Roman Identity and Lived Religion: Baptismal Art in Late Antiquity 1009408658, 9781009408653

Table of contents :
Copyright page
Dedication
Contents
Figures
Acknowledgements
Preface
Abbreviations
Introduction
The Absence of Christian Iconography and the Presence of Roman Cult and Culture in the Baptismal Com
The Use of Non-Christian Imagery in Baptisteries
The Conversion of a Personification
Coda
References
Index

Citation preview

Roman Identity and Lived Religion

Christianity is considered prevalent when it comes to defining the key values of late antique society, whereas ‘feeling connected to the Roman past’ is commonly regarded as an add-on for cultivated elites. This book demonstrates the significant impact of popular Roman culture on the religious identity of common Christians from the fifth to the seventh century in the Mediterranean world. Baptism is central to the formation of Christian identity. The decoration of baptisteries reveals that traditional Roman culture persisted as an integral component of Christian identity in various communities. In their baptisteries, Christians visually and spatially evoked their links to Roman and, at times, even pagan traditions. A close examination of visual and material sources in North Africa, the Iberian Peninsula, and Italy shows that baptisteries served roles beyond mere conduits to Christian orthodoxy.

  is Postdoctoral Researcher and Lecturer in the Department of Art History and Art Collection at the University of Göttingen. She was Curator of the international exhibition ‘Imagining the Divine. Art and the Rise of World Religions’ held at the Ashmolean Museum, Oxford, in 2017–2018, Postdoctoral Researcher on ‘Global Horizons in Pre-Modern Art’ at the University of Bern in 2018–2021, and Fellow at the RomanIslam Center of the University of Hamburg in 2021–2022.

      Series Editors JAŚ ELSNER, University of Oxford SIMON GOLDHILL, University of Cambridge CONSTANZE GÜTHENKE, University of Oxford MICHAEL SQUIRE, University of Cambridge Founding Editors SUSAN E. ALCOCK JAŚ ELSNER SIMON GOLDHILL The Greek culture of the Roman Empire offers a rich field of study. Extraordinary insights can be gained into processes of multicultural contact and exchange, political and ideological conflict, and the creativity of a changing, polyglot empire. During this period, many fundamental elements of Western society were being set in place: from the rise of Christianity, to an influential system of education, to long-lived artistic canons. This series is the first to focus on the response of Greek culture to its Roman imperial setting as a significant phenomenon in its own right. To this end, it will publish original and innovative research in the art, archaeology, epigraphy, history, philosophy, religion, and literature of the empire, with an emphasis on Greek material. Recent titles in the series: Christianity, Philosophy, and Roman Power Lea Niccolai Greek Declamation and the Roman Empire William Guast The Death of Myth on Roman Sarcophagi: Allegory and Visual Narrative in the Late Empire Mont Allen The Lives of Ancient Villages: Rural Society in Roman Anatolia Peter Thonemann Roman Ionia: Constructions of Cultural Identity in Western Asia Minor Martin Hallmannsecker Late Hellenistic Greek Literature in Dialogue Jason König and Nicolas Wiater Oppian’s Halieutica Emily Kneebone

Roman Identity and Lived Religion Baptismal Art in Late Antiquity   University of Göttingen

Shaftesbury Road, Cambridge CB2 8EA, 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 Cambridge University Press & Assessment, a department of the University of Cambridge. We share the University’s mission to contribute to society through the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781009408653 DOI: 10.1017/9781009408677 © Stefanie Lenk 2025 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 & Assessment. When citing this work, please include a reference to the DOI 10.1017/9781009408677 First published 2025 Printed in the United Kingdom by TJ Books Limited, Padstow Cornwall A catalogue record for this publication is available from the British Library. A Cataloging-in-Publication data record for this book is available from the Library of Congress ISBN 978-1-009-40865-3 Hardback Cambridge University Press & Assessment 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. Printed with the kind assistance of Siblings Boehringer Ingelheim Foundation for the Humanities.

To EoF

Contents

List of Figures [page ix] Acknowledgements [xv] Preface [xvii] List of Abbreviations [xxx]

Introduction [2] Rethinking Christian Identity: Multiple Identities [8] The Baptistery as a Place of Christian Identity Construction [12] ‘Lived Religion’ and the Study of Visual and Material Culture [15] Roman Visual and Material Culture in the Baptismal Sphere [18] Structure and Geographical Scope [22]

1 The Absence of Christian Iconography and the Presence of Roman Cult and Culture in the Baptismal Complex of Cuicul, Numidia [25] 1.1 The Baptistery of Cuicul [28] 1.2 The Christian Complex of Cuicul: A Baptismal Site and a Centre of Pilgrimage? [43] 1.3 Dating the Christian Complex of Cuicul: A Nicene Memorial Site [52] 1.4 On Allegorical Interpretations and Visual Memories: The Floor Mosaic of the Baptistery [60] 1.5 Christian Attitudes towards Roman Cult and Culture at the Turn of the Fifth Century in North Africa [65] 1.6 Possible Traces of Roman Religious Practice in Cuicul’s Christian Complex [67] 1.7 Conclusion [75]

2 The Use of Non-Christian Imagery in Baptisteries [79] 2.1 The Circus and the Baptistery: Henchir el Koucha [80] 2.1.1 The Archaeological Evidence [80] 2.1.2 The Motif of the Facing Horses: A Pictorial Formula between Christian Salvation and the Roman Circus [88] 2.1.3 Baptismal Liturgy, the Circus, and Christian Identity [95] 2.2 Mythological Subject Matter between Belief and Practice: Myrtilis Iulia (Mértola) and Milreu [102] 2.2.1 Myrtilis Iulia (Mértola): The Archaeological Evidence [105] 2.2.1.1 The Discovery of Font II: How Plausible Is the Identification of Font I as a Baptistery? [112]

vii

viii

Contents

2.2.1.2 The North African Mosaic Workshop [118] 2.2.1.3 The Mosaic Decoration of the Platform as Evidence of Baptismal Use? [124] 2.2.1.4 The Christian and Graeco-Roman Bellerophon Scene [127] 2.2.2 Milreu: The Archaeological Evidence [131] 2.2.2.1 The Pre-Christian Archaeological Evidence [131] 2.2.2.2 The Christian Phase of the Monument and the Baptismal Space [137] 2.2.2.3 Reusing Roman Material Heritage: Pragmatics or Intention? [147] 2.3 Conclusion [153]

3 The Conversion of a Personification: The River Jordan in Ravenna [155] 3.1 The Two Personifications [156] 3.2 The Orthodox and Arian Baptisteries of Ravenna and Their Dome Mosaics [163] 3.3 The Creation of a New Iconography: The Personification of Jordan [172] 3.4 Chrysologus’s Reinterpretation of the River Jordan: East Meets West [182] 3.5 The Personifications of Jordan in the Orthodox and Arian Baptisteries: Responses to Chrysologus and Representations of Antiquity [187] 3.6 Conclusion [193]

Coda [194] References [201] General Index [245]

Figures

Lars . Ramberg, installation PALAST DES ZWEIFELS on the Palace of the Republic, 8
42 m, Berlin, 26 January–15 May 2005 [page xviii] 0.2 Franco Stella, western façade of the Humboldt Forum, Berlin, 2023 [xxv] 0.3 Arwed Messmer, remnant of the wall between Rathausbrücke and the former site of the Palace of the Republic, Berlin (graffiti: The GDR never existed), 2008 [xxvi] 0.4 Glass ballot box of the Volkskammer, 1989, displayed at Humboldt Forum, Berlin, 2021 [xxviii] 0.5 Klaus Wittkugel, Peter Rockel, element from the signage system of the Palace of the Republic, VEB Leuchtenbaukombinat Leipzig, 1975, displayed at Humboldt Forum, Berlin, 2021 [xviii] 0.6 Sites discussed in this book (printed in regular script) [1] 1.1 Cuicul (Djémila), baptistery seen from the south, late fourth to first half of the fifth century [26] 1.2 The North African provinces after Diocletian’s provincial reorganization [29] 1.3 Ground plan of the church complex of Cuicul [30] 1.4 Plan of the baptismal complex of Cuicul with vestibules and bath [31] 1.5 Cuicul, baptismal chamber, facing the western exit to the bath [32] 1.6 Cuicul, ambulatory of the baptistery [33] 1.7 Cuicul, baptismal font covered by a ciborium, seen from the west [34] 1.8 Cuicul, vault of the ciborium from underneath [35] 1.9 3D-model of the baptismal complex of Cuicul with vestibules and bath [36] 1.10 Cuicul, bath adjacent to the baptistery; frigidarium in the foreground, main entrance to the baptistery in the background [37] 1.11 Cuicul, detail of the mosaic floor of the ambulatory [38] 1.12 Cuicul, detail of the mosaic floor around the font [39] 0.1

ix

x

List of Figures

1.13 1.14 1.15a 1.15b 1.16 1.17 1.18 1.19 1.20a 1.20b 1.21 1.22 1.23 1.24 1.25 1.26 1.27 1.28a 1.28b 1.28c 1.29 1.30 2.1 2.2 2.3 2.4

Cuicul, mosaic pavement of the font [40] Cuicul, schematic drawing of the decoration of the font [41] Cuicul, mosaic panel at the western entrance to the baptistery [42] Cuicul, detail of the mosaic panel at the western entrance [42] Drawing of the subterranean passageway connecting the south and north church, Cuicul [44] Cresconius inscription from the south church of Cuicul, Museum of Djémila [45] Plan of the Alexander church of Tipasa [48] Cuicul, latticework from the south church [53] Detail of the mosaic pavement of the south church of Cuicul, Museum of Djémila [54] Detail of the mosaic pavement of the south church of Cuicul, Museum of Djémila [55] Details of the mosaic pavement of the north church of Cuicul [56] Detail of the mosaic of the toilet of Venus of the maison de l’âne of Cuicul, Museum of Djémila [63] Mosaic from the centre of the peristyle of the maison d’Amphitrite of Cuicul, Museum of Djémila [64] Cuicul, Great Bath, sculpted dolphin at the entrance to the latrines [64] Cuicul, conched apse in the east of the baptistery [70] Plan of the baptistery and bath of the Great Basilica of Tipasa [71] Plan of the west church of Thamugadi (Timgad) [71] Thamugadi, baptistery of the west church, c. 400 [73] Thamugadi, detail of the font’s mosaic [73] Thamugadi, detail of the font’s mosaic [74] Baptismal font of Kelibia, sixth century, Musée National du Bardo, Tunis [76] Baptismal font of Bekalta before its restoration, sixth century, Musée Archéologique de Sousse [77] Plan of the church of Henchir el Koucha, fifth to sixth century. Later additions are marked in red [81] Henchir el Koucha, mosaic floor of the baptistery [82] Henchir el Koucha, detail of the central panel of the mosaic floor of the baptismal chamber [84] Drawing of the southern and northern panel of the baptismal chamber of Henchir el Koucha, based on photography on the web page http://archeologiechretienne.ive.org/?p¼723 (last accessed 15 August 2023) [84]

List of Figures

2.5a Henchir el Koucha, detail of a pair of horses facing a plant on the mosaic pavement of the baptismal chamber [86] 2.5b Drawing of the pair of horses on the mosaic pavement, Henchir el Koucha [86] 2.6 Henchir el Koucha, detail of the mosaic pavement of the central nave of the basilica [87] 2.7 Tomb mosaic of Lollianus, western necropolis of Thabraca, sixth century, Musée National du Bardo, Tunis [90] 2.8 Detail of the floor mosaic of the house of Sorothus, Hadrumetum, late second to early third century, Musée Archéologique de Sousse [91] 2.9 Detail of the floor mosaic of the house of Sorothus, Hadrumetum, late second to early third century, destroyed [92] 2.10 Floor mosaic of a pair of horses facing a modius of the maison du paon, Carthage, second half of the fourth century, Musée National du Bardo, Tunis [93] 2.11 Mosaic of a pair of horses facing a modius of the bath of Fundus Bassianus, Sidi Abdallah, end of fourth to beginning of fifth century, Musée National du Bardo, Tunis [93] 2.12 Mosaic of pair of horses from Lahmimine, sixth century [94] 2.13 Historical photography of a mosaic depicting a circus scene from Capsa (Gafsa), sixth century, today Musée National du Bardo, Tunis [97] 2.14 Plan of the platform and adjacent buildings of Myrtilis Iulia (Mértola) [106] 2.15 Plan of the presumed baptismal complex I of Myrtilis [107] 2.16 Myrtilis, portico of the platform [108] 2.17 Myrtilis, assumed baptistery I [109] 2.18 Myrtilis, fragment of the mosaic pavement of the assumed baptistery I depicting Bellerophon killing the chimaera [110] 2.19 Myrtilis, fragment of the mosaic pavement of the assumed baptistery I depicting two-petaled rosebuds inscribed in a scale pattern [111] 2.20 Myrtilis, baptistery II, fifth or sixth century [113] 2.21 Plan of baptistery II of Myrtilis [114] 2.22 Myrtilis, baptistery II, fragment of a fresco on plaster, 10 cm height
16 cm width
4 cm depth, sixth to seventh century [115] 2.23 Myrtilis, mosaic of the portico, detail of pair of lions facing a stone pine and wild animals [119] 2.24 Myrtilis, mosaic of the portico, detail of a rider carrying a falcon and various animals [120]

xi

xii

List of Figures

2.25a 2.25b 2.26 2.27 2.28

2.29 2.30

2.31 2.32

2.33 2.34 2.35 2.36 2.37 2.38a 2.38b 2.39

2.40

2.41

Hergla, mosaic of the basilica, panel G of the left aisle [121] Drawing of panel G of the basilica of Hergla [121] Hergla, mosaic of the basilica, panel C of the right aisle [122] Mosaic of a pair of lions facing a stone pine from Carthage, fourth century, Musée National du Bardo, Tunis [123] Mosaic of Achilles riding the centaur Chiron from Belalis Maior (Henchir el Faouar), end of fifth to sixth century, Musée National du Bardo, Tunis [123] Mount Nebo, Diaconicon baptistery, mosaic of a hunting scene, c. 530 [125] Detail of the mosaic floor of the baptistery of Valentia (Valence), sixth to seventh century, Musée d’Art et d’Archéologie de Valence [126] Milreu/Estói, late Roman monument, first half of the fourth century [132] Detail of the plan of the excavation of Milreu under Estácio da Veiga (1877). The hexagonal basin is located in the centre of the monument (D) [133] Plan of the villa and the monument of Milreu [134] Milreu, mosaic frieze of the podium on the north-western side [135] Milreu, view of the monument from the north-west with the remains of the baptismal chamber [137] Milreu, baptismal font above a grave, second half of the fifth or early sixth century [138] Brick fragment from Milreu, fifth to sixth century, Museu Nacional de Arqueologia, Lisbon, inv. no. 997.17.1 [140] Milreu, wall of the podium at the height of the former baptismal chamber [141] Milreu, fragment of the mosaic frieze depicting a curled tail at the height of the former baptismal chamber [142] Drawing of a mosaic fragment of a triton and a sea lion at Milreu, made during the excavation of Estácio da Veiga in 1877, Archive of the Museu Nacional de Arqueologia, Lisbon, inv. no. 25b [142] Photography by A.M. Xavier de Meireilles (1890s) of the mosaic fragment on a wall west of the monument of Milreu. Archive of the Museu Nacional de Arqueologia, Lisbon, inv. no. Foto25B [143] Milreu, detail of the mosaic frieze on the eastern side of the podium [144]

List of Figures

2.42 Milreu, detail of the mosaic frieze on the north-eastern front side of the monument [145] 2.43 Mosaic fragment of the forepart of a boat from Milreu, Museu Nacional de Arqueologia, Lisbon, inv. no. 18701 [146] 2.44 Water conduit with carvings depicting sea life, sixth to seventh century, Museo del Arte y la Cultura Visigoda, Mérida, inv. no. 28 [148] 2.45 Floor mosaic of the destroyed church of San Miguel, second half of the second century, Museu d’Arqueologia de Catalunya, Barcelona, inv. no. 19042 [150] 2.46 Fragment from the mosaic pavement of the baptismal font of the villa of Montinho das Laranjeiras, first half of the fourth century, Museu Nacional de Arqueologia, Lisbon, inv. no. 18754 [151] 3.1a Ravenna, cupola mosaic of the Arian Baptistery, c. 500 [157] 3.1b Ravenna, detail of the cupola mosaic of the Arian baptistery [157] 3.2a Ravena, cupola mosaic of the Neonian baptistery San Giovanni in Fonte, c. 451–473 [158] 3.2b Ravenna, detail of the cupola mosaic of the Neonian baptistery [158] 3.2c Ravenna, Neonian baptistery, detail of the personification of Jordan [159] 3.3 Floor mosaic depicting the mask of Oceanus from Aïn Temouchent, end of the fourth century, Musée National des Antiquités et des Arts Islamiques, Algiers [159] 3.4 Detail of the drawing of the dome mosaic of the Neonian baptistery, published in Giovanni Ciampini’s Vetera Monimenta (1699), 235 [166] 3.5 Drawing of the dome mosaic of the Neonian baptistery with indications of the restoration work by Felice Kybel. [168] 3.6 Throne of Maximian, panel depicting the baptism of Christ, sixth century, Museo Arcivescovile, Ravenna [173] 3.7 Column drum depicting the baptism of Christ from Constantinople (?), attended by two angels and the River Jordan, sixth century, Archaeological Museum Istanbul [173] 3.8 Rome, Santa Maria Antiqua, sarcophagus depicting a personification of Neptune, c. 270 [174] 3.9 City gate sarcophagus, left-hand panel depicting the ascension of Elijah, late fourth century, Musée du Louvre, inv. no. Ma 2980.1 [175]

xiii

xiv

List of Figures

3.10 City gate sarcophagus, left-hand panel depicting the ascension of Elijah, late fourth century, Sant’Ambrogio, Milan [176] 3.11 Rome, Arch of Titus, detail of the personification of Jordan, c. 70 [177] 3.12 Cyrenaica (Qasr Libya), mosaic floor panel depicting the personification of the river Gehon [178] 3.13 Mosaic pavement of the baptistery of Mariana (Corsica), late fourth century, Musée de Mariana, inv. no. 2018.2.4 [178] 3.14 Ohrid, floor mosaic of the baptistery, fifth century [179] 3.15 Panel of the Werden Casket depicting the baptism of Christ, fifth century or Carolingian copy of a late antique model, Victoria and Albert Museum, inv. no. 2009CA7202 [180] 3.16 Patera depicting the triumph of Attis and Cybele with two river personifications at the bottom from Parabiago, second half of the fourth century, Civico Museo Archeologico di Milano, inv. no. A 0.9.14264 – St. 5986 [182] 3.17 Madaba, church of the Apostles, detail of the floor mosaic depicting the bust of Thalassa, 578–579 [192]

Acknowledgements

This book started out as a PhD thesis which was finished at Wolfson College, Oxford, in 2019. It is one of the outcomes of a research project which has fundamentally changed my thinking about what academic work should look like. Empires of Faith, a Leverhulme-funded project (2013–2018) run by the University of Oxford and the British Museum, set out to find new ways of incorporating material culture into the study of religion. I believe that we have made a good start in this respect. I know for certain that Empires of Faith has helped me grow in many unexpected ways. I am deeply indebted to my colleagues Philippa Adrych, Nadia Ali, Robert Bracey, Belinda Crerar, Katherine Cross, Dominic Dalglish, Jaś Elsner, Maria Lidova, Georgi Parpulov, and Rachel Wood. Many hands have helped with this book. My greatest debt is to my supervisors, Jaś Elsner and Gervase Rosser, who have accompanied this and many other projects at the time of writing with great energy and unwavering optimism. The patience with which Jaś supported this publication went far beyond the usual measure and was only surpassed by his insistence that I write the preface. Decisive for the conception of the book up to its final version were the inspiration and critical feedback of Gunnar Brands, Ine Jacobs, Stefan Trinks, and, of course, Marco Meyer. I am deeply grateful to each of them. Many friends and colleagues have discussed with me earlier drafts over the years, in Oxford as well as in Bern, Hamburg, and Göttingen. The suggestions of Beat Brenk, Craig Clunas, Dominic Dalglish, Antony Eastmond, Markus Fiebig, Beate Fricke, Joachim Helfer, Theresa Holler, Zumrad Ilyasova, Andrea Kölbel, Jaecki Lindenau, Meekyung MacMurdie, Corinne Mühlemann, Meseret Oldjira, Saskia Quené, Giacomo Savani, Margaret Scarborough, Jean-Michel Spieser, Nancy Thebaut, the late Mark Whittow, and Edward Zychowicz-Coghill have greatly improved the final version. The responses from the series editors of the Greek Culture in the Roman World series and the anonymous reader of Cambridge University Press were incredibly helpful in shaping the book into its final form. The editorial and production staff guided the production of the book with great care, and I would like to thank in particular Katie

xv

xvi

Acknowledgements

Idle, Nicola Maclean, Michael Sharp and Titus Prathish. Thanks are also due to many others for their generosity in giving advice and offering support in various ways: Philippa Adrych, Achim Arbeiter, Jess Bailey, Fathi Bejaoui, Antonia Bosanquet, Amel Bouder, Barbara Bruderer Eichberg, Jane Chick, Katherine Cross, Jon Cubas Díaz, Nathan Dennis, Elizabeth Fentress, Alberto Ferreiro, Ivan Foletti, Taher Ghalia, Marina Grout, Stefan Heidemann, Vladimir Ivanovici, Kurt Keller, Nathalie Klinck, Clemens Kruse, Susanne Lange, Tomas Lehmann, Heike and Thomas Lenk, Virgílio Lopes and the Campo Arqueológico de Mértola, the archivists of the Museu Nacional de Arqueologia in Lisbon, Manfred Luchterhandt, Anastasia Moskvina, Sabine Panzram, Karla Pollmann, Annemarie Schantor, Stephan Schneider, Hugo Shakeshaft, Gesa Steinbömer, Veronika Tvrzníková, and Bryan Ward-Perkins. A generous doctoral stipend from the Leverhulme Trust made the writing of this book possible. Thanks to the assistance of the Leverhulme Trust, the British Museum, the Oxford Centre for Byzantine Research, the Santander Academic Travel Award, and Wolfson College, I was able to travel to the sites of my case studies with my incomparable travel companions Andrea Kölbel and Margaret Scarborough and attend several conferences. I think back particularly fondly to the Oxford study group which made it through Algeria in 2017. A fellowship at the Center for Advanced Study ‘RomanIslam – Center for Comparative Empire and Transcultural Studies’ at the University of Hamburg from September 2020 to February 2021 provided me with a great environment to revise the manuscript. The printing of the book has been made possible by the generous support of the Art History Department of the University of Göttingen, as well as the Siblings Boehringer Ingelheim Foundation for the Humanities.

Preface

ZWEIFEL. The bold white letters of the German word for ‘doubt’ capture our attention even from a distance (Figure 0.1). They rest heavily on the worn, partially dismantled façade. However, the complex as a whole is robust enough not to look like it is being squeezed under the pressure of the neon signs. ‘Doubt’ seems to be rooted in the compact building’s base by the four slender white bars climbing up the façade. In their simple clarity, the word and the building complement each other, forming a colossal architectural sculpture. The sculptural amalgam casts doubt on the tranquillity of the fleeting, casual movements in the foreground of the image. The building seems to be rearing up in its final phase of existence. By the same token, it is indexed by a keyword – labelled and assigned to a corner of history which is doubtful and irretrievable. The installation Palace of Doubt by Lars . Ramberg in 2005 marked the end of East Berlin’s Palast der Republik (Palace of the Republic). A decade and a half after German reunification, the ‘people’s house’ and seat of the puppet parliament of the former German Democratic Republic, closed in 1990, began to be torn down in 2006. Ramberg, a Norwegian citizen and resident of Berlin since 1998, did not work on the Palace as a passionate fighter for or against the preservation of the building nor as an initiate to the building’s past in East Berlin. The doubt Ramberg captures is reflective of the debate about the heritage status of socialist architecture. The installation, however, also speaks of the uncertainty of those who did not grow up with the palace, when the two Germanys were an unquestionable fact of life.1 Palace of Doubt addresses not least those to whom the palace is not a lived memory but a soon-to-beerased plane of projection. What to take on board from the mixed feelings and diverging beliefs attached to the socialist palace of the people at the time of its imminent disappearance? How to fill the void? The historic period and the art and architecture covered in this book are very different to this example of an intervention of a democratic state with 1

Holfelder 2008.

xvii

xviii

Preface

Figure 0.1 Lars . Ramberg, installation PALAST DES ZWEIFELS on the Palace of the Republic, 8
42 m, Berlin, 26 January–15 May 2005. © Lars . Ramberg.

a free-market economy into socialist state architecture. If we take a step back and look from a distance, however, a modern-day perspective may help us to discern a theme we otherwise might have overlooked under the dust of time which covers late antique art. At the turning point from antiquity to the Middle Ages, the Christians discussed in this book felt Roman – by virtue of being Christian. This book demonstrates that Christians in the western Mediterranean adorned their baptisteries with decorations which underlined the importance of GraecoRoman culture at the key sites of Christian initiation. The newly baptized were introduced to a Christian universe in which Graeco-Roman culture was integral. Counterintuitively, the aspects of Graeco-Roman traditions evoked in these baptisteries did little to support Christian norms and values. At times, they were even in conflict with Christian teachings. However, in the experience of those who commanded the interior decoration of these baptisteries, the imagery and the feelings triggered by the

Preface

decoration were evidently consistent with their Christian identity. The intuition that being part of the Graeco-Roman world was closely tied to being Christian – as evidenced in sites which represented a supreme celebration of the triumph of the Christian faith – had a good chance of being passed on to future generations of Christians. One may wonder that fifth- to seventh-century Christians felt strongly enough about traditions such as chariot racing or mythology to integrate them visually into the ceremonial admission of catechumens and converts. With Christianity safely established in the hearts and minds of a great majority, one may ask, could such imagery have had any more significance than reflecting some whim of a few fanciful artists and commissioners? One may also question whether looking into art and architecture is a promising way at all of learning something about these Christians, especially what they felt loyal towards and what gave them purpose. I will address these questions in the methodological part of the introduction and throughout this book. But a concern about the ways in which art keeps a repudiated past alive, and indeed how this is relevant to matters of communal identity, is not purely historical. When stripped of specific contextual factors, it comes down to a more general question: do generations born and raised after the breakdown of a political or religious system or other socially binding organizations identify with something they may only know from hearsay, something which their parents or their ancestors lived with and cherished? And if they do, to what extent do we accept the identity which derives from their identification with the past as real? Does the new system which has apparently subsumed the past encompass their entire sense of personal identity and cultural affiliation? I became interested in this question when I began reflecting on my own East German heritage. I was born in 1988 in East Berlin, about a year before the Wall came down. I belong to the first generation of Germans raised in a reunited Germany. I remember my childhood having taken place in a blur of rotten grey façades, but it was also increasingly spotted with the candy-coloured merchandise through which communist nostalgia is marketed to this day. The children I grew up with followed many different career paths and took varied decisions about how to lead their lives; a lot of their life choices would have been impossible in the political system in which our parents grew up. But there is one thing many of us have in common: we feel East German. The thirtieth anniversary of the fall of the Wall in 2019 fostered a renewed interest in the relevance of East German identity today. The public discussion about the long-term effects of reunification mirrors the growing

xix

xx

Preface

attention given to identity politics around the globe more broadly and is another jigsaw piece in the fevered debate about the usefulness of ‘identity’ as a concept.2 Public discourse has allocated a special role to the postcommunist-transition children (Nachwendekinder), today in their late twenties and thirties. It is now on this generation to tip the scales in the thirty-year struggle to accomplish German reunification for good. Expectations that the GDR-born citizens of the new federal states should fully integrate into a reunited Germany have long been given up. Those born after unification, however, are viewed with hope by those who want to put an end to East German particularism. A 2019 survey comparing the commitment towards the German state of about 2,200 residents from the old and the new federal states born between 1989 and 2000 painted a pessimistic picture.3 Even in 2019, more than one young person in five living on the territory of the former GDR (i.e. in one of the new federal states) identifies more strongly as East German than as German.4 Even more telling is the ratio of those who believe that it still makes a difference whether you are from the East or the West. Around 57 per cent of young people from the old federal states do not think that it makes a difference. In East Germany, only 33 per cent agree. So, for those born post-GDR, a generation which never lived under a socialist dictatorship evidently feels impacted by their origin from the East, not infrequently in ways which define their identity. In my own experience, this feeling has been met with incomprehension from my West German friends, be they older or close to my age. Indeed, differences in unemployment, income, property, and productivity still exist between West and East, in some regions substantially so, but none of them have left visible traces on my own biography. So why should someone like me feel this way at all? East German identity is largely a post-socialist phenomenon stemming from the interaction between East and West Germans. In 1990, 32 per cent of the former citizens of the GDR described themselves as East German; in 1992 it was 60 per cent.5 Of course, successful or failed attempts of East German parents to integrate into the system of the Federal Republic left a mark on their children, as did the social environment they grew up in.6 2 4

5 6

3 See, for instance, Appiah 2018; Jullien 2021; Remotti 2007. Faus, Storks 2019. In comparison, only 8 per cent of young Germans from the old federal states identify primarily as West German. Faus, Storks 2019, 26–29. Ahbe 2013, 27. The effects of socialization for constructing an East German identity are best explored for the generation of the ‘Wendekinder’, that is, those born between 1975 and 1985. See Lettrari, Nestler and Troi-Boeck, 2016; Hacker et. al., 2012.

Preface

However, the effect which today’s dynamics between East and West play in the preservation of East German identity cannot be overestimated. A qualitative survey of persons born between 1990 and 1995 in the old and new federal states found that those born in the East after the fall of the GDR still experience downgrading and othering because of where they are from.7 Although young citizens of the new federal states tend to perceive East German identity as a dated concept, disparagement of today’s ‘East Germany’ or of the ‘communist’ generation of their parents, whether voiced by West German peers or in the press and German society more broadly, makes them feel more East German.8 Given that a majority of the young generation in the West do not see any difference in where a person is born, the question arises as to why the East continues to be denigrated. What does this help to achieve? Martin Sabrow distinguishes three principal ways in which Germans remember the GDR today.9 The predominant way, and the one sanctioned by unified Germany, is the commemoration of the GDR as a dictatorship in which citizens turned into offenders and victims until internal resistance caused the downfall of the state. In addition to this, there are two more ways of commemoration, both of which are mainly practised in East Germany. A small minority focuses on the early days of the GDR, remembering it as a morally and politically legitimate alternative to the capitalist Federal Republic in the wake of the devastation of the Second World War. Widely practised among the former citizens of the GDR is another type of commemoration which Sabrow calls ‘commemoration of adjustment’.10 This third way of remembering questions the neat separation between biographies and regime. It starts from the daily effort to lead a dignified life amidst the constraints and all too often illegitimate demands of a dictatorship. It reports memories of self-assertion, of the readiness to collaborate with the regime, and of the pride in what was achieved. This last way of remembering is typically performed at the kitchen table of East German households, although it is rarely voiced through public channels. When I was going to school, the small handful of history classes dedicated to the German separation were entirely filled with only one type of commemoration: remembering and rejecting a hated dictatorship. Indeed, these lessons are essential and far too infrequent when measured by the societal impact they ought to have. However, they were also the prelude to the experience of encountering only one normative and valid 7 10

8 9 Kubiak 2018, 2020. Kubiak 2020, 195–196. Sabrow 2010. The German term is Arrangementgedächtnis’. Sabrow 2010, 17.

xxi

xxii

Preface

narrative about the GDR throughout Germany – except inside the homes of former GDR citizens, where history suddenly increases in complexity. My generation grew up living between two separate spheres of historical narration, irreconcilably divided between ‘public opinion’ and the stories reported by family and friends. The claim to normativity is monopolized by the West. Discourse analysis finds East Germans to be regularly marked as deficient in that they are deemed to deviate from a presupposed German norm.11 This dynamic of a normative West and a deficient East helps to restate the moral superiority of today’s Federal Republic. It is apparently even still useful for the younger generations of the old federal states, who continue to speak the language of downgrading the East despite the fading differences between East and West. Part and parcel of the dominant narrative of twentieth-century German history is that it ended in unity. Dissatisfaction with the form which this unity took – as is still expressed in the territory of the former GDR – has no place in this narrative. On the twenty-fifth anniversary of German unification in 2015, German president Joachim Gauck struck a reassuring tone: ‘The differences have become smaller and, especially among the young generation, they have actually completely vanished.’12 The implication here is that the unity criticized by the generation born and raised in the GDR has been established through their children because young East and West Germans do not differ from each other. The prevalent discourse about the East thus leaves the post-GDR generation with two choices: to either feel alienated by what is perceived as a persistent downgrading of East Germany and to maintain an East German identity out of defiance or to let go of any claim to their specific descent and upbringing so as to become living proof of the successes of reunification. The price for the greater goal of unity is to acknowledge that East German identity must be firmly locked in the closet of the past. These experiences have modulated my work on late antique Christianity, and in particular on the stance it took towards art and material culture ‘from the past’. What were the motivations for integrating pre-Christian iconography and relics in Christian structures? Were these pieces of stone and segments of mosaic actually considered to belong to a lost past when

11 12

Roth 2008. ‘Die Unterschiede sind kleiner geworden und besonders in der jungen Generation, da sind sie doch eigentlich gänzlich verschwunden.’ www.bundespraesident.de/SharedDocs/Reden/DE/ Joachim-Gauck/Reden/2015/10/151003-Festakt-Deutsche-Einheit.html?nn=1891680, last retrieved 16 June 2023.

