patrick? Bury St. Edmunds: Boydell 1985 [1999], xvii–xxxi. Eubank, Nathan, Wages of Cross-Bearing and Debt of Sin: The Economy of Heaven in Matthew’s Gospel. Berlin: De Gruyter 2013. Evans, John K., War, Women and Children in Ancient Rome. London: Routledge 1991. Evans Grubbs, Judith, ‘Child Enslavement in Late Antiquity and the Middle Ages’, in D. Eltis, S. Engerman, C. Perry, and D. Richardson (eds.), The Cambridge Companion World History of Slavery, Vol. 2. Cambridge: Cambridge University Press 2021, 155–84. ‘Infant Exposure and Infanticide’, in J. Evans Grubbs, T. Parkin, and R. Bell (eds.), The Oxford Handbook of Childhood and Education in the Classical World. New York: Oxford University Press 2013, 83–107. ‘The Dynamics of Infant Abandonment: Motives, Attitudes and (Unintended) Consequences’, in K. Mustakallio and C. Laes (eds.), The Dark Side of Childhood in Late Antiquity and the Middle Ages. Oxford: Oxbow 2011, 21–36. ‘Church, State, and Children: Christian and Imperial Attitudes toward Infant Exposure in Late Antiquity’, in A. Cain and N. Lenski (eds.), The Power of Religion in Late Antiquity. Farnham: Ashgate 2009, 119–31. ‘Abduction Marriage in Antiquity: A Law of Constantine (CTh IX 24.1) and Its Social Context’, Journal of Roman Studies 79 (1989): 59–83. Evelyn-White, Hugh, ‘Introduction’, in Ausonius, I. Cambridge: Harvard University Press 1988a, vii–xliii. Ausonius, II. Cambridge: Harvard University Press 1988b. Fabbrini, Fabrizio, La manumissio in ecclesia. Milano: A. Giuffrè 1965. Fabiano, John, ‘Builders and Integrated Associations in Fourth-Century CE Rome: A New Interpretation of AE 1941, 68’, Journal of Late Antiquity 12.1 (2019): 65–87. Faulkner, Neil, The Decline and Fall of Roman Britain. Stroud: Tempus 2000. Fear, Andrew, ‘St. Patrick and the Art of Allusion’, in R. Morello and A. D. Morrison (eds.), Ancient Letters: Classical and Late Antique Epistolography. Oxford: Oxford University Press 2007a, 325–47. ‘War and Society’, in P. Sabin, H. van Wees, and M. Whitby (eds.), The Cambridge History of Greek and Roman Warfare Volume II: Rome from the Late Republic to the Late Empire. Cambridge: Cambridge University Press 2007b, 424–58. Feissel, Denis, Documents, droit, diplomatique de l’Empire romain tardif. Paris: Association des amis du Centre d’histoire et civilisation de Byzance 2010. Feldman, Allen, Formations of Violence: The Narrative of the Body and Political Terror in Northern Ireland. Chicago: University of Chicago Press 1991. Feldman, Fred, Distributive Justice. Oxford: Oxford University Press 2016.

Bibliography

329

Feldmeier, Reinhard, The First Letter of Peter: A Commentary on the Greek Text. Waco, TX: Baylor University Press 2008. Fernández, Damián, Aristocrats and Statehood in Western Iberia, 300–600 C.E. Philadelphia: University of Pennsylvania Press 2017a. ‘Statehood, Taxation and State Infrastructural Power in Visigothic Iberia’, in C. Ando and S. Richardson (eds.), Ancient States and Infrastructural Power: Europe, Asia, and America. Philadelphia: University of Pennsylvania Press 2017b, 243–71. Fiema, Zbigniew T., ‘Historical Context’, in J. Frösén, A. Arjava, and M. Lehtinen (eds.), The Petra Papyri I. Amman: American Center of Oriental Research 2002, 1–4. Fiema, Zbigniew T., Chrysanthos Kanellopoulos, Tomasz Waliszewski, and Robert Schick, The Petra Church. Amman: American Center of Oriental Research 2001. Finley, Moses I., The Ancient Economy. Berkeley: University of California Press 1999. Ancient Slavery and Modern Ideology. Princeton, NJ: Markus Wiener Publishers 1998. Finn, Richard, Almsgiving in the Later Roman Empire. Oxford: Oxford University Press 2015. Fisher, Greg, ‘Crisis, Provincial Historiography, and Identity in Sub-Roman Britain’, in E. D. Digeser and R. M. Frakes, Religious Identity in Late Antiquity. Toronto: Edgar Kent 2006, 150–91. Fitzgerald, John T., ‘Theodore of Mopsuestia on Paul’s Letter to Philemon’, in D. François Tolmie (ed.), Philemon in Perspective: Interpreting a Pauline Letter. Berlin and New York: Walter de Gruyter, 2010, 333–63. Fitzgerald, William, Slavery and the Roman Literary Imagination. Cambridge: Cambridge University Press 2000. Flechner, Roy, Saint Patrick Retold: The Legend and History of Ireland’s Patron Saint. Princeton, NJ: Princeton University Press 2019. ‘Patrick’s Reasons for Leaving Britain’, in F. Edmonds and P. Russell (eds.), Tome: Studies in Medieval Celtic History and Law in Honour of Thomas Charles-Edwards. Woodbridge, UK: Boydell 2011, 125–33. Fleming, Robin, Britain after Rome: The Fall and Rise. Oxford: Oxford University Press 2010. Fletcher, Richard, The Barbarian Conversion from Paganism to Christianity. Berkeley: University of California Press 1997. Fontaine, Jacques, La naissance de la poésie dans l’occident chrétien. Paris: Études Augustiniennes 1981. Forness, Philip M., Preaching Christology in the Roman Near East: A Study of Jacob of Serugh. Oxford: Oxford University Press 2018. Förster, Richard (ed.), Scriptores physiognomonici Graeci et Latini vol. 2: Physiognomonica anonymi, Pseudopolemonis, Rasis, Secreti secretorum Latine, Anonymi Graece, fragmenta, indices continens. Leipzig: Teubner 1893. Foucault, Michel, Discipline and Punish: The Birth of the Prison, A. Sheridan (trans.). Harmondsworth: Penguin 1979.

330

Bibliography

Fraade, Steven D., From Tradition to Commentary: Torah and Its Interpretation in the Midrash Sifre to Deuteronomy. Albany: State University of New York Press 1991. ‘Sifre Deuteronomy 26 (ad Deut. 3:23): How Conscious the Composition?’ Hebrew Union College Annual 54 (1983): 245–301. Franz, Kurt, ‘Slavery in Islam: Legal Norms and Social Practice’, in R. Amitai and C. Cluse (eds.), Slavery and the Slave Trade in the Eastern Mediterranean (c. 1000–1500 CE). Turnhout: Brepols 2017, 51–141. Fraser, James E., ‘St. Patrick and Barbarian Northern Britain in the Fifth Century’, in F. Hunter and K. Painter (eds.), Late Roman Silver: The Traprain Treasure in Context. Edinburgh: Society of Antiquaries of Scotland 2013, 15–27. Freedman, Paul, The Origins of Peasant Servitude in Medieval Catalonia. Cambridge: Cambridge University Press 1991. Frenkel, Luise M., Theodotus of Ancyra’s Homilies at the Council of Ephesus. Leuven: Peeters 2016. Frier, Bruce W. (ed.), The Codex of Justinian. A New Annotated Translation, with Parallel Latin and Greek Text. Cambridge: Cambridge University Press 2016. Frösén, Jaakko, Antti Arjava, and Marjo Lehtinen (eds.), The Petra Papyri I. Amman: American Center of Oriental Research 2002. Fürst, Alfons, Origenes: Grieche und Christ in römischer Zeit. Stuttgart: Hiersemann 2018. Fynn-Paul, Jeffrey, ‘Empire, Monotheism and Slavery in the Greater Mediterranean Region from Antiquity to the Early Modern Era’, Past and Present 205 (2009): 3–40. Gaca, Kathy L., ‘The Andrapodizing of War Captives in Greek Historical Memory’, Transactions of the American Philological Association 140.1 (2010): 117–61. Gaddis, Michael, There Is No Crime for Those Who Have Christ: Religious Violence in the Christian Roman Empire. Berkeley: University of California Press 2005. Gamauf, Richard, ‘Sklavenkinder in den Rechtsquellen’, in H. Heinen (ed.), Kindersklaven – Sklavenkinder. Schicksale zwischen Zuneigung und Ausbeutung in der Antike und im interkulturellen Vergleich. Stuttgart: Franz Steiner 2012, 231–60. García Moreno, Luis, ‘From Coloni to Servi. A History of the Peasantry in Visigothic Spain’, Klio 83 (2001): 198–212. Garnsey, Peter, ‘Introduction’, in A. Bowen and P. Garnsey (eds. and trans.), Lactantius: Divine Institutes, Translated Texts for Historians 40. Liverpool: Liverpool University Press 2003, 1–55. Ideas of Slavery from Aristotle to Augustine. Cambridge: Cambridge University Press 1996. ‘Philo Judaeus and Slave Theory’, Scripta Classica Israelica 13 (1994): 30–45. Social Status and Legal Privilege in the Roman Empire. Oxford: Clarendon 1970. Garnsey, Peter and Caroline Humfress, The Evolution of the Late Antique World. Cambridge: Orchard Academic Press 2001.

Bibliography

331

George, Michele (ed.), Roman Slavery and Roman Material Culture. Toronto: University of Toronto Press 2013. Geroussi-Bendermacher, Eugenia, ‘Propriété foncière et inventaire d’esclaves: Un texte inédit de Perissa (Thera) tardo-antique’, in V. Anastasiadis and P. Doukellis (eds.), Esclavage antique et discriminations socio-culturelles. Bern: Lang 2005, 325–58. Giacchero, Marta, Edictum Diocletiani et Collegarum de pretiis rerum venalium in integrum fere restitutum e Latinis Graecisque fragmentis, Vol. 1. Genova: Istituto di storia antica e scienze ausilirie 1974. Giardina, Andrea, ‘Esplosione di Tardoantico’, Studi Storici 40.1 (1999): 157–80. Gielen, Marlis, ‘Der 1. Petrusbrief und Kaiser Hadrian: zur Frage der zeitgeschichtlichen Einordnung des 1. Petrusbriefes’, Biblische Zeitschrift 57.2 (2013): 161–83. Gil, Juan (ed.), ‘Formulae Visigothicae’, in J. Gil (ed.), Miscellanea Wisigothica, 2nd ed. Seville: Publicaciones de la Universidad de Sevilla 1991, 69–112. Gilli-Elewy, Hend, ‘On the Provenance of Slaves in Mecca during the Time of the Prophet Muhammad’, International Journal of Middle East Studies 49 (2017): 164–8. Gilliot, Claude, ‘Mujā hid’s Exegesis: Origins, Paths of Transmission and Development of a Meccan Exegetical Tradition in Its Human, Spiritual and Theological Environment’, in A. Görke and J. Pink (eds.), Tafsıˉ r and Islamic Intellectual History: Exploring the Boundaries of a Genre. Oxford: Oxford University Press 2014, 63–111. Glancy, Jennifer A., ‘Body Work: Slavery and the Pauline Churches’, in D. A. Pargas and F. Roş u (eds.), Critical Readings on Global Slavery, Vol. 2. Leiden: Brill 2018, 427–76. ’Slavery in Acts of Thomas’, Journal of Early Christian History 2.2 (2012): 3–21. ‘Slavery and the Rise of Christianity’, in K. Bradley and P. Cartledge (eds.), The Cambridge World History of Slavery, Vol. 1: The Ancient Mediterranean World. Cambridge: Cambridge University Press 2011, 456–81. ‘Christian Slavery in Late Antiquity’, in R. Hörmann and G. Mackenthun (eds.), Human Bondage in the Cultural Contact Zone: Transdisciplinary Perspectives on Slavery and Its Discourses. Münster: Waxmann 2010, 63–80. Slavery in Early Christianity. New York: Oxford University Press 2002. Goffman, Erving, Encounters: Two Studies in the Sociology of Interaction. Harmondsworth: Penguin Academic 1961. Golden, Mark, ‘Pais, « Child » and « Slave »’, L’antiquité classique 54 (1985): 91–104. Goldenberg, David M., ‘Racism, Color Symbolism, and Color Prejudice’, in M. Eliav-Feldon, B. H. Isaac, and J. Ziegler (eds.), The Origins of Racism in the West. Cambridge: Cambridge University Press 2009, 88–108. The Curse of Ham: Race and Slavery in Early Judaism, Christianity, and Islam. Princeton, NJ: Princeton University Press 2005. Goldman, Max L., ‘Ethnicity in Writers of Physiognomica’, in R. F. Kennedy and M. Jones-Lewis (eds.), The Routledge Handbook of Identity and the