Preface

they were used in the fifth, sixth, and seventh centuries? Or have we historians been misled by the norm-constructing alliance of late imperial church and state, which claimed in a gesture of triumphalism a neat separation of past from present? How similar are the phenomena of the late antique church and the united Federal Republic, given that both have an interest in denying later-born generations certain identities seen as inappropriate or illicit? Is not the church’s repeated stigmatization of continuing Roman traditions (such as public games and festivals) which could not be brought into line with Christian ethos and beliefs structurally similar to the oblivion towards an ongoing identification with East Germany in the contemporary Federal Republic? If there is any value in my proposition, then the relics ‘from the past’ give testimony to a lived reality, an involvement with Graeco-Roman tradition which is more than simply lip service, and a commemoration which may include the wish to participate in what is being remembered. The relics might tell of an active identification with what has officially been called the ‘past’ but is in fact still held in esteem and kept alive – in memory and in practice. A Romano-Christian identity in late antiquity might be more complex, more indebted to the ‘past’ than is often assumed. But how can we know? As historians of societal change, we search for moments in which a new idea takes hold, for processes which reveal a decline in commitment to a past truth, or for acts of resistance against something new. At times, and to our great dissatisfaction, the ways in which our objects of study show either clear signs of reorientation or the perseverance of old attitudes are often elusive. Change and tenacity are not always rooted in conscious decisions and are not necessarily externalized. I myself would find it difficult to explain why I feel East German, and that explanation would shift depending on who my audience might be. How strongly do I react to the public discourses in contemporary Germany, which I have described here? How do I factor in the models I take from friends and family? Were there key experiences which made me form my sense of self? Further, I might ask whether I normally perform my identity in ways that a later historian would be able to trace. In the post-Roman world, identity research typically uses one of two approaches to gain information on how men and women identified as Roman.13 The first route leads via textual records of an official or a private 13

See most recently Pohl et al. 2018. For further literature on the subject, see the Introduction, ###.

xxiii

xxiv

Preface

nature (depending on the happenstance of what survives) and examines conscious external or internal ascriptions of certain qualities. The second route is concerned with material culture related to daily practices and looks for habitual ways of performing identity. Art and architecture, however, have rarely been exploited.14 Archaeologists who dominate the study of late antique art and material culture rightly fear the risks of confining identity research to elitist viewpoints if they focus on objects and architecture affordable only for the upper echelons of society. Moreover, the focus on the practices of daily life reveals a research paradigm which locates identity in the habitual and the repeated as opposed to the particular and the extraordinary. Public spaces, however, sometimes have the potential to entangle the habitual and the extraordinary. They can be both a location where everyday life takes place and the focal point for the public expression of a shared belonging. If public spaces are vested with collective memory, they have the power to unite society. Such lieux de mémoire are particularly prone to reinterpretation and reinvention. As they represent the larger community, they inevitably come to reflect social change, too.15 At times, a discord in the community literally tears them to the ground. In such spaces, art and architecture may be significant to all strata of society. To illustrate my point, let us return once more to Germany’s history-making of the recent past. The Palace of the Republic is one such lieu de mémoire, even in spite of its demolition between 2006 and 2008.16 In 1950, the newly founded GDR demolished the damaged Royal Palace in Berlin’s historic centre and installed a parade ground in its place. Between 1973 and 1976, the Palace of the Republic (Figure 0.1) was built there instead. From the outset, the Palace served two functions. On the one hand, it hosted the Volkskammer, the GDR’s rubber-stamp parliament. On the other hand, and spatially separated from the parliament hall, was the most-visited meeting and event space of the entire East German state. Party congresses of the Socialist Unity Party, state receptions, and meetings of the trade union federation, the youth organization Freie Deutsche Jugend (Free German Youth)(FDJ), and other national associations were all held at the Palace. For the vast majority of the time, however, the Palace was a place of culture and leisure.17 Open to the public from 10 am to

14 16

17

15 Pitts 2007, fig. 5. Nora 1989. See, for instance, Wefing 2009 for an example of the inclusion of the Palace of the Republic in a volume on the ‘Erinnerungsorte’ of the GDR. Mürbe 2007, 114.

Preface

midnight every day, the Palace offered cultural and commercial activities of all kinds, including international concerts, restaurants where you could find seating at the last minute (something quite exceptional in the GDR), bowling lanes, a discothèque, and so forth. From 1976 up until its closure in 1990, it received a total of 60 million guests – an average of 12,300 visitors daily.18 Longer than the building’s operational phase, however, was the period of shutdown, when the Palace became the key architectural bone of contention of the reunited Germany. In March 1990, the first freely elected Volkskammer convened in the building. A few months later, it voted for the GDR’s accession to the Federal Republic. Because of the asbestos used in its construction, the Palace was closed for health reasons in September 1990. The removal of the asbestos meant the building was reduced to its shell. In 2003, parliament finally decided that it should be demolished altogether. In the meantime, a plan was afoot to rebuild the former Prussian Royal Palace on the very site, reviving the monument which the socialist Palace had replaced. Named the Humboldt Forum, the new building, clad on three of four sides in a reconstruction of the façade of the destroyed Royal Palace, opened the doors to its several museums and event spaces in 2021 (Figure 0.2).19

Figure 0.2 Franco Stella, western façade of the Humboldt Forum, Berlin, 2023. Photography by the author.

18

Neumann (ed.) 2019, 223.

19

Stiftung Humboldt Forum im Berliner Schloss, Wolter 2020.

xxv

xxvi

Preface

Architecture’s capacity to embody communal cultures of memory and belonging could hardly be better illustrated than by the heated debate over what to do with the Palace after 1990. While I was growing up in Berlin, the immense ruin loomed large right in the middle of what I then took to be the centre of Berlin – a fallacy in a thoroughly divided city with two centres. Two sorts of memories clashed violently against each other. To the citizens in West Germany, the Palace epitomized socialist dictatorship. To many former citizens of the GDR who had frequented the Palace, the space was a site of private memories. They had spent time there with family and friends, not least at transformative moments in their lives like a wedding celebration or the coming-of-age party known as Jugendweihe. Under the threat of demolition, the Palace to many of them came to signify the value of their personal history.20 Replacing the communist Palace with its royal predecessor was deemed tantamount to a new orthodoxy annihilating the past of East Germans (Figure 0.3).

Figure 0.3 Arwed Messmer, remnant of the wall between Rathausbrücke and the former site of the Palace of the Republic, Berlin (graffiti: The GDR never existed), 2008. © Arwed Messmer.

A final turn in the years preceding the demolition of the Palace saw initiatives to revive the ruined building. However, the protagonists of this movement did not take sides for or against the building’s imminent destruction; rather, they wanted to take leave of the Palace in a selfconscious manner. Between 2004 and 2005, over 900 events took place 20

Mürbe 2007, 115–117.

Preface

on the site, including rubber boat tours through the flooded basement, a concert by the famous West Berlin punk band Einstürzende Neubauten (‘Collapsing new buildings’), and Ramberg’s installation Palace of Doubt. The non-partisan temporary use of the building was organized by a third group, that of a younger generation which may or may not have spent its childhood in the GDR.21 Ramberg’s installation (Figure 0.1), the graffiti announcing the inexistence of the GDR (Figure 0.3), and the ultimate demolition of the structure and subsequent reinvention of the site (Figure 0.2) are all signs of an active identification with a particular view of the past which has consequences for the present. Different cultures of memory and belonging have been inscribed on the plot of ground of the Palace. Not all of these interventions signal full accord with the view promoted by the state. Signs like these can be fruitfully examined for research into matters of historical identity. The newly built Humboldt Forum has a difficult heritage to manage. The place for ‘the arts and sciences, for exchange, diversity and a multiplicity of voices’ has the task of negotiating mutually exclusive cultures of memory.22 In 2016, the Foundation Humboldt Forum hosted a roundtable discussion on the question of how and what to remember of the Palace at the Forum.23 Since then, some of the ideas of the researchers, politicians, curators, and contemporary witnesses have been implemented at the site, notably the exhibits of the glass ballot box of the first freely elected Volkskammer (Figure 0.4) and an element of the signage system which once helped guide people through the Palace (Figure 0.5). These objects are cased and labelled, contained, and historicized. The Humboldt Forum deliberately avoids the trap of re-enactment, which risks signalling any fraternization with the Palace. What the Humboldt Forum cannot avoid is that it stands literally on the foundations of the Palace of the Republic. A democratic antipode to the Palace, its roots are possibly better located in the tradition of antique fora than in the idea of a ‘people’s house’ developed by a nineteenth-century labour movement and taken up in Soviet times.24 However, its aspiration to be a point of identification for the nation as well as a cosmopolitan site where society meets reveals that the Forum wants to be the better Palace. How German society, old and young, will use, react to, and unite or disintegrate in the building remains to be seen.

21 22

23

Reinbold, Novak 2007. For a self-portrayal of the Humboldt Forum, see https://www.humboldtforum.org/en/, last retrieved 16 June 2023. 24 Stiftung Humboldt Forum im Berliner Schloss, Prokasky, Maaß (eds.) 2017. Sölch 2019.

xxvii

xxviii

Preface

Figure 0.4 Glass ballot box of the Volkskammer, 1989, displayed at Humboldt Forum, Berlin, 2021. Photography by the author.

Figure 0.5 Klaus Wittkugel, Peter Rockel, element from the signage system of the Palace of the Republic, VEB Leuchtenbaukombinat Leipzig, 1975, displayed at Humboldt Forum, Berlin, 2021. Photography by the author.

Preface

Spaces where a community reflects on itself – like the plot of ground which gave the basis for the Royal Palace, the Palace of the Republic, and the Humboldt Forum – are fertile sources for research on complex and conflicting identities. In my search for Roman identity in the post-Roman western Mediterranean, I also study a type of space which was constructed as central to the identity formation of the citizens of a renewed state: the late antique baptistery. In the introduction, I will explain why personal and communal identities in late antiquity were vividly negotiated in the baptistery. I start from the assumption that the histories of the baptisteries I study are similarly complicated and ambivalent as the history of the Palace. The long-term impact of the thousand-year-old Roman Empire is infinitely larger than that of a failed state which lasted forty years. In the fifth to seventh century, no less than in contemporary Germany, we encounter art and architecture as signs of active identification with traditions officially claimed to belong to the past. By examining these signs carefully, my aim is to give post-Roman generations a voice, which they expressed within the heart of their local Christian communities, even if the church suppressed identification with aspects of Graeco-Roman culture deemed un-Christian. One finding of this book is that the baptisteries usually alluded to popular Graeco-Roman culture – the everyday experience of Roman social life. Post-Roman Christians and the post-GDR generation seem to have this wish in common: to acknowledge the impact which the living memory and the lived practice of an everyday life ‘from the past’ has on the here and now.

xxix

Abbreviations

AE Ambrose, De sacr. Augustine, Conf. Augustine, Contra ep. fund. Augustine, De civ. Dei. Augustine, Enarr. in ps. Basil, Hom. in hex. C. Th. CIL Clement of Alexandria, Paed. Cyril of Jerusalem, Myst. Cat. Gregory of Nyssa, Catechet. Mag. Gregory of Tours, Glor. conf. Hesiod, Theog. Ildefonsus, De cogn. bapt. Isidore, De eccl. off. Isidore, Etym. John of Damascus, De fid. orthod. Justin Martyr, Apol. LCI Lib. Pont. LIMC Martin of Braga, De corr. rust. Origen, Hom. Jos. PCBE Piazenca Pilgrim, Itin. xxx

L’Année épigraphique Ambrose, De sacramentis Augustine, Confessiones Augustine, Contra epistulam Manichaei quam vocant fundamenti Augustine, De civitate Dei Augustine, Enarrationes in psalmos Basil, Homiliae in hexameron Codex Theodosianus Corpus Inscriptionum Latinarum Clement of Alexandria, Paedagogus Catecheses mystagogicae Gregory of Nyssa, Oratio Catechetica Magna Gregory of Tours, Liber in gloria confessorum Hesiod, Theogonia Ildefonsus, De cognitione baptismi Isidore, De ecclesiasticis officiis Isidore, Etymologiae John of Damascus, De fide orthodoxa Justin Martyr, Apologia Lexikon Christlicher Ikonographie Liber Pontificalis Lexicon Iconographicum Mythologiae Classicae Martin of Braga, De correctione rusticorum Origen, Homiliae in Joshuam Prosopographie chrétienne du Bas-Empire Piazenca Pilgrim, Itinerarium Antonini Placentini

List of Abbreviations

PL PLRE Quodvultdeus, Lib. prom. et praed. Dei Reg. eccl. Carthag. excerpta Salvian, De Gubern. Dei Tertullian, De spec. Theodoret, Jerem.

Patrologia Latina Prosopography of the Later Roman Empire Liber de promissionibus et praedictionibus Dei Registri ecclesiae Carthaginensis excerpta Salvian, De Gubernatione Dei Tertullian, De spectaculis Theodoret, Explanatio in Jeremiam

xxxi

Figure 0.6 Sites discussed in this book (printed in regular script). © Jon Cubas-Díaz.

Introduction

Orthodox Roman Christians (Nicene orthodox in the West and Chalcedonian orthodox in the East) established themselves in the late Roman world, a time of great social and political transition, and flourished in the Migration Period of the fifth and sixth centuries. It is widely accepted that Nicene Christians in the West saw themselves as partakers in the Roman world, dissociating themselves from the allegedly non-Roman migrating peoples who still adhered to paganism or non-orthodox Christian faiths, in particular Arianism. Arian migrants, on the other hand, claiming orthodoxy for themselves, certainly did not always agree with being stigmatized as non-Roman.1 For historians, however, the lifespan of the compound ‘Roman’ and ‘orthodox’ is a limited one. At the threshold of the early Middle Ages, ‘Barbarian beliefs’ were eroded by the increasing acculturation of migrants. Their progressive inclusion into the Nicene church changed the character of the church into a global one, no longer bound to Roman identity.2 In what sense did Christians in the post-Roman world consider themselves to be Roman exactly? Was their Romanness, as much scholarship on post-Roman Christianity often assumes, a function of their Christianness? Was Romanness a bundle of independent meanings which Christians subscribed to? This book interrogates material sources to better understand what implications the association of Christianity and Romanness had for the religious identity of Christians in the fifth, sixth, and seventh centuries. It presents evidence of Christian communities who integrated traditional Graeco-Roman cultural practices into their vision of Christianity and at times broadened the concept of Christianness beyond orthodoxy. The art and material culture of the period show us that Christians treated GraecoRoman cultural practices, some of which were considered un-Christian by 1

2

2

Hen 2018, 65–67; Pohl 2018, 23–24; Pohl 2013, 22; Greatrex 2000, 277. Regional perspectives are offered in Conant 2015, 191–192; Whelan 2018; Muhlberger 1992, 36. For Byzantium, see Stouraitis 2018, 131–132. Pohl 2013, 22–24; Greatrex 2000, 278. See further on the interplay between religion and imperial identities Gantner, Pohl, and Payne 2012.

Introduction

high-ranking church representatives, as part of Christian culture.3 Their identity as Romans (or aspirations to this identity) influenced Christians’ understandings of what counted as proper and orthodox ways of performing Christianity. Hence, we should reconsider the conventional belief that late antique Romans were primarily Christians. The evidence presented in this book suggests an alternative perspective: that Christian identity could be a product of Roman identity. By embracing Christianity, neophytes also joined the community of the inheritors of the Graeco-Roman cultural koine. While determining whether individuals were more motivated by one or the other incentive for baptism may often be challenging, if not impossible, the analysis of visual and material culture brings this question to the forefront. In response to the dominant approach of dealing with questions of continuity and rupture with the classical past in late antique Christianity by analysing texts, I will shift the focus to what an analysis of material and visual culture can contribute to the study of Christian religiosity. Here I follow an approach to religion that privileges lived practice over doctrinal ambitions. While the ‘lived religion’ approach has been productive, especially in the field of Roman religion, the study of late antique Christian art has not yet taken sufficient advantage of this and other approaches which take the exploration of non-elitist, local, and individual perspectives as seriously as the incommensurably better-studied testimonies of the church fathers.4 I focus my argument on the decoration of Mediterranean baptisteries, which were built, maintained, and refurbished between the fifth and seventh centuries – spaces which were instrumental to the formation of Christian religious identity in late antiquity. Under the guidance of their ecclesiastical leaders, Christian communities created spaces and celebrated baptismal ceremonies in them, crafting a vision of Christianity in which Graeco-Roman visual culture was an intrinsic component. Often, the material evidence does not allow us to distinguish between baptisteries used by local Roman or Romanized populations and those used by migrants. Where it does, however, it is clear that Graeco-Roman cultural practices were welcomed in the religious lives of both.

3 4

On the Christian invention of ‘paganism’, see Kahlos 2007, 18–26 and 93–112. On the role of individuality as a conceptual framework in Roman history and archaeology, see Rüpke 2016; Rüpke and Spickermann 2012; Raja 2012; Rebillard and Rüpke 2015. Recent publications on the impact of regional traditions on continuity and rupture with the Roman past include Mugnai, Nikolaus, and Ray 2016; Bredekamp and Trinks 2017; Stouraitis 2018.

3

4

Introduction

All the case studies are situated in the western half of the Mediterranean. One might question why the book focuses on the Roman West, especially when the artistic examples mentioned relate to a visual culture prevalent in Graeco-Roman antiquity throughout the Mediterranean and beyond. Shifting the focus to lesser-known examples of baptismal art in North Africa and on the Iberian Peninsula aims to challenge the popular view of the Byzantines – the Rhōmaîoi – as the self-proclaimed guardians of ancient Roman heritage, which sets the idea of a politically stable Byzantium apart from the so-called post-Roman West.5 Without denying Byzantium’s strong identification with the Roman past, the persistence of Roman identity clearly extends across and indeed beyond the entire Mediterranean region. despite the progressive political disintegration of the western Roman Empire. Also the art, architecture, and liturgy of baptisteries in the West exhibit similar tendencies in the use and reuse of elements of Graeco-Roman culture as do those of the Byzantine East. One of the striking results of the self-construction of the Roman Empire as a Graeco-Roman entity was that elements of Graeco-Roman culture serve as evidence in identifying signs of Roman identity.6 I delve into questions of the interrelation of Roman identity and material culture in more detail below. For the moment, suffice it to say that, in a broad sense, I believe Glen Bowersock hit the nail on the head when contemplating the nature of what he called ‘Hellenism’ in late antiquity. In Byzantium and beyond, notably across the Byzantine lands of north Africa, southern Spain, and Italy, reconquered for Constantinople under the emperor Justinian in the mid sixth century, local communities expressed their unique traditions through language, myth, and imagery rooted in Greek culture. As demonstrated by Bowersock, these cultural amalgams were distinctly local yet allowed communication with other parts of the Helleno-Roman world.7 Similarly, we should be cautious in attributing both local and universal characteristics to Graeco-Roman visual culture. It is possible that the same visual motif adorning a baptistery in Byzantium and another one in the post-Roman West could be interpreted similarly, with the only distinction being its cultural affiliation identified as ‘Hellenic’ in one place and as ‘Roman’ in the other. This does not negate the motif’s complex genesis

5

6

7

On Byzantine identity as Roman, see Kaldellis 2007, 2019; Whalin 2021. Most recently on Byzantine identity, see Stewart, Parnell, and Whately 2022. On the complex interplay of Graeco-Roman and Roman identities in the Roman Empire, see Veyne 2005. Bowersock 1990, esp. 9.

Introduction

over centuries of artmaking in the entangled Mediterranean. The subjectivity with which people, rooted in different localities, made sense of visual culture simply needs acknowledgement. Given that this book predominantly explores evidence from the West, I will refer from now on to Roman (visual and material) culture unless specifically addressing Greek traditions. The case studies in this book form only a small part of the entirety of preserved late antique baptismal art in the western Mediterranean. Late antique baptismal art is particularly rich in modern Tunisia and Italy and can also be found in Algeria, on the Iberian Peninsula, and in southern France. The research for this book brought to light a total of sixty-three western Mediterranean baptismal decorations either still in situ or attested in texts containing information about the location and iconography of the imagery. I discuss six of these in depth (Figure 0.6): the baptisteries of Cuicul (modern-day Djémila in Algeria), Milreu, and Myrtilis Iulia (modern-day Mértola) in southern Portugal; the baptistery of Henchir el Koucha in Tunisia; and the Orthodox and Arian baptisteries of Ravenna. The sixty-three examples can be divided into three categories: (a) eight examples depict Christian narrative scenes from the New Testament or holy figures; (b) thirty-two examples show ornamental, floral, or animal decorations which allude to the psalms or are combined with Christian symbols or inscriptions; and (c) twenty-three examples show ornamental, floral, or animal scenes which cannot be univocally identified with Christianity. Group (c) comprises most of the case studies, while the baptisteries of Ravenna belong to group (a). Numerically, the subset of the six cases discussed in this book is almost equal to group (a), the entirety of preserved baptismal narrative scenes in the western Mediterranean. Furthermore, I tackle only a fraction of the baptismal mosaics rooted in Roman visual culture, many of which form part of group (c).8 8

I have taken care to list all late antique decorated baptisteries in the western Mediterranean, be they still extant or simply recorded by modern scholarship or ancient sources, to the best of my knowledge. This list, however, is almost certainly incomplete, and will provide future readers with ample opportunity for improvement. The forthcoming volume Baptisteries of the Early Christian World, edited by Robin M. Jensen and Nathan Dennis, will offer greater clarity on the extant corpus of late antique baptisteries. Category (a) North Africa: St Cyprian in Carthage, Ad Aquas (Borj Sebbalat el Bey) Italy, France: Naples, Primulacium, Orthodox Baptistery of Ravenna, Arian Baptistery of Ravenna, Catacomb of San Ponziano, Rome Iberian Peninsula: Baptistery II in Myrtilis Iulia (Mértola) Category (b) North Africa: Bekalta, Bir Ftouha, Bou Achir, Bou Smir, Hadjara-Mengouba, Henchir Errich, Henchir Hakaima, Hammam Lif, Henchir B’ghil, Henchir Messaouda, Henchir Sokrine,

5

6

Introduction

The study of late antique baptismal art has produced a rich literature. The groundwork in the field was laid between the late nineteenth and the mid twentieth centuries, when attempts were made to establish an overview of early baptismal art as a unitary totality and define general iconographic trends.9 The majority of research in the second half of the twentieth century has consisted of national inventories and typologies of baptismal architecture and art.10 More recently, a number of studies concerned with local specificities have added nuance to earlier generalist accounts.11 Interest in baptismal ritual and liturgy and their origins has also increased significantly in recent years.12 Iconographical studies have emphasized the interconnections between baptismal liturgy and art and have placed a new emphasis on baptizands’ bodily experience of decorated baptismal spaces.13 The most recent general accounts of baptismal art are Robin M. Jensen’s monographs Living Water: Images, Symbols, and Settings of Early Christian Baptism (2011) and Baptismal Imagery in Early Christianity: Ritual, Visual, and Theological Dimensions (2012).14 Jensen is concerned with bringing

9 10

11

12 13

14

Henchir Zembra, Kelibia, La Skhira, Henchir el Koucha, Oued Zit, Oumcetren, Basilica of St Vitalis in Sufetula (Sbeitla), Seynane, Baptistery I of Sfax, Baptistery II of Sfax, Sidi Abich, Basilica II in Sidi Jdidi, Sidi Mansour, Thuburbo Maius, west church of Thamugadi (Timgad), Basilica I in Uppena (Henchir Chigarnia) Italy, France: Albenga, Cividale del Friuli, Grado, Lateran Baptistery in Rome, Pontenove di Bedizzole Category (c) North Africa: Bît-el-Assa, Basilica I in Bulla Regia, Basilica III in Carthage, Chott Menzel Yahia, Basilica I in Dermech, Cuicul (Djémila), El-Erg, Hippo Regius, Ksar el Hadouch, Basilica III in Mactaris (Makhtar), Oued Ramel, the Church of Bellator in Sufetula (Sbeitla), the Great Basilica in Tipasa Italy, France: Aquae Sextiae (Aix-en-Provence), Aquileia, Cimitile, Isola Comacina, Mariana on Corsica, Massilia (Marseille) Iberian Peninsula: Baptistery I of Myrtilis Iulia (Mértola), Milreu, Montinho das Laranjeiras, Baptistery I in Barcino (Barcelona) Strzygowski 1885; de Waal 1896; de Bruyne 1957; Stern 1957; Fausone 1982. On the western Mediterranean, see Iturgaiz 1967, 1968; Lassus 1970; Gui, Duval, and Caillet 1992; Baratte et al. 2014; Buhler 1975; Gandolfi 2001; Bisconti 2001. Comprehensive overviews can be found in Khatchatrian 1962 and Ristow 1998. For example, Morfino 2011; Schneider 2011; Beltrán de Heredia Bercero and Godoy Fernández 2017; Lück 2018; Brandt 2012, 2016; Ghalia 2016. For a recent substantial contribution see Caseau and Orlandi 2024. The design and decoration of Italian baptisteries, in particular, has received special attention: see Marcenaro 2007; Brandt 2006a, 2006b; Bierbaum 2014; Weinryb 2002; Barber 2018; Ferri 2013; Croci 2019. Johnson 1999; Spinks 2006; Ferguson 2009; Hellholm et al. 2011. Bruderer Eichberg 2003; Foletti and Romano 2009; Foletti 2009; Apostolos-Cappadona 2011; Ivanovici 2014, 2016, 19–123; Morfino 2011; Dennis 2017, 2018. Jensen 2011b, 2012.

Introduction

the study of baptismal art together with the study of figurative baptismal language in scripture, liturgy, and the writings of the church fathers. Jensen’s thoroughly researched work, which provides a model for the study of baptismal art, concentrates on material evidence from the late antique Roman West, like the present study. Preserved instances of narrative imagery are rare in the Roman West and can usually only be found in the most elaborate baptismal settings. Many spaces are composed of images of birds, fish, other animals, shells, flowers, trees, vases, and water. Geometric patterns are often used in tandem with figurative imagery and sometimes dominate the baptismal space. Jensen sees sacramental or paradisiacal meanings expressed in most of the baptismal decorations which are free from narrative – a conclusion shared by a large part of the previously mentioned scholarship.15 While I am far from contesting this view, it strikes me that the scholarship reaching this conclusion has often taken the massive body of empirical data from different contexts in the East and the West as evidence of an ideal, theologically conceptualized totality. The quantitative evidence of preserved baptismal decorations hardly backs scholars’ claims about the generally orthodox character of baptismal imagery. On the contrary, the numbers suggest that Christian adoptions of Graeco-Roman visual culture, which is bare of one-dimensional Christian significance, shaped late antique baptismal art considerably. As foundational and important as the search for baptismal iconography’s scriptural models from the Bible or the church fathers is, it also risks drawing a picture of a hermetic Christianity bent exclusively on orthodoxy. The universalist core assumption all too rarely grants baptismal imagery the potential to be experimental and discursive within a fissile, complex, and diverse range of Christianities. The case studies presented in this book have been selected to nuance the common assumption that fifth- and sixth-century baptismal art generally endorses and promotes what scholarship would identify as orthodox Christian practice and belief. The results of the present study do not apply to the entirety of late antique baptismal art and do not constitute an alternative reading of baptismal art altogether. Yet, they counterbalance the prevalent vision of what the rite of baptism and decorations of baptisteries were meant to achieve. They also give substance to the growing conviction that late antique Christian identities were multi-faceted, bound to local Roman tradition, and could deviate from official Christian doctrine.

15

Jensen 2012, 1; on paradisiacal imagery specifically, 177–213. Jensen establishes the sacramental model in Jensen 2000, 84–88.

7

8

Introduction

Rethinking Christian Identity: Multiple Identities In recent decades, the traditional binary opposition of pagan and Christian has been extensively questioned.16 Historians and archaeologists have stressed that, throughout history, Christians experienced their religion in less clear-cut ways than the writings of Christian apologists suggest.17 Instead, many scholars assume that there was a certain flexibility in Christians’ self-conceptions, in regard to both the sheer multitude of confessions, factions, and heresies (Nicene, Arian, Donatist, Syrian, Egyptian, Pelagian, Nestorian, Manichaean, and others), and the readiness or unwillingness to adhere to Christianity’s claim to exclusivity.18 Scholarship which focuses in detail on how Christians reconciled ongoing pagan traditions and the exclusivity of the Christian religion is, however, still a minority concern.19 The popular view that in the fourth and fifth centuries Roman society developed a secular realm which was independent of the religious realm – an argument prominently advocated by Robert Markus – has slowly begun to be criticized.20 Critics of this theory stress that, at this time, Christians were more integrated into environments still marked by Roman customs and institutions than Markus allowed for.21 They also seek to deconstruct late antique notions of non-negotiable divides between pagans and Christians as discursive constructs used to establish a common identity among Christians.22 In his contribution on Christian identities in North Africa from the third to the fifth centuries, historian Éric Rebillard, using sociological theories of identity formation, has challenged the view that the behaviour of North African Christians was predominantly determined by their Christianity.23 Instead, he champions an approach which acknowledges 16

17

18

19

20 22

23

Many influential studies operate with this opposition. For a prominent example, see Mac Mullen 1986. On the impact of Christian apologetics on researchers’ understandings of Christian religiosity, see Brown 1997, 633–636. This claim, however, has often been limited to only the fourth and fifth centuries. See Kahlos 2007, 31–38; Rebillard 2012, 95; Cameron 2011, 176–177; O’Donnell 1979; Bowersock 1990, 5–6; Trombley 1993–1994, 147–168; Fowden 1998; Sandwell 2007, 4; Busine 2015. Some important contributions are Salzman 2002; Kahlos 2007; Sandwell 2007; Rebillard 2012; Rebillard and Rüpke 2015. 21 Markus 1990. See for instance Rebillard 2012, 90. Kahlos 2007; Sandwell 2007; Lander 2016. On Christian identity construction, see also Harrison et al. 2014. Rebillard 2012, 3. Rebillard’s contribution forms part of a wide field of scholarship on late antique religious identity, which is dominated by historians. See for instance the shared volume Frakes and DePalma Digeser 2006. Art historical scholarship of late antiquity is also

Rethinking Christian Identity: Multiple Identities

the internal plurality of individuals and allows for the multiple social roles (‘multiple identities’ in Rebillard’s terminology) that a person ‘activates’ at any given time.24 Rebillard considers being a Christian only one of many social roles a person could play in their lifetime. Rebillard’s approach is individualistic insofar as it questions the legitimacy of using ‘internally homogeneous and externally bounded groups’ – which is how many late antique Christian writers often used the term ‘Christians’ – as categories of study.25 Rebillard argues that North African Christians had plural identities based on ‘category memberships [social identities] such as ethnicity, religion and occupation’.26 Further, following sociologist Don Handelman, he holds that these category memberships were arranged laterally, meaning that a particular social identity will prevail in one situation, while a different identity will prevail in another. This lateral arrangement is opposed to hierarchical category memberships in which a certain category (religion, for instance) predominates and determines an individual’s behaviour over other categories.27 A sermon of Augustine of Hippo (354–430) can be used as an illustration: ‘There are plenty of bad Christians who pore over astrological almanacs, inquiring into and observing auspicious seasons and days.’28 Augustine is speaking about people who actively combine different identities. In a lateral category membership, this behaviour can seem perfectly plausible, while in a hierarchical category membership (like the one Augustine advocates), this kind of astrological interest is illicit and must be stigmatized (‘bad Christians’).29 Rebillard’s plea for lateral category membership in late antique Christian identities thus introduces a model for thinking about religious affiliation, which allows for more refined interpretations of religious identifications than the schematic categorizations ‘Christian’, ‘pagan’, ‘semi-Christian’, and so on. Rebillard draws our attention to the possibility that Christian identity was accompanied by other identities which could (but did not have to) take centre stage.

24

25 27 29

increasingly using the concept of identity. See for instance Garipzanov, Goodson, and Maguire 2017; Thomas 2016. On the usefulness of the concept of identity for studying material culture, see Pitts 2007; Versluys 2008. For social identity theory Mead 1934, 2015; Jenkins 1996 and for a critique thereof Brubaker and Cooper 2000. On multiple identities, see Burke 2003a, 167–224; Burke and Stets 2009, 130–154; Josselson and Harway 2012; Settles and Buchanan 2014. 26 Rebillard 2012, 2. See Brubaker 2004, 164. Rebillard 2012, 4. 28 Handelman 1977, 191. Augustine, Enarr. in ps., 40 (41). Quoted in Rebillard 2012, 72. Rebillard 2012, 72.