332

Bibliography

Environment in the Classical and Medieval Worlds. London: Routledge 2015, 62–74. Goldmann, Emil, Beiträge zur Geschichte der germanischen Freilassung durch Wehrhaftmachung. Breslau: Marcus 1904. Gordon, Matthew S., The Breaking of a Thousand Swords: A History of the Turkish Military of Samarra (AH 200–275/815–889 CE). New York: SUNY Press 2001. Gordon, Matthew S. and Kathryn A. Hain (eds.), Concubines and Courtesans: Women and Slavery in Islamic History. Oxford: Oxford University Press 2017. Gordon, Murray, Slavery in the Arab World. Lanham, MD: Rowman & Littlefield 1989. Grabar, André, L’empereur dans l’art byzantin: recherches sur l’art officiel de l’empire d’Orient. Paris: Les Belles lettres 1936. Graiver, Inbar, Asceticism of the Mind: Forms of Attention and SelfTransformation in Late Antique Monasticism. Toronto: Institute of Mediaeval Studies 2018. Graumann, Lutz Alexander, ‘Children’s Accidents in the Roman Empire: The Medical Eye on 500 Years of Mishaps in Injured Children’, in C. Laes and V. Vuolanto (eds.), Children and Everyday Life in the Roman and Late Antique World. London and New York: Routledge 2017, 267–86. Grégoire, Henri (ed.), Recueil des inscriptions grecques chrétiennes de l’Asie Mineure. Paris: Leroux 1922. GRETA, 7th General Report on GRETA’S Activities. General report, VII. Strasbourg: Council of Europe 2018. Grey, Cam, Constructing Communities in the Late Roman Countryside. Cambridge: Cambridge University Press 2011a. ‘Slavery in the Late Roman World’, in K. Bradley and P. Cartledge (eds.), The Cambridge World History of Slavery: Volume 1, The Ancient Mediterranean World. Cambridge: Cambridge University Press 2011b, 482–509. ‘Contextualizing Colonatus: The Origo of the Late Roman Empire’, Journal of Roman Studies 97 (2007): 155–75. ‘Salvian, the Ideal Christian Community, and the Fate of the Poor in FifthCentury Gaul’, in M. Atkins and R. Osborne (eds.), Poverty in the Roman World. Cambridge: Cambridge University Press 2006, 168–82. Grieser, Heike, ‘Loskauf II (Gefangener)’, Reallexikon für Antike und Christentum 23 (2010): 510–26. Griffin, Svetla S. and Ilaria Ramelli (eds.), Lovers of Souls and Lovers of Bodies: Philosophical and Religious Perspectives in Late Antiquity. Washington, DC: Harvard Center for Hellenic Studies forthcoming. Gruen, Erich S., Rethinking the Other in Antiquity. Princeton, NJ: Princeton University Press 2011. Gustafson, Mark, ‘The Tattoo in the Later Roman Empire and Beyond’, in J. Caplan (ed.), Written on the Body: The Tattoo in European and American History. Princeton, NJ: Princeton University Press 2000, 17–31. Guyot, Peter, Eunuchen als Sklaven und Freigelassene in der griechisch-römischen Antike. Stuttgart: Klett-Cotta 1980.

Bibliography

333

Haarer, Fiona K. (ed.), with R. Collins, K. J. Fitzpatrick-Matthews, S. Moorhead, D. Petts, and P. J. Walton. AD 410: The History and Archaeology of Late and Post-Roman Britain. London: Society for the Promotion of Roman Studies 2014. Haas, Christopher, Alexandria in Late Antiquity: Topography and Social Conflict. Baltimore, MD: Johns Hopkins University Press 1997. Hanson, Richard P. C., The Life and Historical Writings of St. Patrick. New York: Seabury 1983. Saint Patrick: Confession et Lettre à Coroticus, Sources Chrétiennes 249. Paris: Cerf 1978. Harmless, J. William, ‘Monasticism’, in S. Ashbrook Harvey and D. Hunter (eds.), The Oxford Handbook of Early Christian Studies. Oxford: Oxford University Press 2008, 493–520. Harnack, Adolf von, The Mission and Expansion of Christianity in the First Three Centuries, Vol. 1, J. Moffatt (ed. and trans.). London: Williams and Norgate 1908. Harper, Kyle, ‘The Transformation of Roman Slavery: An Economic Myth?’, Antiquité Tardive 20 (2012): 165–72. Slavery in the Late Roman World, AD 275–425: An Economic, Social, and Institutional Study. Cambridge: Cambridge University Press 2011. ‘Slave Prices in Late Antiquity (and in the Very Long Term)’, Historia 59 (2010): 206–38. ‘The Greek Census Inscriptions of Late Antiquity’, Journal of Roman Studies 98 (2008): 83–119. Harrill, J. Albert, ‘Slavery and Inhumanity: Keith Bradley’s Legacy on Slavery in New Testament Studies’, Biblical Interpretation 21 (2013): 506–14. ‘The Metaphor of Slavery in the Writings of Tertullian’, Studia Patristica 42 (2006a): 385–90. Slaves in the New Testament: Literary, Social, and Moral Dimensions. Minneapolis, MN: Fortress, 2006b. The Manumission of Slaves in Early Christianity. Tübingen: Mohr Siebeck 1995. Harris, Bruce Fairgray, ‘Bithynia: Roman Sovereignty and the Survival of Hellenism’, Aufstieg und Niedergang der römischen Welt 2.7.2 (1980): 857–901. Harris, William V., Restraining Rage: The Ideology of Anger Control in Classical Antiquity. Cambridge, MA: Harvard University Press 2001. ‘Demography, Geography and the Sources of Roman Slaves’, Journal of Roman Studies 89 (1999): 62–75. Harvey, Susan Ashbrook, ‘Sacred Bonding: Mothers and Daughters in Early Syriac Hagiography’, Journal of Early Christian Studies 4.1 (1996): 27–56. Hayes-Healy, Stephanie, ‘Saint Patrick’s Journey into the Desert: Confessio 16–28 as Ascetic Discourse’, Archivium Hibernicum 59 (2005): 237–59. Heather, Peter, Goths and Romans, 332–489. Oxford: Clarendon 1991. ‘The Anti-Scythian Tirade of Synesius’ De Regno’, Phoenix 42 (1988): 152–72. Heather, Peter and John Matthews, The Goths in the Fourth Century. Liverpool: Liverpool University Press 1991.

334

Bibliography

Heinen, Heinz, ‘Die „Bissula“des Ausonius oder die Kunst der Romanisierung’, in M. Weinmann-Walser (ed.), Historische Interpretationen: Gerold Walser zum 75. Geburtstag. Stuttgart: Franz Steiner 1995, 81–96. Herren, Michael W., ‘Patrick, Gaul, and Gildas: A New Lens on the Apostle of Ireland’s Career’, in S. Sheehand, J. Findon, and W. Follett (eds.), Gablanach in Scelaigecht: Celtic Studies in honour of Ann Dooley. Dublin: Four Courts 2013, 9–25. ‘Mission and Monasticism in the Confessio of Patrick’, in D. Ó Corráin, L. Breatnach, and K. McCone (eds.), Sages, Saints and Storytellers: Celtic Studies in Honour of Professor James Carney, Maynooth Monographs 2. Maynooth: An Sagart 1989, 76–85. Herrmann-Otto, Elisabeth, Ex ancilla natus: Untersuchungen zu den “hausgeborenen” Sklaven und Sklavinnen im Westen des Römischen Kaiserreiches. Stuttgart: Franz Steiner 2014. ‘Kindsein im römischen Reich’, in H. Heinen (ed.), Kindersklaven – Sklavenkinder. Schicksale zwischen Zuneigung und Ausbeutung in der Antike und im interkulturellen Vergleich. Stuttgart: Franz Steiner 2012, 171–202. Ex Ancilla Natus. Untersuchungen zu den ‘hausgeborenen’ Sklaven und Sklavinnen im Westen des Römischen Kaiserreiches. Stuttgart: F Steiner 1994 Herz, Zachary, ‘The Effect of Bankruptcy Law on Roman Credit Markets’, Business & Bankruptcy Law Journal 7 (2015): 207–49. Hezser, Catherine, Review of Ramelli 2016b. Journal of Religion 99.3 (2019): 395–7. ‘Greek and Roman Slaving in Comparative Ancient Perspective: The Level of Integration’, in S. Hodkinson, M. Kleijwegt, and K. Vlassopoulos (eds.), The Oxford Handbook of Greek and Roman Slaveries, online ed. Oxford: Oxford University Press 2016, 1–23, online: https://bit.ly/3BVcsbC. ‘Der Loskauf von Sklaven und Kriegsgefangenen im antiken Judentum’, in H. Grieser and N. Priesching (eds.), Gefangenenloskauf im Mittelmeerraum: Ein interreligiöser Vergleich. Hildesheim: Georg Olms 2015, 3–23. Jewish Slavery in Antiquity. Oxford: Oxford University Press 2005. Hickey, Todd M., ‘Aristocratic Landholding and the Economy of Byzantine Egypt’, in R. S. Bagnall (ed.), Egypt in the Byzantine World, 300–700. Cambridge: Cambridge University Press 2007, 288–308. Higham, Nick J., ‘From Sub-Roman Britain to Anglo-Saxon England: Debating the Insular Dark Ages’, History Compass 2 (2004): 1–29. ‘Literary Evidence for Villas, Towns and Hillforts in Fifth-Century Britain’, Britannia 25 (1994): 229–32. Hilberg, Isidor (ed.), Sancti Eusebii Hieronymi: Epistulae. Vienna: Tempsky 1913. Hillner, Julia, ‘Monks and Children: Corporal Punishment in Late Antiquity’, European Review of History 16.6 (2009): 773–91. ‘Die Berufsangaben und Adressen auf den stadtrömischen Sklavenhalsbändern’, Historia 50 (2001): 193–216. Hoffmann, Dietrich, Das Spätrömische Bewegungsheer und die Notitia Dignitatum, Vol. 1. Düsseldorf: Rheinland-Verlag 1970.

Bibliography

335

Das Spätrömische Bewegungsheer und die Notitia Dignitatum, Vol. 1. Düsseldorf: Rheinland-Verlag 1969. Honigman, Sylvie, ‘Abraham in Egypt: Hebrew and Jewish-Aramaic Names in Egypt and Judea in Hellenistic and Early Roman Times’, Zeitschrift für Papyrologie und Epigraphik 146 (2004): 279–97. Hood, Allan B. E., St. Patrick: His Writings and Muirchú’s Life. Totowa, NJ: Rowman & Littlefield 1978. Hoogendijk, Francisca A. J., ‘Byzantinischer Sklavenkauf’, Archiv für Papyrusforschung und verwandte Gebiete 42.2 (1996): 225–34. Hopkins, Keith, ‘Elite Mobility in the Roman Empire’, Past & Present 32 (1965): 12–26. ‘Eunuchs in Politics in the Later Roman Empire’, The Cambridge Classical Journal 9 (1963): 62–80. Horn, Cornelia, ‘Children in Oriental Christian and Greek Hagiography from the Early Byzantine World (ca. 400–800 CE)’, in R. Aasgaard, C. Horn, and O. M. Cojocaru (eds.), Childhood in History: Perceptions of Children in the Ancient and Medieval Worlds. London: Routledge 2018, 174–92. Howlett, David, ‘Ex saliva Scripturae meae’, in D. Ó Corráin, L. Breatnach, and K. McCone (eds.), Sages, Saints and Storytellers: Celtic Studies in Honour of Professor James Carney, Maynooth Monographs 2. Maynooth: An Sagart 1989, 86–101. Hübner, Sabine, ‘Adoption and Fosterage in the Ancient Eastern Mediterranean’, in J. Evans Grubbs and T. Parkin (eds.), The Oxford Handbook on Childhood and Education in the Classical World. Oxford: Oxford University Press 2013, 510–31. Hunt, Edward D., ‘Melania the Younger’, in T. Whitmarsh (ed.), Oxford Classical Dictionary. Oxford: Oxford University Press 2016, online: https://bit.ly /2Xons1d9. ‘Eunuch’, in S. Hornblower and A. Spawforth (eds.), The Oxford Classical Dictionary. 3rd ed. Oxford: Oxford University Press 1996, 569. Hunt, Peter, Ancient Greek and Roman Slavery. Malden: Wiley Blackwell 2018. Hunter, Fraser and Kenneth Painter (eds.), Late Roman Silver: The Traprain Treasure in Context. Edinburgh: Society of Antiquaries of Scotland 2013. Ibn al-Kutiya, Historia de la conquista de España de Abenalcotía el cordobés, Vol. 2. Julián Ribera (ed.). Madrid: M. Rivadeneyra 1926. Ilian, Bridget L. Forgive Us Our Debts: Financial Metaphors for the Moral Life in the Gospel of Matthew. Doctoral thesis, Lutheran School of Theology at Chicago 2014. Isaac, Benjamin, The Invention of Racism in Classical Antiquity. Princeton, NJ: Princeton University Press 2004. Isbell, Harold, ‘Decimus Magnus Ausonius: The Poet and His World’, in J. W. Binns (ed.), Latin Literature of the Fourth Century. London: Routledge 1974, 22–57. Jacquemard-Le Saos, Catherine, ‘Introduction’, in Querolus. Les Grincheux: Comedie de la Petite Marmite. Paris: Belles Lettres 1994, vii–lxix. James, Edward, Britain in the First Millennium. London: Bloomsbury 2001.

336

Bibliography

Jefferson, Lee M. and Robin M. Jensen (eds.), The Art of Empire: Christian Art in Its Imperial Context. Minneapolis, MN: Fortress Press 2015. Johnson, Scott F. (ed.), The Oxford Handbook of Late Antiquity. Oxford: Oxford University Press 2012. Johnston, Elva, ‘Ireland in Late Antiquity: A Forgotten Frontier?’ Studies in Late Antiquity 1 (2017): 107–23. Jones, Arnold Hugh Martin, The Later Roman Empire, 284–602: A Social Economic and Administrative Survey. Baltimore, MD: Johns Hopkins University Press 1992 [1973, 1964]. The Criminal Courts of the Roman Republic and Principate. Oxford: Blackwell 1972. ‘Census Records of the Later Roman Empire’, Journal of Roman Studies 43 (1953): 49–64. Jones, Michael E., The End of Roman Britain. Ithaca, NY: Cornell University Press 1996. Joshel, Sandra R., Slavery in the Roman World. Cambridge: Cambridge University Press 2010. Joshel, Sandra R., and Lauren Hackworth Petersen, The Material Life of Roman Slaves. Cambridge: Cambridge University Press 2017. Jurasz, Izabela, ‘Carpocrate et Epiphane: chrétiens et platoniciens radicaux’, Vigiliae Christianae 71 (2017): 134–67. Kaçar, Turhan, ‘Early Christianity in the Black Sea: An Examination of the Literary Evidence’, in M. Dogan-Alparslan, M. Alparslan, H. Peker, and Y. G. Ergin (eds.), Colloquium Anatolicum Anadolu Sohbetleri VII. Istanbul: Türk Eskiçağ Bilimleri Enstitüsü 2008, 197–216. Kahlos, Maijastina, ‘Handmaid’s Tale in a Fourth-Century Format: Ausonius and Bissula’, Jahrbuch für Antike und Christentum 63, Forthcoming. Kartzow, Marianne B., The Slave Metaphor and Gendered Enslavement in Early Christian Discourse: Double Trouble Embodied. London: Routledge 2018a. ‘Slave Children in the First-Century Jesus Movement’, in R. Aasgaard, C. Horn, and O. M. Cojocaru (eds.), Childhood in History: Perceptions of Children in the Ancient and Medieval Worlds. London and New York: Routledge 2018b, 111–26. ‘Reproductive Capital and Slave Surrogacy: Thinking about/with/beyond Hagar’, in A. H. Grung, M. B. Kartzow, and A. R. Solevåg (eds.), Bodies, Borders, Believers: Ancient Texts and Present Conversations. Eugene, OR: Wipf & Stock 2015, 396–409. ‘“Asking the Other Question”: An Intersectional Approach to Galatians 3:28 and the Colossian Household Codes’, Biblical Interpretation 18 (2010): 364– 89. Keegan, Peter, ‘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; Buffalo; London: University of Toronto Press 2013, 69–98. King, P. D., Law and Society in the Visigothic Kingdom. Cambridge: Cambridge University Press 1972.

Bibliography

337

King, Wilma, Stolen Childhood: Slave Youth in Nineteenth Century America. Bloomington: Indiana University Press 1995. Kinzig, Wolfram, In Search of Asterius: Studies on the Authorship of the Homilies on the Psalms. Göttingen: Vandenhoeck & Ruprecht 1990. Kiria, Ekaterina, ‘Die vierte Bitte des Vaterunsers bei Gregor von Nyssa’, in E. Moutsoulas (ed.), Jesus Christ in St. Gregory of Nyssa’s Theology: Minutes on the Ninth International Conference on St. Gregory of Nyssa. Athens: Eptalofos 2005, 653–67. Klein, Richard, Die Haltung der kappadokischen Bischöfe Basilius von Caesarea, Gregor von Nazianz und Gregor von Nyssa zur Sklaverei. Stuttgart: Steiner 2000. Die Sklaverei in der Sicht der Bischöfe Ambrosius und Augustinus. Stuttgart: Franz Steiner 1988. Klingshirn, William, ‘Charity and Power: Caesarius of Arles and the Ransoming of Captives in Sub-Roman Gaul’, Journal of Roman Studies 75 (1985): 183–203. Koch, John T., ‘The Early Chronology for St Patrick (c.351–c.428): Some New Ideas and Possibilities’, in J. Cartwright (ed.), Celtic Hagiography and Saints’ Cults. Cardiff: University of Wales Press 2003, 102–22. Konstan, David, Review of Ramelli 2016b. The Classical World 111.2 (2018): 275–6. Roman Comedy. Ithaca, NY: Cornell University Press 1983. Korpela, Jukka, ‘The Baltic Finnic People in the Medieval and Pre-Modern Eastern European Slave Trade’, Russian History 41.1 (2014): 85–117. Krawiec, Rebecca, ‘Reading Abraham in the White Monastery: Fathers, Sources, and History’, in D. Brakke and S. Davis (eds.), From Gnostics to Monastics: Studies in Coptic and Early Christianity in Honor of Bentley Layton. Leuven: Peeters 2017, 455–74. ‘Asceticism’, in S. Ashbrook Harvey and D. Hunter (eds.), The Oxford Handbook of Early Christian Studies. Oxford: Oxford University Press 2008, 764–85. Shenoute and the Women of the White Monastery. Oxford: Oxford University Press 2002. Kreuzsaler, Claudia, ‘Loan Contracts Serving Other Purposes’, in J. Keenan, J. Manning, and U. Yiftach-Firanko (eds.), Law and Legal Practice in Egypt from Alexander to the Arab Conquest. A Selection of Papyrological Sources. Cambridge: Cambridge University Press 2014, 265–74. Krueger, Derek, Liturgical Subjects: Christian Ritual, Biblical Narrative, and the Formation of the Self in Byzantium. Philadelphia: University of Pennsylvania Press 2014. Laes, Christian, Children in the Roman Empire. Outsiders Within. Cambridge: Cambridge University Press 2011. ‘Child Slaves at Work in Roman Antiquity’, Ancient Society 38 (2008): 235–83. ‘Child Beating in Roman Antiquity: Some Reconsiderations’, in K. Mustakallio, J. Hanska, H.-L. Sainio, and V. Vuolanto (eds.), Hoping for Continuity: Childhood, Education and Death in Antiquity and the Middle Ages. Rome: Institutum Romanum Finlandiae 2005, 75–89.

338

Bibliography

‘Desperately Different? Delicia Children in the Roman Household’, in D. L. Balch and C. Osiek (eds.), Early Christian Families in Context: An Interdisciplinary Dialogue. Grand Rapids; Cambridge: Eerdmans 2003, 298–324. Laeuchli, Samuel, Power and Sexuality: The Emergence of Canon Law at the Synod of Elvira. Philadelphia, PA: Temple University Press 1972. Lagarrigue, Georges (trans.), Salvien de Marseille, Ouvres II: Du gouvernment de Dieu. Paris: Les Editions du Cerf 1975. Lambert, David, History and Community in the Works of Salvian of Marseille. DPhil Thesis. University of Oxford, Oxford 2002. ‘The Barbarians in Salvian’s De Gubernatione Dei’, in S. Mitchell and G. Greatrex (eds.), Ethinicity and Culture in Late Antiquity. London: Duckworth 2000, 103–15. ‘The Uses of Decay: History in Salvian’s De Gubernatione Dei’, Augustinian Studies 30 (1999): 115–30. Lambertz, Max, Die Griechischen Sklavennamen. Vienna: Lambertz 1907. Lampe, Geoffrey W. H., A Patristic Greek Lexicon. Oxford: Oxford University Press 2010. Lang, T. J., ‘Disbursing the Account of God: Fiscal Terminology and the Economy of God in Colossians 1,24–25’, Zeitschrift für die Neutestamentliche Wissenschaft und die Kunde der älteren Kirche 107.1 (2016): 116–36. Lareau, Annette, and Erin McNamara Horvat, ‘Moments of Social Inclusion and Exclusion: Race, Class, and Cultural Capital in Family-School Relationships’, Sociology of Education 72 (1999): 37–53. Lavan, Myles, Slaves to Rome: Paradigms of Empire in Roman Culture. Cambridge: Cambridge University Press 2013. Layton, Bentley (ed. and trans.), The Canons of Our Fathers: Monastic Rules of Shenoute. Oxford: Oxford University Press 2014. Leander, Hans, Discourses of Empire: The Gospel of Mark from a Postcolonial Perspective. Atlanta, GA: Society of Biblical Literature 2013. Lee, A. D., War in Late Antiquity: A Social History. Oxford; Malden, MA: Blackwell 2007. Leemans, Johan, Geert Roskam, and Josien Segers (eds.), John Chrysostom and Severian of Gabala: Homilists, Exegetes and Theologians. Leuven: Peeters 2019. Leipoldt, Johannes (ed. and trans.), Sinuthii Archimandritae: Vita et Opera Omnia, Vol 4. Paris: Imprimerie Nationale 1913. Lekkas, Georgios, Liberté et progrès chez Origène. Turnhout: Brepols 2001. Lendon, Jon E., ‘Roman Honor’, in M. Peachin (ed.), The Oxford Handbook of Social Relations in the Roman World. Oxford: Oxford University Press 2011, 377–403. Lenski, Noel, ‘Peasant and Slave in Late Antique North Africa, c. 100–600’, in R. Lizzi Testa (ed.), Late Antiquity in Contemporary Debate. Cambridge: Cambridge Scholars Publishing 2017, 113–55. ‘Violence and the Roman Slave’, in G. G. Fagan and W. Reiss (ed.), The Topography of Violence in the Greco-Roman World. Cambridge: Cambridge University Press 2016, 275–98.

Bibliography

339

‘Constantine and Slavery: Libertas and the Fusion of Roman and Christian Values’, Atti Dell’Accademia Romanistica Costantiniana 19 (2011a): 235– 60. ‘Captivity and Romano-Barbarian Interchange’, in R. W. Mathisen and D. Shanzer (eds.), Romans, Barbarians, and the Transformation of the Roman World: Cultural Interaction and the Creation of Identity in Late Antiquity. Aldershot: Ashgate 2011b, 185–98. ‘Captivity and Slavery among the Saracens in Late Antiquity (ca. 250–630)’, Antiquité Tardive, 19 (2011c): 237–66. ‘Schiavi armati e formazione di eserciti privati nel mondo tardoantico’, in G. Urso (ed.), Ordine e sovversione nel mondo greco e romano: Atti del convegno internazionale, Cividale del Friuli, 25–27 settembre 2008. Pisa: ETS 2009, 145–75. ‘Captivity, Slavery and Cultural Exchange between Rome and the Germans from the First to the Seventh Century CE’, in C. M. Cameron (ed.), Invisible Citizens: Captives and Their Consequences. Salt Lake City: University of Utah Press, 2008, 80–109. ‘Servi Publici in Late Antiquity’, in J.-U. Krause and C. Witschel (eds.), Die Stadt in der Spätantike – Niedergang oder Wandel? Akten des internationalen Kolloquiums in München am 30. Und 31. Mai 2003. Stuttgart: Steiner 2006, 335–57. ‘Slavery in the Byzantine Empire’, in D. Eltis, S. Engerman, C. Perry, and D. Richardson (eds.), The Cambridge World History of Slavery, Vol. 2: 500–1420 CE. Cambridge: Cambridge University Press 2021, 453–81. Leroy, François J., L’homilétique de Proclus de Constantinople. Vatican City: Biblioteca Apostolica Vaticana 1967. Leveau, Philippe, ‘The Western Provinces’, in W. Scheidel, I. Morris and R. Saller (eds.), The Cambridge Economic History of the Greco-Roman World. Cambridge: Cambridge University Press 2007, 651–70. Lewis, Bernard, Race and Slavery in the Middle East: An Historical Enquiry. Oxford: Oxford University Press 1992. Liebeschuetz, J. H. W. G., ‘Goths and Romans in the Leges Visigothorum’, in J. H. W. G Liebeschuetz (ed.), East and West in Late Antiquity: Invasion, Settlement, Ethnogenesis and Conflicts of Religion. Leiden: Brill 2015, 89–104. Barbarians and Bishops: Army, Church, and State in the Age of Arcadius and Chrysostom. Oxford: Oxford University Press 1990. Liebs, Detlef, Römische Jurisprudenz in Gallien (2. bis 8. Jahrhundert). Berlin: Duncker & Humblot 2002. ‘Sklaverei aus Not im germanisch-römischen Recht’, Zeitschrift der SavignyStiftung für Rechtsgeschichte 118 (2001): 286–311. Lightfoot, J. B. and Harmer, J. R., The Apostolic Fathers. Revised Greek texts with introductions and English translations. Berkeley, CA: Apocryphile Press, 2004 (originally published in 1891). Linder, Amnon, ‘The Legal Status of the Jews in the Roman Empire’, in S. T. Katz (ed.), The Cambridge History of Judaism Volume 4: The Late RomanRabbinic Period. Cambridge: Cambridge University Press 2006, 128–73.