9

10

Introduction

What would lateral category memberships have meant in practice for Christians of the fifth to seventh centuries? Scholars of identity formation agree that actors commonly adopt more than a single role in social situations.30 For a late antique Christian, this could, for instance, mean that she identified as a Christian community member, daughter, wife, patron of the arts, and so forth, and that in many situations more than one role was activated. She would have had to negotiate these roles, as they mutually determined her actions. A caveat is in order here. Rebillard, questioning the status of Christianness as the most prevalent identity category, is primarily concerned with lay Christianity in North Africa at the turn of the fifth century. In the context of this study, however, the actors examined – namely Christian communities building and using baptisteries – comprise both clerics and lay members who were spread across different locations and moments in the fifth to seventh centuries. Over the course of roughly three centuries, lay and ordained Christians differed in the degree to which their Christianness determined other aspects of their identities. In very general terms, we can expect that the advance of Christianization in these centuries also increased the salience of Christianness for the construction of personal identities. This book is not concerned with identifying situations in which the identity category of Christianness was not ‘activated’ or was subordinate to others. On the contrary, we may assume that Christians would have been particularly aware of their Christian identity when visiting baptismal spaces. Nevertheless, the theory of multiple identities is still relevant to my argument, as operating with multiple identity categories can prevent the trivialization of Christians’ relationships with other aspects of their lives. For instance, iconographies and monuments like the ones discussed in this book have traditionally been seen as exemplary of the ‘Christianization’ of Roman society. While such observations are not untrue, they do not take into account that Christian works of art may have been intended to strengthen more than one identity. Stating that a work of art reflects the Christianization of a region without any further specification suggests that the work of art was principally understood in Christian terms. Any other ways in which it might have mattered fall off the radar. Effectively, what the work of art might reveal about the identity of its makers and recipients is decided before any serious consideration is given to the question of what its purpose was. In the language of identity theory,

30

Stryker 1968; Turner 1978.

Rethinking Christian Identity: Multiple Identities

we could say that scholarship takes for granted a hierarchical category membership in which Christians did not care for certain objects beyond their Christian value. Identity theory proposes that individuals seek to establish congruence between different, more or less salient identities.31 How challenging this task is depends on how mutually exclusive these identities are. The more common ‘meanings’ that different identities share, the more sustainably they can be activated at the same time. Where identities share many ‘meanings’, verifying one identity will help verify the other. If, for instance, an individual maintains the role of an engineer, and the same individual maintains the role of a dedicated citizen, then planning his city’s new bridge will help him verify both identities. If, however, two identities have oppositional meanings, they cannot be verified at the same time. The person will likely re-identify in order to re-establish the congruence of their multiple identities.32 As we saw in the beginning when we pondered the link between ‘Roman’ and ‘orthodox’, we have no difficulty in acknowledging that Christian identity and Roman identity were considered congruent enough to be acted out together, that is, that they shared enough ‘meanings’ to verify one another. We accept that those who lived with Roman traditions were Christians as a matter of course. Likewise, we agree that Christian spaces projected a vision of Christianity which was, for instance, inclusive of Roman visual and material culture. On top of this, what this book proposes is that the constant negotiation of the lateral category memberships Romanness and Christianness affected the religious identity of Christians. Acting out several identities together, this book argues, had an impact on how each of them was defined and nuanced. In other words, what being a Christian meant to Christians depended on how they enacted their Roman identity. We will explore different nuances of this interplay across the late antique western Mediterranean. As a starting point, we will encounter a Christian community in today’s Algeria which, material culture suggests, emphasized its belonging to traditional Roman culture over that of its Christianness (Chapter 1). Further case studies indicate that some Christian communities in the late antique Roman West visualized their religiosity in terms that Christian authorities considered unorthodox (Chapter 2). These visualizations,

31

Burke 2003b; Stryker 2000; Stryker and Burke 2000, 289–291.

32

Burke 2003b, 197–200.

11

12

Introduction

however, might well have been in line with individual understandings of orthodoxy. Finally, I suggest by means of visual analysis that Christian communities in Ravenna deliberately established an ancestral relationship between Christianity and the Graeco-Roman world. Simultaneously, they associated the latter with antiquity, a connection that needed to be surpassed (Chapter 3). The practice of simultaneously identifying with both Christianity and Romanness in baptismal spaces in particular had potentially long-term consequences for entire communities’ understandings of Christianity. As will be argued in the next section, baptism was experienced as constitutive of the construction of Christian identity and could affect future generations’ ideas of what it meant to be Christian.

The Baptistery as a Place of Christian Identity Construction In this book, I propose the study of decorations of baptismal spaces as a means to examine Christian identity construction. Made for all who wished to be baptized, baptismal art and architecture are, according to the missionary logic of conversion, meant for everyone.33 They are, as it were, situated dogmatically at the very heart of the church. That the sacrament initiating the profession of Christian truth could have taken place in sight of imagery with the potential to undermine Christian doctrine would thus require further explanation. Indeed, pre-Christian imagery is occasionally found in a variety of Christian communal spaces in late antiquity, and all these occurrences deserve attention. The baptismal space, however, has a special status which requires particularly careful examination. This special status derives from where the baptismal rite is situated semiotically in the life of a Christian. Baptism affects the spiritual rebirth of the baptizand as a member of the body of Christ and is the precondition for eternal life in heaven.34 Thus, only the baptized could receive Christian burial.35 In life too, the sacrament was a momentous turning point. All but the baptized were excluded from Eucharistic prayers and communion.36 With baptism came the understanding of the true nature of Christ. The Christian mysteries, the nature 33

34

35 36

On conversion in the Roman Empire, see Goodman 1994; McLynn, Papaconstantinou, and Schwartz 2015. On the sacrament of baptism as effecting a change in the religious identity of Ambrose and Augustine, see Aasgaard 2011. Brown 1997, 657. This is well attested for Augustinian North Africa. Burns Jr. and Jensen 2014, 203.

The Baptistery: Christian Identity Construction

of the Trinity, and the meaning of the Eucharist were revealed in the process of preparation for, or retrospective explanation of, the baptismal rite.37 At the same time, the official profession of Christian faith was another unprecedented requirement in the life of a Christian. Textual evidence suggests that the late antique baptismal ceremony was preceded by a period of catechesis, in addition to many other ceremonies of spiritual cleansing, exhortation, and inner contemplation. The creed was learned by heart as part of the catechetical lessons. Augustine, who is the prime source for baptismal liturgy in the Latin-speaking world of the fifth century, made catechumens profess the creed individually the day or night before their baptism.38 The introduction of the credal recitation, it has been argued, had implications for the faith of the catechumens. Through it, their faith was meant to change from a personal commitment to Christ into a belief in the body of doctrines necessary for the baptismal ceremony.39 During the ceremony itself, at least where the Nicene rite was followed, the baptized were immersed three times in the name of the Trinity, progressing from the name of the Father and the Son to the Holy Spirit. This had to be affirmed each time by the baptized.40 Augustine himself stressed the importance of the neophyte’s profession of faith for the efficacy of the sacrament: ‘What is the baptism of Christ? The bath of water in the word. Take away the water; there is no baptism. Take away the word; there is no baptism.’41 Augustine describes the baptismal experience quite literally as an immersion in Christian doctrine. Individuals arguably perceived baptism as transformative and identitychanging. In the fifth and sixth centuries, adult baptism, which required a clear awareness of the sacrament as a choice made by the individual, was practised alongside child baptism and appears to have prevailed in some areas.42 Iberian council texts, for instance, characterize baptizands generally as infants in the sixth century at the earliest. The archaeological evidence suggests an even later date for the predominance of child baptism in this region.43 Identity-changing experiences are, however, not only

37 38

39 41 42

43

On baptismal catechesis and ceremonies, see Ferguson 2009; Saxer 1988. Ferguson 2009, 784. For an overview of Augustine’s theology of baptism, see Patout Burns 2011. On the Augustinian ceremony, see also Ferguson 2009, 785–786. 40 Bradshaw 2006, 107. Ferguson 2009, 638. Augustine, Expositio evangelii secundum Johannem, 15.4. Jeremias 1958, 102–114; Kretschmar 1970, 744–750; Nagel 1980, 111–118; Cummins 1994. An archaeological perspective is offered in Schneider 2011. For North Africa, see Harmless 2014, 111–123; Meer, Battershaw, and Lamb 1961, 349–353. Schneider 2011; esp. 1703, 1713–1714. See also n. 408 in Chapter 2.

13

14

Introduction

confined to adult baptism. Where sponsors stood in for children, they vouched for them, taking on the task of helping them become faithful Christians. The sacrament had also an effect on the self-concept of others besides the baptizand. The baptismal ceremony effected the transition from the catechumenate, the state of adhering to Christ, to fidelis, the state of being faithful. Technically both the catechumens and the faithful were Christians, but the role which Christianity played (or was supposed to play) in one’s life changed considerably through baptism. This is reflected in the demanding procedures necessary for becoming a fidelis, as opposed to those required to become a Christian. The most outspoken witness we have for the admittance to the catechumenate after the fourth century is again Augustine.44 In a letter from c. AD 403, Deogratias, a deacon of the church of Carthage, asked Augustine to instruct him on how to best relate the good news to someone eager to become a Christian. Augustine’s reply, De catechizandis rudibus, provides insight into the missionary work which started from the first encounter between the cleric and the prospective Christian.45 This first meeting did not take long. The cleric instructed the candidate about subjects such as the unity of the Old and New Testament, the ideals of the love of God and the love of one’s neighbour, and the necessity of living in accordance with Christian morals.46 The candidate was then asked if they believed in these things and wished to put them into practice. If the answer was positive, the candidate received the signatio, a sign of the cross on the forehead, and was henceforth a Christian.47 The entry to the catechumenate was relatively undemanding, particularly when compared to the more thorough and rigorous examination which candidates in previous centuries had to undergo. According to the Apostolic Tradition of Hyppolitus, catechumens spent three years in attendance before admission to baptism was granted.48 When and where exactly the Apostolic Tradition was written – before AD 235 in Rome, or written and compiled by different authors between the second and fourth 44

45 47

48

Other preserved treatises on the catechumenate in the fourth and fifth centuries do not describe how candidates were instructed. Dujarier 1979, 91–93. 46 Harmless 2014, 131–133; Dujarier 1979, 93–94. Burns Jr. and Jensen 2014, 202. There is debate over which other ceremonies might have accompanied this admittance to the catechumenate. The giving of salt seems to have played a role as a pre-baptismal practice in North Africa; the exsufflatio a blowing for purposes of exorcism and the laying of hands upon the candidate are also reported from other parts of the Mediterranean. Burns Jr. and Jensen 2014, 202–203; Harmless 2014, 174, n. 157; Dujarier 1979, 92; Meer, Battershaw, and Lamb 1961, 354. Dujarier 1979, 48 and 91–106.

‘Lived Religion’

centuries – is a subject of debate.49 Augustine, in contrast, did not specify any minimum period of preparation required for catechumens to become fideles.50 While catechumens were invited to attend sermons like everyone else, we do not have any evidence that their spiritual journey was supervised before they enrolled in the baptismal preparation process held during Lent.51 Regular catechetical lessons, examinations of candidates’ preparedness, and fasting and abstinence from a variety of things are recorded for the Lenten preparation only.52 For Augustine’s time, a procedure designed to help with the process of inner transformation required by baptism is testified to only after catechumens had signed up for baptism. As a consequence, it seems that popular opinion held that it was necessary to make a distinction between the seriousness of catechumens and that of the faithful, as Augustine himself reports: ‘Let him be, let him do as he likes; he is not baptized yet.’53 The sacrament of baptism was the once-in-a-lifetime transition from the old life into the new Christian one and the official celebration of admission into the circle of the faithful. It was also the first instance of a public profession of belief in Christian doctrines. In late antiquity, the ceremony of baptism was enacted as a central event in the formation of Christian identity – perhaps even more so than in the early days of Christianity due to the fading importance of the catechumenate. The baptismal space is a place of Christian identity construction.54 In this regard, the decision of some communities to surround neophytes during this rite of passage with imagery recalling a cultural sphere which Christian orthodoxy demanded be left behind must have surely been significant. But how can we make sense of visual and material culture to better understand how Christians constructed their identity?

‘Lived Religion’ and the Study of Visual and Material Culture Students of late antique baptism and baptismal art find themselves in a dilemma: catechetical and mystagogical literature seems to promise the most immediate and detailed glimpses into a remote ritual practice, yet it 49

50 52 53 54

For diverging opinions, see the debate between Brent 1995; Bradshaw et al. 2002; Bradshaw 2002; Stewart 2015. 51 Dujarier 1979, 94; Burns Jr. and Jensen 2014, 202. Burns Jr. and Jensen 2014, 203. Burns Jr. and Jensen 2014, 204–205. Augustine, Conf. 1.11.18. See also Rebillard 2012, 65–66. The impact of the baptismal space on the identity transformation of a neophyte is also discussed in Day 2018.

15

16

Introduction

also provides us with the normative and universalist perspectives of Christian writers which we originally sought to pit local customs against. To counterbalance this, the case studies in this book focus on gaining information from the material culture on site. In doing so, I try to avoid taking works of art as expressions of a preconceived shared understanding of religious values. Instead, I start from the assumption that imagery and space are constitutive of both the creation and negotiation of group identities within (religious) groups.55 The formation of a group’s religious identity is a discursive process which needs reconfirmation through cultural practices, for example, language or the use of material culture.56 Images, architectural settings, places, and situatedness within landscape all give visibility to the characteristics and aims of a given community. These visual and spatial settings, created at some point in time, enter an interactive process of transformation, replacement, and reuse. People react to the buildings in a given place by formulating visions of what their religious environment should be like. At the same time, their notions are predefined and limited by the built environment. Imagery and architecture are not just receptive of the community’s ideas about their religious environment; they also form the community. Shaping, living with, and altering built environments are all part of the lived practice of religion. This book is inspired by efforts in the study of religion and archaeology to establish ‘lived religion’ as a category of investigation.57 This ongoing project reflects the wish to free the study of religion from a fixation on belief as the defining quality of religion, which ultimately derives from a Christian, and especially Protestant, bias.58 The ‘lived religion’ approach takes the actual experiences and practices of religious persons as key indicators for what religion is. This is not synonymous with disregarding the religious beliefs a person might hold; it merely shifts the emphasis of our observation to the individual and situational character of all religious experience and practice.59 That said, ‘lived religion’ does not 55

56 57

58

59

Neuner, Klammer, and Fricke 2011. An important early study on the impact of space on processes of communal identity construction of late antique Christians is Yasin 2009. Bourdieu 1991, esp. 220–223. Material culture is accepted as a crucial component in the field of lived religion. See for instance Gasparini et al. 2020; Orsi 2005, 73–109; McGuire 2008, 97–118; Vásquez 2011; Morgan 2010b; Droogan 2013; Meyer 2015; Houtman and Meyer 2012; Albrecht et al. 2018. On the history of Christian faith, see Morgan 2015. On the Christian bias towards belief, see Lopez Jr. 1998; McGuire 2008, 14; Droogan 2013, 21–28. On the way in which Protestant biases shaped the study of early Christianity, see Smith 1990, esp. 1–53. In this sense, religious belief itself can be understood as a religious practice. Morgan 2010a, 4–5.

‘Lived Religion’

project complete individuality onto religious experience; rather, it accepts that ‘people construct their religious worlds together’, for instance by performing rituals or building and using religious spaces.60 The study of Christian identity construction stands to benefit from a material focus.61 However, while the attention to objects, images, spaces, clothing, or food has begun to transform academic discourse in the study of religion, methodological approaches which can do justice to this ‘mute’ body of evidence are still in their infancy.62 Especially Chapters 1 and 2 contribute to this area of research. They interrogate the material and visual culture of the sites studied in this book for information about the lived religion of the communities who built and used the sites. The visual analysis of the baptismal decoration plays a key role in this investigation. The chapters examine how access, circulation patterns, and sightlines predetermined the experience of baptismal decorations; how ecclesiastical and profane buildings flanking the baptisteries extended the ways baptismal space was used; and how the reuse and reframing of pre-existing structures helped shape the perception of the baptismal space.63 Chapter 3 is an iconographical study which draws on textual analysis. As in the previous chapters, I will privilege the site-specific evidence. At the outset, the limitations of researching the interplay between art, architecture, and identity formation must be clearly stated. Studying the conditions under which Roman material culture occurred in baptisteries informs us about the sensory frames provided for individual responses but cannot give us insight into users’ reactions. As individual responses to material culture are not discernible, the study of material culture thus forces us to talk about an anonymous collective of recipients. Moreover, the archaeological evidence provides ample material for sharpening our understanding of religious practices, but it is difficult to isolate religious beliefs from practice.64 These limitations, however, should not discourage

60

61

62 63

64

McGuire 2008, 14. In the same vein, ritual studies have questioned the common understanding of a ritual as a tool for the transmission of beliefs, replacing it with models that emphasize the impact of the act of performance on the spiritual and emotional experiences of individuals and amongst groups. See for instance Smith 1987; Seligman et al. 2008. The study of identities through the study of material culture and art has long been identified as a promising area of research: Fowler 2010; Grahame 1998; Raja and Rüpke 2015, 22–23; Laurence and Berry 1998; Insoll 2011, esp. 958; Graves-Brown, Gamble, and Jones 1996. Morgan 2016, 14–15. Exemplary in their discussion of perception and movement are Leatherbury 2020; Yasin 2016; Ardeleanu 2018. Rebillard 2015, Woolf 2015, 474. For an overview of archaeological approaches to religion, see Raja and Rüpke 2015.

17

18

Introduction

inquiry into religious practices by way of material culture. How religiosity is expressed is highly significant for the religious experience of individuals, and this is certainly no less valid where religious practices deviated from what might be considered normative today.65 But how can an analysis of baptismal art of pre-Christian origin help us understand how Christians in the fifth to seventh centuries reconciled Roman and Christian identities?

Roman Visual and Material Culture in the Baptismal Sphere The period of investigation overlaps with the cultural and social changes which marked the disintegration of the late western Roman Empire.66 The establishment of Barbarian kingdoms in the western Mediterranean, the Visigothic Kingdom on the Iberian Peninsula, the Vandal Kingdom in North Africa, and the Ostrogothic Kingdom in Italy led to processes of diversification across the interconnected Mediterranean.67 Ethnically diverse societies transformed the Roman world, increasingly blurring the boundaries between indigenous and migrating populations. As a result, the role which ethnicity played for the identity of migrating peoples has received much attention.68 What constituted Roman identity in the post-Roman western Empire, however, is a question which has only recently begun to be asked with fervour.69 Walter Pohl suggests differentiating between different modes of late antique Roman identification which can but do not have to determine Roman identity.70 Pohl proposes categories as diverse as urban, political, legal and civic, military, territorial, imperial, cultural, religious, and ethnic Roman identities.71 This view allows Romanness to mean many different things in different places and for different peoples.72 It also helps increase our awareness of features of Romanness which might have been identifiable as Roman longer than others. 65 66

67

68

69

70

Raja and Weiss 2015, 141. On the transformations of the late antique Roman West, see Pohl 1997; Ward-Perkins 2005, 80–102; Smith 2005; Halsall 2007; Heather 2007; Wickham 2009. Abulafia 2011, 226–238; Horden and Purcell 2000. A picture of the disintegration of the Mediterranean region long before the Barbarian era is drawn in Hingley 2005, 91–120, esp. 91–92. See for instance Amory 1997a; Pohl and Reimitz 1998; Geary 1999; Gillett 2002; Barnish and Marazzi 2007; Gantner, Pohl, and Payne 2012; Pohl and Heydemann 2013; Buchberger 2017. On the initial results of the ERC project ‘Social cohesion, identity and religion in Europe, 400–1200’ that was held at the University of Vienna from 2011–2016, see Pohl et al. 2018. 71 72 Pohl 2018, esp. 9–26. Pohl 2018, 9–26. See also Greatrex 2000.

Roman Visual and Material Culture

This study is predominantly concerned with an aspect of Roman identity often considered exceptionally long-lived: cultural Romanness, or more precisely the use of Roman art and material culture. While political and legal definitions tend to be discussed as being increasingly irrelevant from the fifth century onwards, Roman literary and artistic culture are driving factors for identifying with Romanness in late antiquity.73 A shared education, literature, and language are the markers of Roman identity most commonly referred to by scholars, with special attention given to elites, including Barbarian elites.74 Upper social strata not only cherished classical art in the domestic sphere but are also seen as the motor behind the assimilation of Roman pre-Christian iconography in Christian art.75 It should be pointed out that Roman culture is explicitly not treated as an elite phenomenon in this study.76 Instead, by culture I mean any continuously evolving process of a group of people engaged in creating a shared identity.77 This can include fleeting expressions of Romanness, such as ways of behaving publicly and structuring social life, including dining, festivals, and dress. The art and architecture dealt with in this book clearly targeted all Christians irrespective of their social standing.78 Profound classical learning was barely required to relate to the iconographies discussed. Deciding what defines material culture as ‘Roman’ in post-imperial Rome, however, remains a challenge.79 What do we consider? The techniques of production, style, and subject matter, or the ambition and civic, ethnic, or cultural background of the craftsman, of the commissioner, of the users?80 To go with all of these criteria would mean losing any clear 73 74

75

76

77

78 79

80

Woolf 1998, 240–249; Merrills and Miles 2010, 88–90. See for instance Pelikan 1993; Humphries and Gwynn 2010; Cameron 2004. For ‘Germanic’ use of ‘Roman’ culture Hen 2007 and n. 347 in Chapter 2. Classical studies include Weitzmann 1960, 43–68; Hanfmann 1980, 75–98; Grabar 1961, 31–54, esp. 41–42. For a more recent study, see Uytterhoeven 2009, 332–335. The capacity of art and material culture to transgress private and public spheres is increasingly becoming a concern to scholars; see for instance Thomas 2016. Elitist concepts of culture are omnipresent in the history of the late Roman world. See for instance Hingley 2005, 69–71; Cameron 1993, 131–141. On Lusitania specifically, see Jorge 1999. On non-elite culture, see Clarke 2003, 73–94; Burrus 2005; Toner 2009, 185–197. Fresh perspectives on the subject can be found in Grig 2017; Tannous 2018. Assmann 1992, 134; Howes 1996, 155–157. On culture as a process and a problematization of the status of material culture in such a definition of culture, see Tomlinson 1991, 4–5. On the social stratification of late antique churchgoers, see Rebillard 2018; MacMullen 1989. A comprehensive study on this subject is forthcoming: Hubert Fehr, Walter Pohl, and Philipp von Rummel (eds.), The Transformation of Romanness: Archaeological Perspectives (400–800 AD). See also Webster 2001; Rummel 2013. Pohl 2013, 32.

19

20

Introduction

sense of what the term ‘Roman’ implies in the utterly Romanized late antique societies. From a certain point of view, we may argue that virtually every Christian building addressed in this book is more or less rooted in Roman art and architecture and could therefore be labelled ‘Roman’. Indeed, the church’s crucial role in the continuity and transformation of classical Graeco-Roman heritage in the post-Roman period is well established.81 However, with this loosest possible sense of the term in mind, little is won if we want to obtain insights into the self-concept and the lived religion of late antique Christians from the material and visual culture they used. Miguel Versluys has lucidly advocated abandoning any notion of Roman art as the style of a nation-state or identifying what is Roman in a given artefact solely on the basis of stylistic or material properties. Such an approach does not do justice to the great connectivity character of the Mediterranean and Near Eastern pre-modern cultures, a character which led the successor culture of Rome to develop an immensely aggregative approach to art, nor does it account for the contextual construction of meaning. With the adaptation of the Mediterranean cultural koine for the specific purposes of local contexts, formerly attributable styles, materials, iconographic elements, and so forth lost original meanings and geographical associations and received new ones.82 Instead, Versluys champions an approach to art and material culture which investigates the ‘cultural affiliations’ a style or element carries, that is, those aspects which constitute their belonging to a given culture or cultures in the eyes of the beholder.83 Cultural affiliations are not inherent properties. Individual or collective attributions and ways of handling imagery and objects endow them with meaning, and cultural significance is one part of this.84 The interpretative level of cultural connotations has come to 81

82 84

A concise summary of the historiography and prevalent themes in this fertile field of research can be found in Angenendt 2009. For a critical reassessment of Angenendt’s use of the concept of ‘primary’ and ‘secondary’ religion, see Uehlinger 2021; Lenk 2021. 83 Versluys 2015, 158. Versluys 2015, 154–155. On the attribution of value and meaning to objects, imagery, and architecture, see Appadurai 1986, esp. 3–5; Jones 2000, 29–34; Elsner 2012. In the study of art bound up with liturgical practice, Staale Sinding-Larsen’s processual understanding of the production of meaning has been groundbreaking. An image in a liturgical space will leave space for different interpretations in the course of a given liturgical action; moreover, its significance will change in different ceremonies over the ecclesiastical year. Since a whole array of meanings can be applied to one image, the image potentially embodies them all. Hence, the image will evoke the many different ways it can be read, and the many different ritual actions connected with it, even when these actions are not executed. In consequence, the image embraces a complex range of meanings that can shape ritual processes and is itself shaped by them. See Sinding-Larsen 1984, 36.

Roman Visual and Material Culture

prominence in the context of art-historical globalization studies.85 How viewers associate objects or images with a particular culture – thereby defining them as foreign or familiar, as representative of something outside or within their own culture, or both – is a question highly relevant to the study of the material cultures of interconnected societies. However, it is worthwhile to use the concept of cultural connotations in the field of late antique art more broadly, as this concept is not only of interest in border zones but also crucial in times of cultural transition. In this book, only such cultural affiliations with Rome which can be relatively safely assumed to have existed in the minds of many late antique Christians will serve as initial indicators that baptismal imagery could have been associated with ‘Roman’ culture. I am thinking of imagery alluding to institutions and practices which, in late antiquity, are still strongly anchored in Roman tradition. More precisely, these are circus games, bathing culture, and mythology. The selection of these institutions and practices depends entirely on the characteristics of the imagery found to adorn baptismal spaces in the western Mediterranean region. I will ask about the role such imagery played in a given baptismal space, and how the Christian community dealt with their Roman heritage more broadly. I will further examine the contexts in which similar imagery was displayed outside of the Christian realm, and whether late antique viewers could have been acquainted with these other settings. As we shall see, many of the Roman traditions which baptismal art alluded to were either still actively practised or relatively fresh in the memory of fifth- to seventhcentury Christians. Granted, ascribing indiscriminately monolithic cultural affiliations to the objects of study would defeat the purpose, as it would not differ from ascribing one-dimensional orthodox Christian meanings to them. Therefore, in each case study I will explore which other possible meanings the respective imagery might have held beyond its cultural affiliation with Rome, both on a local and on a more global level. Moreover, as viewers’ ability to detect Roman cultural affiliations arguably decreased over time, at later stages of a baptistery’s use we need to differentiate, if possible, between consciously undertaken maintenance of Roman material culture and preservation for practical reasons (see Chapter 2). By investigating to what degree Christian communities associated baptismal art not only with Christian but also with traditional Roman culture, 85

On cultural connotations as an investigative category, see Rujivacharakul 2011, 15–19; Versluys 2015, 141–143.

21

22

Introduction

we will encounter material and visual culture which is in a more or less open conflict with the customs and values of the Christian church. Confronted with images of this kind in a church space, uncertainty about their religious significance arises. Was an image considered Christian when placed in a Christian religious setting? Was it associated with the past? Was this past understood as Christian, Roman, or pagan? Were Romanness and Christianness essentially considered the same thing? Did an image lose its affiliation with the past when it was integrated into the Christian sphere? A central question of the study is whether baptismal art confirmed Roman and Christian identities simultaneously. To put it simply, did baptismal art encourage baptizands to think that by entering the circle of the faithful, they also confirmed that they were Romans? Another guiding question is whether Christians’ affirmation of and identification with Roman popular culture had an impact on their definition of what it meant to be a Christian. Is it possible that baptismal art, at times, was less orthodox than we think due to Christians’ indebtedness to Roman culture? At various points in this book, I will answer both of these questions in the affirmative. Additionally, I will discuss the baptismal imagery of the Orthodox baptistery of Ravenna as an example of a carefully crafted orthodox response to such questions by a member of the clerical elite.

Structure and Geographical Scope Each chapter of this book illuminates a different aspect of affiliation with Roman culture in the baptismal space: the use of Roman stock imagery and the contemporaneous absence of explicitly Christian imagery (Chapter 1), the occurrence of so-called pagan iconographies (Chapter 2), and the integration of antique personifications in the guise of Roman deities within Christian narrative scenes (Chapter 3). The diversity of the Christian communities tackled in this book shows that the integrative force of Romanness appealed to rural and urban Christians alike, to Christians of a range of confessions, and potentially to people of different ethnic backgrounds as well. Chapter 1 reinterprets the natural imagery on the floor mosaic of the baptistery of Cuicul in the Roman province of Numidia (today Djémila in north-eastern Algeria) as bare of any univocally Christian messages. Cuicul’s Christian community was indebted to traditional Roman ways of life and ritual practice and maintained elements of this tradition in their religious life as Christians. The chapter argues that the Christian

Structure and Geographical Scope

community was so profoundly Roman that its initiatory space did not need to be differentiated from other Roman institutions by its interior decoration. The desire to be distinguishable from non-Christian architecture through specifically Christian symbolism seems simply not to have existed. At the peak of what is known as the Donatist controversy, the Nicene community of Cuicul chose to blend in with traditional private and civic Roman visual culture. Chapter 2 examines three Christian communities in southern Lusitania (today Algarve and Baixo Alentejo in Portugal) and Africa Proconsularis (today north-eastern Tunisia) of which two or possibly all three used baptismal iconographies pertaining to Graeco-Roman myths and civic cultural practices. The chapter demonstrates that the iconographies alluded to Roman cultural practices that were rejected by ecclesiastical elites. They provided avenues for expressing Christian identity while still allowing an association with traditional Roman culture. Through baptismal art, the Christian communities handed down a local understanding of Christian identity which deviated from the orthodox mainstream. The ethnic identities of the Christian communities cannot be determined for any of the communities. The extent to which the communities were affected by the arrival of the Vandals or Byzantines in the province of Africa Proconsularis or by the Visigoths in Lusitania is likewise unclear. In any case, fundamental shifts in the political landscape evidently did not hinder them from feeling connected to Roman culture. Chapter 3 re-examines the probably best-studied late antique baptismal work of art, the cupola mosaic of the Orthodox baptistery in Ravenna (c. 451–473), and the cupola mosaic modelled after it in the Arian baptistery nearby (c. 500). The chapter focuses on the representation of the River Jordan at Jesus’ baptism in the form of a Graeco-Roman river personification reminiscent of a river god. The chapter argues that the Nicene bishop Neon was inspired by a sermon of his predecessor Chrysologus, when he chose to personify the River Jordan in the shape of an antique river deity revering Christ. This type of representation fitted well with Chrysologus’ interpretation of the river as a convert to Christianity. However, by visualizing the sermon in this way, Neon added a new layer of meaning to Chrysologus’ textual template. The cupola mosaic no longer showed only the conversion of the River Jordan but also a representative of pre-Christian antiquity accepting the greater truth of Christianity. The visual rendition of the same scene in the Arian baptistery accentuates Jordan’s classical appearance in the guise of a river god, nourishing the

23

24

Introduction

suspicion that Ravenna’s Arian commissioners were particularly keen to comment on the beneficial connection between Christianity and the classical world. It seems that the Arian baptistery surpassed the Orthodox model in its praise of antiquity. This new reading of Ravenna’s dome mosaics makes them outstanding cases of conscious visual commentaries on Christianity’s inclusive relation to the pre-Christian Roman past. Throughout the book, comparisons with eastern Mediterranean art and architecture and the demonstration of interdependence with developments in the eastern church will draw out how the baptisteries discussed here are – beyond all local particularisms – nourished by a shared GraecoRoman culture which transcends the late antique and early medieval East and West. A case in point is of course Ravenna, the western capital of the still united Roman Empire, where the orientation towards Constantinople is observable in virtually every domain, including art and politics.86 Ravenna’s baptismal art is an exemplary illustration of this tendency, as the creation of the hitherto unattested iconography in the cupola mosaic of the Orthodox baptistery is best explained by its indebtedness to exegetical literature written in Greek.87 However, even baptisteries created in the far West of the Mediterranean show similar tendencies. For example, the closest comparison to the Numidian habit of attaching bathhouses to baptisteries, which will be discussed at length in the next chapter, is to be found at the pilgrimage site of Abu Mena in Egypt.88 Furthermore, the mosaics covering the cryptoporticus of the supposed baptismal complex I in Mértola, Portugal, which is geographically closer to the Atlantic than to the Mediterranean Sea, share their closest iconographic resemblance with the mosaic floor of the Diaconicon baptistery on Mount Nebo in Jordan (Figure 2.29).89 Such parallels may at times be coincidental. However, strong evidence that the mosaicists in Mértola originally came from Byzantine North Africa, possibly via the Balearic Islands, is an important reminder of the extensive interconnectedness of the Mediterranean.

86 89

See most recently Herrin 2021. See 124 in Chapter 2.

87

See 182–187 in Chapter 3.

88

See 72 in Chapter 1.

1

|

The Absence of Christian Iconography and the Presence of Roman Cult and Culture in the Baptismal Complex of Cuicul, Numidia

North Africa, in particular the territories which make up today’s Tunisia and, to a lesser extent, Algeria, provides us with the largest number of late antique baptismal fonts paved with mosaics in the Roman West.1 In my attempt to explore what baptismal art and architecture can tell us about the lived religion of Christian communities, I will start with the baptistery of Cuicul (Figure 1.1) in the Roman province of Numidia (modern Djémila) in what is today north-eastern Algeria. Cuicul’s baptistery, built between the late fourth and the first half of the fifth century, exemplifies well the large amount of late antique baptismal art which employs only sparsely, if at all, imagery identifiable as Christian. Considering the impact baptism had on Christian individuals and the coherence and self-perception of Christian communities, it must be asked what motivated the choice of motifs lacking clear Christian significance. It should pique our curiosity that fourth- and fifth-century mosaic decorations of churches and baptisteries in North Africa are, in comparison to other regions of the Roman Empire, particularly conservative and limited in their use of imagery, and most do not depict Christian symbols like crosses or biblical narrative.2 Both the large mosaic ensembles covering church floors and common privately funded tomb-mosaics share a repertoire of motifs almost identical with the imagery of a generally beneficial and lucky nature, which was so popular in public and private buildings of the late third and fourth centuries in Roman North Africa.3 Vines, palms, flowers, fruit, trees, kantharoi, and animals (especially birds and fish) were taken over directly from non-Christian mosaics into the new genre of church decorations. The mosaic decoration of the Christian complex of Cuicul is a case in point. The double church of Cuicul is rich in figurative imagery, yet univocally Christian 1

2

Lassus 1970; Gui, Duval, and Caillet 1992; Baratte et al. 2014; Morfino 2011. The wealth of baptismal imagery has led to many iconographical studies, including Palazzo 1992; Courtois 1955b; Février 1984; Perler 1964; Duval and Lazreg 1995; Cintas and Feuille 1952; Duval 1980. 3 Dunbabin 1978, 188–195. On beneficial symbols, see Dunbabin 1978, 165–172.