340

Bibliography

Lingas, Alexander, ‘The Liturgical Place of Kontakion in Constantinople’, in K. Akentyev (ed.), Liturgy, Architecture, and Art in the Byzantine World: Papers of the XVIII International Byzantine Congress (Moscow, 8– 15 August 1991) and Other Essays Dedicated to the Memory of J. Meyendorff. St Petersburg: Vizantinorossika 1995, 50–7. Llewellyn Ihssen, Brenda, Review of Ramelli 2016b. Journal of Theological Studies 69.1 (2018): 318–20. Lloyd-Jones, Hugh, ‘Artemis and Iphigeneia’, Journal of Hellenic Studies 103 (1983): 87–102. Long, A. A., and D. N. Sedley, The Hellenistic Philosophers Vol. 1: Translations of the Principal Sources with Philosophical Commentary. Cambridge: Cambridge University Press 1987. López, Ariel G., Shenoute of Atripe and the Uses of Poverty: Rural Patronage, Religious Conflict and Monasticism in Late Antique Egypt. Berkeley: University of California Press 2013. Loscertales de G. de Valdeavellano, Pilar (ed.), Tumbos del Monasterio de Sobrado de los Monjes. Madrid: Dirección General del Patrimonio Artístico y Cultural & Archivo Histórico Nacional 1976. Luckritz Marquis, Christine, Death of the Desert: Monastic Memory and the Loss of Egypt’s Golden Age. Philadelphia: University of Pennsylvania Press 2022 MacDonald, Margaret Y., The Power of Children: The Construction of Christian Families in the Greco-Roman World. Waco, TX: Baylor University Press 2014. ‘Reading the New Testament Household Codes in Light of New Research on Children and Childhood in the Roman World’, Studies in Religion 41.3 (2012): 376–87. MacMullen, Ramsay, ‘Late Roman Slavery’, Historia 36 (1987): 359–82. Malamud, M., Rutilius Namatianus’ Going Home: De reditu suo. Translated and with an Introductory Essay. London: Routledge 2016. Mandatori, Gianluca, Review of Ramelli 2016b. Augustinianum 67 (2017): 264– 70. Mander, Jason, Portraits of Children on Roman Funerary Monuments. Cambridge: Cambridge University Press 2013. Mann, William E., ‘Augustine on Evil and Original Sin’, in E. Stump and N. Kretzmann (eds.), The Cambridge Companion to Augustine. Cambridge: Cambridge University Press 2001, 40–8. Marguerat, Daniel, Paul in Acts and Paul in His Letters. Tübingen: Mohr Siebeck 2013. Marin, Emilio, et al. (eds.), Salona IV: Inscriptions de Salone chrétienne, IVe-VIIe siècles. 2 Vols. Rome/Split: Ecole Française de Rome & Arheološki muzej Split 2010. Martin, Brice L., Christ and the Law in Paul. Leiden: Brill 1989. Martin, Dale B., ‘Introduction’, in D. B. Martin and P. Cox Miller (eds.), The Cultural Turn in Late Ancient Studies: Gender, Asceticism, and Historiography. Durham, NC: Duke University Press 2005, 1–24.

Bibliography

341

Slavery as Salvation: The Metaphor of Slavery in Pauline Christianity. New Haven, CT: Yale University Press 1990. Marx, Benedikt, Procliana. Untersuchung über den homilitischen Nachlass des Patriarchen Proklos von Konstantinopel. Münster: Aschdorff 1940. Mathews, Thomas F., The Clash of Gods: A Reinterpretation of Early Christian Art. Princeton, NJ: Princeton University Press 1995. Matthews, John, The Journey of Theophanes: Travel, Business, and Daily Life in the Roman East. New Haven, CT: Yale University Press 2006. Western Aristocracies and Imperial Court. AD 364–425. Oxford: Clarendon Press 1990. Maxwell, Jaclyn, 2018, ‘Sermons’, in S. McGill and E. J. Watts (eds.), A Companion to Late Antique Literature. Hoboken, NJ: Wiley Blackwell 2018, 343–58. McCormick, Michael, ‘Slavery from Rome to Medieval Europe and Beyond: Words, Things, and Genomes’, in J. Bodel and W. Scheidel (eds.), On Human Bondage: After Slavery and Social Death. Oxford: Wiley-Blackwell 2017, 249–64. Origins of the European Economy: Communications and Commerce A.D. 300–900. Cambridge: Cambridge University Press 2001. McCoskey, Denise E., Race: Antiquity and Its Legacy. London: I.B. Tauris 2012. McDermott, Scott, ‘Epistola Christi usque ad ultimum terrae: The Confession of St. Patrick’, The Ancient World 35 (2004): 3–38. McDonald, David A., ‘Poetics and the Performance of Violence in Israel/ Palestine’, Ethnomusicology 53.1 (2009): 58–85. McKeown, Niall, ‘Had They No Shame? Martial, Statius and Roman Sexual Attitudes towards Slave Children’, in S. Crawford and G. Shepherd (eds.), Children, Childhood and Society. Oxford: Archaeopress 2007a, 57–62. ‘The Sound of John Henderson Laughing: Pliny 3.14 and Roman Slaveowners’ Fear of Their Slaves’, in A. Serghidou (ed.), Fear of Slaves – Fear of Enslavement in the Ancient Mediterranean. Besançon: Presses universitaires de Franche-Comté 2007b, 265–79. The Invention of Ancient Slavery? London: Duckworth 2007c. McLuhan, Elizabeth, ‘“Ministerium servitutis meae”: The Metaphor and Reality of Slavery in Saint Patrick’s Epistola and Confessio’, in J. Carey, M. Herbert, and P. Ó Riain (eds.), Studies in Irish Hagiography: Saints and Scholars. Dublin: Four Courts 2001, 63–71. McLynn, Neil, ‘Paulinus the Impenitent: A Study of the Eucharisticos’, Journal of Early Christian Studies 3.4 (1995), 461–86. Meouak, Mohamed, Saqaliba, eunuques et esclaves à la conquête du pouvoir: Géographie et histoire des élites politiques «marginales» dans l’Espagne umayyade. Helsinki: Academia Scientiarum Fennica 2004. Miles, Richard (ed.), Constructing Identities in Late Antiquity. London: Routledge 1999. Miller, Nathaniel, ‘Warrior Elites on the Verge of Islam: Between Court and Tribe in Early Islamic Poetry’, in S. Svärd and R. Rollinger (eds.), Cross-Cultural Studies in Near Eastern History and Literature. Münster: Ugarit-Verlag 2016, 137–71.

342

Bibliography

Millett, Martin, The Romanization of Britain: An Essay in Archaeological Interpretation. Cambridge: Cambridge University Press 1990. Mínguez Fernández, José María (ed.), Colección diplomática del Monasterio de Sahagún, Colección ‘Fuentes y estudios de historia leonesa’, no. 17. León: Centro de Estudios e Investigación San Isidoro 1976. Mohrmann, Christine, The Latin of St. Patrick: Four Lectures. Dublin: Dublin Institute for Advanced Studies 1961. Montero, Roman, All Things in Common: The Economic Practices of the Early Christians. Eugene, OR: Resource 2017. Morel, Pierre-Marie, ‘Épicure et les biens matériels, ou la pauvreté bien tempérée’, in É. Helmer (ed.), Richesse et pauvreté chez les philosophes de l’Antiquité. Paris: Vrin 2016, 111–24. Moriarty, Rachel, ‘Human Owners, Human Slaves: Gregory of Nyssa, Hom. Eccl. 4’, Studia Patristica 27 (1993): 62–99. Mottahedeh, Roy, Loyalty and Leadership in an Early Islamic Society. Princeton, NJ: Princeton University Press 1980. Mouritsen, Henrik, The Freedman in the Roman World. Cambridge: Cambridge University Press 2011. Muehlberger, Ellen (ed. and trans. ), ‘Shenoute of Atripe, Abraham, Our Father’, in E. Muehlberger (ed.), The Cambridge Edition of Early Christian Writings, Vol. 2: Practice. Cambridge: Cambridge University Press 2017, 178–89. Mynors, Roger A. B. (ed.), C. Plini Caecili Secundi: Epistularum Libri Decem, Oxford Classical Texts. Oxford: Oxford University Press 1963. Nehlsen, Hermann, Sklavenrecht zwischen Antike und Mittelalter: Germanisches und römisches Recht in den germanischen Rechtsaufzeichnungen. Göttingen: Musterschmidt 1972. Nerney, D. S., ‘A Study of St. Patrick’s Sources: II. The Doctrine of Divine Vocation’, Irish Ecclesiastical Record 72 (1950): 14–26. ‘A Study of St. Patrick’s Sources: I. Introduction. Relation of Scripta S. Patricii to the Corinthian Epistles’, Irish Ecclesiastical Record 71 (1949): 497–507. Niquet, Heike. Monumenta virtutum titulique. Senatorische Selbstdarstellung im spätantiken Rom im Spiegel der epigraphischen Denkmäler. Stuttgart: Steiner 2000. Nöldeke, Theodor, The History of the Qurʾā n, W. H. Behn (trans.). Leiden: Brill 2013 [Geschichte des Qorâns (second revised edition, 2 volumes, 1909– 1919)]. Norris, Richard A., ‘Augustine and the Close of the Ancient Period of Interpretation’, in A. J. Hauser and D. F. Watson (eds.), A History of Biblical Interpretation Vol. 1: The Ancient Period. Grand Rapids: Eerdmans 2003, 63–79. Ó Corráin, Donnchadh, Liam Breatnach, and Kim McCone (eds.), Sages, Saints and Storytellers: Celtic Studies in Honour of Professor James Carney. Maynooth: An Sagart 1989. O’Donnell, James, ‘Salvian and Augustine’, Augustinian Studies 14 (1983): 25–34. O’Loughlin, Thomas, Discovering Saint Patrick. New York: Paulist 2005.

Bibliography

343

Oakman, Douglas E., Jesus, Debt, and the Lord’s Prayer. Eugene, OR: Cascade 2014. Olberg, Gabriele von, Die Bezeichnungen für soziale Stände, Schichten und Gruppen in den Leges Barbarorum. Berlin: De Gruyter 1991. Orchard, Andy, ‘“Audite Omnes Amantes”: A Hymn in Patrick’s Praise’, in D. N. Dumville and L. Abrams (eds.), Saint Patrick, A.D. 493–1993. Woodbridge, UK: Boydell 1993, 153–73. Origo, Iris, ‘The Domestic Enemy: The Eastern Slaves in Tuscany in the Fourteenth and Fifteenth Centuries’, Speculum 30 (1955): 321–66. Orlandi, Silvia, ‘Gli ultimi sacerdoti pagani di Roma: analisi della documentazione epigrafica’, in P. Brown and R. Lizzi Testa (eds.), Pagans and Christians in the Roman Empire: The Breaking of a Dialogue. Berlin: LIT 2011, 425–66. Orlandi, Tito, Paolo di Tamma Opere: Introduzione, Testo, Traduzione e Concordanze. Rome: C.I.M. 1988. Osiek, Carolyn, ‘The New Testament Teaching on Family Matters’, HTS Theological Studies 62.3 (2006): 819–43. Otten, Willemien, ‘The Long Shadow of Human Sin: Augustine on Adam, Eve and the Fall’, in B. Becking and S. Hennecke (eds.), Out of Paradise: Eve and Adam and Their Interpretations. Sheffield: Sheffield Phoenix 2010, 29–49. Ouzoulias, Pierre, L’économie agraire de la Gaule: aperçus historiographiques et perspectives archéologiques (Doctorat-Archéologie). Université de FrancheComté, Besançon 2006. Papaconstantinou, Arietta, ‘Notes sur les actes de donation d’enfant au monastère thébain de Saint-Phoibammon’, Journal of Juristic Papyrology 32 (2002a): 83–105. ‘ΘEIA OIKONOMIA. Les actes thébains de donation d’enfants ou la gestion monastique de la pénurie’, Mélanges Gilbert Dragon 14 (2002b): 511–26. Pargas, Damian Alan, ‘From the Cradle to the Fields: Slave Childcare and Childhood in the Antebellum South’, Slavery and Abolition 32.4 (2011): 477–93. Pasierowska, Rachael, ‘Up from Childhood: When African American Enslaved Children Learned of Their Servile Status’, Slavery and Abolition 27.1 (2016): 94–116. Patraka, Vivian M., Spectacular Suffering: Theatre, Fascism, and the Holocaust. Bloomington: Indiana University Press 1999. Patterson, Orlando, Slavery and Social Death: A Comparative Study. Cambridge, MA: Harvard University Press 1982. Peppard, Michael, The Son of God in the Roman World: Divine Sonship in Its Social and Political Context. New York; Oxford: Oxford University Press 2011. Petruccione, John, ‘The Persecutor’s Envy and the Martyr’s Death in Peristephanon 13 and 7’, Sacris Erudiri 32.2 (1991): 69–93. Pharr, Clyde, The Theodosian Code and Novels, and the Sirmondian Constitutions. A Translation with Commentary, Glossary, and Bibliography. Union, NJ: The Lawbook Exchange 2001. Phillips, William D., Slavery in Medieval and Early Modern Iberia. Philadelphia: University of Pennsylvania Press 2013.