25

26

The Absence of Christian Iconography

Figure 1.1 Cuicul (Djémila), baptistery seen from the south, late fourth to first half of the fifth century. Photography by the author.

iconography is absent.4 The double church shares this quality with some of the largest North African churches, such as Hippo Regius, Dermech in Carthage, Thamugadi (Timgad), and Theveste (Tebessa).5 A certain unwillingness to depict sacred images on floors might explain the phenomenon in parts.6 For example, tomb-mosaics in Basilica B in Sitifis (Sétif ) demonstrate how, for the best part of the fifth century, Christians chose stock imagery of a generic lucky nature to decorate tombs. The first Christian symbol, a chi-rho with alpha and omega, adorns a tomb only in 471.7 However, Christian symbols were by no means avoided consistently, as monograms, alphas and omegas, and crosses can be found in early church spaces and baptisteries.8 For example, the baptismal font of the west church at Thamugadi (Figure 1.28a), an early fifth-century baptismal structure from Numidia, displays, amongst geometric and vegetal ornaments, what appear to be short-armed Greek crosses on the six corners of the upper 4 5 6 7

A detailed description of the motifs is provided in Monceaux 1923, 95–96. Dunbabin 1978, 188–195. On North Africa, see Février 1965, 152; Dunbabin 1978, 194; Dunbabin 2014. 8 Février 1965, 87–88, 152. Dunbabin 1978, 190.

The Baptismal Complex of Cuicul

interior steps (Figure 1.28b) and christograms in the same positions one step below (Figure 1.28c). A chi-rho is depicted on one of the outer sides (Figure 1.28a).9 A limited number of standardized biblical narrative scenes – deer drinking from a fountain (Psalm 42), for instance, as well as rare examples of more inventive imagery (e.g. the ark of Noah on the baptismal font of Kelibia) – can also occasionally be found in North African churches and baptisteries (Figure 1.29). Chapter 2 includes a particularly innovative example of North African baptismal imagery.10 Should we hold the decline of the mosaicists’ craft in late antique North Africa responsible for the poverty in Christian pictorial inventiveness, as Katherine Dunbabin has suggested?11 Dunbabin herself seems not fully satisfied by this explanation, as she points to the experimental atmosphere which still led to highly original mosaic pavements in domestic contexts as late as the early fifth century.12 In this chapter, I seek to explain the pictorial conservativism of Cuicul’s baptistery by drawing attention to the lived religious reality of the community which built and used it between the end of the fourth and the first half of the fifth century. In 1965, Paul-Albert Février suggested that the Christians of Sitifis who refrained from Christian symbolism on funerary mosaics wished to follow the customs of their time without distinguishing themselves simply because they were Christians.13 Février dealt with the choices of individuals in private matters. Could the same conclusion be drawn regarding the choice of church officials in relation to ecclesiastical infrastructure? I will argue that the ambiguous character of Cuicul’s baptismal art is grounded in the Christian community’s regard for and identification with Roman traditions. The community’s alignment with Roman worship and culture outweighed the desire to visually distinguish themselves from opposing dissident communities, commonly known as Donatists, as we shall see. The Catholic church complex of Cuicul employed a visual language in the baptistery which was highly accessible and widely accepted at the expense of visual specificity. An in-depth study of a particular community like the one discussed in this chapter aids in comprehending the unique religiopolitical circumstances which influenced the selection and preservation of a specific type of decoration. It helps us realize that a decoration with global appeal and relatively vague symbolism may still serve a distinct local purpose. 9 10 12

Gui, Duval, and Caillet 1992; Leglay 1957. Dunbabin 1978, 190–193. See 88–95 in Chapter 2. 13 Dunbabin 1978, 194. Février 1965, 152.

11

Dunbabin 1978, 195.

27

28

The Absence of Christian Iconography

I will first describe the baptistery in the context of the Christian complex of Cuicul and take a stance on the issue of the construction date, suggesting that it was built in the late fourth or first half of the fifth century. The chapter posits that the baptistery played a crucial role in the complex’s function as a Catholic pilgrimage site and as a centre for constructing communal identity in opposition to dissidents. I will then recap the dominant interpretations of Cuicul’s baptismal iconography based on the writings of the church fathers and examine their usefulness for reconstructing the viewing experiences of the time. In the second half of the chapter, I will discuss what the architecture and material culture of the Christian complex of Cuicul can tell us about the Christian community’s openness towards Roman, non-Christian ritual practices. The archaeological evidence suggests that Cuicul’s baptismal architecture provided for prebaptismal purification rituals reminiscent of ritual purification in Roman religious practices. The findings should encourage us to abandon an exclusively Christian framework for understanding Cuicul’s baptismal imagery. The Christian community of Cuicul was thoroughly indebted to Roman culture and rituals, and saw no need for either binary differentiation or total transformation. The art and architecture of the baptistery drew upon various traditions firmly rooted in Roman Numidia during the late fourth and early fifth centuries, occasionally pushing the boundaries of orthodoxy. It is possible that it was this traditional character which held the most potential for the Christian community in their pursuit of establishing Nicene dominance in Numidia.

1.1

The Baptistery of Cuicul

The baptistery of Cuicul is one of the most elaborate and best-preserved baptisteries in North Africa. Since its excavation in 1922, however, it has received little scholarly attention and since Algerian independence has suffered from a lack of archaeological investment. Cuicul lies in the mountains of the Tell Atlas, in the north-western part of the Roman province of Numidia, at an altitude of around 850 m, on the route between Sitifis and Cirta (Constantine).14 The city was founded as a colony for veterans possibly under Nerva or Trajan at the turn of the second century and 14

On Djémila’s urban development, see Sears 2007, 55–58; Lepelley 1979–1981, vol. 2, 402–415; Février 1996; Blanchard-Lemée 1975. See also Blas de Roblès and Sintès 2003, 89–124.

The Baptistery of Cuicul

Figure 1.2 The North African provinces after Diocletian’s provincial reorganization. © Jon Cubas-Díaz.

was inhabited until at least the seventh century.15 Cuicul was located at the border to the province Mauretania Sitifensis, which had been cut out of Mauretania Caesariensis during the reign of Diocletian (Figure 1.2). The city was under Vandal rule from c. 430 onwards, except for an interlude between 442 and c. 455, when it briefly fell back to the Roman Empire.16 In the fourth and fifth centuries, Cuicul experienced a vivid phase of construction and restoration of public and private buildings.17 A substantial Christian complex which supplanted earlier structures in the newly developed southern district of Cuicul was also constructed at this time.18 The carefully planned complex consists of two churches set parallel to each other and facing eastwards, a chapel, the baptistery, as well as an entrance hall, an alley, courts, and living quarters. Apart from the smaller north church (known since 1840), the main part of the complex was excavated in 1921/22 by Albert Ballu and Paul Monceaux.19 The

15 16

17 18 19

On Cuicul in the Byzantine period, see Février 1966, 87; Février 1996, 674–676. On the Vandal presence in Cuicul, see Chastagnol and Duval 1974, 91–94; Lepelley 1979–1981, vol. 2, 402–403. On the interlude of 442–455, see Courtois 1955a, 175–185. Sears 2007, 56–58; Février 1996, 674; Sintès and Rebahi 2003, 299. Christern 1976, 137–144; Gui, Duval, and Caillet 1992, 92–96. See also Duval 1989. Ballu 1921; Ballu 1923; Monceaux 1922, 1923. Beyond the excavation reports, literature on the baptistery is sparse. The baptistery is usually dealt with in the context of North African

29

30

The Absence of Christian Iconography

Figure 1.3 Ground plan of the church complex of Cuicul. Christern (1976), 138, fig. 27a.

documentation of the excavation has been criticized for its inaccuracy and inconsistency.20 A revised ground plan of the site made by Jürgen Christern in 1976 is to date the basis for the little research that has been done on the church complex (Figure 1.3). Ballu’s ground plan indicates several earlier structures built on the terrain in the west which have not been examined in detail, since the northern, southern, and western surroundings of the complex have not been properly excavated.21

20

21

Christian art more generally. See Morfino 2011; Ristow 1998, no. 75; Gui, Duval, and Caillet 1992, 95; Khatchatrian 1982, 83; Gessel 1981, 55–56; Lassus 1970; Leglay 1957; Allais 1938, 59–62; Grégoire 1938. For a more recent discussion, see Dennis 2020, 26–30. The excavators also discovered a building of a shape similar to the baptistery on a lower level west of the chapel. Monceaux 1922, fig. 5. This building features a circular wall with niches carved in the inside. The excavators speculatively identified it as a previous baptistery. Ballu 1923, 28. See for example Février 1996, 667, n. 56. The ground plan of the church complex of Cuicul is published in Ballu 1921, 216. See Ballu 1923, 25; Christern 1976, fig. 72.

The Baptistery of Cuicul

Figure 1.4 Plan of the baptismal complex of Cuicul with vestibules and bath. © Sadi Maréchal.

Cuicul’s baptistery is one of the rare examples of a free-standing baptistery in North Africa (Figure 1.4).22 Visitors would approach the baptistery from an alley flanked by porticoes (Figure 1.1). They would first reach the monumental portal of the larger, five-aisled south church, then the smaller, three-aisled north church, in front of which the baptistery is located. Westwards of the porticoed alley, in front of the south church, a small three-aisled chapel sits slightly out of line with the rest of the neatly structured complex. The circular baptistery, with a diameter of 12.15 m, consists of a baptismal chamber (Figure 1.5) with an interior diameter of 5.16 m and an ambulatory (Figure 1.6). The original building is preserved to a maximum height of 2 m.23 The quadrangular baptismal font in the centre of the inner chamber has a side length of 97 cm and a depth of 75 cm and can be entered via two steps from all four sides. It contains two drains for the flow of water and is covered by a ciborium (Figure 1.7).24 The ciborium is quite rare, as most late antique examples no longer exist.25 Four fluted columns with Corinthian capitals carry a stone vault which was found broken but 22 23

24

25

Duval 1989, 364. Christern 1976, 138, n. 36. I have not been able to establish the exact date of the reconstruction of the roof. See Gessel 1981, 56, for an approximate date not too long before 1981. Monceaux 1923, 99. On the water circulation, see the plan in Thébert 2003, pl. LXXXI, 3. On the reconstruction of the ciborium, see Gessel 1981, 56; Leglay 1957, 402–403. A fully preserved ciborium can be found in Cividale del Friuli. The ciborium of the baptistery of El Bovalar in Catalunya (today Museu de Lleida) is preserved in parts.

31

32

The Absence of Christian Iconography

Figure 1.5 Cuicul, baptismal chamber, facing the western exit to the bath. Photography by the author.

preserved in all parts, and which has been reconstructed.26 The lower side of the vault has a loop in the centre, possibly for affixing a lamp (Figure 1.8). Further loops were added on the upper side of the vault, above the capitals, and could also have been fixtures for lamps.27 The vaulted ambulatory is structured by semi-circular niches of a depth of around 50 cm; the inner wall consists of twelve such niches, the outer one of twenty-four (Figure 1.6).28 Two more niches of this kind were 26 27

28

Ballu 1923, 21. A depiction of a similar installation can be found on the apse mosaic showing an altar canopy at the monastery of Mor Gabriel, Tur’ Abdin, Anatolia, c. 512. See Lafli and Bru 2013, fig. 14a. I owe this comparison to Vladimir Ivanovici. Monceaux 1922, 404. It has been suggested that the niches, inserted at a height of 47 cm, may have provided seating space. Ballu 1923, 19.

The Baptistery of Cuicul

Figure 1.6 Cuicul, ambulatory of the baptistery. Photography by the author.

placed at the north and south of the baptismal chamber (Figure 1.5). The find of the cement frame of an oculus of 25 cm by 27 cm nourished the excavators’ assumption that the building had been lit from above.29 During the restoration, the ambulatory was vaulted, and the baptismal chamber covered by a cupola. The baptistery’s original height and type of roofing cannot be determined with certainty.

29

On the basis of the find of the cement frame, Monceaux suggests an opening in the middle of the cupola. Monceaux 1922, 404. Ballu proposes albeit with no evident reason that eight square oculi of 25 cm length were placed in the vaulted ceiling of the ambulatory. Ballu 1923, 20. Finally, Allais assumes that occuli were placed both in the vaulted gallery and in the cupola and were glazed, as glass fragments were found on the site. Allais 1938, 60. Without any further documentation provided by the excavators, the function and installation of the oculus remain unclear.

33

34

The Absence of Christian Iconography

Figure 1.7 Cuicul, baptismal font covered by a ciborium, seen from the west. Photography by the author.

The baptistery’s central rotunda sits inside of a slightly irregular quadrangular complex (Figures 1.4 and 1.9) consisting of two representative vestibules in the east and a small bath in the west. In the east, two entrances lead via staircases downstairs into the vestibules of the baptistery. Turning sharply to the right or left, the inner circular zone of the baptistery could be entered via two narrow doorways. The south-eastern vestibule is placed directly in front of the main entrance of the recessed north church (Figure 1.3). From this vestibule, one entered straight ahead into the bath. Adjacent to the north-eastern vestibule is a small room of unclear function.30 The most prominent entrance to the baptistery is the western 30

Allais suggested that the apsidal space within the rooms was the consignatorium, where catechumens professed their faith before the baptismal rite: Allais 1938, 61. Alternatively, it has also been suggested that the small chapel adjacent to the bath could have functioned as

The Baptistery of Cuicul

Figure 1.8 Cuicul, vault of the ciborium from underneath. Photography by the author.

entrance, which is designed as an apsidal niche and links the bath with the baptistery (Figure 1.10). A straight vista passes from here over the ciborium into a conched apse in the east (Figure 1.7). The small bath does not seem to have ever held any other function.31 It consists of a frigidarium in the central hall, latrines in the south, and a northern part which contained three basins which could be heated (Figure 1.4).32 The bath could also be reached from an independent entrance in the west.33 The richly decorated baptistery is comparatively well preserved. On the exterior of the rotunda, ashlar arranged in the form of crosses with short arms is integrated into the opus latericium (Figure 1.1). The interior walls of the ambulatory and baptismal chamber would have been covered with plaster, and traces of paint were still visible in the niches of the ambulatory

31

32 33

consignatorium. However, due to a slope between the chapel and the baptistery, the structures around the chapel have disappeared up to the cave level, making it impossible to deduce the exact relation of the chapel and the baptismal complex. Lassus 1965, 598; Lassus 1970, 250–251; Février 1996 no 53. Thébert 2003, 203. Février has raised the possibility that the bath was not in use anymore at the time the baptistery was built, albeit without providing any arguments. Février 1996, 664. Thébert 2003, 202–203. Cf. Maréchal 2020, 349–350. See also Ballu 1923, 22–23. A further entrance in the south leads to the latrines.

35

36

The Absence of Christian Iconography

Figure 1.9 3D-model of the baptismal complex of Cuicul with vestibules and bath. © Zamani, Cultural Heritage Documentation Project, University of Cape Town.

at the time of excavation.34 The excavators left no record of any fragments of a potential mosaic or painted decoration of the ceiling. Shells carved in stucco decorate the vaults of the niches.35 A series of small pilasters with capitals and consoles runs between the niches (Figure 1.6). The intricate pattern of the ambulatory floor mosaic has a star motif; between the stars, there are crosses pattées filled with ivy leaves and swastikas (Figure 1.11). The baptismal rotunda contains figurative decoration. Around the loop in the stone vault of the ciborium is a carving of an eight-pointed star in shallow relief (Figure 1.8). The rotunda’s floor mosaic makes the biggest visual impact. A marine scene with a large variety of sea animals is framed by a meander at the outer border (Figure 1.12).36 The white ground is marked with dark blue parallel lines, which suggest moving water, but is also structured by four

34 36

35 Monceaux 1922, 404; Ballu 1923, 19. Monceaux 1922, 402–403. The mosaic decoration has been mentioned in passing in Monceaux 1922, 403; Allais 1938, 61; Gessel 1981, 56; Drewer 1981, 545. Only the mosaic inscriptions have been explored in any depth: Dennis 2020, 29; Pflaum and Dupuis 2003, 868–869; Monceaux 1923, 106–107 and on the marine scene 99; Grégoire 1938.

The Baptistery of Cuicul

Figure 1.10 Cuicul, bath adjacent to the baptistery; frigidarium in the foreground, main entrance to the baptistery in the background. Leschi (1953), 51.

ornamental shrubs kept in baskets at the corners of the font, as well as by four jewelled kantharoi filled with water on the font’s four sides. Pairs of stylized dolphins frame the kantharoi, while the majority of the other animals are loosely placed and depicted naturalistically. In addition to diverse kinds of fish, there are crabs, starfish, octopuses, shrimps, sea urchins, molluscs, water flies, and tortoises. The ambition of the mosaicists seems to have been to represent the entirety of marine life. The potted shrubs, however, evoke a cultivated and domestic sphere. They, and the kantharoi, are taken from a metaphorical set of imagery generally associated with notions of plenty.37 Today, the floor of the almost quadrangular baptismal font depicts a pattern of eight red and blue fish on a white background. Of the fish, half are swimming along the sides of the font, while the remaining four are facing a small white swastika framed in a red roundel in the middle of the basin (Figure 1.13). In contrast to the colourful design of the floor of the rotunda, the more stylized mosaic of the basin uses a moderate selection of colours. A white inscription on a red background, with letters measuring about 15-cm high, runs along three of the four sides of the basin’s 37

The potted shrubs are classical compositional elements and are attested in North African domestic decorations. Compare for instance the pavement of oecus XIV at the House of Nicentius in Thuburbo Maius from the late fourth century today preserved in the Musée du Bardo. Ben Abed-Ben Khader et al. 1985a, plate XX, fig. 40a; Yacoub 1996, 78. For the translation of such separating elements into church spaces see Öğüş-Uzun 2010.

37

38

The Absence of Christian Iconography

Figure 1.11 Cuicul, detail of the mosaic floor of the ambulatory. Photography by the author.

border.38 A drawing of the inscription published by Monceaux shortly after the discovery of the baptistery in 1923 reveals that less than half of the central image had survived to the twentieth century (Figure 1.14).39 The mosaic was heavily worked over in a later restoration, the date of which is not documented. The original components can be distinguished with the naked eye because they are darker in tone. The preserved part of the inscription reads: ‘EMPUS ERITOMNESIN FONTE’. It has been plausibly completed to tempus erit omnes in fonte [renasci] – ‘the time will come when all [will be reborn] in the fountain’.40 The design of the inscription prescribes the reading direction. The baptized would have been able to read the inscription fully only when facing the conched apse (Figure 1.10) in the east. The text starts in the east (EMPUS), where the first half of the line is missing, continues on the southern side (ERITOMNESIN) and finishes on 38 40

39 Monceaux 1922, 405. Monceaux 1923, 107, n. 21. This reading goes back to Monceaux, who suggested that the missing part should be completed as follows: [Gentes t]empus erit omnes in fonte [renasci]. The addition of gentes in the sense of ‘heathens’ is, however, only speculative and not as crucially necessary for the logic of the phrase as the addition of a verb renasci is. The addition is inspired by Matthew 28:19: Docete omnes gentes, baptizantes eos . . . Monceaux 1922, 406, n. 1; Pflaum and Dupuis 2003, 869, no. 8281. Nathan Dennis has recently demonstrated that the inscription is a paraphrase of Vergil’s third Eclogue. Dennis 2020, 29.

The Baptistery of Cuicul

Figure 1.12 Cuicul, detail of the mosaic floor around the font. Photography by the author.

the northern side (FONTE), where the second half of the line is missing. The western side shows a wave pattern. Therefore, the setting of the inscription indicates that the neophyte must have entered the font from the west (Figure 1.7). Unfortunately, Monceaux’s drawing (Figure 1.14) records the text but not the design of the mosaic. However, it is noted in the drawing that the central panel depicted fish, and indeed, in the surviving fragment, two fish

39

40

The Absence of Christian Iconography

Figure 1.13 Cuicul, mosaic pavement of the font. Photography by the author.

swim towards the centre of the mosaic with another one swimming in between them. The swastika in the middle of the panel was added by the twentieth-century restorers, probably in emulation of the swastikas integrated into the ornamental floor of the ambulatory (Figure 1.11).41 The font’s schematic design, including a centrally placed cross (rather than a swastika), is an unprovable, but not implausible addition of the restorers. At the latest in the sixth century, fonts from the provinces of Africa Proconsularis (Figure 1.29) and Byzacena (Figure 1.30) were frequently decorated with a large, centrally placed cross or a chi-rho;42 at times, a dove or a lamb would replace the cross.43 However, it seems unlikely that Monceaux would have left a cross or swastika out of his report. Even if there had been a cross or another Christian symbol in the original state, due to the mosaic’s design it could not have been much larger than today’s swastika, 41

42

43

A small swastika is also depicted in one of the panels of the mosaic floor of the south church, which is today exhibited in the museum of Djémila (Figure 1.17). Due to its singularity in an otherwise regular ornamental pattern, it cannot be excluded that this swastika is a modern addition. Baratte et al. 2014: Kelibia/Demna 165–169, Bekalta 227–228, Hammam Lif 150–152, El Kenissia 229–230, Bou Smir 233, baptistery I of Sfax 237, Basilica of St Vitalis in Sufetula 386–391. Baratte et al. 2014: Sidi Mansour (lamb) 235–236, Oued R’mel (dove) 190–191.

The Baptistery of Cuicul

Figure 1.14 Cuicul, schematic drawing of the decoration of the font. Monceaux (1923), 107, cat. 21.

which measures about 15 cm. The comparatively small symbol would have been imperceptible from most places in the baptismal chamber because it would have been positioned 75 cm below the floor level (Figure 1.5). To draw a preliminary conclusion, judging from the remaining material evidence, univocally Christian imagery was scarce or even absent entirely. Granted, we cannot exclude the possibility that the ceiling and walls were adorned with paintings, even though the excavators found traces of paint only in the niches in the ambulatory. Also, liturgical instruments and ephemeral decorations like wall hangings and lamps likely altered the atmosphere (though the uncertainty about their visual qualities leaves us with speculation only).44 Nevertheless, the considerable extant interior alludes to a luxurious profane Roman interior design (discussed further below) with few – if any – specifically Christian messages. Furthermore, it is noteworthy that the extant inscriptions in the central rotunda do not reveal unambiguous Christian messages either. Two mosaic fields with circular inscriptions in their centre have survived in the east and west of the font. Only a fragment of the western one, placed prominently in front of the entrance to the bath, still exists (Figures 1.15a and 1.15b). The 44

Late antique textual references to ephemerals in ecclesiastical spaces are largely missing. See de Blauuw and Dolezalová 2019. On the rich evidence from Rome, see de Blauuw 2020.

41

42

The Absence of Christian Iconography

Figure 1.15a Cuicul, mosaic panel at the western entrance to the baptistery. Photography by the author.

Figure 1.15b Cuicul, detail of the mosaic panel at the western entrance. Photography by the author.

The Christian Complex of Cuicul

inscription is written in white on a red background and red on a white background within a double circle which has a diameter of 37 cm.45 Today, the inscription of the outer circle reads ‘ADDUETNLVMIN’; in the inside circle, a single ‘N’ can be distinguished, along with fragments of another two or three letters. In 1922, the excavators suggested the partial reading in lumin[e].46 The phrase ‘[ACCEDITE] AD D[E]V[M] ET I[N L]LUMIN [AMINI]’ – ‘Come to God and be enlightened’, which draws on Psalm 34:5, was subsequently suggested.47 However, this reading is not only based on many assumptions, it also disregards the fact that deciphering the message would have been possible only by the initiated. The corresponding inscription in front of the apse is entirely illegible today. A third mosaic inscription lies almost hidden beside the south-western column of the ciborium. The enigmatic inscription in red on white background, without vowels, reads: ‘TP T SPT Q’.48 The excavators also recorded a graffito which had been engraved with a pointed instrument into the plaster in the wall south of the central rotunda: ‘+ I[n] n[omin]e D[omi]ni, ame[n]. Passassius lector in D[e]o bibat’ – ‘In the name of the Lord, amen. May the lector Passasius live in God.’49 As it was not part of the original design, this inscription raises questions concerning the occasion on which it was produced. The popular use of formulations like ‘Vivas in Deo’ in North African funerary inscriptions is unlikely to be applicable here, as the mosaic floor has not been disturbed by inhumations. Rather, the inscription resembles commemorative graffiti like the inscription ‘Paulus in Deo bibat’, found on a column in Sitifis.50

1.2

The Christian Complex of Cuicul: A Baptismal Site and a Centre of Pilgrimage?

We need to look at the Christian complex of Cuicul at large (Figure 1.3) to understand the baptistery’s function within it. The complex has an unusual element: a 90 m long subterranean passageway which runs roughly 3 m below the north and the south church and provides access to a series of 45 47 48 49

50

46 Pflaum and Dupuis 2003, 868, no. 8278a. Monceaux 1922, 403. The Clementine Vulgate Ps. 33:6 has Accedite ad eum et illuminamini: Grégoire 1938, 590. Monceaux 1922, 404. Now in the museum at Cuicul. Monceaux takes Passassius for a variety of the relatively common North African Christian name Paschasius. Monceaux 1922, 405. Pflaum and Dupuis 2003, 869, no. 8283 suggest Pallasius. The replacement of ‘b’ for ‘v’ is common in North African late antique script. Monceaux 1908, 287.

43

44

The Absence of Christian Iconography

Figure 1.16 Drawing of the subterranean passageway connecting the south and north church, Cuicul. Christern (1976), 139, fig. 27b.

crypts (Figure 1.16).51 This passageway is accessible via flights of stairs in the northern aisles of both the north church and the south church; due to a lack of excavation, it is unclear whether an additional access existed at the southern end. The two most prominent crypts are apsidal and are located underneath the apses of the north and the south church, which the crypts imitate in their dimensions. In the western walls, niches of different forms face the crypts.52 The crypts contain the remains of bases of unclear purpose.53 The passageway might have played a central role in the complex’s use. Christern argued that the passageway had a memorial function. He based his argument on another monument in the south church, the Cresconius inscription (Figure 1.17). The inscription indicates the ambition to use the church as a memorial site for deceased bishops. It is also relevant for the dating of the Christian district, which will be discussed in the next section. The inscription, placed in the choir of the south church, west of the fenced altar space, is a panegyric verse on the deeds of Cresconius, a bishop of Cuicul. The verse first praises the wonderful buildings (Hic ubi tam claris laudantur moenia tectis) at the site and tells us that Cresconius prevented his legitimate predecessors (iustos . . . priores) from being forgotten by relocating them to this place of worship. The following phrases describe united Christians (in unam congeriem) who come to the place of the bishops’ tombs to pray, chant hymns, and receive salvation by a sacrament which involves being anointed with chrism (sacramento dei medicinam sumere c[r]ismae). The last part of the inscription is the epitaph of Cresconius, which explicitly states that Cresconius was a Catholic, that is, an adherent of the Nicene church.54

51 53 54

52 Christern 1976, 140; Duval and Février 1991. Christern 1976, 140. Christern 1976, 140 and 142–143. Cf. Gui, Duval, and Caillet 1992, 95. On the ‘Catholic-Donatist’ conflict within the North African church, see 57–59 in this chapter. On the labels ‘Catholic’ and ‘Donatist’, the latter of which is an ex-post creation, see Shaw 1992.

The Christian Complex of Cuicul

Figure 1.17 Cresconius inscription from the south church of Cuicul, Museum of Djémila. Photography by the author.

Here where the city with its famous buildings is praised you find shining gables and holy altars. These are not the work of noblemen; but through the glory of such great achievement the name of its creator Cresconius resonates throughout the centuries. People are happy that with his honourable works he put his rightful predecessors in their beautiful positions; [these] who repose took out of sight permanently are now shining brightly resting underneath a beautiful altar. They are happy that their gathered crown shines, because this honourable protector has wisely fulfilled the task with spirit. And from everywhere a generation of Christians comes together, fulfilled by the wish to see themselves united in praise of God and happy to step over the holy threshold. Everyone is happy to sing the holy songs and stretch out the hands in order to receive through the sacrament of God the medicine of the chrism. Cresconius was born to serve the laws themselves and the altars, and he was anointed with honours in the Catholic church. He was the protector of modesty, he devoted himself to charity and peace, and through his teaching the countless people of Cuicul are flourishing. Friend of the poor people, he

45

46

The Absence of Christian Iconography

dedicated himself to every charity and never denied this heavenly task. His soul dwells serenely while his body rests in peace, expecting the coming resurrection, crowned by Christ, so that he will become a fellow of the saints in the homestead of the heavenly kingdom.55

Based on the inscription, Christern proposed that the passageway housed the translated remains of Cuicul’s bishops, who were venerated like martyrs, and thereby argued for Cuicul’s role as an important pilgrimage site.56 The complex comprised not only a substantial ecclesiastical infrastructure of two churches, a baptistery, and a chapel but also several facilities required for maintaining the daily functions of communal life, for example, the large latrines in the north-east of the district, the small bath adjacent to the baptistery, and the southern quarters eastwards and westwards of the alleyway (Figure 1.3). Various functions have been ascribed to these quarters, including living quarters for humans (in the case of the south-western parcel, the bishop’s residence), housing for animals, or administrative buildings.57 Troughs, used for all sorts of communal collection and distribution of goods, often in relation to the charitable work of North African churches, were located in the south-eastern quarters.58 The question of the nature of the veneration of bishops in Cuicul is complicated, however, by the fact that the Cresconius inscription is in large parts a copy and compilation of two inscriptions in the Alexander church

55

56 57

58

Pflaum and Dupuis 2003, 873, no. 8299: ‘Hic ubi tam claris laudantur moenia tectis,/ culmina quod nitent sanctaque altaria cernis,/ non opus est procerum, sed tanti gloria facti/ Cresconi rectoris ovat cum saecula nomen,/ quibus honorificos eum ostendente labores/ iustos in pulcra sede gaudent locasse priores/ quos diuturna quies fallebat posse videri,/ nunc luce profulgent subnixi altare decoro,/ collectamque suam gaudent florere coronam,/ animo quod sollers implevit custos honestus./ undique se visendi studio Cristiana decurrit/ aetas in unam congeriem deo dicere laudes/ liminaque sancta pedibus contingere laeta;/ omnis sacra canens manus porrigere gaudet/ sacramento dei medicinam sumere c[r]ismae./ Cresconius legibus ipsis et altaribus natus, honoribusque in ecclesia catolica unctus,/ castitatis custos, caritatis pacique dicatus,/ cuius doctrina floret innumera pleps Cuiculitana./ Pauperum amator elemosine deditus omni/ cui numquam defuere unde opus celeste fecisset/ huius anima refrigerat corpus in pace quiescit/ resurrectione[m] expectans futuram in Cristo corona/ consors ut fiat sanctis in sede regni celestis.’ I thank Clemens Kruse and Kurt Keller for their kind help with the translation of this text. See also Cagnat 1922, XXVI–XXXII; Albertini 1934, 413–415; Christern 1976, 140–141; Masaro 2014, 188–195; Dennis 2020, 26–29. Christern 1976, 253–254. Similarly Grabar 1946, 448–450. As episcopal palaces have a great degree of variation, the Algerian inventory of church monuments refrains from identifying the south-western quarter as an episcopal palace. Gui, Duval, and Caillet 1992, 92–103. On the identification of the quarter as an episcopal palace, see Müller-Wiener 1989, 695; Duval 1989, 348. For the issue of typology Real 2003, on North Africa esp. 224–228. Christern 1976, 137. On the troughs, see Bockmann 2013, esp. 153.

The Christian Complex of Cuicul

at Tipasa (Figure 1.18) in the province of Mauretania Caesariensis.59 The text is divided into two segments and was located in the nave of the Alexander church, which sits next to one of the two principal Christian cemeteries in Tipasa; one segment describes the relocation of Alexander’s predecessors, and the other is the epitaph of Bishop Alexander himself.60 The two inscriptions have not been dated with certainty but, based on palaeographic analysis, seem to belong to the fourth century. A late fourth- or early fifth-century dating is also supported by other epigraphy in the church.61 Therefore, the Cresconius inscription is either an adaptation of the Alexander inscriptions, or the Cresconius and Alexander inscriptions are both derived from the same shared model. This model would have been adapted to the specific site: in the Cresconius inscription, the name of the bishop and the city were changed, as were a few other phrases.62 The inscriptions and their spatial set-ups in Tipasa and Cuicul indicate that the two sites differed in how the bishops were laid to rest. The starting point of both inscriptions is the translation of the relics of the deceased to

59

Albertini 1934, XXVII–XXIX; Christern 1976, 140–142; Pflaum and Dupuis 2003, 873, no. 8299. On the Alexander church and the Alexander inscription, see Dennis 2020, figs. 1, 2; Ardeleanu 2018, esp. 475–489; Daugherty 2015, 208–231; Gui, Duval, and Caillet 1992, 32–35; CIL 8, Suppl. 3, 20903, 20905; Pachtère 1911, 95–96, n. 397. For a roughly contemporary comparison of the practice of sharing inscriptions for the decoration of churches and baptisteries, see letter 32 by Paulinus of Nola to Sulpicius Severus. In the letter, Paulinus offers Sulpicius the epigrams of the churches of Nola and Fundi to use in the church of Primuliacum. Paulinus of Nola and Skeb 1998, vol. 2, 747–803, esp. 772.

60

CIL 8, Suppl. 3, 20903 : ‘Hic ubi tam claris laudantur moenia tectis/ culmina quod nitent sanctaque altaria cernis/ non opus est procerum set tanti gloria facti/ Alexandri rectoris ovat per saecula nomen,/ cuius honorificos fama ostendente labores / iustos in pulcrham sedem gaudent locasse priores/ quos diuturna quies fallebat posse videri/ nunc luce praefulgent subnixi altare decoro/ collectamque suam gaudent florere coronam/ animo quod sollers implevit custos honestus./ Undiq[ue] visendi studio crhistiana aetas circumfusa venit/ liminaque sancta pedibus contingere laeta;/ omnis sacra canens, sacramento manus porrigere gaudens.’ CIL 8, Suppl. 3, 20905 : ‘Alexander episcopu[s l]egibus ipsis, altaribus natus,/ aetatibus honoribusque in aeclesia catholica functus,/ castitatis custos karitati pacique dicatus,/ cuius doctrina floret innumera plebs Tipasensis,/ pauperum amator, aelemosinae deditus omnis, cui numquam defuere unde opus caeleste fecisset/ huius anima refrigerat, corpus hic in pace quiescit/ resurrectionem expectans futuram de mortuis primam/ consors ut fiat sanctis in possessione regni caelestis.’ Christern 1976, 142, n. 48. Ardeleanu dates Alexander’s restructuring of the church to the late fourth or early fifth century based on an analysis of all the inscriptions found in the Alexander church. Ardeleanu 2018, 479. For a detailed comparison, see Pflaum and Dupuis 2003, 873–874; Albertini 1934, XXVII–XXIX.

61

62

47

48

The Absence of Christian Iconography

Figure 1.18 Plan of the Alexander church of Tipasa. Gui, Duval, and Caillet (1992), vol. 2, pl. XXXVI, n. 1. © Jean-Pierre Caillet.

the church (or a new site within it): ‘[these] who long repose took out of sight permanently are now shining brightly, resting upon a beautiful altar’. It is unclear whether the formulation meant that the episcopal remains had been exhibited on the Eucharistic altar (subnixi altare). In the Alexander church (Figure 1.18), the inscription commemorating the translation of the bishops’ remains is located in front of the sanctuary, which is elevated by 90 cm. Nine sarcophagi are embedded in the sanctuary. The extant fragment of a mosaic inscription covering the tombs reads ‘sanctu[s] [Al]exand

The Christian Complex of Cuicul

[er]’. ‘Altar’ in this case probably referred to the elevated platform marking the sanctuary.63 The bodies, which had possibly first been buried in the nearby cemetery, had been translated to the sanctuary and were commemorated in the inscription. By analogy, we should take care not to take the inscription’s mention of an altar too literally in Cuicul: the bishops’ remains could have been placed underneath altars in the crypts, or the bases could have marked the places where their remains were located in the crypts.64 In fact, one of the Cresconius inscription’s textual adaptations of the Alexander inscription indicates that the burials were not placed close to the inscription. Alexander’s epitaph reads corpus hic in pace quiescit (‘his body rests here in peace’), while Cuicul’s mosaicists omitted the word hic (‘here’). Considering the great care with which the designers of the Cresconius inscription altered the model for their needs (as we will see below), a likely explanation for the omission is that the location of the tombs was intentionally not expressed as being nearby. The Cresconius inscription indicates that Bishop Cresconius wanted to raise Cuicul to a memorial site for his predecessors to venerate them like martyrs.65 Furthermore, the similarity to the inscription at Tipasa suggests that Cresconius acted on a model which had been implemented by at least one other bishop in North Africa. If we grant the description of the tombs’ public veneration more weight than mere poetic trope, we could conclude that the patrons of the newly built gravesites hoped to attract pilgrims on a supra-regional level. In any case, it is plausible that the infrastructure of Cuicul’s Christian district would have allowed for a regular, if perhaps only small, inflow of visitors. Was the baptistery conceived of as an integral part of the episcopal memorial site? The monolithic arrangement of the Christian complex, well-suited for a pilgrimage site, suggests that this might be the case. The prevalent view is that the north and south churches were built contemporaneously: the consistent building technique seen in the foundations of both churches as well as the subterranean passageway, which does not have any structural alterations, do not allow for any other conclusion.66 Cuicul is,

63 64

65

66

Christern 1976, 142–143. The term altare subnixi has led to several speculations. For a translation as ‘upon the altar’, see Christern 1976, 142, n. 51. For a translation as ‘underneath the altar’, see Brandenburg 1995, 89; Frend 1997, 126–127. On the interpretation of ‘collectamque suam gaudent florere coronam’ as a reference to the crowns of martyrdom, see Brandenburg 1995, 89; Frend 1997, 127. Février 1996, 667; Février 1966, 88–89; Christern 1976, 143–144; Chastagnol and Duval 1974, 88–94.

49

50

The Absence of Christian Iconography

therefore, the only double church in North Africa for which a uniform building phase has been made plausible.67 We can further assume that the baptistery, which can be entered directly from the north church, was constructed during the same building phase.68 Beyond the arrangement of the complex, the careful wording of the Cresconius inscription also suggests that baptism played an elevated role at Cuicul. Let us begin with the wording of the Alexander inscription. The segment about the translation of the bishops’ bodies describes how the sacrament is administered to the faithful: ‘Everyone is happy to sing the holy songs and stretch out the hands in order to receive the sacrament’ (omnis sacra canens, sacramento manus porrigere gaudens). This passage, in all likelihood, refers to the Eucharist. The believers are congregated in the church and receive the sacrament with outstretched hands. The description applies best to that moment of the communion when the host is received.69 The Cresconius inscription has two more verses than the Alexander inscription; they are in parts clumsily constructed and can convincingly be interpreted as being interventions written for the specific situation at Cuicul.70 One of them adds to the passage above and arguably changes its meaning: ‘Everyone is happy to sing the holy songs and stretch out the hands in order to receive through the sacrament of God the medicine of the chrism’ (omnis sacra canens, manus porrigere gaudet/ sacramento, Dei medicinam sumere c[r]ismae). The mention of chrism as a component of the sacrament is astonishing.71 Anointment with chrism

67

68

69

70

71

Comparable compounds are located in the Church of Bellator and in the Basilica of St Vitalis in Sufetula, Bulla Regia, and Sabratha III-IV: Duval 1989, 386. Older scholarship assumed a gap of time between the construction of north and south church because of stylistic differences of the churches’ mosaics. The baptistery was considered to have been built at the same time as one of the two churches. See for instance Albertini 1934, 415. Pflaum and Dupuis 2003, 874. On the sacrament of the Eucharist in North Africa in Augustine’s time, see Burns Jr. and Jensen 2014, 261–287. Pflaum and Dupuis 2003, 874 suggests the grammatically correct reading of ut per sacramentum Dei medicinam sumant chrismatis and points to the fact that the addition of this verse resulted in the omission of the Virgilian imitation circumfuga venit, still attested in the Alexander inscription, and a likely constituent of the original model. On references to classical literature in the Cresconius inscription, see Masaro 2014, 190–193; Dennis 2020, 29. Albertini’s and Monceaux’s initial reading wrongly reconstructs the defect word as ‘c [h]isma’. Albertini 1922, 30–31; Monceaux 1922, 397, followed by Daugherty 2015, 235. Daugherty leaves unclear what exactly a ‘medicine of schism through the sacrament of God’ would consist of. Meanwhile, it has been established that the correct reconstruction is ‘c[r]isma’, as the narrow lacuna does not allow for the addition of the letter ‘H’ (Figure 1.17). Février 1966, 90; Pflaum and Dupuis 2003, 873–874; Dennis 2020, n. 23. The late antique shift in gender from neuter to female and in declension type originally chrisma, -atis from the first to the third

The Christian Complex of Cuicul

plays no liturgical role in the sacrament of the Eucharist; with few exceptions it is confined to baptismal use only in the early church.72 Previous translations have preferred less literal translations of ‘chrism’, for example ‘the sign of Christ’.73 These have the disadvantage of reducing the textual adaptation to embellishment – a wordy addition which obscures the meaning so clearly graspable in the Alexander inscription. It is difficult to imagine why the designers of the Cresconius inscription would have gone to the trouble of altering the pre-existent model if there was no additional meaning intended. Instead, we should assume that the intervention purposefully changed the meaning of the inscription. In Augustinian North Africa, the post-immersion anointment with chrism is essential to the sacrament of baptism, because only this rite, combined with the bishop’s laying on of hands, conferred the Holy Spirit.74 The Cresconius inscription most likely celebrated (or expressed the hope of establishing) the baptismal ceremonies, which were held in conjunction with the pilgrimage to Cuicul.75 Another alteration in the Cresconius inscription occurs in the following line: the omission of the letter ‘f’ in the verse honoribusque in ecclesia catolica [f]unctus (Figure 1.17). Cuicul’s deviation of honoribus unctus from the more common expression honoribus functus, attested in Tipasa, has so far been held to be a mistake. The verse has been mostly translated along the lines of ‘he died [functus] with honours in the Catholic church’.76 Taken literally, however, the Cresconius inscription reads, ‘he was anointed [unctus] with honours in the Catholic church’. Given that the inscription appreciates the anointment with chrism, and indirectly the sacrament of baptism, the ‘f’ was likely omitted deliberately. The reformulation

72

73

74 75

76

declension is well attested for chrisma and a series of other Greek neutral loanwords, for example, dogma. Stotz 1998, 25–26. I thank Kurt Keller for bringing this to my attention. Rouwhorst 2019, 347, 360–362. Beyond its baptismal use, chrism is also attested by a few sources to have been used in the early church for the anointing of the dead as well as for the anointing of altars. Rouwhorst 2019, 367. See further Hofmeister 1948, 19–20, 132–136. Christern 1976, 141 has ‘durch Gottes Sakrament und das Zeichen Christi des Heiles teilhaftig zu werden’; Masaro 2014, 190 suggests ‘protendere le mani per la santa Eucaristia (sacramento), di assumere la medicina dello spirito di Dio’. See, in contrast, Nathan Dennis’s translation: ‘to receive by the sacrament of God the medicine of chrism’. Dennis 2020, 27. Burns Jr. and Jensen 2014, 202; Rouwhorst 2019, 351–353, 360–362; Ferguson 2009, 786. Instead of Cresconius’ predecessors, Jensen suggests that several local saints who are however not identified with certainty were the target of Cuicul’s pilgrimage. For baptismal practices at pilgrimage sites, with a particular eye on North Africa, see Jensen 2011a, esp. 1682. Cf. also Kötting 1980, esp. 120, 192, 256, 318; Castelfranchi 1995. Cf. Monceaux 1922, 397; Pflaum and Dupuis 2003, 873; Masaro 2014, 190. In contrast Daugherty 2015, 236.

51

52

The Absence of Christian Iconography

emphasizes the value of baptism for the adoption of the correct (Nicene) faith, exemplified by the exemplary baptism of the bishop himself. In summary, Bishop Cresconius established a memorial site for his predecessors in the Cuicul complex. The complex was exemplary in its cohesive layout, and exceptional in the planned construction of a double church. The insinuation that pilgrims would visit the site, implied in the Cresconius inscription, gives a hint as to the role Cresconius imagined Cuicul playing within and beyond the region. Both the inscription within and the elaborate, spacious, free-standing baptistery itself emphasize the importance of Cuicul as a baptismal centre.

1.3

Dating the Christian Complex of Cuicul: A Nicene Memorial Site

The Cresconius inscription is also central to the debate about the dating of the Christian complex. Two bishops of Cuicul named Cresconius have been identified: the first Cresconius took part in the Conference of Carthage of 411, which tried to overcome the Donatist schism in North Africa; the second Cresconius (or Crescens, depending on the reading) attended the Council of Constantinople in 553.77 The majority opinion on the dating of the Cresconius inscription is that it refers to the first Cresconius, attested to in 411.78 If this is correct, then the inscription would have originated in the first half of the fifth century – but certainly after 411, since Cresconius had died by the time the inscription was made. The existence of the sixth-century Bishop Cresconius, however, makes it possible that the inscription was produced after the Byzantine reconquest of North Africa (533–534).79 The dating of the inscription therefore has considerable implications for our understanding of the Christian complex. The double church of Cuicul was part of a larger group of ecclesiastical and profane sites built by the same workshops, most of them near Theveste (Tebessa) in Numidia.80 Cuicul is the most distant of the sites attributed to the Tebessa group but displays many of the group’s characteristic features in terms of architecture and decoration. The south church features a typical combination of columns and piers, and a piece of latticework, also found in the south church, conforms both in style and motif to the architectural 77 78 79

Février 1966, 89. For a summary of the arguments, see Masaro 2014, 188–195, esp. 193–195. 80 Février 1966, 89–92. Christern 1976, 156, fig 32.

Dating the Christian Complex

Figure 1.19 Cuicul, latticework from the south church. Photography by the author.

sculpture of the Tebessa group (Figure 1.19).81 The Tebessa group has been dated from the last years of the fourth century to the first decades of the fifth.82 Doubts about the validity of the dating have been raised by some scholars, but no alternative suggestion has been widely accepted.83 Stylistic analysis has made a sixth-century date for the architectural sculpture of the Tebessa group a viable hypothesis.84 The closest parallels with the group’s work can be found in pre-Justinianic Constantinople, more specifically in Hagia Sophia and the Church of St Polyeuctos. Christiane Strube has argued that Byzantine workshops influenced the local builders of the Tebessa group in the sixth century, but whether this happened before or after the Justinianic conquest of North Africa remains uncertain.85 In Cuicul, only one example of latticework (Figure 1.19) can be attributed to the workshop of the Tebessa group: a remaining fragment 81 82

83 85

Christern 1976, 153–154, pl. 62 and 155, fig. 30. It must be pointed out, however, that the Cresconius inscription belongs to the evidence used to justify this dating. Christern 1976, 215–225. 84 For diverging opinions, see Baratte 2018, 193; Strube 1996. Strube 1996. Strube 1996, 452–453.

53

54

The Absence of Christian Iconography

shows a laurel wreath in the centre of which sits a cross with arms ending in petals and leaves. The combination of vegetabilized objects with nonvegetal ornamentation is also characteristic of early sixth-century Constantinopolitan stonemasonry.86 However, the piece may well have been a later addition from a refurbishment of the south church. As discussed above, the north and south churches were built contemporaneously. Nevertheless, stylistic differences have led some scholars to assume that the north church preceded the south church.87 Both churches were once covered with floor mosaics showing geometrical and figurative decorations (Figures 1.20a and 1.20b). The less sophisticated execution of the mosaics of the south church, notably the lack of three-dimensionality in the depiction of animals, has been interpreted as a sign of later construction.88 But our knowledge about the periodization of North African late antique mosaics is not sufficient to rely on stylistic differences only. We should not exclude the possibility of several workshops working side by side, all the more since the effort seems to have been made to align various aspects of the design throughout the complex. The eye-catching colour scheme of the inscriptions (either white letters on red background or vice versa) is consistently applied in both churches and the baptistery.

Figure 1.20a Detail of the mosaic pavement of the south church of Cuicul, Museum of Djémila. Leschi (1950), 98, fig. 58. 86 88

87 Strube 1996, 433–438. See n. 179 in this chapter. Pachtère 1911, 70; Monceaux 1922, 388–389 and 394–398; Monceaux 1923, 111. A more critical response to the older scholarship can be found in Février 1966, 87–89; Christern 1976, 140–142.

Dating the Christian Complex

Figure 1.20b Detail of the mosaic pavement of the south church of Cuicul, Museum of Djémila. Leschi (1950), 82, fig. 49.

Moreover, several elements of the two churches’ mosaics point independently from each other to a shared construction date in the late fourth or first half of the fifth century. In the north church, seven votive inscriptions have been preserved (Figure 1.21).89 They demonstrate that Cuicul’s Christian buildings profited from private euergetism, partially by the same families. The donors mentioned are all part of Cuicul’s upper class. Four were viri clarissimi (formally members of the ordo senatorius) and fulfilled different official roles, probably for the province of Numidia.90 Their inscriptions are located in the northern aisle, separated from the inscriptions of three donors of lower ranks that are located in the central nave.91 These comprise two viri honesti (municipal magistrates), and a former sacerdos provinciae. The donors’ insistence on a variety of official titles shows that Roman administration was still functional and acts as evidence of a wealthy elite who supported municipal building projects. Euergetism of this scale conforms to private beneficence in Cuicul at large, and North African cities more generally, in the late fourth and early fifth 89

90

91

Chastagnol and Duval 1974. On the inscriptions, see Ravoisié 1846, pl. 52 and pl. 53; Monceaux 1923, 103–104, a selection in CIL VIII, 8344–8348. See also Yasin 2009, 118–119. These are two tribunes, a princeps, and an ex-princeps. Since Cuicul was located on the border of Numidia and the neighbouring province Mauretania Sitifensis, Cuicul’s provincial affiliation is not entirely clear. Chastagnol and Duval 1974, 91. See also Février 1967. Février 1967, 210; on the tensions between these groups, see Lepelley 1979–1981, vol. 2, 412–413.

55

56

The Absence of Christian Iconography

Figure 1.21 Details of the mosaic pavement of the north church of Cuicul. Ravoisié (1846), pl. 53.

centuries.92 A similar density of officials commemorated as public benefactors, however, is unprecedented in sixth-century North Africa.93 Moreover, the families attested in the inscriptions are not traceable beyond the middle of the fifth century. The family that can be traced back the longest is the family of the Pomponii or Ponponii.94 The vir honestissimus Ponponius Rusticus, benefactor of the north church, is arguably a relative of Pomponius Pudentianus, flamen perpetuus, who was amongst the donors of Cuicul’s civic basilica, which replaced the Temple of Frugifer between 364 and 367.95 An epitaph in an extra-urban church east of the city documents the clarissima femina Pomponia Rusticula as having lived from 436 to 452.96 92 93

94 96

On the donation of civic monuments in Cuicul, see Lepelley 1979–1981, vol. 2, 404–408. On the development of benefactions, see the overviews of urban development in North Africa in Lepelley 1979–1981, vol. 1, 298–318; Sears 2007; Leone 2013; Baratte 2018. 95 Février 1966, 88; CIL 8, 8347. Février 1967, 210. Février 1967, 207–214; AE, année 1967, no. 595. On the church where the epitaph was found, see Allais 1967. Février’s suggestion that Ponponia was the daughter of Ponponius Rusticus has

Dating the Christian Complex

While the execution of the mosaics of the south church is inferior to those of the north church, the mosaics of the south church (Figure 1.17) have strong similarities to geometric patterns in two mosaics from Sitifis, dated to the late fourth or early fifth century.97 Moreover, the content and language of the Cresconius inscription specifically suggest that it was composed at a time of conflict between members of the Nicene church, who called themselves Catholics, and dissidents, usually referred to as Donatists.98 The conflict had its beginnings in the early fourth century when, in the aftermath of the Christian persecution by the tetrarchy, the question of how to deal with renegade priests had alienated hardliners from those who adopted an integrative approach. Although the dissidents, many of whom were in Numidia, held the same views on most aspects of doctrine, liturgy, and ethics as the Catholics, the divide about the value of Christian discipline in the face of persecution continued to cost lives for over a century.99 One of the few significant doctrinal changes in the dissident church concerned the question of who was allowed to administer the sacrament of baptism.100 To the dissidents, baptisms administered by apostates and their successors were invalid. Catholics who wanted to enter the dissident church were hence rebaptized – a change of doctrine which was unacceptable to the Nicene church.101 Since the positions of dissidents and Catholics were the same in most other respects, sectarian rhetoric became all the more important for the foundation and consolidation of the two Christian communities.102 This use of sectarian rhetoric can be studied in

97

98

99 100 101 102

been largely supported: Chastagnol and Duval 1974, 91; AE, 1967, no. 595; PLRE, vol. 2, Pomponia Rusticula; PCBE, vol. 1, 1016, Ponponius Rusticus. A combination of two compositions of differently coloured triangles resulting in a quadrangle and a cross-pattern were found both in Cuicul’s south church and in Sitifis. Christern 1976, 144, n. 55. Elizabeth Fentress maintains that the works were done by the same workshop. Fentress 1988, 327. On the epitaph in Basilica B, see Février 1965, 86, fig. 87, fig. 136. The epitaph lay next to an epitaph dated to 410 and was most probably part of a large group of epitaphs made in the church between 389 and 410. Février 1965, 40. The small bath in the north-eastern quarter of Sitifis and the mosaic depicting the triumph of Venus have been dated to the late fourth or early fifth century based on stylistic and stratigraphic analyses. Février, Gaspary, and Guéry 1970, 50–59. At the Conference of Carthage 411, Donatist bishops rejected the label ‘Donatist’ but referred to themselves as Christians rather than Catholics: Shaw 1992, esp. 21–22. See also Augustine, Contra ep. Fund. 4,5–5,6. On the difficulties of differentiating between Catholic and Donatist churches on an archaeological basis, see Sears 2007, 34–37; Leone 2016. Frend 1952. More recent contributions include Tilley 1997; Shaw 2004, 2011; Miles 2016. Burns Jr. and Jensen 2014, 195–201; Edwards 2016, 112–115; Viciano i Vives 2017, 200–202. Gaumer 2016, 103–146. For an overview Ferguson 2009, 795–803. Shaw 2011, 409–489.

57

58

The Absence of Christian Iconography

the Cresconius inscription. The inscription was based on a textual model which was widely disseminated in North Africa and was adapted to the specific situations of various sites. We have already seen that two of the modifications in the Cresconius inscription alluded to the ‘Catholic’ rite of baptism. They express that the sacrament was to crown a pilgrim’s journey to Cuicul and stress that the venerated Bishop Cresconius himself had been baptized in the correct Catholic manner. However, in both Tipasa and Cuicul, the installation of the inscriptions was first and foremost connected to a translation and renewed veneration of the episcopal predecessors of Alexander and Cresconius respectively. One verse in particular distinguishes the Cresconius inscription from the version of Tipasa. The verse describes the visitors to the bishops’ tombs as being ‘united in praise of God’ (in unam congeriem deo dicere laudes). According to Augustine, the formulation Deo laus was a notorious catchword of the dissidents.103 Here, it was assimilated to suit the purposes of the adverse party, who placed it ostentatiously with in unam congeriem. The phrase can be read as a claim about the unique legitimacy of Cuicul’s church community: only a united (i.e. Catholic) congregation is capable of praising God.104 This adaptation shows that the patron of Cuicul’s inscription aimed for a greater rhetoric escalation than the patron in Tipasa did, but the impetus leading to the inscriptions was arguably a shared one: Bishop Cresconius reported at the Conference of Carthage in 411 that the dissident bishop of Cuicul had died.105 Given that at least one deceased dissident bishop is attested for Cuicul, one wonders where his tomb was located at the time of the translation of the predecessors of Cresconius? The inscriptions of Cresconius and Alexander agree that the translation targeted their ‘rightful predecessors’ (iusti priores), that is, only the Catholics amongst the bishops. From the renewed veneration of Cuicul’s Catholic bishops suffered the memory of the neglected dissident bishop. In Tipasa, for that matter, the Nicene church was predominant, but even there a small dissident community is attested.106 Whoever wrote the model of the inscriptions had intended to kill two birds with one stone. On the one hand, they had hoped to increase the appreciation of Catholic leaders and their status as role models and sources of a shared sense of identity 103

104

105

Augustine, ep. 108,5,14. On the validity of Augustine’s assessment, see Sears 2007, 35. See also Frend 1952, 307–308; Masaro 2014, 193. Albertini 1934, XXX–XXXI; Pflaum and Dupuis 2003, 874. A similar formulation is attested in a basilica near Ksar el Kelb that contained the memoria of Marculus, in which some wish to see the dissident martyr Marculus. It reads: Deo Laudes Hic Omnes Dicamus. Cayrel 1934, 131–132. A recent discussion of dissident memoriae can be found in Leone 2016. 106 Lancel 1974, 12,1. Sears 2016, 135.

Dating the Christian Complex

among the Nicene community. On the other hand, the text supported the practice of consolidating episcopal remains which had previously lain scattered – or worse, were buried in confusing proximity to dissidents’ burial places – at one and the same cemetery. Urban space was divided into Catholic and marginalized dissident places, and segregated zones were superimposed onto the entangled structures of urban life. The fact that two versions of the inscription were found at such a distance from each other only underlines the fact that Nicene leaders shared strategies to intensify the sense of belonging of their respective communities. That the Cresconius inscription was installed during such a time of intensified conflict makes sense. Although dissident communities were still present under Vandal and Byzantine rulership, the controversy certainly reached its peak in the early fifth century.107 After the Vandal occupation of Carthage in 439 and the progressive establishment of Arian Christian communities in North Africa, the Nicene church shifted its attention to the new, forceful enemy. Anti-dissident rhetoric and narrative decreased considerably from this point onwards.108 Taking all the evidence together, it is most plausible that Cuicul’s double church and baptistery were constructed in the late fourth or first half of the fifth century – most likely before the Vandal invasion of the early 430s but potentially also during the period of Roman recapture (442–c. 455). To what extent the Vandals were able to enforce the Vandal anti-Nicene policy beyond Carthage is a matter of dispute.109 A Catholic building project of the scale and importance of Cuicul seems unlikely to have been initiated under Vandal rule. What seems certain is that, even if the complex was not constructed as a memorial site, after 411 it did gain a memorial function which served to deepen Catholic communal identity on a supraregional level. This means that baptismal services at the baptistery of Cuicul were decidedly charged. The baptistery was not merely a place to become a faithful Christian; it was a place where Christians would be initiated into the rightful Catholic belief under the protection of the venerated episcopal defenders of the Nicene Creed. The second half of this chapter revisits the design and layout of the baptismal complex in light of this ecclesio-political significance. We have already established that specifically Christian iconography was scarce if not absent in a large part of the decoration. How does this square with the interests of the Nicene patrons and church community?

107 109

Conant 2016; Arbi Nsiri 2015; Frend 1952, 305–306. Bockmann 2013, 130–131.

108

Conant 2016, 350.

59

60

The Absence of Christian Iconography

1.4

On Allegorical Interpretations and Visual Memories: The Floor Mosaic of the Baptistery

Depictions of sea life were popular in North African churches and baptismal spaces. They are commonly held to be allegories representing divine creation, paradise, and baptismal water.110 The absence of unambiguously Christian imagery has therefore not been considered problematic. For Cuicul specifically, it has been suggested that the (reconstructed) floor of the font depicted fish as symbols of Christian souls and the surrounding sea as baptismal water.111 This interpretation is indeed perfectly viable within the logic of allegory. In North African Christian writings, water was often taken to represent the ‘living water’ which nourished the fish living in it or to symbolize the worldliness from which the fish needed to be freed. By the early third century, Tertullian (c. AD 150/160–c. 220) had coined the image of Christian souls swimming like fish in salvific baptismal water: ‘We little fish, after the image of our Ichthys Jesus Christ, are born in the water, nor otherwise than swimming in the water are we safe.’112 He was followed by Cyprian of Carthage (c. 200/210–258), who also represented water as being salvific baptismal water.113 Augustine, by contrast, evoked water as an allegory for sinfulness: ‘But if you join the initial letters of these five Greek words, Ιησοῦς Χριστὸς Θεοῦ υἱὸς σωτήρ, which mean, “Jesus Christ the Son of God, the Saviour”, they will make the word ἰχθύς, that is “fish”, in which word Christ is mystically understood, because He was able to live, that is, to exist, without sin in the abyss of this mortality as in the depth of waters.’114 However, as Henry Maguire has shown, allegorical interpretations open up as many problems as they solve.115 In an attempt to systematize the 110

111

112 113 114

115

Especially in the sixth century, aquatic scenery grew in popularity on mosaic church floors, including on epitaphs. For a selection of examples, see Ghalia 1998, 44–48. On the interesting example at El-Ouara, see also Baratte and Bejaoui 2001, 1491. Concise overviews of the most frequent interpretative models are offered in Brandenburg 1983; Bisconti 2000, 252–258. I have not been able to consult Smati 2001. For the earliest attested example (c. 315), with an overview of further literature, see Lehmann 2010. Drewer 1981, esp. 545. On the roughly contemporary aquatic mosaic floor in the front part of the baptistery of the Great Basilica of Tipasa, see Dennis 2020, 29, and 25, fig. 9. Tertullian, De Baptismo 1. Cyprian, ep. 63.8.1-9.1. See also Burns Jr. and Jensen 2014, 184. Augustine, De civ. Dei 18,23. Here Augustine follows a tradition testified in Clement of Alexandria, Paidagogos 3:11: ‘. . . Jesus, Saviour of the Human race, Shepherd, Husbandman, Helm, Bridle, Heavenly Wing of the allholy flock, Fisher of men who are saved, catching the chaste fishes with sweet life from the hateful wave of a sea of vice’; Jerome, Hom. 92: ‘. . . by the word of God are lifted out of the abysmal waters of this world like so many fish’; Ambrose, De sacr. 3,3: ‘So even for you this world is a sea. It has diverse floods, heavy waters, severe storms. And do you be a fish, that the water of the world may not submerge you’. Maguire 1987.

Allegorical Interpretations and Visual Memories

allegorical frameworks which early Byzantine depictions of the natural world could have drawn on, Maguire drew attention to a strong tradition in late antique church writing which favoured the literal sense of depictions of nature.116 He focused on the sermons about and commentaries on the Hexameron – the first six days of the world – in order to show how all sorts of positions were taken by the church fathers, from exclusively literal readings to deeply allegorical stances on the meaning of God’s creation. Included in a vast corpus of literal descriptions is the influential commentary on the Hexameron by the bishop of Caesarea Basil the Great (330379).117 The sermons, preached in 378, were soon imitated by Ambrose of Milan (c. 340–397), and later by Augustine.118 When Basil considers the creation of the beasts and plants on the fifth day, he says: Those who cannot accept the scriptures in their common sense, say that water is not water, but some other substance; the words plants and fish they interpret as seems good to them; the creation of the reptiles and wild beasts they explain by twisting [the sense] according to their own suppositions, like interpreters of dreams who give whatever meaning they wish to the phantasies that have appeared during sleep. As for me, when I hear grass spoken of, I think of grass: and so also with plants, fish, wild animals, and domestic animals. I accept everything at face value.119

Basil emphasizes that the creation conveys an idea of its Creator’s greatness. Maguire argues that we should read some of the mosaic church floors of the fifth and sixth century in the same way.120 Basil’s ‘face-value approach’ offers an alternative way of understanding Cuicul’s baptismal floor as nothing more than a celebration of God’s creation. More importantly, it exposes the structural weakness of purely allegorical readings, which presume that something other than what is depicted should be expressed. The problem with allegory is that it is impossible to disprove it. The meaning of the depiction depends on the meaning which the allegory ascribes to it, and the visual evidence becomes secondary to the interpretation.121 As a consequence, discrepancies between a depiction and its ascribed allegorical meaning tend to fall off the radar. Starfish, water fly,

116 118 119 121

117 Maguire 1987, 5–40. Maguire 1987, 31–32. Maguire 1987, 31. On the sermon’s influence, see Giet 1949, 70–71. 120 Basil, Hom. In hex. 9,1, transl. follows Maguire 1987, 31–32. Maguire 1987, 32–40. On this problem, see Raeck 1992, 9. Maguire does not fully succeed in escaping this problem when he writes that the text ‘can reveal the thought processes that lie behind the work of art, even if the text itself was not known either directly or indirectly to the artist’. Maguire 1987, 1.

61

62

The Absence of Christian Iconography

and crustaceans are clearly ill-suited to represent ‘Christians swimming in the living water’. There is nothing about these animals which would draw the viewers’ attention to the baptismal significance of water, either. When challenged in this way, the allegorizer must withdraw to the most generic interpretation possible: a depiction of the wealth of creation, be it worldly or paradisiacal. Such an interpretation of the floor mosaic perpetuates pre-Christian interpretations of such scenes as images of plenty and a happy life.122 An additional weakness of allegorical readings is that the interpretation privileges elite clerical voices only, as it is only their allegorical literature which is accessible to us. A fixation on a depiction’s allegorical sense considers viewpoints of the ecclesiastical elite as normative for a lay understanding of Christian art. Finally, the allegorical approach entails a selection bias: artworks which allow symbolic interpretation are often preferred as objects of study.123 The study of Christian art thus becomes an illustrated confirmation of the knowledge scholars have already generated from the established textual canon. The consequence is a selfperpetuating fixation on the authoritative voices of the church fathers. To be clear, I am not denying Cuicul’s baptismal mosaic an allegorical Christian meaning.124 Arguably, ideas about the wealth of God’s creation and promises of paradise would have crossed the minds of late antique viewers. But there were other associations, too, like visual memories. Aquatic scenery which depicted diverse species in a naturalistic manner had exceptionally high recognition value in late Roman North Africa. Depictions of marine life, whether as its own subject matter or in combination with a wide range of other themes – including fishing scenes as well as depictions of sea deities, sea monsters, Venus, Europa, or Ulysses – commonly embellished Roman villas and fountains, pools, and baths, and outlived many other subjects.125 Cuicul’s preserved mosaics are a good example of this: the marine subject matter exceeds all other figurative themes which were depicted in profane settings.126 Depictions of sea life roughly contemporary to those of the baptistery were displayed in many of Cuicul’s most wealthy premises: the maison de l’âne (Figure 1.22), the

122 124

125

126

123 Brandenburg 1983, 252–255. See for example Maguire 1987, 2. Regarding the inventiveness with which Christians did reinterpret classical art, see most recently Jacobs 2020. Dunbabin 1978, 125–130; Brandenburg 1983; Yacoub 1995, 149–168; Blanchard-Lemée 1996; Smati 2001; Ghalia 2006. Cf. Blanchard-Lemée 1975.