344

Bibliography

Pipes, Daniel, Slave Soldiers and Islam: The Genesis of a Military System. New Haven, CT: Yale University Press 1981. ‘Mawlas: Freed Slaves and Converts in Early Islam’, Slavery and Abolition 1 (1980): 132–77. Pliego Vázquez, Ruth, La Moneda Visigoda, Vol. 1: Historia Monetaria del reino visigodo de Toledo (c. 569–711). Seville: Secretario de publicaciones 2009. Pool, Jeff B., ‘No Entrance into Truth Except through Love: Contributions from Augustine of Hippo to a Contemporary Christian Hermeneutic of Love’, Review & Expositor 101.4 (2004): 629–66. Popovic, Alexandre, La révolte des esclaves en Iraq au IIIe/IXe siècle. Paris: Librairie Orientaliste Paul Geuthner 1976. Priesching, Nicole and Heike Grieser (eds.), Theologie und Sklaverei von der Antike bis in die frühe Neuzeit. Hildesheim: Olms 2016. Pudsey, April, ‘Children’s Cultures in Graeco-Roman Egypt’, in L. Grig (ed.), Popular Culture in the Ancient World. Cambridge: Cambridge University Press 2017, 208–34. ‘Children and Families in Late Roman Egypt: Family and Everyday Life in Monastic Contexts’, in C. Laes, K. Mustakallio, and V. Vuolanto (eds.), Children and Family in Late Antiquity: Life, Death and Interaction. Leuven: Peeters 2015, 215–34. ‘Children and Families in Roman Egypt’, in J. Evans Grubbs, T. Parkin, and R. Bell (eds.), The Oxford Handbook of Childhood and Education in the Classical World. New York: Oxford University Press 2013, 484–509. ‘Nuptiality and the Demographic Life Cycle of the Family in Roman Egypt’, in C. Holleran and A. Pudsey (eds.), Demography and the Graeco-Roman World: New Insights and Approaches. Cambridge: Cambridge University Press 2011, 60–98. Pudsey, April and Ville Vuolanto, ‘Being a Niece or Nephew in an Ancient City. Children’s Social Environment in Roman Oxyrhynchos’, in C. Laes and V. Vuolanto (eds.), Children and Everyday Life in the Roman and Late Antique World. London and New York: Routledge 2017, 79–96. ‘Children’s Urban Environments in an Ancient City: Social and Physical Realities.’ Childhood in the Past 14.2 (2021), 161–176. Puente, Cristina de la, ‘Free Fathers, Slave Mothers and Their Children: A Contribution to the Family Structures in al-Andalus’, Imago Temporis: Medium Aevum 7 (2013): 27–44. Qian, Ailin, ‘Delights in Paradise: A Comparative Survey of Heavenly Food and Drink in the Quran’, in S. Günther and T. Lawson (eds.), Roads to Paradise: Eschatology and Concepts of the Hereafter in Islam. Brill: Leiden 2017, I, 251–70. Quirós Castillo, Juan Antonio, The Archaeology of Early Medieval Villages in Europe. Documentos de Arqueología e Historia 1. Bilbao: Universidad del País Vasco 2009. ‘Las Aldeas de los historiadores y de los arqueólogos en la alta edad media del norte peninsular’, Territorio, Sociedad y Poder 2 (2007): 65–86. Rambault, Nathalie, Homélies sur la Résurrection, l’Ascension et la Pentecôte. Paris: Les Éditions du Cerf 2013.

Bibliography

345

Ramelli, Ilaria L. E., ‘Soma (Σῶμα)’, in Das Reallexikon für Antike und Christentum, 30. Stuttgart: Hiersemann 2021, 814–47. ‘Religion and Science in Gregory of Nyssa: The Unity of the Creative and Scientific Logos’, Marburg Journal of Religion 22.2 (2020), https://doi.org/10 .17192/mjr.2020.22.8299. ‘Autobiographical Self-Fashioning in Origen’, in M. R. Niehoff and J. Levinson (eds.), Self, Self-Fashioning and Individuality in Late Antiquity: New Perspectives. Tübingen: Mohr Siebeck, 2019, 273–92. ‘Reply to J. Albert Harrill’s Review of Social Justice and the Legitimacy of Slavery, by Ilaria L. E. Ramelli’, Journal of Early Christian History 8.1 (2018a): 88–90. ‘Philo as One of the Main Inspirers of Early Christian Hermeneutics and Apophatic Theology’, Adamantius 24 (2018b): 276–92. ‘Gregory of Nyssa’s and Evagrius’s Biographical and Theological Relations: Origen’s Heritage and Neoplatonism’, in I. Ramelli (ed.), in collaboration with K. Corrigan, G. Maspero, M. Tobon, and M. Vinzent, Evagrius between Origen, the Cappadocians, and Neoplatonism. Leuven: Peeters 2017, 165–231. ‘The Sentences of Sextus and the Christian Transformation of Pythagorean Asceticism’, in A. B. Renger and A. Stavru (eds.), Pythagorean Knowledge from the Ancient to the Modern World. Wiesbaden: Harrassowitz 2016a, 151–62. Social Justice and the Legitimacy of Slavery: The Role of Philosophical Asceticism from Ancient Judaism to Late Antiquity. Oxford: Oxford University Press 2016b. The Christian Doctrine of Apokatastasis: A Critical Assessment from the New Testament to Eriugena. Leiden: Brill 2013. ‘Gregory of Nyssa’s Position in Late-Antique Debates on Slavery and Poverty and the Role of Ascetics’, Journal of Late Antiquity 5 (2012): 87–118. ‘The Birth of the Rome-Alexandria Connection: The Early Sources on Mark and Philo, and the Petrine Tradition’, Studia Philonica Annual 23 (2011a): 69–95. ‘Ierocle Neostoico in Stobeo: I καθήκοντα e l’evoluzione dell’etica stoica’, in G. Reydams-Schils (ed.), Thinking through Excerpts: Studies on Stobaeus. Turnhout: Brepols 2011b, 537–75. ‘Origen’s Anti-subordinationism and Its Heritage in the Nicene and Cappadocian Line’, Vigiliae Christianae 65 (2011c): 21–49. ‘Philosophical Allegoresis of Scripture in Philo and Its Legacy in Gregory of Nyssa’, Studia Philonica Annual 20 (2008): 55–99. ‘Sources and Reception of Dynamic Unity in Middle and Neoplatonism’, in J. Casteigt (ed.), Die Quellen der Idee der dynamischen Einheit im Johannesevangelium forthcoming a. ‘Mysticism and Mystic Apophaticism in Middle and Neoplatonism’, in A. Wilke (ed.), Constructions of Mysticism as a Universal: Roots and Interactions Across the Borders. Wiesbaden: Harrassowitz forthcoming b. Rankin, David, ‘Tertullian and the Imperial Cult’, in Studia Patristica 34 (2001): 204–16.

346

Bibliography

Rapp, Claudia, ‘Late Antique Metaphors for the Shaping of Christian Identity: Coins, Seals and Contracts’, in H. Amstutz, A. Dorn, M. Müller, M. Ronsdorf, and S. Uljas (eds.), Fuzzy Boundaries. Festschrift für Antonio Loprieno. Hamburg: Widmaier 2015, 727–44. ‘Safe-Conducts to Heaven: Holy Men, Mediation and the Role of Writing’, in Ph. Rousseau and E. Papoutsakis (eds.), Transformations of Late Antiquity. Essays for Peter Brown. Farnham: Ashgate 2009, 187–203. Rathbone, Dominic, Economic Rationalism and Rural Society in Third-Century A.D. Egypt: The Heroninos Archive and the Appianus Estate. Cambridge: Cambridge University Press 1991. Rawson, Beryl, Children and Childhood in Roman Italy. Oxford: Oxford University Press 2003. Reid, Jennifer K., ‘Patrician and Augustinian Ideas of the “Inner Man”’. Journal of Medieval Latin 20 (2010): 16–37. Richard, Marcel, Asterii sophistae commentariorum in Psalmos quae supersunt. Oslo: Brogger 1962. Richardson, Kristina, ‘Singing Slave Girls (Qiyan) of the ʿAbbasid Court in the Ninth and Tenth Centuries’, in G. Campbell, S. Miers, and J. C. Miller (eds.), Children in Slavery through the Ages. Athens, Ohio: Ohio University Press 2009, 105–18. Rio, Alice, Slavery after Rome 500–1100. Oxford: Oxford University Press 2017. Riviere, Yann, ‘Recherche et identification des esclaves fugitifs dans l’empire romain’, in J. Andreau and C. Virloivet (eds.), L’information et la mer dans le monde antique, Collection de l’Ecole Française de Rome 297, Rome: École française de Rome 2002, 115–96. Robbins, Derek, Bourdieu and Culture. London: Sage 2000. Roberts, Michael, The Jeweled Style. Poetry and Poetics in Late Antiquity. Ithaca, NY: Cornell University Press 1989. ‘The Treatment of Narrative in Late Antique Literature: Ammianus Marcellinus (16.10), Rutilius Namatianus and Paulinus of Pella’, Philologus 132 (1988): 181–95. Robinson, David M., ‘Ancient Sinope: Second Part’, American Journal of Philology 27.3 (1906): 245–79. Robinson, Olivia F., The Criminal Law of Ancient Rome. London: Routledge 1995. Rogers, Justin, ‘Origen in the Likeness of Philo’, Studies in Christian–Jewish Relations 12 (2017): 1–13. Roller, Matthew B., Constructing Autocracy: Aristocrats and Emperors in JulioClaudian Rome. Princeton, NJ: Princeton University Press 2001. Rosivach, Vincent J., ‘Enslaving “Barbaroi” and the Athenian Ideology of Slavery’, Historia 48 (1999): 129–57. Roth, Ulrike, ‘Slavery and the Church in Visigothic Spain: The Donation and Will of Vincent of Huesca’, Antiquitè Tardive 24 (2016): 433–52. Thinking Tools: Agricultural Slavery Between Evidence and Models. London: University of London/Institute of Classical Studies 2007. The Curse of Thistle and Thorn: Child Slavery in Ancient Rome, in preparation.

Bibliography

347

Rotman, Youval, Byzantine Slavery and the Mediterranean World. Cambridge, MA: Harvard University Press 2009. Les esclaves et l’esclavage: De la Méditerranée antique à la Méditerranée médiévale, VIe–Xie siècles. Paris: Les Belles Lettres 2004. Rousseau, Philip (ed.), A Companion to Late Antiquity. Malden, MA: WileyBlackwell 2012. Roymans, Nico and Marenne Zandstra, ‘Indications for Rural Slavery in the Northern Provinces’, in N. Roymans and T. Derks (eds.), Villa Landscapes in the Roman North: Economy, Culture and Lifestyles. Amsterdam: Amsterdam University Press 2011, 161–77. Rüdiger, Ulrich, ‘Die Anaglypha Hadriani’, Antike Plastik 12 (1973): 161–73. Rüpke, Jörg, On Roman Religion: Lived Religion and the Individual in Ancient Rome. Ithaca, NY: Cornell University Press 2016. Rupprecht, Arthur W., ‘Attitudes on Slavery among the Church Fathers’, in R. N. Longenecker and M. C. Tenney (eds.), New Dimensions in New Testament Study. Grand Rapids: Zondervan 1974, 261–77. Russell, Donald A. and Nigel G. Wilson (eds. and trans.), Menander Rhetor: A Commentary. Oxford: Clarendon 1981. Rustomji, Nerina, ‘Beauty in the Garden: Aesthetics and the Wildā n, Ghilmā n, and Ḥ uˉ r’, in S. Günther and T. Lawson (eds.), Roads to Paradise: Eschatology and Concepts of the Hereafter in Islam. Brill: Leiden 2017, I, 295–307. Saint-Laurent, Jeanne-Nicole Mellon, ‘Syriac Hagiographic Literature’, in D. King (ed.), The Syriac World, Routledge Worlds. London: Routledge 2019, 339– 54. ‘Images de femmes dans l’hagiographie Syriaque’, in A. Binggeli (ed.), L’hagiographie syriaque, Études Syriaques 9. Paris: Geuthner 2012, 201–24. Saller, Richard P., ‘The Hierarchical Household in Roman Society’, in M. L. Bush (ed.), Serfdom and Slavery: Studies in Legal Bondage. London: Longman 1996, 112–29. ‘Corporal Punishment, Authority, and Obedience in the Roman Household’, in B. Rawson (ed.), Marriage, Divorce, and Children in Ancient Rome. Oxford: Clarendon 1991, 144–65. Salway, Benet, ‘Mancipium rusticum sive urbanum: The Slave Chapter of Diocletian’s Edict on Maximum Prices’, in U. Roth (ed.), By the Sweat of Your Brow: Roman Slavery in Its Socio-economic Setting. London: Institute of Classical Studies 2010, 1–20. Samellas, Antigone, ‘The Anti-usury Arguments of the Church Fathers of the East in Their Historical Context and the Accommodation of the Church to the Prevailing “Credit Economy” in Late Antiquity’, Journal of Ancient History 5.1 (2017): 134–78. Samson, Ross, ‘Slavery, the Roman Legacy’, in J. Drinkwater and H. Elton (eds.), Fifth-Century Gaul: A Crisis of Identity? Cambridge: Cambridge University Press 2002, 218–27. ‘Slavery, the Roman Legacy’, in J. Drinkwater and H. Elton (eds.), FifthCentury Gaul: A Crisis of Identity? Cambridge: Cambridge University Press 1992, 218–27.

348

Bibliography

‘Rural Slavery, Inscriptions, Archaeology and Marx. A Response to Ramsay MacMullen’s “Late Roman Slavery”’, Historia 38.1 (1989): 99–110. Sanford, Eva M. (trans.), Salvian, The Presbyter, Writings. Baltimore, MD: Catholic University of America Press 1947. Sartre, Jean-Paul, Search for a Method. New York: Alfred A. Knopf 1963. Satlow, Michael, ‘Giving for Return: Jewish Offerings in Late Antiquity’, in D. Brakke, M. Satlow, and S. Weitzman (eds.), Religion and the Self in Antiquity. Bloomington: Indiana University Press 2005, 91–108. Scheidel, Walter, ‘The Roman Slave Supply’, in K. Bradley and P. Cartledge (eds.), The Cambridge World History of Slavery, Volume 1: The Ancient Mediterranean World. Cambridge: Cambridge University Press 2011, 287–310. ‘Reflections on the Differential Valuation of Slaves in Diocletian’s Price Edict and in the United States’, Münstersche Beiträge zur antiken Handelsgeschichte 15.1 (1996): 67–79. Scheidel, Walter, Ian Morris, and Richard Saller (eds.), The Cambridge Economic History of the Greco-Roman World. Cambridge: Cambridge University Press 2007. Scholten, Helga, Der Eunuch in Kaisernähe: Zur politischen und sozialen Bedeutung des praepositus sacri cubiculi im 4. und 5. Jahrhundert n. Chr. Frankfurt am Main: Lang 1995. Schröder, Ingo W., and Bettina E. Schmidt, ‘Introduction: Violent Imaginaries and Violent Practices’, in B. E. Schmidt and I. W. Schröder (eds.), Anthropology of Violence and Conflict. London: Routledge 2001, 1–24. Schroeder, Caroline T., ‘The Labor of Children in Early Egyptian Monasteries’, paper presented at the Annual Meeting of the Society of Biblical Literature. Denver, CO, November 2018. Schwartz, Marie Jenkins, Born in Bondage: Growing up Enslaved in the Antebellum South. Cambridge, MA; London: Cambridge University Press 2010. Scott, Sarah, Art and Society in Fourth-Century Britain: Villa Mosaics in Context, Oxford School of Archaeology Monographs 53. Oxford: Oxford University School of Archaeology 2000. Segal, Alan F. , ‘Conversion to Judaism’, in L. R. Rambo and C. E. Farhadian (eds.), The Oxford Handbook of Religious Conversion. Oxford: Oxford University Press 2014, 578–97. Segal, Erich, Roman Laughter. The Comedy of Plautus. New York: Harper & Row 1968. Segal, Ronald, Islam’s Black Slaves: The Other Black Diaspora. New York: Farrar, Straus and Giroux 2001. Serfass, Adam, ‘Slavery and Pope Gregory the Great’, Journal of Early Christian Studies 14.1 (2006): 77–103. Sessa, Kristina, Daily Life in Late Antiquity. Cambridge: Cambridge University Press 2018. Shaner, Katherine, Enslaved Leadership in Early Christianity. Oxford: Oxford University Press 2018.

Bibliography

349

Shanzer, Danuta, ‘“Iuvenes vestri visiones videbunt”: Visions and the Literary Sources of Patrick’s Confessio’, Journal of Medieval Latin 3 (1993): 169–201. Shaw, Brent D., Sacred Violence: African Christians and Sectarian Hatred in the Age of Augustine. Cambridge: Cambridge University Press 2011. ‘Body/Power/Identity: Passions of the Martyrs’, Journal of Early Christian Studies 4.3 (1996): 269–312. ‘The Family in Late Antiquity: The Experience of Augustine’, Past and Present 115 (1987): 3–51. Sherwin-White, A. N., The Letters of Pliny: A Historical and Social Commentary. Oxford: Clarendon 1966. Sideris, Georges, ‘“Eunuchs of Light”: Power, Imperial Ceremonial and Positive Representations of Eunuchs in Byzantium (4th–12th Centuries)’, in S. Tougher (ed.), Eunuchs in Antiquity and Beyond. London: Classical Press of Wales and Duckworth 2002, 161–75. Sigismund-Nielsen, Hanne, ‘Slave and Lower-Class Roman Children’, in J. Evans Grubbs, T. Parkin, and R. Bell (eds.), The Oxford Handbook of Childhood and Education in the Classical World. Oxford: Oxford University Press 2013, 286–301. ‘Alumnus: A Term of Relation Denoting Quasi-Adoption’, Classica et Mediaevalia 38 (1987): 141–88. Silver, Morris, ‘Bondage by Contract in the Late Roman Empire’, International Review of Law and Economics 54 (2018): 17–29. Sinai, Nicolai, ‘The Qur’anic Commentary of Muqā til b. Sulaymā n and the Evolution of Early Tafsıˉ r Genre’, in A. Görke and J. Pink (eds.), Tafsıˉ r and Islamic Intellectual History: Exploring the Boundaries of a Genre. Oxford: Oxford University Press 2014, 113–43. Sivan, Hagith, Ausonius of Bordeaux: Genesis of a Gallic Aristocracy. London: Routledge 1993. Skinner, Alexander, ‘Political Mobility in the Later Roman Empire’, Past & Present 218 (2013): 17–53. Smallwood, Mary E., The Jews under Roman Rule: From Pompey to Diocletian: A Study in Political Relations. Leiden: Brill 2001. Smodlaka Kotur, Antonija, ‘Alumni. Legal Status in Roman Dalmatia (the Dalmatian Evidence)’, Revue Internationale des Droits de l’Antiquite 41 (1994): 391–409. Snyder, Christopher A., An Age of Tyrants: Britain and the Britons A.D. 400–600. Stroud: Sutton 1998. Sogno, Cristiana, ‘Barbarians as Spectacle: The Account of an Ancient “Embedded Reporter” (Symm. Or. 2.10–12)’, in R. W. Mathisen and D. Shanzer (eds.), Romans, Barbarians, and the Transformation of the Roman World: Cultural Interaction and the Creation of Identity in Late Antiquity. Aldershot: Ashgate 2011, 133–41. Solin, Heikki, ‘Zur Herkunft der römischen Sklaven (zugleich eine Ergänzung von M. Bang, die Herkunft der römischen Sklaven, 1910)’, in H. Heinen (ed.), Menschenraub, Menschenhandel und Sklaverei in antiker und moderner Perspektive: Ergebnisse des Mitarbeitertreffens des Akademievorhabens

350

Bibliography

Forschungen zur antiken Sklaverei (Mainz, 10. Oktober 2006). Stuttgart: Franz Steiner 2008, 99–130. Die Griechischen Personennamen in Rom: Ein Namenbuch. 2., völlig neu bearbeitete Aufl. Berlin: De Gruyter 2003. Die Stadtrömischen Sklavennamen: Ein Namenbuch. 1. Teil, Lateinische Namen; 2. Teil, Griechische Namen; 3. Teil, Barbarische Namen; Indices. Stuttgart: Franz Steiner 1996. Sommar, Mary, ‘Ecclesiastical Servi in the Frankish and Visigothic Kingdoms’, Zeitschrift der Savigny-Stiftung für Rechtsgeschichte: Kanonistische Abteilung 127 (2010): 57–79. Sorabji, Richard. Emotion and Peace of Mind: From Stoic Agitation to Christian Temptation. Oxford: Oxford University Press 2000. Sotgiu, Giovanna, ‘Un collare di schiavo rinvenuto in Sardegna’, Archeologia classica 25–6 (1973–4): 688–97. Stanfill, Jon, ‘Embracing the Barbarian: John Chrysostom’s Pastoral Care of the Goths’, Ph.D. Dissertation, Fordham University, New York 2015. Stanton, G. R., ‘Τέκνον, παῖς and Related Words in Koine Greek’, in B. G. Mandilaras (ed.), Proceedings of the XVIII International Congress of Papyrology, Athens 25–31 May 1986, Vol. I. Athens: Greek Papyrological Society 1988, 463–80. Ste Croix, Geoffrey E. M. de, The Class Struggle in the Ancient Greek World from the Archaic Age to the Arab Conquests. Ithaca, NY: Cornell University Press 1998 [1983]. Stein, Peter, ‘Was There a Slave Trade in Southern Arabia in Pre-Islamic Times?’, Manuscript of the Month 9 (2015), online: www.manuscript-cultures.unihamburg.de/mom/2015_09_mom_e.html. Sterrett, J. R. S., An Epigraphical Journey in Asia Minor, Vol. 2. Boston, MA: Damrell and Upham 1888. Stevenson, Jane, ‘Literacy in Ireland: The Evidence of the Patrick Dossier in the Book of Armagh’, in R. McKitterick (ed.), The Uses of Literacy in Early Medieval Europe. Cambridge: Cambridge University Press 1990, 11–35. Stewart, Randall (ed. and trans.), Sortes Astrampyschi, Vol. 2. Munich: Saur 2001, 114–39. Stöckl, Heidi, Ligia Kiss, Jobst Koehler, Dung Thuy Dong, and Cathy Zimmerman, ‘Trafficking of Vietnamese Women and Girls for Marriage in China’, Global Health Research and Policy 2 (2017), online: www.researchgate.net/publication/320288694. Stone, Michael E., Adam’s Contract with Satan: The Legend of the Cheirograph of Adam. Bloomington: Indiana University Press 2002. Straus, Jean A., L’Achat et la vente des esclaves dans l’Égypte romaine. Contribution papyrologique à l’étude de l’esclavage dans une province orientale de l’empire romain, Archiv für Papyrusforschung und verwandte Gebiete, Beiheft 14. München: K.G. Saur 2004. Swancutt, Diana M., ‘Still Before Sexuality: “Greek” Androgyny, the Roman Imperial Politics of Masculinity and the Roman Invention of the Tribas’, in T. Penner and C. Vander Stichele (eds.), Mapping Gender in Ancient Religious Discourses, Biblical Interpretation 84. Leiden: Brill 2007, 11–61.