Allegorical Interpretations and Visual Memories

63

Figure 1.22 Detail of the mosaic of the toilet of Venus of the maison de l’âne of Cuicul, Museum of Djémila. Photography by the author.

maison d’Amphitrite (Figure 1.23), and the maison d’Europe.127 The sculpture of a dolphin, still in situ, frames the entrance to the latrines of the Great Bath of Cuicul (Figure 1.24), and is a reminder of the great applicability and omnipresence of such stock imagery also in civic contexts.128 The visual culture of Cuicul’s baptistery is of the very same sort as that traditionally used in Cuicul’s elite private households and public institutions. We need to include this frame of reference if we want to achieve a plausible reconstruction of what visitors to the baptistery would have experienced. Observing the similarity between the baptistery’s mosaic floor 127

128

Blanchard-Lemée 1975, for ‘The toilet of Venus’ 61–84, for ‘Amphitrite’ 106–119, pl. XXIX, for ‘Fish and wild animals’ 121–124, for ‘The abduction of Europe’ 143–147, pl. XXXV, fig. a. For an early example of marine scenery in public baths, see for instance the mosaic of Oceanus from the Baths of Themetra (Chot Meriam), beginning of the third century, today at the Musée Archéologique de Sousse. Yacoub 1995, 231, fig. 117a shows a complete montage of the fragmented pieces of the mosaic.

64

The Absence of Christian Iconography

Figure 1.23 Mosaic from the centre of the peristyle of the maison d’Amphitrite of Cuicul, Museum of Djémila. Blanchard-Lemée (1975), pl. XXXI, fig. b.

Figure 1.24 Cuicul, Great Bath, sculpted dolphin at the entrance to the latrines. Photography by the author.

Christian Attitudes towards Roman Cult and Culture

and the prevalent visual culture of Cuicul and many other Roman cities would have had an impact on the onlookers. It would have given them an impression of the Nicene church’s openness towards Roman culture. The baptismal liturgy made the neophytes break with their old lives to immerse themselves fully in their new roles as fideles. Nevertheless, the principal artistic theme shaping the rite in Cuicul does not evoke rupture with the habitual, nor a wholly new or transformative idea or atmosphere.

1.5

Christian Attitudes towards Roman Cult and Culture at the Turn of the Fifth Century in North Africa

Christians’ attachment to Roman culture in North Africa can be observed in further instances both within and outside the context of baptismal complexes. Before taking a closer look, we should recollect the historical situation in which this happened. The turn of the fifth century was a time of great religious transformation in North Africa, which is well expressed in the hopeful statement of the inscription of Cuicul’s font, ‘the time will come when all [will be reborn] in the fountain’. Paganism certainly continued to exist, even though the precise nature of Roman religion after its formal prohibition in 391 by a decree of Theodosius I is notoriously hard to determine.129 Secret worship and domestic rituals are only rarely traceable, and, on top of this, relevant archaeological evidence often remains unstudied. Indirect evidence for paganism, however, can be found in Christian expressions of concern about pagan practices, in North Africa most prominently by Augustine.130 The Dolbeau sermon 26 (from the series of Augustinian sermons discovered by François Dolbeau in 1990) serves as an apt example.131 Augustine mentions the secret continuation of initiatory rites into pagan cults in communities only recently Christianized, as well as his concern about Christians attending spectacles and magical and astrological sessions.132 In particular, the Tractatus contra paganos, given on the first day of January sometime between 400 and 404, gives insight into Augustine’s worries about the continuation of pagan worship. 129

130 132

C. Th. 16,10,10 issued on 24 February 391. On the difficulties of interpreting the archaeological evidence of private religious practices in late antiquity, see Sfameni 2014. On domestic religion in the ancient world more generally, see Bodel and Olyan 2008; Zimmermann and Öhler 2017. 131 Markus 1990, 117. For an annotated French translation, see Gilson 1999. Augustine, Dolbeau 26,28. See Lepelley 1998, 398.

65

66

The Absence of Christian Iconography

Exceptionally long and intended to keep listeners away from the Roman New Year celebration, the Tractatus describes Neoplatonic spiritual exercises and the purifying rites of mystery cults as ongoing.133 Augustine also affirmed that, in early fifth-century North Africa, the divide between the Christian lower classes and a pagan aristocracy had not been overcome.134 The strong affiliation of the Roman aristocracy with paganism was problematic for Augustine because its influence on the broader public was significant. Augustine reports a phrase used in his diocese in Hippo: ‘If this noble became Christian, nobody would stay pagan.’135 Imperial and private anti-pagan measures worsened relationships between Christians and pagans and disrupted the public order. A wave of conflicts began when the Theodosian decree was enforced in North Africa from 399 onwards. Gaudentius and Jovius, two delegates sent by Emperor Honorius, destroyed temples and statues in Carthage, providing a model for other Christians in Carthage and elsewhere.136 Christian iconoclasm continued until at least the middle of the fifth century.137 For instance, the destruction of a statue of Venus in a public bath in Carthage was ordered by Bishop Quodvultdeus in 434 because a girl had been possessed by a spirit while contemplating the deity.138 Christianization, meanwhile, proceeded to transform the African landscape.139 However, rural Africa Proconsularis and Byzacena are thought to have been thoroughly marked by Christian infrastructure only after the Byzantine reconquest; Christian traces in Numidia are even more patchy.140 Despite these changes, North Africa’s Roman – and notably also pagan – heritage continued to shape the landscape, provoking different reactions and ways of dealing with it among Christians. Five months after the destruction of temples in Carthage in 399, Honorius ordered that the temples be left intact once they had been emptied of illicit 133 135 136

137

138 139

140

134 Lepelley 1998, 401–406. Augustine, Dolbeau 26,59. Lepelley 1998, 408, n. 42, on Augustine, Ps. 54,13. Riggs 2001, 294. The event is recorded in Augustine, De civ. Dei, 18,54. See further Lepelley 1998, 400, n. 17. An overview of Christian attitudes towards pagan statues in North Africa is provided in Leone 2013, 121–187. Lepelley 1994, 5. Thébert 1983, 115. On the Christianization of North Africa from an archaeological perspective Duval 1989. A comprehensive study of rural Byzantine churches in North Africa is missing to date. A start has been made by the work of Fathi Bejaoui and François Baratte: Baratte and Bejaoui 2001; Baratte et al. 2014. Overviews of fifth- to seventh-century church buildings can be found in Ghalia 2002; Ghalia 2016. On Byzacena specifically, see Duval 1972, 1127–1179. On rural North Africa more generally, see Rummel 2016.

Traces of Roman Religious Practice

things.141 Similar orders reached the provinces, both in Africa and elsewhere, throughout the fifth century. Archaeological evidence suggests that many of the surviving African temples were closed and fell into decay.142 Yet, in 407/408, another law followed that advised the reuse of former temples.143 A canon issued by the provincial church council of Africa Proconsularis in 401 in Carthage took a stance on these developments. Interestingly, the African clerics considered the artistic properties of the temples as a valid category in their decision making for or against the persecution of the pagan heritage.144 The bishops broadly advised the destruction of temples but were ready to make exceptions for artistically outstanding buildings; temples in rural areas and on private property were defined as targets for destruction because they were bare of ornaments. Implicitly, the bishops recognized the aesthetic value of pagan architecture as a reason for its preservation.145 Negotiating the grey area between cult and culture continued to be a challenge for the Christianizing North African society and the wider Mediterranean of the fifth and sixth centuries. In the last section of this chapter, I demonstrate that Cuicul’s Christian community was rooted in traditions originally connected to Roman emperor worship and Roman religious practice. The community made sure that these traditions lived on within the frame of Christian religious practice, both within and beyond the baptismal context.

1.6

Possible Traces of Roman Religious Practice in Cuicul’s Christian Complex

In this section, I discuss two incidents of practices connected to Roman religion and emperor worship in Cuicul’s Christian complex. For the first, we return to the seven donor inscriptions of the north church (Figure 1.21), where a member of the city’s Christian elite positioned himself as a preserver of the Roman administration and the imperial cult (or rather what was left of it). One of the inscriptions attests Tulius Adeodatus Sacerdotalis Votum Comp(levit) – ‘the sacerdotalis Tulius Adeodatus 141 143 144

145

142 C. Th. 16, 10, 18 issued on 20 August 399. Leone 2013, 29; Sears 2007, 229. C. Th. 16,10,19,2 issued on 15 November 408; 407. On the dating, see Pharr 2001, 458, n. 96. Reg. eccl. Carthag. excerpta, canon 58: ‘. . . templa eorum, quae in agris vel locis abditis constituta nullo ornamento sunt, iubeantur omnino destrui’; Munier 1974, 196. See also Lepelley 1994, 6. On the impact of aesthetics specifically on Roman construction law, see Geyer 1993 and on urban planning Jacobs 2013; Dally 2003.

67

68

The Absence of Christian Iconography

fulfilled his vow’.146 The title sacerdotalis does not refer to a priestly office; rather, it is a truncated version of sacerdotalis provinciae, which designates a former sacerdos provinciae, the president of the concilium of the province who was elected on a yearly basis.147 The provincial council held a meeting once a year during the celebrations in honour of Rome and the first emperor, Augustus. The council’s duty was to represent the wishes of the province through decrees and ambassadors and to assess the work of governors and other administrators sent by the court. The presiding sacerdos resided in the capital of the province throughout the year and was charged in particular with preparing the sacrifices, public games, and festival ceremonies.148 The inscription proves that the Christian community considered the office of the sacerdos provinciae to be meaningful and prestigious around or after 400. The inscription is not unique in North Africa, allowing for the possibility that the institution of the provincial council was maintained in the fifth century as well.149 What can be said with certainty is that one of the benefactors of the Catholic community of Cuicul had a central function in a traditional Roman institution invested in organizing state festivals. In the late fourth century, Emperor Julian (360–363) had attempted to make the sacerdos provinciae head of all pagan priests in the province, and it is unlikely that the memory of the priestly duties the office used to perform faded overnight.150 The phenomenon of allowing Christians to hold priestly offices related to the imperial cult is better known with respect to the similarly long-lived flamen perpetuus.151 In the early fourth century, the church council of Elvira in Hispania permitted Christians to take up this office as long as they did not perform sacrifices or profit financially from it. The flamen perpetuus had traditionally carried out sacrifices and ceremonies in honour of

146 147

148 149

150 151

CIL 8, 8348; Pachtère 1911, 70, n. 298. Due to the unstable borders between the two provinces Numidia and Mauretania Sitifensis, it is a subject of speculation whether Tulius Adeodatus presided over the united provinces or was responsible for Numidia only. See Chastagnol and Duval 1974, 91–94. Chastagnol and Duval 1974, 110–111. Lepelley 1979–1981, vol. 1, 362–369. An amendment dated to 455, in the Vandal period, mentions Maximinus sacerdotalis of the united provinces of Numidia and Mauretania Sitifensis. Another late document might be the undated epitaph of a certain Astius Dinaminus sacerdotalis provinciae Africae from Ammaedara in Byzacena. Lepelley 1979–1981, vol. 1; Chastagnol and Duval 1974, 100–113. Chastagnol and Duval 1974, 116. On the continuous use of the title in North Africa, see Lepelley 1979–1981, vol. 1, 362–364.

Traces of Roman Religious Practice

the emperor and his deceased predecessors.152 An inscription of flamen perpetuus Astius Mustellus from Ammaedara in Byzacena still attests this title in 525/526. Moreover, the inscription raises doubts about the purely secular reputation of the office, even at this late point in time, as it calls Astius Mustellus a flamen perpetuus Christianus – the emphasis on his identity as a Christian only appears necessary if the office still entailed pagan overtones.153 My second point concerns a preparatory stage of the baptismal rite itself. Let us think back to the portal leading from the baptismal rotunda directly into the bath (Figures 1.9 and 1.10).154 Why is by far the most prominent entrance into the baptistery orientated towards the bath? The most logical explanation for the particular arrangement seems to be that it enabled the community to perform a washing in the bath annexe on the way to the font. I argue that Cuicul’s Christian community likely saw a need for bodily purification before the ultimate rite of purification, much in line with traditional Roman religious practice. Let us look firstly at the entrance (Figures 1.4 and 1.9). When a baptismal procession entered through the main entrance opposite the north church, the group could best proceed forward straight ahead. The procession would have passed the apsidal doorway leading to the bath and turned right to arrive at the large entrance to the baptistery in the west. The alternative entrances into the baptistery from the vestibules are both narrow and have a steep angle, making them inconvenient for the passage of groups. Rather, they are suitable for individuals helping to administer the ceremony. When the baptizands turned eastwards, they would have looked straight onto the font and the conched apse behind it (Figure 1.7). At the moment of baptism, they entered the font from this position.155 This vista is symbolically charged, as Cuicul’s bishop arguably expected the baptizands on the font’s other end, in front of the conched apse in the east (Figure 1.25). Augustine’s liturgy at Hippo may serve as a point of reference for Cuicul’s unknown baptismal liturgy.156 In Hippo, two rites preceded the first Eucharist, which the newly baptized celebrated together with the community.157 The first was the dressing of the baptizands. Importantly, 152 153 154 155 156 157

Chastagnol and Duval 1974, 106–110. Kahlos 2007, 38, n. 109 and on the inscription and its context, 232; Lepelley 2002, 276–277. On the baptismal bath, see Thébert 2003, 202–203; Maréchal 2020, 349–350. See 38–39 in this chapter. On the variety in liturgical practice in North Africa: Augustine, ep. 55,19. Brons 2017, 99–100; Burns Jr. and Jensen 2014, 209–211.

69

70

The Absence of Christian Iconography

Figure 1.25 Cuicul, conched apse in the east of the baptistery. Photography by the author.

the second was the anointing with chrism consecrated by the bishop and the imposition of hands by the bishop. As we have seen, the Cresconius inscription alludes to this stage in the baptismal ceremony; it is in this stage that the Holy Spirit is conferred. A special room for this purpose, the consignatorium, might have once existed.158 However, another possibility is that the baptizands were anointed at the apse after having left the font. Why, then, was this circulation pattern chosen in Cuicul? It appears that the architects in charge of constructing Cuicul’s baptistery sought to create a vista connecting the ‘well of life’ – potentially the place of the anointing with chrism – and the bath. Cuicul’s is not the only baptistery with an annexed bath in the area of today’s Algeria. The baptisteries of both the Great Basilica at Tipasa (Figure 1.26) and of the so-called Donatist church in Thamugadi (Figures 1.27 and 1.28a–c) are both connected to small-scale baths.159 In Tipasa, the bath is adjacent to the baptistery;160 in Thamugadi and Cuicul the baptistery and the bath are so closely interconnected that

158 159

160

See n. 143 in this chapter. Maréchal 2020, 202 and 209. It is far from certain that the church belonged to the dissidents. A reassessment of the arguments can be found in Rebillard 2015, 433. Thébert 2003, 226–227; Maréchal 2020, 354.

Traces of Roman Religious Practice

Figure 1.26 Plan of the baptistery and bath of the Great Basilica of Tipasa. © Sadi Maréchal.

Figure 1.27 Plan of the west church of Thamugadi (Timgad). Christern (1976), 272, fig. 38a.

71

72

The Absence of Christian Iconography

the baptistery can only be accessed via the bath.161 Baths in church complexes with baptisteries have also been suggested for Hippo, Rusguniae, Bulla Regia, and the Basilica of St Vitalis in Sufetula (Sbeitla).162 The phenomenon of North African baptismal bath annexes has not yet been studied in detail, and both systematic archaeological surveys and interpretations of their usage are missing.163 It is likely, however, that baths had a function in the baptismal rite in Cuicul, Thamugadi, and Tipasa.164 The only extant source for pre-baptismal cleansing in North Africa is again Augustine, who wrote to Januarius (c. 400) on the practice of bathing on the Thursday before Easter: . . . many or almost all persons in many places were in the habit of bathing on that day. . . . If you ask how the custom of bathing arose, no more reasonable explanation occurs to me than that the bodies of those to be baptized had become foul during the observance of Lent, and they would be offensive if they came to the font without bathing on some previous day. This day was especially chosen for it, on which the Lord’s Supper is annually commemorated, and, because it was permitted for those about to be baptized, many others wished to join with them in bathing and relaxing the fast.165

In this text, Augustine describes bathing as a practice common among neophytes when they broke the fast on Maundy Thursday after the period of strict observance called for by Lent. During Lent, the neophytes in Hippo fasted by eating only once a day and abstained from eating meat, drinking wine, engaging in sexual relations, and attending the theatre. Additionally, they did not visit the public baths during this time. The Lenten period was moreover structured by a series of purifications and catechetical lessons like exorcisms and the teaching of the creed and the Lord’s Prayer.166 The baptismal rite itself took place during the Easter vigil on Saturday night.

161 162 163

164 165 166

Thébert 2003, 252–253; Maréchal 2020, 352–354. Duval 1989, 389; Lassus 1970, 251; Lassus 1965, 597–598; Leglay 1957, 706. An overview is offered in Thébert 2003. The baths are mentioned in passing in Morfino 2011, 32, 64. See also Blas de Roblès and Sintès 2003, 116. Two significantly larger bath complexes were part of the Egyptian pilgrimage site of Abu Mena. See Grossmann 1986, 20–23; Frankfurter 1998, 292–293. Given their size, it seems implausible that the baths were predominantly used on Maundy Thursday as Grossmann suggests. For the long history of the ecclesiastical and monastic use of baths, see Stasolla 2002a, 2002b. On the conversion of baths into churches and Christian positions towards bathing culture in the West, see Jiménez Sánchez and Sales Carbonell 2004. Lassus 1970, 251; Blas de Roblès and Sintès 2003, 116. Augustine, ep. 54,7,9-10, transl. Parsons (ed.) 1951, 259. Burns Jr. and Jensen 2014, 205–206.

Traces of Roman Religious Practice

73

Figure 1.28a Thamugadi, baptistery of the west church, c. 400. Médiathèque du patrimoine et de la photographie, PbMMF054133

Figure 1.28b Thamugadi, detail of the font’s mosaic. Photography by the author.

74

The Absence of Christian Iconography

Figure 1.28c Thamugadi, detail of the font’s mosaic

It is noteworthy that Augustine describes the washing on Maundy Thursday as a pragmatic necessity without giving it any significance in relation to the spiritual purification of the body which had taken place in the weeks before. The fact that large crowds attended the bath on an important day in the Christian calendar can also be interpreted in various ways. The archaeological evidence suggests that the Christians of Cuicul, Tipasa, Thamugadi, and possibly other areas placed great importance on the existence of a bath as part of the religious infrastructure.167 Prebaptismal washings like the one on Maundy Thursday could easily have been perceived by the neophytes as another ritual step towards the ultimate cleansing.168

167

168

Though there is little evidence that shared bathing was not the norm in late Roman North Africa, a separate Christian bath would also have allowed for the separation of the sexes. For an assessment of North African bathing practices, see Sears 2007, 86. One of the two bath complexes at Abu Mina is a double bath, potentially built in this way for the same purpose of dividing the sexes: Grossmann 1986, 20–23. See further with a focus on eastern sources Bady and Foschia 2014. An alternative option would have been the post-baptismal washing of the feet, administered by Augustine on Maundy Thursday, against the decision of the council of Elvira. Brons 2017, 91. Augustine’s testimony, however, makes clear that foot washing in close relation with the

Conclusion

Whatever meaning washing held for individual Christian communities (pragmatic washing or ritual cleansing), it seems safe to say that prebaptismal purification made particular sense for a population which was still strongly influenced by Roman religious practices that emphasized the need for purification before the performance of sacred actions. Preliminary purification by washing belonged to the standard practices of much of Roman religious worship – be it in civic temples or in healing or oracular sanctuaries, in mystery religions like Mithras worship, or in magical practices.169 As paradoxical as it seems, the architectural solution found in Cuicul’s baptismal complex is best explained by assuming a prebaptismal washing custom to have preceded the ultimate cleansing of the baptism. The Maundy Thursday washing does not exclude the possibility of another purification directly before the baptismal ceremony. In any case, the layout and creation of sightlines in the baptismal complex would have kept the memory of any pre-baptismal washing alive at the moment of baptism.

1.7

Conclusion

Cuicul’s Christian community was indebted to traditional Roman ways of life and of ritual practice, and they maintained elements of these in their religious lives as Christians. This was certainly true for members of the city’s elite, but it also applied to society at large. Discussing the reasons for using motifs taken from the natural world in church buildings, Maguire states that by ‘[b]orrowing much of the iconography of secular abundance and pleasure, the Christian authorities converted the good life dominated by the late Roman aristocrats into a good life that was controlled by the church’.170 Maguire shows how, despite harsh critiques of domestic luxury

169

170

baptismal ceremony was contested in North Africa. Where it was administered, it happened in the week following the baptism or even later: ‘. . .many, however, have not accepted this [foot washing] as a custom, lest it should be thought to belong to the ordinance of baptism; and some have not hesitated to deny it any place among our ceremonies. Some, however, in order to connect its observance with the more sacred associations of this solemn season, and at the same time to prevent its being confounded with baptism in any way, have selected for this ceremony either the eighth day itself, or that on which the third eighth day occurs, because of the great significance of the number three in many holy mysteries.’ Augustine, ep. 55, 33, transl. Burns Jr. and Jensen 2014, 212. Ferguson 2009, 25–37; Hellholm et al. 2011, 41–154. On the role of water in Roman Mithras worship in particular, see Adrych et al. 2017, 76–78. Maguire 1999, 253.

75

76

The Absence of Christian Iconography

Figure 1.29 Baptismal font of Kelibia, sixth century, Musée National du Bardo, Tunis. Photography by the author.

by clerics, the furnishings of ecclesiastic buildings adapted precisely the models the clerics criticized. The church presented itself as heir to the socio-economic and political power held by the Roman aristocracy and claimed its succession by means of artistic imitation. The baptismal mosaic of Cuicul exemplifies Maguire’s point well, though I suggest going a step further. Cuicul’s aquatic scenery (as indeed much of the natural imagery which Maguire has in mind) was not restricted to use by the aristocracy; rather, it had the character of stock imagery for multi-purpose use in a variety of both domestic and public contexts. The point of reference of the baptismal decoration must be conceived of in broad terms. The baptistery is decorated – and functions – in a distinctively Roman way. The best illustration of how the baptistery functioned in a Roman way is through the potential preliminary washing preceding the rite of baptism – akin to Roman religious worship. It was left to the interpretation of the viewers and the guidance of the clerics to reconcile a pre-baptismal washing ritual with its non-Christian

Conclusion

Figure 1.30 Baptismal font of Bekalta before its restoration, sixth century, Musée Archéologique de Sousse. Photography by the author.

roots, as well as the multi-purpose nature of the baptismal decoration, with Christian interpretations. Evidently, under the shifting conditions of urban life, Roman infrastructure and architecture continued to frame the public life of a society which was in the process of Christianizing. The Conference of Carthage of 411, for instance, took place in the Baths of Gargilius.171 Cuicul’s Christian community was so profoundly Roman that its initiatory space did not need to be differentiated from other Roman institutions by its interior decoration. The wish to become distinguishable through specifically Christian visual messages seems not to have existed. This holds true at a time of intense conflict between the Nicene and the dissident church. The close reading of the Cresconius inscription reveals the efforts made by Cuicul’s Nicene bishopric to downplay the previous presence of a dissident bishop in Cuicul and establish a memorial site and pilgrimage for the city’s Nicene bishops. Furthermore, it has been

171

On the continuous use and refurbishment of baths, see Thébert 2003, 418–421; Rummel 2016, 113; Conant 2015, 53; Sears 2007, 86. Vandal rulers also took much care to restore the baths. See Miles 2005, 110–111.

77

78

The Absence of Christian Iconography

demonstrated that Cuicul’s baptismal service was another practice used to reinforce the exclusive authority of the Catholic church. The inscription explicitly refers to the power of the baptismal rite in fostering a communal sense of (Nicene) identity in Cuicul. In this aspect, the inscription differs from the nearly identical Alexander inscription in Tipasa, which does not make any mention of the baptismal rite. However, there is nothing to suggest that the Nicene community watered down its position towards Roman visual culture in order to create differences to interior designs used in dissident communities. On the contrary, it appears that the Nicene community harnessed the unifying power of Roman art and ritual in the construction of the baptistery. Becoming a faithful ‘Catholic’ in Cuicul meant affirming Roman traditions. Furthermore, during a period of intense conflict between the Nicene and the dissident church of Numidia, the utilization of traditional Roman, nonChristian imagery and the presence of a bath suitable for pre-ritual cleansing in the baptistery might reveal more than a general commitment to traditional forms of Roman worship and culture. These choices could also be interpreted as signs that the church offered the Christian community of Cuicul the opportunity to integrate selected pre-Christian Roman customs into the baptismal rite in order to strengthen that community’s attachment to the church. Though there cannot be certainty on this point, it seems to me to be a plausible hypothesis. Had we stopped investigating the baptismal decoration once we had established its Christian readings according to the bible and the church fathers, or had we not permitted ourselves to wonder about the reasons for Roman reminiscences in Cuicul, for example because Augustine and others claimed them to be irrelevant, we would not have arrived at this new avenue of possibilities. Complementing historical scholarship with a comprehensive analysis of visual and material evidence, surpassing the validation of textual knowledge through tangible sources, has been a prerequisite for this excursion into the lived religion of Cuicul’s Christian community.

2

|

The Use of Non-Christian Imagery in Baptisteries

In this chapter, we shift our attention to non-Christian baptismal iconography, by which I mean iconography, the origins of which lie outside the Christian religion and which does not conform with theological positions on correct Christian conduct.1 The use of such imagery in baptisteries requires us to explore the images’ relevance and function within and beyond a Christian message. The three case studies in this section are drawn from the late antique mosaic tradition which connected North African mosaic workshops with those on the Iberian Peninsula, fostering the transmission of artistic techniques and visual repertoires.2 The chapter starts with the fifth- or sixth-century rural baptistery of Henchir el Koucha in Africa Proconsularis in the northern part of present-day Tunisia, whose decoration alludes to the tradition of the Roman circus. The discussion then moves to the Roman province of Lusitania in what is today southern Portugal, where the assumed baptistery of Myrtilis Iulia (now Mértola) in the Alentejo contains an image of the Graeco-Roman mythological hero Bellerophon. The third case study is the baptistery of the villa rustica at Milreu/Estói in the Algarve, also in the province of Lusitania, which is attached to an earlier Roman building covered with a mosaic frieze showing a marine thiasus scene. While the so-called water temple of Milreu has been the subject of scholarship for over a century, excavations of the other two case studies are recent. The lack of textual sources which could provide insights into the significance of the imagery for a baptismal purpose is itself revealing. Our investigation focuses on a realm which lies on the outskirts of Orthodox Christian tradition yet is still part of the official Christian infrastructure. 1

2

In the course of this chapter, I refrain from calling such imagery ‘pagan’ for the reasons outlined in the Introduction. See 8–12. On the artistic exchange in various crafts between North Africa and the Iberian Peninsula, particularly in Lusitania and Baetica, see Schlunk and Hauschild 1978, 35–73, esp. 38–39, 46–47, 54–55, 57–58, 77. See also Duval 1971, 381–395. For an overview on North Africa’s trade relationship with the Iberian Peninsula, see Wickham 2005, 709–713. On Baetica and Lusitania in particular, see Carr 1999; Quaresma 2010.

79

80

The Use of Non-Christian Imagery

It is within this realm that our typically loquacious Christian sources fall silent. In Chapter 1, I criticized the tendency to cease investigating the purpose of imagery which appears to deviate from Christian values once a ‘Christianized’ meaning has been established.3 The necessity to give due consideration to material evidence and defend its historical value against the dominance of normative Christian writings and the associated historical scholarship becomes even more imperative when the material and textual evidence are in direct conflict. This is particularly evident in the case of the baptistery of Henchir el Koucha, where the visual message conveyed by the baptismal decoration contradicts the formulas for neophytes to express which are prescribed by conventional baptismal liturgy. A noticeable disparity also exists between Christian standards of proper behaviour and the local artistic practices observed at Milreu and Myrtilis Iulia. When these Christian communities chose to (re)employ mythological themes as a source for baptismal decorations, I argue that they simultaneously violated the accepted Christian code of ethical conduct. The chapter will also provide us with an opportunity to discuss the extent to which the study of extant material remains can help us comprehend the motivations behind the use of unorthodox Christian imagery. If we allow the visual evidence to take precedence, can we draw closer to the values and ideas cherished by the Christian communities that created, utilized, and preserved these images which were in tension with their professed beliefs?

2.1

The Circus and the Baptistery: Henchir el Koucha

2.1.1 The Archaeological Evidence Our first example brings us to the flourishing rural landscape of Africa Proconsularis, an area where many of the North African baptismal fonts have been preserved.4 The baptistery of Henchir el Koucha confronts us with the question of how to make sense of an iconography derived from the world of the Roman ludi, and in particular of the chariot races in the circus or hippodrome. Many leaders of the church were fervently opposed to games held in the circus, and even some baptismal liturgies explicitly warned against them. The iconography of the pair of facing horses nevertheless connected the sphere of the circus with that of the baptistery. When

3

See 60–65 in Chapter 1.

4

The most recent comprehensive account is Baratte et al. 2014.

The Circus and the Baptistery

Figure 2.1 Plan of the church of Henchir el Koucha, fifth to sixth century. Later additions are marked in red. Baratte et al. (2014), 82, fig. 64-24. © Fathi Bejaoui.

baptismal salvation needed to be articulated visually, the Christians of Henchir el Koucha found inspiration in the visual culture of the games. In 2012, a modest church (Figure 2.1) was discovered at the site of Henchir el Koucha near the village of Oued Zarga, 75 km south-east of Carthage. Following an excavation by the Institut National du Patrimoine under the direction of Fathi Bejaoui, the results have so far been published in two short articles, which offer only a few photographic illustrations.5 The pavement of this 25 m-long, north-oriented church with a counter-apse is largely covered in mosaics, to which I will return shortly.6 The site’s mosaics are being currently restored.7 Based on ceramic finds, the three-aisled church was built at the beginning of the fifth century at the latest and was considerably transformed throughout its history.8 Initially, only the northern apse existed. It was flanked by two small rooms from which a longitudinal room 5 7 8

6 Baratte et al. 2014, 81–84; Bejaoui 2016, 1682–1685. Bejaoui 2016, 1683. I thank Fathi Bejaoui for the kind information. The excavators suggest sixth-century dates for the later phases. Baratte et al. 2014, 84.