Bibliography

351

Swift, Ellen, Jo Stoner, and April Pudsey, A Social Archaeology of Roman and Late Antique Egypt: Artefacts of Everyday Life. Oxford: Oxford University Press, forthcoming. Taylor, Joan, Pythagoreans and Essenes. Leuven: Peeters 2004. Tejerizo García, Carlos, ‘“The End of the World as We Know It”: Post-Imperial Social Landscapes in North-Central Iberia (5th–6th Centuries)’, Archeologia Medievale 43 (2016): 383–97. ‘La Arqueología de las aldeas altomedievales en la cuenca del Duero (ss. V– VIII): Problemas y perspectivas’, Debates de Arqueología Medieval 3 (2013): 317–34. TeSelle, Eugene, ‘Toward an Augustinian Politics’, Journal of Religious Ethics 16.1 (1988): 87–108. Thomas, Charles, ‘Palladius and Patrick’, in A. MacShamráin (ed.), The Island of St. Patrick: Church and Ruling Dynasties in Fingal and Meath, 400–1148. Dublin: Four Courts 2004, 13–37. Christianity in Roman Britain to AD 500. Berkeley: University of California Press 1981. ‘Saint Patrick and Fifth-Century Britain: An Historical Model Explored’, in P. J. Casey (ed.), The End of Roman Britain. Oxford: British Archaeological Reports 1979, 81–101. Thompson, Edward A., Who Was St. Patrick? Bury St. Edmunds: Boydell 1985 [1999]. The Goths in Spain. Oxford: Clarendon 1969. Thompson, F. Hugh, The Archaeology of Greek and Roman Slavery. London: Duckworth 2003. Thür, G., ‘Folter’, Reallexikon für Antike und Christentum 8 (1972): 101–12. Thurmond, David L., ‘Some Roman Slave Collars in CIL’, Athenaeum 82.72 (1994): 459–93. Tobon, Monica, Review of Ramelli 2016b, Classical Review 68.1 (2018): 126–28. Tomás-Faci, Guillermo and José Carlos Martín-Iglesias, ‘Cuatro documentos inéditos del monasterio Visigoda de San Martín de Asán (522–586)’, Mittellateinisches Jahrbuch 52 (2017): 261–86. Trimble, Jennifer, ‘The Zoninus Collar and the Archaeology of Roman Slavery’, American Journal of Archaeology 120.3 (2016): 447–72. Tuor-Kurth, Christina, Kindesaussetzung und Moral in der Antike: Jüdische und Christliche Kritik am Nichtaufziehen und Töten neugeborener Kinder. Göttingen: Vandenhoeck & Ruprecht 2010. United Nations, Office of the High Commissioner for Human Rights, Protocol to Prevent, Suppress and Punish Trafficking in Persons Especially Women and Children, Supplementing the United Nations Convention against Transnational Organized Crime, United Nations 2000, online: https://bit.ly /3zKtbMY. Urban, Elizabeth, ‘The Identity Crisis of Abu Bakra: Mawla of the Prophet, or Polemical Tool?’ in P. M. Cobb (ed.), The Lineaments of Islam: Studies in Honor of Fred McGraw Donner. Brill: Leiden 2012, 121–49.

352

Bibliography

Van der Bergh, Ronald, ‘A Thematic and Chronological Analysis of the Reception of Ananias and Sapphira’, Journal of Early Christian History 7.2 (2017): 1–16. van Rompay, Lucas, ‘Romanos le mélode: un poète syrien à Constantinople’, in J. den Boeft and A. Hilhorst (eds.), Early Christian Poetry: A Collection of Essays. Leiden: Brill 1993, 283–96. Velázquez, Isabel, ‘Baselicas multas miro opera construxit (VSPE 5.1.1). El valor de las fuentes literarias y epigráficas sobre la edilicia religiosa en la Hispania Visigoda’, Hortus artium medievalium 13 (2007): 261–9. Vera, Domenico, ‘Questioni di storia agraria tardoromana: Schiavi, coloni, villae’, Antiquité Tardive 20 (2012): 115–22. ‘Essere “schiavi della terra” nell’Italia tardoantica: le razionalitá di una dipendenza’, Studia Historica 25 (2007): 489–505. Vergote, J., ‘Folterwerkzeuge’, Reallexicon für Antike und Christentum 8 (1972): 112–41. Verlinden, Charles, L’Esclavage dans l’Europe Médiévale. Brugge: De Tempel 1955. ‘L’Esclavage dans le monde ibérique médieval’, Anuario de Historia del Derecho Español 11 (1934): 283–448. Vidovic, Goran, ‘Dish to Cash, Cash to Ash: The Last Roman Parasite and the Birth of a Comic Profession’. MA Thesis in Medieval Studies, at the Central European University (Budapest) 2009. Vives, José (ed.), Concilios Visigóticos e Hispano-Romanos. Barcelona: Consejo Superior de Investigaciones Científicas, Instituto Enrique Flórez 1963. Vlassopoulos, Konstantinos, ‘Finley’s Slavery’, in D. Jew, R. Osborne, and M. Scott (eds.), M. I. Finley: An Ancient Historian and His Impact. Cambridge: Cambridge University Press 2016, 76–99. Von Schöneman, Katja, ‘“Confine Your Women!” Diachronic Development of Medieval Sunnıˉ Exegeses on the Creation of Woman’, MA thesis at the University of Helsinki 2018. Vuolanto, Ville, ‘Experience, Agency, and the Children in the Past. The Case of Roman Childhood’, in C. Laes and V. Vuolanto (eds.), Children and Everyday Life in the Roman and Late Antique World. London; New York: Routledge 2017, 11–24. Children and Asceticism in Late Antiquity: Continuity, Family Dynamics and the Rise of Christianity. London: Routledge 2015a. ‘Children and Work. Family Strategies and Socialisation in Roman Egypt’, in K. Mustakallio and J. Hanska (eds.), Agents and Objects – Children in Premodern Europe. Institutum Romanum Finlandiae: Rome 2015b, 97–111. ‘Family Relations and the Socialisation of Children in the Autobiographical Narratives of Late Antiquity’, in L. Brubaker and S. Tougher (eds.), Approaches to the Byzantine Family. Birmingham: Ashgate 2013, 47–74. ‘Selling a Freeborn Child: Rhetoric and Social Realities in the Late Roman World’, Ancient Society 33 (2003): 169–207. Ward-Perkins, Bryan, ‘Four Bases from Stratonikeia: A (Failed) Attempt to Christianize the Statue Habit’, in A. Busine (ed.), Religious Practices and

Bibliography

353

Christianization of the Late Antique City (4th–7th cent.). Leiden: Brill 2015, 179–87. Watson, Alan (ed.), The Digest of Justinian, Vol. 4. Philadelphia: University of Pennsylvania Press 1998. Webb, Peter, Imagining the Arabs: Arab Identity and the Rise of Islam. Edinburgh: Edinburgh University Press 2016. Weedon, Chris, Feminist Practice and Poststructuralist Theory. Oxford: Blackwell 1987. Weiss, Alexander, Sklave der Stadt: Untersuchungen zur öffentlichen Sklaverei in den Städten des Römischen Reiches. Stuttgart: Steiner 2004. Wemple, Suzanne F., ‘Consent and Dissent to Sexual Intercourse in Germanic Societies from the Fifth to the Tenth Centuries’, in Angeliki E. Laiou (ed.), Consent and Coercion to Sex and Marriage in Ancient and Medieval Societies. Washington, DC: Dumbarton Oaks 1993, 227–43. Wessel, Susan, Passion and Compassion in Early Christianity. Cambridge: Cambridge University Press 2016. Westergren, Andreas, ‘The Monastic Paradox: Desert Ascetics as Founders, Fathers, and Benefactors in Early Christian Historiography’, Vigiliae Christianae 72 (2018): 283–317. Westerholm, Stephen and Martin Westerholm, Reading Sacred Scripture: Voices from the History of Biblical Interpretation. Grand Rapids: Eerdmans 2016. Westermann, William L., The Slave Systems of Greek and Roman Antiquity. New York: American Philological Society 1955. Wheeler, Sandra M., Lana Williams, Patrick Beauchesne, and Tosha L. Dupras, ‘Shattered Lives: Evidence of Physical Child Abuse from Ancient Egypt’, International Journal of Paleopathology 3.2 (2013): 71–82. Whitehead, Neil L., ‘Introduction: Cultures, Conflicts, and the Poetics of Violent Practice’, in N. L. Whitehead (ed.), Violence. Santa Fe: School of American Research Press 2004, 3–24. ‘On the Poetics of Violence’, in N. L. Whitehead (ed.), Violence. Santa Fe: School of American Research Press 2004, 55–78. Whittaker, Charles R., ‘Rural Labour in Three Roman Provinces’, in Land, City and Trade in the Roman Empire. Aldershot: Variorum 1993a, II 73–99. ‘Circe’s Pigs: From Slavery to Serfdom in the Later Roman World’, in Land, City and Trade in the Roman Empire. Aldershot: Variorum 1993b, V 89–120. Wickham, Chris, Framing the Early Middle Ages: Europe and the Mediterranean 400–800. Oxford: Oxford University Press 2005. ‘Marx, Sherlock and Late Roman Commerce’, The Journal of Roman Studies 78 (1988): 183–93. Widow, Gonzalo Leterier, ‘¿Qué son los principios de la doctrina social de la Iglesia?’ Revista Theologica Xaveriana 67 (2017): 85–111. Wieber, Anja, ‘Eine schwarze Kindheit und Jugend – autobiographisches Schrifttum von Sklavinnen im 19. Jhdt. im Vergleich zu Lebensbedingungen antiker Sklavinnen’, in H. Heinen (ed.), Kindersklaven – Sklavenkinder. Schicksale zwischen Zuneigung und Ausbeutung in der Antike und im interkulturellen Vergleich. Stuttgart: Franz Steiner 2012, 261–83.

354

Bibliography

Wightman, Edith M., ‘Peasants and Potentates: An investigation of Social Structure and Land Tenure in Roman Gaul’, American Journal of Ancient History 3.2 (1978): 97–128. Wilken, Robert L., The Christians as the Romans Saw Them. New Haven, CT: Yale University Press 1984. Williams, Wynne (ed.), Pliny: Correspondence with Trajan from Bithynia (Epistles X, 15–121), Aris & Phillips Classical Texts. Liverpool: Liverpool University Press 1990. Wilmott, T., ‘The Late Roman Transition at Birdoswald and on Hadrian’s Wall’, in T. Wilmott and P. Wilson (eds.), The Late Roman Transition in the North: Papers from the Roman Archaeology Conference, Durham 1999, BAR British Series 299. Oxford: Archaeopress 2000, 13–23. Winterbottom, Michael (ed. and trans.), Gildas: The Ruin of Britain and Other Works. London: Phillimore 1978. Witschel, Christian, ‘Zur Situation im römischen Africa während des 3. Jahrhunderts’, in K.-P. Johne, T. Gerhardt, and U. Hartmann (eds.), Deleto paene imperio Romano: Transformationsprozesse des Römischen Reiches im 3. Jahrhundert und ihre Rezeption in der Neuzeit. Stuttgart: Franz Steiner 2006, 145–221. Wood, Ian, ‘The Final Phase’, in M. Todd (ed.), A Companion to Roman Britain. Malden, MA: Blackwell 2004, 428–42. ‘Continuity or Calamity: The Constraints of Literary Models’, in H. Drinkwater and J. Elton (eds.), Fifth Century Gaul: A Crisis of Identity? Cambridge: Cambridge University Press 1992, 9–18. ‘The Fall of the Western Empire and the End of Roman Britain’, Britannia 18 (1987): 251–62. Wood, Philip, ‘We Have No King but Christ’: Christian Political Thought in Greater Syria on the Eve of the Arab Conquest (c. 400–585). Oxford: Oxford University Press 2010. Yates, Roy, ‘Colossians 2, 14: Metaphor of Forgiveness’, Biblica 71.2 (1990): 248–59. Yavetz, Zvi, Slaves and Slavery in Ancient Rome. New Brunswick: Transaction Publishers 1988. Zaccaria, Claudio, ‘Fidelissimus servus. Considerazioni sul rapporto servopadrone (testimonianze aquileiesi)’, in M. Dondin-Payre and N. Tran (eds.), Esclaves et maîtres dans le monde romain: Expressions épigraphiques de leurs relations. Rome: École française de Rome 2016, 185–213. Zeumer, Karl, ‘Geschichte der Westgotischen Gesetzgegung IV’, Neues Archiv der Gesellschaft für ältere deutsche Geschichtskunde 26 (1901): 91–149. ‘Geschichte der Westgotischen Gesetzgegung III’, Neues Archiv der Gesellschaft für ältere deutsche Geschichtskunde 24 (1899): 571–630. Ziegler, Joseph, ‘Physiognomy, Science, and Proto-Racism 1200–1500’, in M. Eliav-Feldon, B. H. Isaac, and J. Ziegler (eds.), The Origins of Racism in the West. Cambridge: Cambridge University Press 2009, 181–99.