81

82

The Use of Non-Christian Imagery

Figure 2.2 Henchir el Koucha, mosaic floor of the baptistery. Baratte et al. (2014), 83, fig. 65-24. © Fathi Bejaoui.

behind the apse could be accessed. The longitudinal room gave way to an adjacent room of around 8 m by 3.6 m, flanking the church on its northeastern side (a subsequent blocking of the passage is marked in red on the plan).9 The excavators suggest that the baptismal font was inserted into this room during a later phase of construction (Figure 2.1).10 As so often in Africa Proconsularis, the baptismal setting is astonishingly lavish for rural construction.11 A quadrilobed font of a depth of 66 cm, covered initially by a ciborium and furnished with a heated drainage system, took up the northern half of the baptismal chamber (Figure 2.2).12 In connection with the installation of the font, a mosaic floor decoration covering the entirety of the baptistery was also installed. Quadrilobed fonts like the one in Henchir el Koucha are the most common font type in North African baptisteries overall.13 Traditionally, they have been thought to date mostly from after the Byzantine reconquest of North Africa (533–534); it is assumed that a large wave of baptismal constructions spread across rural North Africa after the reconquest.14 However, several quadrilobed fonts also date to the fourth or fifth century.15 The find of an epitaph near the

9 11 12 13 14 15

10 Baratte et al. 2014, 82. Baratte et al. 2014, 82–84. For an overview, see Morfino 2011. The basin occupies 4 m
3.8 m of the chamber. Baratte et al. 2014, 82. Morfino 2011, 49–52, fig. 23. On the Christianization of rural North Africa, see Bowes 2008, 162–178. Dated to somewhere between the middle of the fourth and the early fifth centuries are Belalis Maior II, Sidi Ferruch, Cuicul, and Chott Menzel Yahia. Dated to the fifth century is Ksar

The Circus and the Baptistery

quadrilobed baptismal font of El-Erg (Byzacena) dated to the reign of Thrasamund in 517 is further evidence that the quadrilobed form was also in use during the Vandal period in North Africa (429–534). As the excavators point out, an early sixth-century dating of Henchir el Koucha’s font is no less plausible than a dating in the later sixth century under Byzantine rulership.16 With traditional biases about the insignificant role of the Vandal period for the construction of North African churches increasingly dwindling, the once-fixed chronology – which automatically attributed elaborate baptismal fonts to the ‘culturally advanced’ Byzantines, often without sufficient stratigraphic evidence – no longer holds.17 At some point in the history of the baptismal chamber, the passage from the longitudinal room behind the apse to the baptismal chamber was blocked off using blocks of limestone (Figure 2.1, marked in red). The blocks cover a section of the mosaic floor in the baptismal chamber. When the entrance was blocked, the mosaic floor was already in place. Another entrance was carved into the south of the annexe so that the baptismal chamber received an entrance independent from the church.18 Another substantial alteration was the installation of a counter-apse in the south of the church (marked in red). North African counter-apses like this one are thought to date to the Byzantine period, which would make this the latest alteration to the ensemble.19 The mosaic programme is, on the one hand, a particularly innovative and unusual contribution to the large corpus of North African mosaic baptismal decorations but also poorly executed. On the other hand, it was marked by remarkable coherence compared to the mosaic decoration of the basilica, which was transformed substantially by the insertion of more and more epitaphs. Visitors entering the baptismal chamber from the south faced two large mosaic panels, framed by a vine scroll interspersed by circles with radiating rays filled with crosses (Figure 2.2). The southern panel is filled with a shield ornament pattern, while the larger northern panel is unique in its design (Figures 2.3 and 2.4).20 The northern panel connects the baptistery’s centre with the area around the font and is

16 17

18 19

20

Belezma, and to the late fifth century Acholla and perhaps Meninx. Another fifteen quadrilobed fonts are dated to the sixth century. Morfino 2011, 52, n. 142. Baratte et al. 2014, 84 and 341. Late fifth-century construction dates have been attested for the Bir el Knissia basilica in the south of Carthage, and for basilicas at Bir Messaouda and Carthagenna. See Bockmann 2013, 250. On mosaics of the Vandal period, see Ghalia 2009. Baratte et al. 2014, 82. Baratte et al. 2014, 84. On the North African and Iberian model of the counter-apse, see Duval 1971; Ulbert 1978. Bejaoui 2016, 1684. See also Ghalia 2016, 315–316, fig. 6.

83

84

The Use of Non-Christian Imagery

Figure 2.3 Henchir el Koucha, detail of the central panel of the mosaic floor of the baptismal chamber. Bejaoui 2016, 1689, fig. 12. © Fathi Bejaoui.

Figure 2.4 Drawing of the southern and northern panels of the baptismal chamber of Henchir el Koucha, based on photography on the web page http://?archeologiechretienne? .ive.org/?p¼723 (last accessed 15 August 2023). © Jörg Denkinger.

The Circus and the Baptistery

entirely orientated towards the font. Divided into two halves, with the central half facing the font and the northern one surrounding it, the only position from which the panel can be fully appreciated is near the font. In the centre of the northern panel, a source of water symbolizing the rivers of paradise streams from the area closest to the font and nourishes a pair of deer (Figure 2.4). This is a perfectly conventional baptismal motif which was in wide use in North Africa and alluded to Psalm 42:1 – ‘As a deer pants for flowing streams, so pants my soul for you, O God.’21 Psalm 42 was, according to Augustine of Hippo (354–430), chanted during the baptismal ceremony.22 The pair of drinking deer is set within a larger scene: underneath a dense grouping of stone pines, the bust of a young, beardless man with dark hair appears behind the deer on the left. Above the pine trees, a blond man, draped in a garment covering him up to his knees, hovers at the top of the scene. In the airy sphere, he is framed by two birds and several flowers with red buds. Floating despite having no wings, the man holds in his left hand a long, fork-like object with approximately six prongs on a handle ending in a slight curve.23 Both the male bust and the flying man are unparalleled in North African ecclesiastical imagery. The lack of objects of comparison, as well as the mediocre quality of the mosaic, do not permit conclusive interpretations as to what the scene is supposed to represent. The panel’s layout, however, suggests that we are seeing a coherent narrative set amidst woodlands at the source of the rivers of paradise concerning a human and also a heavenly messenger, arguably an angel. Birds and flowers, as well as imagery of plenty and the good life, demarcate the scene on the borders, as was commonly done in sepulchral and ecclesiastical North African decorations. The excavators’ suggestion of identifying the object carried by the angel as an Easter candle (as Easter is the most frequent date of the baptismal

21 22

23

This and all following biblical quotes follow the English Standard Version. For evidence of the pair of deer in North Africa, see Baratte et al. 2014, Henchir el Koucha 81–84, Bou Smir 233, Henchir Messaouda 238–239, Henchir Errich 376. On the mosaic surrounding the baptistery of Bir Ftouha, see Stevens, Kalinovski, and van der Leest 2005, 303–342, esp. 303–308. A deer killing a snake is only depicted in the baptistery of Henchir Messaouda: Puech 1949; Feuille 1949; Baratte et al. 2014, 238–239. Other pairs of deer appear in the baptisteries of Valence (France), Stobi and Ohrid (North Macedonia), Salona (Croatia), Butrint (Albania), and Ras-Siagha on Mount Nebo (Jordan). The motif is not limited to baptismal use in North Africa. See church decorations of Zaouit Sousse in Baratte et al. 2014, 222–223, Henchir Bechouk 68–69, Henchir El Ouara 361–363, El Ounaïssia 363–365. On the iconography at large, see Domagalski 1990; Puech 1949. On the chanting of Psalm 42, see Augustine, Enarr. in ps. 42 (41), 1. In his commentary, Augustine explains further that the thirst of the deer and their habit of killing and eating snakes represent humans’ innate thirst for God and their capacity to destroy sins and vices. Burns and Jensen 2014, 785. Bejaoui 2016, 1684; Baratte et al. 2014, 82.

85

86

The Use of Non-Christian Imagery

Figure 2.5a Henchir el Koucha, detail of a pair of horses facing a plant on the mosaic pavement of the baptismal chamber © Fathi Bejaoui.

Figure 2.5b Drawing of the pair of horses on the mosaic pavement, Henchir el Koucha © Jörg Denkinger.

ceremony) is as tempting as it is questionable.24 While it is true that the baptismal fonts of Sidi Jdidi and Kelibia (Figure 1.29) each display lit candles, the unlit device held by the angel does not resemble any kind of candle prevailing in late antiquity.25 The panel continues to the north, spreading around the font (Figures 2.2 and 2.3). The vegetal border continues on all sides except for the western one. Judging from the photographs available, the wall separating the baptistery from the northern room behind the apse covered parts of the panel. The northern and western sides of the font are decorated with pairs of animals arranged in strict symmetry; the animals in each pair turn to look at one 24 25

Bejaoui 2016, 1684. On lighting devices, see Stiegemann 2001, 205–229; Kötzsche 1986a. Cf. Bouras 1982. On the baptistery of Sidi Jdidi (Africa Proconsularis), see Ben Abed-Ben Khader, Fixot, and Roucole 2011, 313–317, fig. 204 and fig. LIV.

The Circus and the Baptistery

Figure 2.6 Henchir el Koucha, detail of the mosaic pavement of the central nave of the basilica. Bejaoui 2016, 1689, fig. 10. © Fathi Bejaoui.

another, separated by a grape or floral motif. There is no equivalent on the eastern side, which is entirely filled by vine scrolls. The pairs of animals – peacocks as well as birds picking grapes in the north, and horses in the west – follow a pattern of animals facing each other which had its origins in ancient Mesopotamia and mirror the deer facing the fountain in the south.26 The pair of horses (Figure 2.5), adorned with neckbands and each with a raised front leg, face a flowering plant which is no longer identifiable. This scene, in contrast to all others, is not orientated towards the fountain but faces outwards. The likely explanation is that when the mosaic was created (before the entrance from the church into the baptistery was blocked), the horse motif was positioned in this way to welcome those entering the baptistery. As far as we can tell, the baptismal decoration is consistent with the original mosaic programme of the basilica. The floor design has been disturbed by a total of fifteen mosaic epitaphs. These are inserted fairly systematically in the northern and southern rooms behind the apses as well as in front of the apses.27 Apart from these, the excavators describe large panels covering the basilica in vegetal and geometric ornaments.28 The central nave hosts three animal scenes. The southernmost, near the counter-apse, is a pair of deer and a cock surrounded by pines, but an epitaph inserted later caused the erasure of half of the scene.29 Another pair of deer and a palm tree are depicted at the centre of the nave, while in the northernmost animal scene a peacock is following a horse (Figure 2.6); in between the two is a flowering 26

27

28

Danthine 1937. The excavators agree with the identification of the pair of quadrupeds as horses. Bejaoui 2016, 1684. A large number of the epitaphs are adorned with swastikas, rarely found in North Africa, other examples of which have been preserved in the Christian complex of Cuicul. See n. 154 in Chapter 1. See also Bejaoui 2016, 1683. 29 Bejaoui 2016, 1682–1683. Bejaoui 2016, 1689, fig. 11.

87

88

The Use of Non-Christian Imagery

plant.30 The horse is presented in a similar posture to the horses in the baptistery and also wears a neckband. The presumably latest installation, the counter-apse, is adorned with a mosaic which covers a reliquary sunk into the ground which, in 2016, was still undisturbed. On top of the reliquary is a kantharos filled with flowers and grapes. The mosaic of the counter-apse also shows a pair of peacocks facing a cross from which alpha and omega hang.31 A pair of deer is represented in front of the counter-apse.32 The mosaicists carried through with the facing-animal motif as the central figurative element. Judging from the extant mosaic pavement, the narrative culmination point was reached at the baptismal font: animals, humans, and the divine messenger are gathered around the salvific powers of the baptismal water. In contrast to the other paired animals of Henchir el Koucha, there is no tradition of using the facing-horses motif in church spaces.33 Instead, North African ecclesiastical spaces often contain animals with widely recognized Christian allegorical significance. Deer are closely connected to Psalm 42, which played a role in North African baptismal liturgy, while peacocks have been a long-standing symbol of immortality, and the frequently used pairs of sheep may represent the Christian community.34 What interest, then, could the Christian community in Henchir el Koucha have had in adding a pair of horses to the established baptismal iconography?

2.1.2 The Motif of the Facing Horses – A Pictorial Formula between Christian Salvation and the Roman Circus Horses were rarely depicted in ecclesiastical spaces and are virtually absent in baptismal decorations.35 Apart from the pair of facing horses and the peacock following a horse in Henchir el Koucha, only a few North African 30 32 33

34 35

31 Bejaoui 2016, 1683. Baratte et al. 2014, 83. Bejaoui 2016, 1683; Baratte et al. 2014, 84. Another relatively rare motif from Henchir el Koucha is the cock. Other occurrences are known from Thabraca, Haidra, and Bulla Regia. Bejaoui 2016, 1683, n. 22. On paradisiacal and baptismal symbolism, see Ladner 1992, 138–143. The small quadruped holding a laurel wreath that is depicted on the rim of the baptismal font in Basilica II at Sidi Jdidi has not been identified with certainty by the excavators. The fact that the mosaicists took care to depict its hooves with two main digits, even on such a small scale, is strong evidence that it represents a sheep. This apparently went unnoticed by the excavators. Ben Abed-Ben Khader, Fixot, and Roucole 2011, 317, fig. 206. In the eastern Mediterranean, only one example of a representation of horses in a baptistery has come to my attention. The hunting scene in the Diaconicon baptistery in Jordan shows, among a variety of wild animals, two fighting horsemen (Figure 2.29). Another eastern example is the bucolic scene with three horses in the southern annexe of the Arapaj basilica in Albania. See Hidri, Hidri, and Pillinger 2011, 36–41. I thank Ine Jacobs for bringing this example to my attention.

The Circus and the Baptistery

examples of horse imagery exist. At Oued Ramel (Byzacena), in an annexe of the basilica, a destroyed inscription (possibly a dedication panel for the construction of the church) is accompanied by a narrative scene in three registers showing the construction of a building. Amidst various craftsmen in charge of different tasks on the construction site, a pair of horses pulls a cart with a column on it.36 In the realm of private Christian funerary imagery, however, horses sometimes are depicted in their own right. In a tomb-mosaic in a church in Hadjeb-el-Ayoun (Byzacena), a single horse is positioned among flowers between a vase and a dove. On the tomb of Lollianus from Thabraca (Africa Proconsularis) (Figure 2.7) is a depiction of a horse tied to a cypress tree beneath the figure of the Good Shepherd and his sheep; above it are registers with doves and fish.37 Another tombmosaic of the fifth-century deacon Crescentinus at Thabraca features three men on horses in a landscape amidst birds and flowers. Two of the horses face each other in strict symmetry, separated by a tree.38 The depiction of individual horses in funerary contexts has been compared to depictions of the agonistic crown on a number of late antique sarcophagi and tomb-mosaics.39 The agonistic crown likely symbolized the prize which was given to the winners of chariot races in the Roman circus.40 Adding to other triumphal motifs on Christian sarcophagi, such as flower garlands and laurel wreaths, it seems that the connotation of triumph and victory inherent to imagery related to chariot races helped establish the use of such motifs for sepulchral purposes.41 The tripartite motif of two horses separated by an object and facing each other can be traced back to the pictorial repertoire of the hippodrome. Adorned horses, raising their front legs and facing a plant, a cylinder, or a

36

37

38

39

40

Ghalia 2016, 317–319, fig. 7; Dunbabin 1978, 192, pl. LXXIV, fig. 192.; Gauckler 1910, inv. no. 463. Dunbabin 1978, 103. Gauckler 1910, inv. no. 320 and inv. no. 984. The four horses surrounding a cross that Dunbabin identifies in a panel near the altar in the church of El Mouassat have been convincingly identified as sheep due to their enlarged tails, reminiscent of similar examples from Uppenna and Sidi Habich. See Duval 1974, 167. The depiction of hooves with two main digits further indicates that they were intended to represent sheep. Yacoub 1995, 373, fig. 190; Dunbabin 1978, 192; Gauckler 1910, inv. no. 1024. The tree is depicted upside down, arguably to represent a circular movement of the horsemen in the landscape. The centrally placed third horse carries two riders, partially destroyed by repairs. Compare Gauckler 1913, fig. 1024. On elements of horse races as repertoire in Christian sepulchral imagery, see Duval 1990, 137; Duval 1992, 182; Duval 2011, 253–256. Agonistic crowns are further depicted on the tombmosaic of the epitaph of Galliga in Basilica II at Sidi Jdidi. See Ben Abed-Ben Khader, Fixot, and Roucole 2011, figs. X and XI on the epitaph of Paulina in the parish church of Demna from the late fifth century, today on display at the Musée du Bardo. 41 Duval 1990, 137. Dunbabin 1978, 88–108, esp. 103.

89

90

The Use of Non-Christian Imagery

Figure 2.7 Tomb mosaic of Lollianus, western necropolis of Thabraca, sixth century, Musée National du Bardo, Tunis. Photography by the author.

kantharos, were commonly depicted in North African domestic mosaics between the late second and the sixth centuries (Figure 2.8).42 Indeed, over twenty North African examples are known; more than half of them from Africa Proconsularis.43 There is also some evidence for the proliferation of the motif beyond North Africa – certainly on the Iberian Peninsula but probably also in the Greek-speaking domain.44 Initially, the motif of facing horses represented successful racehorses which were proudly depicted by 42 43 44

Yacoub 1994, 252. See also Ennaifer 1983, 1994. Ennaifer 1983, 843, n. 128. Ennaifer adds one more case to the group in Ennaifer 1994. Two horses facing a cylinder, accompanied by a Greek inscription, appear on the hilt of a knife of unknown origin, today in the Museo de Spalato. See Yacoub 1994, 252, n. 19. In Spain, the motif is attested on a second to third century terracotta tile from Osuna (Prov. Sevilla), see Schlunk and Hauschild 1978, 58, pl. 72a; López Monteagudo 1994. On the mosaic pavement of the oecus of the Roman villa of Aguilafuente (Segovia), dated to the beginning of the fourth century, as well as on a mosaic fragment from Itálica, preserved today at the palacio de la Condesa de Lebrija, see López Monteagudo 1994, 354, figs. 17 and 18.

The Circus and the Baptistery

Figure 2.8 Detail of the floor mosaic of the house of Sorothus, Hadrumetum, late second to early third century, Musée Archéologique de Sousse. Photography by the author.

their owners in their estates.45 The earliest dated example from the house of the horse breeder Sorothus in Hadrumetum (Sousse) (Figures 2.8 and 2.9) from the late second or early third century, now partially destroyed, depicted eight pairs of named horses adorned with palms on their heads and colourful harnesses around their necks. The horses are marked with stable marks or the mark SORO/THI on their hindquarters and shoulders, and they surround palm trees in a similar way to other early examples of the motif.46 The imagery of a palm tree combined with a horse has deep roots in North African history, and the presence of these elements in the earliest depictions of racehorses with palm trees may indeed allude to local North African traditions.47 The depiction of a horse (or its head or protome) along with a palm tree can be traced back to the beginning of Punic coinage in the late fifth century BC. These coins, minted in Carthage and initially found in conquered Sicily, featured the horse and palm tree either on the same face 45 46 47

Ennaifer 1983, 825–837. The horse breeder Sorothus and his estate are discussed in Laporte and Lavagne 2006. I thank Stefan Heidemann for bringing this point to my attention.

91

92

The Use of Non-Christian Imagery

Figure 2.9 Detail of the floor mosaic of the house of Sorothus, Hadrumetum, late second to early third century, destroyed. Gauckler, 1914, vol. 2.2, fig. 124.

or on the front and back.48 While it has been established that the horse offered a wide spectrum of symbolic references, both to Baal Hammon, venerated in Punic Carthage, and to Zeus, who lacked a specifically Punic character, the date palm (phoenix in Greek) indeed seems to refer to the Phoenician origin of the Carthaginians.49 The combination of these two elements in Punic coinage was common and long-lasting, with these coins remaining in circulation long after the fall of Punic Carthage in 146 BC.50 During the later third and fourth centuries, a significant change occurred in the central object between the pair of horses, highlighting the motif’s connection to the Roman circus. Frequently, this object took the form of a modius, a cylinder filled with grain which was customarily awarded as a prize to victorious racehorses (Figures 2.10 and 2.11).51 48 50

49 Günther 2015. Günther 2015, 141–143, 147. 51 Huss 2004, 353–356; Acquaro (ed.) 1991. Duval 1992, 199–201.

The Circus and the Baptistery

Figure 2.10 Floor mosaic of a pair of horses facing a modius of the maison du paon, Carthage, second half of the fourth century, Musée National du Bardo, Tunis. Photography by the author.

Figure 2.11 Mosaic of a pair of horses facing a modius of the bath of Fundus Bassianus, Sidi Abdallah, end of fourth to beginning of fifth century, Musée National du Bardo, Tunis. Yacoub (1995), 320, fig. 162.

The modius, along with details like bandaged legs, tails bound up with ribbons (Figure 2.11), or adorning neckbands (Figure 2.8), as well as the use of proper names for the horses – details which mark them as individuals – all suggest that the representations meant actual racehorses.52 52

On the nomenclature of circus horses, see Dunbabin 1978, 95, n. 25.

93

94

The Use of Non-Christian Imagery

Figure 2.12 Mosaic of pair of horses from Lahmimine, sixth century. Ennaifer (1983), 858, fig. 31.

The frequency of the facing horses motif and the related motif of the lone horse pictured in a similar manner are material evidence for the popularity of chariot racing and horse breeding in North Africa, which was among the most prolific breeding regions of the Roman Empire.53 From the fourth century onwards, the motif of the facing horses changed again. Horses after this time are less often identifiable through proper names, and the modius was often replaced by a kantharos.54 Furthermore, the horse motif tends to be represented in proximity to prophylactic or luck-bringing symbols and inscriptions (Figures 2.10 and 2.12).55 These later examples have often been discussed in allegorical terms rather than as naturalistic depictions by scholars. They are seen as luckbringing signs of victory or as part of more mystical narratives built on the

53

54 55

On horse farms in North Africa, see Bourdy 1990, 147. For the motif of the singular horse, see Ennaifer 1983, 825–831. Yacoub 1994, 252. Ennaifer 1983, 836–842. On the mid- to late fifth-century mosaic in Moknine, with its inscription protecting against jealousy, see Yacoub 1994, pl. CLIXI, fig. 1; Yacoub 1995, 334–338. An interesting variation is the fourth-century mosaic of Ellès, in which the horses are replaced by female centaurs and the modius by a figure of Venus. See Dunbabin 1978, 261, plate LX, fig. 152.

The Circus and the Baptistery

cosmological meanings attributed to the circus in antiquity.56 Nevertheless, the transition from the realistic to the allegorical genre was by no means abrupt; some mosaics display features of both traditions (Figure 2.10). Even in the latest representation from Lahmimine (Figure 2.12), dated to the sixth century, the two horses still retain adornments of racehorses. The horses’ tails are bound up with three ribbons, and both are adorned with prominent neckbands.57 The mosaic of Lahmimine combines an allegorical purpose with a reference as to what caused horses to be perceived as victorious and luck-bringing: their ability to succeed in races. The facing horses in the baptistery of Henchir el Koucha (Figure 2.5a and 2.5b), though very stylized, are typologically similar to depictions of racehorses from previous centuries. Like the horses from the house of Sorothus (Figure 2.9), they are unharnessed but adorned with a neckband and possess a long, unbound tail. The position of their legs, which are raised high in a symmetric fashion, is reminiscent of sixth-century depictions from Thabraca (Figure 2.7) or Lahmimine (Figure 2.12). Their location on the baptismal floor is remarkable, as they would have been the first image to be encountered upon entering the baptismal chamber from the basilica. The horses facing a flowering plant lost this position during refurbishment but continued to be bound up in a decorative scheme which emphasized the salvific nature of baptism. The notion of victory implied by the motif in the context of the circus was likely now applied to baptism. It is plausible that the facing horses allude to the triumph which lies in the spiritual rebirth effected by baptism. Thus, the pair of horses became part of the overall message of the baptismal decoration, an expression of joy in the achievement of eternal life as a Christian.

2.1.3 Baptismal Liturgy, the Circus, and Christian Identity The pictorial roots of the facing-horses motif clearly lie in the late antique tradition of initially representing and, from about the fourth century 56

57

Dunbabin 1978, 103–104; Ennaifer 1983, 837. On the different views of the mosaic of Lahmimine, see also Yacoub 1994, 256; Duval 1992, 204–205. On sixth-century cosmological interpretations of the circus, see Stevens 1988. The mosaic, being without provenance, has been dated to the sixth century for stylistic reasons: the lack of a ground line, the exclusive use of lime tesserae, the vivid colour scheme at the expense of subtle transitions of colour as well as the symmetric layout and the pictorial repertoire of kantharoi, birds, and vegetal ornamentation are arguments brought forward for a late date by Mongi Ennaifer and by Mohamed Yacoub, the latter of whom dates the mosaic as late as after the Byzantine reconquest. Ennaifer 1983, 840–842; Yacoub 1994, 251 and 256.

95

96

The Use of Non-Christian Imagery

onwards, allegorizing racehorses. Should we expect late antique viewers in Henchir el Koucha to have been able to recognize the visual inheritance of the circus? The roughly contemporary sixth-century mosaic of Lahmimine (Figure 2.12), in any case, still has elements of traditional racehorse iconography. However, the mosaic lacks any provenance and might have been produced for a private commissioner. In contrast, at Henchir el Koucha, the motif is integrated into a key space of Christian identity formation. Should we assume, then, that the facing-horses motif had already been stripped of its affiliation with the Roman ludi when it was composed for the baptistery? There can be no question that chariot racing was still actively remembered, and in some places still practised, in fifth- and sixth-century Africa Proconsularis. The poetry collection Anthologia Latina offers a host of information on how markedly the circus still occupied the minds of North Africans in the sixth century. Although the Anthologia Latina was not compiled in North Africa, it does contain a rich collection of North African poetry, a sizable portion of which was probably composed for the Vandal court.58 Key topoi are Roman institutions like the circus and the amphitheatre, as well as new Vandal building projects and the social practices connected to them; Christian subjects, by contrast, are sparse.59 A series of ten poems, of which all but one are ascribed to Luxorius, who wrote for the Vandal court during the reign of Hilderic and Gelimir (523–533 AD), strongly suggests that circus games were held in late Vandal Carthage, only 75 km away from Henchir el Koucha.60 Also the circus in Thysdrus (El Djem) in Byzacena may still have been active in the sixth century.61 The Anthologia Latina contains the works of several poets who probably wrote under different rulers in late antique Carthage. Yet, the poets are united in their praise for traditional Roman elite cultural practices

58

59

60

61

On questions of dating and compiling the Anthologia Latina, see Cameron 1996b, 3, n. 8; Miles 2005, 306, n. 3. On the circus, see Stevens 1988. On Christian subjects in the Anthologia Latina Miles 2005, 317–318. The collection of North African poetry in the Anthologia Latina, usually dated to the last days of the Vandal kingdom in 532–534, and the epic poem Johannis, written at the beginnings of the 550s under Byzantine rule by Flavius Cresconius Corippus, stand at the end of a long tradition of North African Latin poetry that incorporated mythological topics with ease. See Cameron 1996b; Miles 2005; Lepelley 2010. Stevens 1988, 177–178. Procopius, De Bello Vandalico, II, 6,6–8 gives testimony to the survival of the circus in this late period. On the late phase of the North African circus and for a discussion of circus imagery, see Jiménez Sánchez 2010, 335–338. On the societal impact of the late antique circus, see further Heucke 1994. Leone 2013, 101.

The Circus and the Baptistery

Figure 2.13 Historical photography of a mosaic depicting a circus scene from Capsa (Gafsa), sixth century, today Musée National du Bardo, Tunis. Médiathèque du patrimoine et de la photographie, APBLL00138N.

such as hunting, feasting, euergetism, and the holding and attending of circus races.62 These were amenities enjoyed by members of the Vandal court as much as by the lay Romano-African elite, and their poetic appraisal might well have been a deliberate attempt by the Vandal rulers to stress the commonalities between these two groups. Scholars have recently placed much emphasis on the interest the Vandal kingdom had in maintaining Roman traditions.63 However, a mosaic depiction of a chariot race from Capsa (Gafsa) (Figure 2.13), possibly from a bath in sixth-century southern Byzacena, confirms that the circus continued to fascinate even far from Vandal Carthage.64

62 63

64

Shackleton Bailey 1982. On recent publications on Vandal continuity of Roman culture, see Conant 2015; Merrills and Miles 2010, 88–108 and 204–227; Miles 2005; an alternative view on the Vandals and the continuity of classical traditions is offered in Bockmann 2013. Dunbabin 1978, 92–93; Yacoub 1995, 305–311, figs. 156a–c.

97

98

The Use of Non-Christian Imagery

Taking the longevity of the practice of horse races in post-imperial North Africa into account, the increasingly allegorical use of the horse motif and its adoption by Christians should not be counted as evidence for the loss of the motif’s cultural affiliation with the circus. To the contrary, its adaptation to new public and private functions in the Christianizing society of the fifth and sixth centuries speaks to its continued relevance to society. As discussed above, the role of the horse motif in Christian contexts is best explained as an allegory of victory or perhaps luck. Evidently, the association of the image of a horse with victory and luck makes most sense in a society which still associated horses with winning a race. Even though many of the extant examples of facing horses have been preserved in the domestic sphere, the motif must have been very widely known for it to have been chosen as the auspicious image on the threshold to the baptistery. In the public sphere, we currently know about the single and facing double horses in the churches of Thabraca (Figure 2.7), Hadjebel-Ayoun, and Henchir el Koucha. Furthermore, the floor mosaics depicting facing horses in Sidi Abdallah, Moknine, and Oued Athmenia (Algeria) decorate baths attached to villas. Such baths are thought in many instances to have been semi-public spaces.65 The motif has also been found on movable objects, such as the hilt of a knife in today’s Spalato and on a terracotta tile from Osuna.66 It is probable that, for fifth- or sixth-century visitors to the baptistery, the associations inspired by the motif would have oscillated between (at least) two poles: baptismal salvation and the Roman circus. By the fourth century, however, horse races had begun to face Christian opposition. In 399 AD, imperial legislation prohibiting all sacrifices in relation to Roman festivals reached North Africa.67 Spectacles such as chariot racing were still permissible, though this did not hinder Christian polemicists from condemning them as pagan.68 Church writings

65

66 67

68

The semi-public or public function of many privately owned baths has often been acknowledged. For North Africa, see Maréchal 2016, 2020, 206–208. On the baths of Fundus Bassianus in Sidi Abdallah and of Moknine specifically, see Maréchal 2020, 349. See also Ennaifer 1983, 835 and 838; Yacoub 1995, 320, fig. 162 and 336, fig. 167. For the horse mosaic in the caldarium of the baths of Pompeianus in Oued Athmenia, see Maréchal 2020, 329–331 and Morvillez 2012. See n. 328 in this chapter. C. Th. 16,10,17, issued on 20 August 399. See also Belayche 2007, 42; Lepelley 1979–1981, vol. 1, 381. On Christian critique of the circus, see Lugaresi 2007. For an overview of religious aspects of Roman spectacles in antiquity, see Zaleski 2014. See also Coleman 1990. On the role of the circus in fourth- and fifth-century ludi, see Jiménez Sánchez 2010, 137–152.

The Circus and the Baptistery

circulating in the western Mediterranean urgently warned against the feelings the circus aroused and its idolatrous implications. The circus was generally deemed an un-Christian institution which needed to be overcome. The two core arguments against the circus made in Tertullian’s De Spectaculis, which the Carthage-born author wrote at the turn of the third century, became the model for subsequent treatises, sermons, and letters on the issue.69 The first argument in De Spectaculis is directed against the circus in particular.70 It concerns the locura or madness which the races induced in spectators. Augustine’s worry that his friend Alypius’s soul was being corrupted by watching the maddening races, or Salvian of Marseille’s account of how the enthralled spectators’ cries in the arena blended in with the cries of the murdered outside as the Vandals conquered Cirta, all evoke the same fear of losing self-control.71 The second argument encompasses all spectacles, accusing them of being idolatrous because they are dedicated to pagan deities. Participants of ludi like the circus were automatically in denial of God. This is the principal argument in Tertullian’s De Spectaculis; it was also picked up by Salvian of Marseille (c. 400–475) and Isidore of Seville (c. 556–636).72 Both the locura argument and the idolatry argument treat circus games as part of a body of cultural practices which were presented in their entirety as being opposed to Christian beliefs and values. Beyond the allegation of idolatry, stereotypical accusations were made against all forms of spectacles: the circus was maddening, the theatre immoral, the amphitheatre cruel.73 Christian writers construed the circus as participating in and even fomenting a culture of dangerous pleasures and shallow entertainments. Conversely, by criticizing the circus, they attacked the cultural practice of spectacles more broadly. The Christian contempt for the ludi even left a mark on the baptismal liturgy, more precisely in the abjuring formula. A formulation used in baptismal liturgies to make neophytes abjure the devil referred to the games. In De Spectaculis, Tertullian gives the first attested record of a version of this formulation: ‘When we enter the water and profess the Christian faith in the terms prescribed by its law, we profess with our 69 71

72

73

70 Jiménez Sánchez 2010, 279 and 286. Tertullian, De spec., 16,1. Augustine, Conf., VI, 7,11–12; Salvian, De Gubern. Dei, VI, 12, 69, 71. On further Christian literature on the subject, see Jiménez Sánchez 2010, 279. Salvian, De Gubern. Dei, VI, 2, 60; Isidore, Etym., XVIII, 27,1 and XVIII, 59. On Isidore and his warning of demons in the ruins of the circus, see García Moreno 2001, 7–8; Mercado Hernández and Sánchez Medina 2001, 225–226. Jiménez Sánchez 2010, 270.

99

100

The Use of Non-Christian Imagery

mouths that we have renounced the devil, his pomp and his angels.’74 Formulations such as pompae diaboli appear several times in De Spectaculis. The meaning of pompa in the baptismal context has been vividly discussed by Tertullian scholars, as pompa evokes an array of meanings including ‘procession’, ‘retinue’, ‘splendour’, and ‘ostentation’. Some interpreters argue that Tertullian was referring specifically to the pompa circi, the procession attached to circus games.75 Indeed, Tertullian adopted the formula ‘the devil, his pomp and his angels’ as an evocation of idolatry, which comprises all ludi. Hence, he concludes, the newly baptized abjures all forms of the games by renouncing ‘the devil, his pomp and his angels’.76 Soon after, Tertullian clarifies that the primary significance of pompa is the procession taking place in the circus: But rather more pompous is the outfit of the games in the circus, to which the name pomp properly belongs. The pomp comes first and shows in itself to whom it belongs, with the long line of images, the succession of statues, the cars, chariots, carriages, the thrones, garlands, robes. What sacred rites, what sacrifices, come at the beginning, in the middle, at the end . . . .77

The term Pompae, as used by Tertullian in the context of the baptismal rite of abjuration, did not necessarily evoke the idolatrous aspects of the circus or of spectacles exclusively. Yet, this valence was clearly an important dimension of the phrase’s overall meaning. Later, Tertullian’s interpretation would remain influential for the exegesis of the abjuring formula. Cyril of Jerusalem’s Mystagogic Catecheses, written in Greek in c. 370 AD, explains the meaning of the phrase ‘I

74

75 76

77

‘Cum aquam ingressi Christianam fidem in legis suae verba profitemur, renuntiasse nos diabolo et pompae et angelis eius ore nostro contestamur.’ Tertullian, De Spec. 4,2 transl. Glover and Rendall (eds.) 1931, 243. On the positions of Waszink and of Köhne, see Waszink 1947, 32–34. Tertullian, De Spec. 4,5: ‘What shall we call the chief and outstanding matter, in which the devil and his pomp and his angels are recognized, rather than idolatry? . . . So, if it shall be established that the whole equipment of the public shows is idolatry pure and simple, we have an indubitable decision laid down in advance, that this profession of renunciation made in baptism touches the public shows too, since they, being idolatry, belong to the devil, his pomp and his angels.’ Transl. Glover and Rendall (eds.) 1931, 242. ‘Sed circensium paulo pompatior suggestus, quibus proprie hoc nomen: pompa praecedens, quorum sit in semetipsa probans de simulacrorum serie, de imaginum agmine, de curribus, de tensis, de armamaxis, de sedibus, de coronis, de exuviis. Quanta praeterea sacra, quanta sacrificia praecedant, intercedant, succedant . . . .’ Tertullian, De Spec. 7,3–5. transl. Glover and Rendall (eds.) 1931, 249.