Index

abandonment of children. See child exposure abduction, 7, 87, 88, 107, 111 Abraham, 137, 161, 305 Acts of Thomas, 110, 125, 126 Adam, 80–4 adscripticii, 225, 226, 244 al-Andalusıˉ, 311 Alaric, 90, 94 Alaric II, 252, 254, 272 al-Bayḍā wıˉ, 311 Alexandria, 118, 167, 216 al-Farrā ʾ, 309 al-Jā ḥiz ̣, 312 al-Qurtubı ̣ ˉ, 311 al-Quˉ tˉ̣ıya, 267 al-Rasʿanıˉ, 310 al-Rā zıˉ, 309, 310, 312, 315 al-Ṭabarıˉ, 302, 309, 311 al-Ṭabarsıˉ, 309, 310, 312, 315 al-Thaʿlabıˉ, 309, 312, 313 alumnus, 236–8 Alypius, 24, 102, 112, 124, 290 Ambrose of Milan, 7, 38, 39, 88, 100, 104, 140, 294 Ammi, rabbi, 145 Ammianus Marcellinus, 89, 289 Amphilochios of Iconium, 77, 78, 79 Anastasios, 73 andrapoda, 196, 197, 202, 203 Antony (the Great), 47 Aristotle, 91, 142, 294 Armagh, 288

Asturii, 260 Athanasius, 47 Augusta Vindelicorum, 228 Augustine of Hippo, 7, 22–5, 39, 42, 72, 87, 88, 91, 95–8, 101, 102, 103, 104, 112, 124, 285, 290, 291, 292, 294 Ausonius, 8, 45, 89, 171, 174, 177, 183, 186 Basil of Caesarea, 7, 46, 47, 55, 57, 72, 88, 99, 104, 294 Basil of Seleucia, 7, 81 Battle of Adrianople, 90, 94 billeting, 108 Bissula, 88, 89 body, 20, 28–9, 30, 31, 36, 37, 63–4, 118, 143, 147, 156, 262 Bordeaux, 45, 184 Braulio, 265 Caesarius of Arles, 295 Canaan, 99–100, 104, 294 Cassian, 6, 44, 51, 56 Cato the Elder, 92 Celsus, 59, 290 census, 153, 200, 212, 215, 220, 224, 226, 227 child exposure, 88, 111, 133, 159, 214, 220, 236, 237, 238, 261, 263, 279 child slavery, 72, 89, 109, 113, 114, 115, 120, 123, 130, 133, 138, 152, 158–60, 198, 201, 207–8, 210–23, 232, 237, 238, 261, 266, 290, 300

355

356

Index

children, selling of, 72, 88, 111, 130, 132, 133, 261, 290, 300 Chindaswinth, 257, 258, 271, 272, 277, 279 chrysargyron, 66, 73 Cicero, 92, 93, 98, 294 Clement, 49 Code of Euric, 253 coloni, 91, 172, 175, 181, 185, 197, 225, 226, 244, 245, 255, 259, 266, 267. See also tenant farmers Confessio, of St. Patrick, 11, 281, 284–8 Constantine, 81, 135, 140, 144, 197, 229, 259, 261, 265, 271, 273, 284, 289 Constantinople, 82, 83, 114 contubernia. See slave marriages and ‘marriages’ Cordoba, 253, 311 Council of Chalcedon, 58 Council of Elvira, 122 Council of Ephesus, 76, 82 Council of Gangra, 6, 45, 57–9 Council of Toledo, 266, 275 Cyprian, 37, 295 Cyril of Alexandria, 81 debt bondage, 66–76, 80, 83, 85, 130–2, 142, 147, 262, 300 decalvatio, 270 delicia children, 89, 238 Diocletian, 225, 229, 245 Doctrina Addai, 110, 116 dominicide, 23, 94, 123, 154 dominus, 34, 40, 178, 182, 184, 235, 236, 238, 239 doulology, 110, 149, 155–60, 181 doulos, 55, 158, 197–8, 199, 204, 208, 209 Early Christian slavery, scholarship on, 5–6, 15–16, 21 Edessa, 107, 108, 109, 110, 111, 115, 116, 117, 120, 126 Egica, 253, 264, 265, 272, 275 eleutheria, 41, 83, 201, 202 emancipation of slaves, 41, 45–7, 50, 57, 59, 74, 130, 139, 193, 201, 203, 209, 221, 234, 243, 245, 258, 264, 266–7, 273–6, 280, 300 Emmelia, 48, 56 epikrisis, 214, 220 Epiphanes, 49

Epistola ad milites Corotici, 11, 282, 285, 286, 287, 292, 295 epoikia, 153, 155, 164, 165 Erwig, 253, 254, 267, 272, 275, 278 Esau, 7, 88, 99–100, 104, 161, 162, 294 Essenes, 59–63 Eucharisticus, 183, 186 Eunomius, 51 eunuch, 312 eunuchs, 11, 46, 89, 240, 246, 299, 309, 313 Euphemia and the Goth, 7, 107–28 Eusebius, 52, 58, 59, 63, 65, 81, 146 Eustathius of Sebaste, 6, 44, 56–9, 65 Evagrius, 6, 44, 51, 52, 59, 61, 64 fear, 22, 23, 24, 35, 37, 97, 112, 123, 152, 217 of violent slaves. See dominicide First Letter of Peter, 20 Florentinus, 95 Formulae Visigothicae, 254 foundlings. See child exposure freed slaves, 202, 203, 229, 230, 232–4, 237, 259, 266, 276. See also emancipation of slaves Fructuosus, 273 Galilee, 136 Gamliel, rabbi, 143 Germanus of Auxerre, 284 ghilmā n lahum, 298, 304–15 Gibbon, Edward, 2 Gildas, 289 Goths, 333 Gratian, 135 Gregory (the Great), 7, 88, 98 Gregory of Nyssa, 6, 44, 45, 46, 47, 49–56, 57, 59–65, 74, 96 habitus, 23, 26–31, 32, 35, 41, 42, 115 Hadrian, 19, 73 Hadrian’s Wall, 285 Hagar, 137, 161 Ham, 7, 88, 99–100, 104, 291, 294 Herodotus, 94 homeborn slaves. See vernae Ibn ʿArabıˉ, 308 Ignatius of Antioch, 146, 158 Ishmael, 137, 161, 162 Isidore of Seville, 254, 268

Index ius gentium, 95, 98 ius naturale, 95 Jacob, 99–100 Jalā l al-Dıˉn al-Maḥallıˉ, 312 Jalā l al-Dıˉn al-Suyuˉ tˉ̣ı, 312 Jerome, 48, 125, 140, 281, 299, 322 Jerusalem, 46, 60, 145 Jesus as a slave, 125, 156, 296 John Chrysostom, 47, 52, 54, 61, 68, 89, 95, 114, 127, 291 Joseph, 99, 100, 305, 307 Josephus, 60–2, 131, 138 Julian, 63, 89 Justin, 192 Justinian, 59, 73, 192, 199, 200, 207–9, 271 Kellis, 160 kidnapping, 87, 102, 111, 112, 119, 124, 127, 146, 262, 263, 270, 289–90, 300 Lactantius, 35, 36, 38, 140 Late Antiquity (as a period for research), 2–4 Leges Visigothorum, 254 Leontios of Constantinople, 75, 81 Leovigild, 252, 272, 277 Lex Romana Visigothorum, 254 libertas, 41, 103, 181, 246 liberti. See freed slaves Life of Malchus, 125, 281, 299 Macarius of Magnesia, 61 Macrina, 47, 48, 51, 52, 56, 57 Magnus Maximus, 283 manumissio in ecclesia, 49, 234, 273 manumission. See emancipation of slaves Marcian, 73 marriage. See slave marriages and ‘marriages’ marriages between slaves and their owners, 102, 137–9, 277, 304. See also sexual abuse Maurice, 192 Mekhilta, 132, 133 Melania the Elder, 45 Melania the Younger, 46 Menander of Laodicea, 89 Milan, 229 Mishnah, 132, 133, 138–9, 141, 145

357

Monica, mother of Augustine, 22, 25 Muhammad, 299, 300, 303, 306, 309, 314 Mujā hid, 308 Muqā til, 308, 310 Nessana, 191–209 Nestorius, 82 Nicomedia, 231 Noah, 99, 291 oiketes, 55, 197, 199, 200, 203–5, 209 Olybrius, 235 Onesimus, 143, 157 Origen, 22, 33, 44, 46, 47, 49, 51, 52, 53, 54, 57, 59, 61, 63, 64, 99, 125 Orosius, 91 orthoposture, 29, 30, 41 Oxyrhynchos, 9, 10, 152, 211–23 pais, paidion, 199–201, 204 Palladius, 45, 167 Panopolis, 164, 166, 168 paradise, 11, 104, 304, 305, 307, 309, 310, 311, 314, 315 paterfamilias, 38, 110, 114, 133, 160, 234, 238, 277 Paul of Tamma, 8, 149, 156–8, 161 Paul the Apostle, 6, 32–3, 45, 77, 95, 143, 156–8, 288, 293, 294, 296 Paula, 43, 48 Paulinus of Nola, 45, 175, 176 Paulinus of Pella, 8, 171, 174, 183–5 peculium, 257, 274, 279 Pelagius, 284 Pentapolis, 215 Petra, 191–209 philanthroˉ pia, 79 Philo, 23, 58, 59–63, 99, 294 Picts, 285, 288, 289, 293, 295 plagiarii. See kidnapping Plato, 49, 91 Platonism, 52, 63 Pliny the Younger, 5, 18–20, 42 Proclus, 7, 82, 83, 85 Prosper of Aquitaine, 288 prostitutes, 133, 139, 242, 268 Ps.-Asterios the Sophist, 81 Pseudo-Joshua the Stylite, 117 Querolus, 8, 171, 174, 177–9, 186

358

Index

Radagaisus, 91 ransom, 87, 88, 90, 145, 146, 299, 301 Reccared, 252, 253 Recceswinth, 254, 257, 262, 272, 275, 279 Romanos the Melodist, 7, 79, 83–5 Rutilius Namatianus, 92 Salona, 231 Salvian of Marseilles, 8, 96, 117, 171, 174, 179–83, 186, 271, 291, 292 Saracens, 125, 299 Savaria, 229 Scotti, 285, 288, 293, 296 Scythians, 87, 93, 94, 101 self-sale, 72, 88, 111, 132, 147, 262, 279 Seneca, 52, 95, 98, 143, 154 Sepulchrum Aureliorum, 233 serfs, serfdom, 1, 2, 171, 244, 251, 276, 280 servi publici (public slaves), 226, 241–2, 258 Severian of Gabala, 81 sexual abuse, 121, 130, 132, 133, 139, 140, 144–5, 222, 237, 257, 263, 270, 271, 295, 303 Shenoute, 8, 149, 156–68 Simeon the Stylite, 47 singing lessons, dangers of, 221 Sinope, 18–19 Sirmium, 229 Sisebut, 271 slave apprentices, 220 slave body. See body slave collars, 225, 234, 242, 301 slave economy, 173 slave marriages and ‘marriages’, 121, 197, 237, 271, 273, 277, 279 slave names, 177, 204–7, 216, 228, 232, 233, 237 slave of Christ, 33, 45, 125, 156–8, 162, 180, 206, 209, 231, 297 slave prices, 199, 201, 207–8, 225, 255, 265 slave trade, 24, 72, 87–91, 101–3, 128, 201, 215–17, 246, 300, 315 slavery, rejection of, 55–65 slaves and military service, 241, 278, 315 slaves, vocabulary for, 158, 175, 197–8, 199, 200, 201, 202, 203, 204, 218, 230, 236, 239, 240, 255, 258, 261, 286, 298, 301–2, 305. See also ghilmā n lahum

Sortes Astrampsychi, 152 Sozomen, 91 Spartans, 39, 40, 94 stereotypes, 7, 21, 101, 117, 122, 125, 127, 176, 179 Stoicism, 23, 33, 44, 61, 62, 95, 98, 140, 143, 147, 177, 292, 296, 302 Stratonikeia, 66, 73, 78 Suevi, 260 Symmachus, 103 Synesius of Cyrene, 7, 87, 88, 93–4, 103 Synod of . . . . See Council of . . . Tacitus, 93 Talmud, 63, 130, 133, 145, 146 Ṭayyā yeˉ , 299 tenant farmers, 149, 153, 155, 198, 225, 256. See also coloni Tertullian, 21, 22, 39–41, 42 Themistius, 90 Theodore of Mopsuestia, 143 Theodoret, 47, 116 Theodosian Code, 254, 262 Theodosius I, 94, 135 Theodotos of Ancyra, 76, 77, 79 Theophilus, 167 Thera, 227, 243, 245 Therapeutae, 59–63, 65 Therasia, 45 Titianus, 228 Tivoli, 226, 242 Toledo, 252, 259, 274 Torah, 129, 145, 147 torture, 19–20, 34, 39, 256, 263, 269, 270. See also violence (in/of slavery) Tosefta, 132, 133, 134, 139, 141, 144, 145 Toulouse, 252 Trajan, 18 Ulfilas, 288 Ulpian, 19, 95 Valens, 90, 252 Valentinian I, 89 Valentinian II, 135 Varro, 92 vernae, 111, 121, 213, 214, 216, 218, 221, 237

Index Via Salaria Vetus, 228 Victoria, 228 violence (in/of slavery), 17–20, 22, 23, 25, 28, 31–42, 118, 126, 140–3, 147, 152, 155, 160, 163, 165, 186, 269. See also sexual abuse Virgil, 92 Wergeld, 260, 262, 264, 269, 276 whip, whipping, 19, 22, 23, 34, 35, 37–40, 256, 268–70, 277

White Monastery, 156 Witiza, 267 Xenophon, 92 Yehoshua, rabbi, 145 Yemen, 300 Yochanan, rabbi, 142, 143 Zeno, 49, 58 Zosimus, 90

359