The Circus and the Baptistery

renounce you, Satan, and all your works, and all his pomp, and all your worship’, which was pronounced by those about to be baptized as they faced westwards, in the direction of Satan. He explicitly refers to the spectacles when discussing pompa: ‘These are the devil’s pomp: a passion for the theatre, horse-races, hunting and all other such vain pursuits. . . . Avoid the races, a mad spectacle which unseats the soul. All these things are the devil’s pomps . . .’.78 While the exact wording was not preserved in the Augustinian rite, the same ritual movements which accompanied the abjuring are attested elsewhere. Before baptism, the candidate turned westwards to renounce Satan and then eastwards to swear allegiance to Christ.79 The ritual of abjuring the devil was standard practice in the baptismal rites, so a similar formulation could have been used in the baptistery of Henchir el Koucha. This usage would have placed neophytes in the unique position of having to abjure the attendance of non-Christian institutions, exemplified by the ludi, in the presence of Christian iconography rooted in the tradition of horse races. Given the strong disapproval of horse racing expressed in Christian writings, it is remarkable that the facing-horses motif made its way into the baptistery. The level of planning necessary to create a design with the narrative coherence specific to Henchir el Koucha’s baptismal decoration suggests that the motif was not chosen by mere coincidence. The mosaic’s originality is most clearly exemplified by the unique addition of humanshaped figures to the more established animal scenes. The position of the facing-horses motif in the (original) entrance area of the baptistery likewise seems too prominent to be a coincidence. That the representation of the horses is comparatively subtle, bare of most (if not all) of the props which would identify them as racehorses, does not necessarily demonstrate oblivion to the motif’s origin – it could also have been a concession to what was felt to be decent in a church. Those who incorporated the facing-horses motif in the baptismal mosaic likely intended to symbolize the victory offered by baptismal salvation. Christian identity was demonstrably salient in Henchir el Koucha. However, the motif carried an additional layer of meaning due to its resemblance to a contemporaneous iconographic tradition of allegorical representations of racehorses associated with the Roman circus. The choice of this motif for the baptistery’s decoration can be seen as an example of visual culture subtly compromising official doctrine. Specifically, in 78 79

Cyril of Jerusalem, Myst. Cat. 1, 4–8., transl. Yarnold and Cyril 2000, 170–171. Burns Jr. and Jensen 2014, 208–209.

101

102

The Use of Non-Christian Imagery

Henchir el Koucha, the liturgy and the baptismal decoration arguably conveyed diverging ideas regarding the extent to which traditional Roman culture should be excluded from Christian life. Baptismal liturgy explicitly rejected the notion that Roman games should have a place in the life of a Christian, while the baptismal decoration was much more inclusive, allowing for an allusion to the popular culture of the games. Horses brought luck – both in the circus and in the church.

2.2

Mythological Subject Matter between Belief and Practice: Myrtilis Iulia (Mértola) and Milreu

Scholars usually problematize Roman mythological imagery integrated into a Christian setting because of the pantheistic world view inherent to Roman mythology.80 The heroes and hybrids who are the subject of this section and the one that follows derive from the same narrative universe as deities, and feature as their companions. A common explanation for the continuous use of mythological imagery in Christian contexts is that it became secularized. It is often held that, by the fifth century, mythological imagery’s prime function was the demonstration of paideia (acquaintance with classical Graeco-Roman culture), which demonstrated cultivation and high social status.81 According to this view, mythological subject matter did not contain anything which contemporary viewers might believe in. It is, however, also acknowledged that the ecclesiastical elite’s attitude to paideia was typically a complex and problematic one.82 If mythological imagery was, in fact, disentangled from religion in late antiquity, would not the main reason for avoiding the use of such imagery in Christian public spaces also be gone? It is apparent from the archaeological evidence that there was a strong divide in the fifth- and sixth-century western Mediterranean between domestic and ecclesiastical visual culture related to mythological subjects. Villa owners throughout the Mediterranean continued to decorate private spaces, furnishings, and equipment with mythological scenery despite most 80 81

82

This section and the following one are extended and revised versions of Lenk 2017. On the demonstration of paideia as a class marker in various artistic contexts, see for instance Raeck 1992, 161; Dunbabin 2012, 2014, 237; Uytterhoeven 2009, 332; Stirling 2005, 138–164. Alternative avenues into the study of the Christian use of mythological topics are offered in Muth 2001; Ferrari 2011. Kaldellis 2007, 120–172. On the Christian parallelization of Hellenic and pagan in the late antique East, see Bowersock 1990, 10–13.

Mythological Subject Matter

likely being Christians. They adapted traditional mythological topics to their needs and interests, which indicates that these themes were very much alive in Christian society.83 By contrast, such imagery is – while not entirely absent – strikingly rare in church interiors and liturgical objects created in the same regions at the same time. Given that mythology was so prominent in lay Christian life, why did it not penetrate official religious Christian spaces more often? It is worth pursuing this question, since it sharpens our understanding about what was at stake when such imagery was employed. Let us assume that church officials had reservations about the feasibility of disentangling mythological imagery entirely from religious beliefs. It would have been wise for them not to take any risks regarding the orthodoxy of the imagery used in churches themselves if Roman deities still mattered in the religious lives of individuals. In the context of the Iberian Peninsula, where the next two case studies are located, this is not an entirely implausible assumption to make. Besides many mentions of cults connected to rivers, forests, and stones in works such as De correctione rusticorum by Martin of Braga (c. 515–580), and works by Isidore of Seville and Braulio of Zaragoza (c. 590–651), sixthcentury testimony of the veneration of gods like Jupiter, Minerva, Mars, and Saturn intimate that worship of Roman deities also persisted.84 The episcopal network of the Iberian Peninsula in the fifth century was among the most thinly spread in the western Mediterranean.85 Bishops had little influence on and only limited access to many Christian communities, which often made their own religious decisions autonomously. To complicate the picture, the political and religious transition of Lusitania to a post-imperial province under Visigothic dominion happened during our period of examination. The Christian communities examined in this and the following section were located in southern Lusitania. The

83

84

85

See for instance Weitzmann 1960; Grabar 1961; Huskinson 1974; Hanfmann 1980; Guardia 1992; Elsner 1995, 251–260; Leader-Newby 2004, 123–171; Elsner 2007; North and Price (eds.) 2011, 691–742; Marcone 2011, 2014; Dunbabin 2012. On the notion of mythological imagery as a dying tradition in late antiquity, see Muth 2001, 116 and 97–98. On the worship of Roman gods, see Garcia González 1979, 665–667. On the Christian literary tradition condemning pagan cults in Hispania, see Madoz y Moleres 1945; Barney et al. (eds.) 2006; Riesco Terrero 1975, 109–115. Hispanic councils repeatedly debated the measures needed to end idolatrous practices; more or less explicitly pagan places of worship were mentioned in the councils of Braga II in 572 (canon I), Toledo III in 589 (canon 16), Narbonne in 589 (canon 14), Toledo XI in 681 (canon 1) and Toledo XVI in 693 (canon 2). See Vives, Marín Martínez, and Martínez Díez 1963; Caballero Zoreda and Sánchez Santos 1990, 436–437. Bowes 2005, 257. On the vital role of the laity in the Iberian Arian church specifically, see Mathisen 2014. See also Sabine Panzram’s forthcoming monograph ‘Christentum ohne Kirche’.

103

104

The Use of Non-Christian Imagery

conquest of the province’s capital Augusta Emerita (Mérida) in 469 assured the permanent presence of the Visigoths in Lusitania, but the political situation remained unstable. Lusitania’s northern territories fell to the Kingdom of the Suebi in the second half of the sixth century. Until the conversion of the Visigothic king Reccared I to the Chalcedonian creed in 587, Visigoths who adhered to Arianism possibly stressed their religious independence from the local Ibero-Roman population. This complicates determining which religious and ethnical identities the Christian communities of Myrtilis Iulia and Milreu held. Attitudes towards Roman mythology potentially differed depending on how Christians identified ethnically, if they did so at all.86 For the Barbarian part of the Iberian population, associating one’s Christianness with the Roman mythological tradition perhaps felt like an adoption, rather than a continuation, of Roman culture.87 But the question remains: why did church officials still reject the use of mythological scenes in Christian religious spaces when mythology no longer promoted a competing religious world view? Mythology was a cultural institution similar to the theatre or the circus in that by ‘practising’ it vocally, visually, or in writing, one enacted cultural practices which Christian authorities considered dangerous. This was not only because they were idolatrous (the problem of assuming the presence of deities besides God) but increasingly because they presented a violation of how these authorities wanted Christians to act (e.g. the fear of locura).88 An essential function of the communal church space, and the baptistery in particular, was to foster and nourish correct Christian ethical conduct. The demonstration of paideia within the ecclesiastical sphere works against any exclusivist Christian code of conduct. Nevertheless, mythological imagery was occasionally used in ecclesiastical spaces, notably in Lusitanian baptisteries. Some Christian communities exposed new generations of Christians to mythological topics during their initiation. The first case study, however, needs to be examined with special care because it is difficult to determine whether the building in question was a baptistery at all.

86 87

88

On Visigothic identity, see Wood 2012, 23–64; Heather 1999. A comprehensive study of Visigothic attitudes towards Roman culture, and mythology specifically, is missing to date. Isidore of Seville’s inclusion of mythological subjects in his history writing demonstrates a sense of Visigothic ownership of Roman and Greek culture and mythology. Merritt Bassett 1976, 288–289. For Visigothic culture see Ferreiro (ed.) 1999. Kaldellis 2007, 120–172.

Mythological Subject Matter

2.2.1 Myrtilis Iulia (Mértola): The Archaeological Evidence The antique Myrtilis Iulia, today Mértola, in the Alentejo, was a centre of mining from pre-Roman times onwards and between the fifth and seventh centuries also a significant centre of commercial trade between Lusitania and the wider Mediterranean.89 In 1877, the city of Mértola was examined by Estácio da Veiga for the first time; his observations are seminal for archaeological studies in Mértola and for Lusitania more generally.90 Mértola’s late antique Christian heritage is considerable. Substantial amounts of Christian funerary inscriptions, found in the basilica at the theatre, the basilica of Rossio do Carmo, and at a mausoleum in the suburbium, give vivid testimony to Myrtilis’s Christian community, which thrived until at least the eight century.91 About a century after da Veiga’s initial excavations, an artificial platform in the north of central Myrtilis began to be studied more extensively.92 The structure is a Roman cryptoporticus which overlooks the river Guadiana and was probably built in the late third or early fourth century. It has a length of 32 m, and the subterranean portion measures 2.7 m in width and has a medium height of 5.8 m.93 It is thought to have served as the foundation of the forum of Myrtilis and also as a subterranean cistern.94 A small civic basilica was also excavated at the end of the nineteenth century in the south of the forum.95 It has been speculated that in late antiquity the platform (Figure 2.14) may have functioned as the site of a bishop’s palace, although no records of a bishop in Myrtilis have been preserved.96 The platform was overbuilt in the Almohad period. The foundations of the Almohad dwellings still exist.97 A number of further excavations in the 1980s and 1990s in the northwestern corner of the platform brought to light a late antique complex of buildings. The complex (Figure 2.15) is extensively studied since 1990 by the Campo Arqueológico de Mértola under the direction of Virgílio Lopes.98 89 91 92 93

94

95 97 98

90 Torres 1998, 7–8; Lopes 2017a, 1379–1380. Lopes and Gómez-Martínez 2006, 271. An overview of the Christian infrastructure of Myrtilis can be found in Lopes 2015. Torres 1982. Torres 1982, 618–619. On the question of dating in regard to the building materials and techniques used, see Lopes 2003, 82–86. The same form of double-use is known from the cryptoporticus at Almuñecar in the province of Granada. Lopes 2003, 86. See also Torres 1982, 617. 96 Lopes 2003, 69. Lopes 2003, 69–71 and 82–86. Lopes and Gómez-Martínez 2006, 274–275, fig. 5. For Islamic Mértola cf. Macías 1996. I thank Virgílio Lopes and the CAM for having kindly shown me the excavation site in June 2014. For Lopes’s thesis on the site, see http://rabida.uhu.es/dspace/handle/10272/8053 last retrieved: 22 Aug 2023.

105

106 Estruturas Romanas IV - V V - VI VI - VII Muralha Medieval 10m 0

Figure 2.14 Plan of the platform and adjacent buildings of Myrtilis Iulia (Mértola). Lopes (2014), 176, fig. 69. © Virgílio Lopes.

Legenda: Mosaicos Pavimentos Argamassados Pavimentos Revestidos a placas de Mármore Muralha Blocos de Granito Muros Romanos Base de Coluna existente Base de Coluna - Sugerida Coluna Porta - Entrada do forum “Arcossólios” Clarabóia do Criptopórtico Entrada do Criptopórtico Galeria A Possivel Limite das estruturas Árvores Existentes Vedação Limite da escavação Cota Base (67.19)

Figure 2.15 Plan of the presumed baptismal complex I of Myrtilis. © Virgílio Lopes.

107

108

The Use of Non-Christian Imagery

Figure 2.16 Myrtilis, portico of the platform. Photography by the author.

A portico covered in mosaics (Figure 2.16) leads to an adjacent room called the ‘northern room’. This room is connected with a rectangular room of around 250 m2, oriented along a north–south axis, in which an octagonal font is centrally placed (Figures 2.15 and 2.17). The font sits in a lowered basin surrounded by an elevated colonnade. A small apse is attached to the colonnade on the eastern side. The basin has a length of 11 m but was reduced in the sixth or seventh century by 1.5 m due to the construction of 60 cm- and 70 cm-high elevations, possibly benches, in front of the northern and southern walls.99 The structure of the font is complex: an octagonal inner space with two deep steps spanning approximately 1 m is

99

The font’s floor lies at a height of 64.5 m and the colonnade at 65.5 m. The floor of the large basin is at a height of 64.94 m. Lopes 2003, 72–73, fig. 46.

Mythological Subject Matter

Figure 2.17 Myrtilis, assumed baptistery I. Photography by the author.

enclosed in a tier of apsidioles.100 Two flights of stairs confine the font westwards and eastwards. They lead into the basin but also create an even level between the font’s rim and the colonnade. Lopes believes them to be later additions because of their comparatively imperfect execution.101 The excavations further revealed a subterranean water conduit as well as fragments of an ascension pipe, found in proximity to the font.102 Late Roman pottery dating to between 320 AD and the end of the sixth century was found on the site.103 A fragment of terra sigillata, discovered in a piece of opus caementicium from the vault of the building’s apse, dates to somewhere between the mid-fifth and mid-sixth centuries.104

100 101 102

103

The maximal depth of the baptistery is 1.07 m. Lopes 2014, 179, n. 23. Lopes 2014, 194–195. Lopes 2003, 69, n. 143. There are no imprints preserved that would indicate a canopy. Lopes 2014, 184–185. 104 Lopes 2014, 198–200 and 183. Lopes 2017b, 132.

109

110

The Use of Non-Christian Imagery

Figure 2.18 Myrtilis, fragment of the mosaic pavement of the assumed baptistery I depicting Bellerophon killing the chimaera. Photography by the author.

The building was initially thought to be a bath with an octagonal basin, possibly set in a palatial context.105 Its identification as a bath has rightfully been questioned because of the lack of hypocausts and any other basins, as well as the unusual construction in which one basin sits within the other. Instead, it has been argued that the building was a baptistery.106 In support of this argument, Lopes points out the accessible structure of the font, several finds of potentially Christian furnishings in the surrounding area, and also the mosaic decoration of the building.107 Lopes tentatively dates the font to the first half of the sixth century by combining the evidence of the ceramic finds with typological comparisons to similar fonts.108 The mosaic fragment of Bellerophon killing the chimaera (Figure 2.18) has been preserved at the south-eastern corner of the colonnade (Figure 2.15). The Graeco-Roman hero Bellerophon is about to spear the

105

106 108

On the identification as a bath, see Torres 1982, 5. On the hypothesis that the building was a palatial structure, see Real 2012, 98. 107 Macías 1996, 54, n. 141; Lopes 2014, 196–197. Lopes 2014, 197. Lopes 2017b, 130; Lopes 2014, 204–207.

Mythological Subject Matter

Figure 2.19 Myrtilis, fragment of the mosaic pavement of the assumed baptistery I depicting two-petaled rosebuds inscribed in a scale pattern. Photography by the author.

lion-headed chimaera, which has a goat’s head on its back and a snakeheaded tail and is spitting fire out of all of its three snouts. The trunk of the chimaera is preserved as is a part of the attacking Bellerophon riding on the harnessed Pegasus. The figures have round, widely opened eyes and supple, full-bodied shapes. The scene is depicted on a white background using a limited palette of white, red, brown, and grey limestone. The tesserae have an average size of 5 mm.109 Only two more small mosaic fragments made of slightly larger limestone tesserae were preserved in the colonnade. They are on the eastern side in proximity to the Bellerophon fragment above the water conduit. A pattern of roses in semicircles can still be seen on the border of the colonnade (Figure 2.19). More centrally placed in the colonnade is the lower end of a flower on a white background facing northward, away from the Bellerophon scene.110 In his functional analysis of the complex surrounding the font, Lopes argues for understanding the portico, the northern room, and the font (with its surrounding colonnade) as parts of the same baptismal complex 109

Lopes 2003, 118.

110

Lopes 2003, 116.

111

112

The Use of Non-Christian Imagery

(Figure 2.15).111 In Lopes’s reconstruction, the catechumens walked through the portico, used the northern room as a waiting area or changing room, and then proceeded to the baptismal area. After the baptism, they received their first communion at the eastern apse and left the baptistery through the north-eastern exit.112 Within the circulation pattern reconstructed by Lopes, the Bellerophon mosaic would have had a rather unassuming position in the south-eastern corner. The colonnaded corridors could have been used, however, for processions or as waiting areas.

2.2.1.1

The Discovery of Font II – How Plausible Is the Identification of Font I as a Baptistery?

Before we can discuss the significance of the imagery of Bellerophon killing the chimaera for the potential baptismal space of Myrtilis, we need to take into account one of Lopes’s excavation discoveries. A second monumental font (Figure 2.20) was unearthed in 2013 only 25 m away from the font on the platform (henceforth called Font I). The newly discovered Font II sits in a building on a small hill south of the platform (Figure 2.14). It has also been identified as a baptismal font.113 Lopes plausibly reconstructs a rectangular baptistery of 262 m2 (Figure 2.21), orientated along an east– west axis, with a rectangular apse in the east and three chambers surrounding the apse.114 The octagonal font sits centrally in the baptismal space. Baptismal candidates reached the floor (with a depth of 1.52 m) via four steps.115 The font’s drain is located at a height of 1.16 m, allowing baptizands to stand in water.116 The font’s maximal size is 4.8 m, which clearly surpasses the other two octagonal fonts in Hispania: Font I from Myrtilis and Baptismal Font I of Barcino (Barcelona).117 Three bases of columns and several column shafts were found in situ on the southern side of the building.118 A limestone and marble pavement is still in place.119

111 112

113 115 117

118

Lopes 2014, 201–202, fig. 86. Lopes 2014, 201. The reconstruction does not conflict with the little available information on Iberian baptismal liturgies. On the first communion after baptism, see Godoy Fernández 1995, 53. On controversies on Iberian baptismal liturgy, see Lopes 2014, 202. More generally, see also Godoy Fernández 1989. 114 On Font II, see Lopes 2014, 250–312; Lopes 2017b, 132–140. Lopes 2017b, 132–134. 116 Lopes 2014, 291–295. Lopes 2014, 259. In the second or third construction phases of Font I of Barcino in the fifth and sixth centuries, the font is octagonal, abolishing a previous cruciform phase. Its interior space measures 3 m, its exterior space 4 m. Beltrán de Heredia and Bonnet 2007, 783 and 788–796. 119 Lopes 2014, 254. Lopes 2014, 254.

Mythological Subject Matter

Figure 2.20 Myrtilis, baptistery II, fifth or sixth century. Photography by the author.

Some surviving painted fresco fragments, provisionally dated to either the sixth or seventh century, give an idea of the interior decoration of the baptistery.120 Most notably, one fragment shows two faces, one crowned by a halo (Figure 2.22). Several more fragments of heads, limbs, drapery, and floral and geometric forms indicate that the baptistery was richly decorated and included a series of human bodies, at least one of them a holy figure. Because of the slight curvature of the frescoes, it is suggested that the fragments were part of a cupola decoration.121 When the font and the interior decoration are taken together, there can be little doubt that the building was used as a baptistery. While datable materials found on the site range from the early second to the seventh century, Lopes dates Font II to the second half of the fifth century onwards.122 He assumes that Font II predates Font I since it is suitably sized for adult baptism, which, Lopes argues, subsequently

120

121

Lopes 2014, 301–308. See also Lopes 2017b, 134. The only other securely attested Iberian painted baptismal decoration is preserved in Baptistery I of Barcino. See Albiol López 2013. On the complicated case of Santa Eulalia de Bóveda, see Rúa 2008; Blanco-Rotea et al. 2009. 122 Lopes 2014, 259–260, figs. 300–308. Lopes 2014, 273 and 261–266.

113

114 Figure 2.21 Plan of baptistery II of Myrtilis. Lopes (2014), 254, fig. 95. © Virgílio Lopes.

Mythological Subject Matter

Figure 2.22 Myrtilis, baptistery II, fragment of a fresco on plaster, 10 cm height
16 cm width
4 cm depth, sixth to seventh century. Lopes (2014), 302, cat. 145. © Virgílio Lopes.

declined in favour of infant baptism.123 This relative dating is problematic since much archaeological evidence suggests that adult and infant baptism co-existed in Lusitania and Baetica throughout the sixth century. It is true that the earliest Iberian council texts to generally characterize baptizands as infants date from the sixth century.124 However, several Lusitanian and Baetican baptismal fonts are attested in the same century, which were significantly deepened, apparently to allow for the immersion of adults.125 Others, built in the sixth or seventh century, were suitable for adult baptism from the beginning.126 The co-existence of small basins attached to baptismal fonts, typically found in Lusitania, is also mostly interpreted as

123 124

125

126

Lopes 2017b, 273. The case for parallel baptisms of adults and children being conducted until the seventh century is argued by Schneider 2011; esp. 1703; Fernandez Alonso 1955, 281; Iturgaiz 1968, 215. For the Iberian baptismal treatises cf. Saxer 1988, 531–566. On the deepening of the fonts of San Pedro de Alcántara, Torre de Palma, and potentially Casa Herrera, see Ulbert 1978, 175–176. On baptisteries in Hispania, see Iturgaiz 1968; Iturgaiz 1967; Godoy Fernández 1989; Schneider 2011. On rural baptisteries, see Ripoll and Velásquez 1999. These are the baptisteries of Alconétar, Casa Herrera, El Gatillo de Arriba, Valdecebadar de Olivenza, S. Pedro de Mérida, La Cocosa, El Germo, S. Pedro de Alcântara, Torre de Palma: Lopes 2014, 282–283.

115

116

The Use of Non-Christian Imagery

provision for contemporaneous adult and child baptisms.127 On the whole, a complete change of rite from adult to infant baptism by the end of the fifth century is unlikely, not least in light of the decentralized structure of the Iberian church. On the basis of its octagonal form, the dating of Font II cannot be narrowed down further than the fifth or sixth century.128 Most western Mediterranean octagonal examples are from these centuries, including Font I from Myrtilis and Font I from Barcino.129 Besides the date of construction of Font II, Lopes concludes that based on all the gathered archaeological data, Fonts I and II were most likely in use contemporaneously.130 The discovery of Font II complicates the identification of Font I as a baptistery. Indeed, there is the possibility that both fonts were used as baptisteries at the same time. Recent scholarship has established that multiple baptisteries in urban centres were fairly common throughout the western Mediterranean.131 However, the spatial situation in Myrtilis is unique, as nowhere else were two large baptisteries, apparently independent from any church building, built in immediate proximity to each other.132 It has been speculated that Myrtilis was the site of a significant, potentially heterodox, religious complex.133 Two Greek funerary inscriptions in Myrtilis (one dated to 544) testifying to two Christians who went by the name Eutyches have led to the speculation that a schismatic community of Eutychians took refuge in Myrtilis and used one of the baptisteries.134 Alternatively, it is not impossible that the two fonts are part of a joint Arian–Nicene complex. Arian and Nicene communities co-existed on the Iberian Peninsula until at least the conversion of Reccared I to Catholicism in the late 580s. However, the near absence of earlier evidence for Arian episcopacies attached to cities in Iberia makes it challenging to gauge how common Nicene and Arian churches operating in parallel really

127

128

129 131 132

133 134

Godoy Fernández 1986, esp. 131–134 with a review of the earlier literature. For a North African parallel, see the baptistery of Henchir B’Ghil. Ghalia 2016, 330–331, fig. 14; Baratte et al. 2014, 98–100. On the spread of octagonal baptisteries, see Brandt 2006b. See also Lopes 2014, 277–279; Ristow 1998; Khatchatrian 1962. 130 See n. 401 in this chapter. Lopes 2017b, 141. Beltrán de Heredia Bercero and Godoy Fernández 2017. The closest comparison are the Arian and Orthodox baptisteries of Ravenna, discussed in Chapter 3, which are several hundred meters apart from each other. Lopes 2014, 771–772; Torres 2015, 362–363; Lopes 2017b, 141–142. Torres 2015, 362–363.

Mythological Subject Matter

were.135 While a heterodox complex cannot be ruled out, the possibility also exists of separate baptisms for both sexes or for children and adults.136 Furthermore, both buildings might have co-existed but could have been used as baptisteries consecutively. Alternatively, the building on the platform could have had several purposes, one of them baptismal. However, we should at least consider the possibility that Font I was not used as a baptismal font at all.137 There is no church securely attested on or anywhere near the forum, but it could have been effaced in the Islamic period: structures underneath the twelfth-century mosque near the platform (Figure 2.14), today a church, give evidence that it stands on several earlier buildings, which could have been a temple and a late antique church.138 The distance between the presumed late antique church and Font I, measuring about 60 m, is unusually long. The find of the ascension pipe in proximity to Font I, moreover, suggests that the font used to be a fountain. On the Iberian Peninsula, only the considerably larger baptismal Font I of Barcino had a similar device.139 Morphologically, Font I’s structure of an octagon surrounded by apsidioles (Figure 2.17) in Myrtilis has no Iberian comparanda at all but is comparable with the font of Emona (Ljubiliana) in Slovenia and the northern Italian fonts of Albenga, Riva Ligure, and Taggia in Liguria, and Aosta in Valle d’Aosta.140 Besides the font’s unusual features, a survey of Christian architectural sculpture on the platform has brought very few results: first, a marble fragment showing a small part of a Greek cross inscribed in a circle, excavated in the possible baptistery, and second, a fragment of a mensa which could have belonged to an altar incorporated into one of the Almohad houses on the platform.141 It has been speculated further that a rectangular imprint, preserved in the pavement of the small apse in the east

135

136 137 138

139

140 141

Only at the Council of Toledo in 589 are both Nicene and Arian bishops from the same cities attested. See Mathisen 2014. The question of the possibility of parallel Arian and Catholic episcopacies in the same bishoprics after the conversion of Reccared I is discussed in Mikat 1984. On the case of the two baptisteries of Barcino, see Beltrán de Heredia Bercero 2017. Lopes 2017b, 141–142. I thank Gunnar Brands and Achim Arbeiter for their advice regarding this question. Lopes 2015, 384; Lopes 2017b, 140. The earliest phase of the mosque is suggested to have been erected after 1157. Ewert 1973, 245. I thank Achim Arbeiter for bringing this to my attention. The interior space of Font I of Barcino measures 3 m. On the font’s fountain, see Beltrán de Heredia and Bonnet 2007, 783, fig. 15. Lopes 2014, 204. Lopes 2014, 197. On the mensa fragment, see Lopes 2014, 212, cat. 29. On the cross fragment, see Torres 1991, 45; Macías 1996, 54, n. 141.

117

118

The Use of Non-Christian Imagery

of the colonnade, is that of an altar.142 The existence of an apse certainly supports a religious purpose of the complex, but none of the finds makes a baptismal use irrefutable. Indeed, Lopes proposes that the complex initially served as a bath and was subsequently adapted for a baptismal use.143 However, as mentioned above, the identification as a bath is problematic.144 It would nevertheless be worth pursuing the possibility that the fountain had a non-religious function; it could also have decorated the peristyle court of a representative residential building. With only two large steps in place, a baptizand would have had to overcome a depth of more than 1 m from the rim of the font to its floor, requiring steps of approximately 50 cm each.145 Indeed, it appears likely that the font was not designed to be climbed down; it might, however, have acquired a baptismal function later in time. I will explore this possibility by returning to the mosaic decoration of the assumed baptistery.

2.2.1.2

The North African Mosaic Workshop

The mosaics of the portico (Figures 2.16, 2.23 and 2.24), preserved on an area of 125 m2, have no parallel in Portuguese mosaic art.146 The long corridor is divided into a series of panels separated by diverse ornamental zones which appear to be thematically connected. The dominant theme is the hunt. A variety of animals are depicted – lions, a stag, a deer, a leopard, a hare, and ducks and other birds (notably an ostrich) – as well as a mounted huntsman carrying a falcon. The figures are again spread on a white background and have similar round, compact forms and large round eyes as the Bellerophon fragment. They are clearly the work of the same workshop. In the panels of the portico, however, the colour scheme is more varied, and a large number of red branches and plants are spread in 142

143

144 145 146

Lopes 2003, 71. The apse, the font, and the basin were decorated as an ensemble, since all were encrusted with alternating white and grey marble plaques (Figure 2.17). Altars are sometimes part of baptismal structures in the western Mediterranean. They were probably used for the first communion after the baptismal ceremony. On this liturgical practice in Iberia, see Godoy Fernández 1995, 53 and Godoy Fernández 1989, 633. On the Iberian Peninsula, a well-known but contested example is Casa Herera, near the Lusitanian capital of Augusta Emerita. See Lopes 2014, 202–203. Lopes suggests that Font I was built inside a previous frigidarium, but concedes that no other basins have been found in the area: Lopes 2014, 766. On the later addition of the stairs leading into the font, see Lopes 2014, 194–195. See 110 in this chapter. The maximal depth of the baptistery is 1.07 m. See Lopes 2014, 179, n. 23. On Myrtilis’s mosaics, Lopes 2014, 363–421; Lopes 2011; Lopes 2005; Lopes 2003, 98–125.

Mythological Subject Matter

Figure 2.23 Myrtilis, mosaic of the portico, detail of pair of lions facing a stone pine and wild animals. Photography by the author.

between the figures. The mosaics are of high quality, demonstrated for instance by the modelling of the stag (Figure 2.23). The outlines of its belly and neck are of the same white colour as the background, creating an illusion of the animal melting with its environment – a visual effect common in nature. The stag’s body is nevertheless perceptibly distinct from its environment thanks to the oval-shaped layout of the tesserae of the upper body, which contrasts with the more evenly arranged tesserae of the background. Several motifs indicate a North African origin of the mosaicists, likely from Africa Proconsularis or Byzacena.147 The sixth-century mosaic pavement of the left aisle of the three-aisled rural basilica of Hergla in Africa Proconsularis, 40 km north of Sousse, is the only other extant example of a mosaic floor depicting a hunting scene to combine the facing lion motif

147

On the possibility of African mosaic workshops in Lusitania, see Lancha 2004. For Myrtilis, see Lopes 2003, 103–117.

119

120

The Use of Non-Christian Imagery

Figure 2.24 Myrtilis, mosaic of the portico, detail of a rider carrying a falcon and various animals. Photography by the author.

and a huntsman carrying a falcon (Figures 2.25a and 2.25b).148 The pair of lions facing an (often umbrella-shaped) tree has several fifth- and sixthcentury parallels in churches in Tunisia and on the Balearic Islands, all thought to have been decorated by North African mosaicists (Figure 2.27).149 Animal hunting scenes, particularly popular in the Near

148 149

For Hergla, see Ghalia 1998; Lopes 2003, 113–114 and 104. Hergla panel G: Ghalia 1998, 68, fig. 21, 58–66 with comparisons to the basilica Es Fornàs de Torelló (Menorca) and the basilica de Isla del Rey. Other North African examples are from the north church of Cuicul (today in the museum of Djémila) and an apsidal mosaic from Carthage (today on display in the Musée du Bardo, Tunis). For Cuicul: Redjel and Tewfik 2010, 23. For Carthage: Yacoub 1994, 117, inv. no. 3574, fig. 126 and Yacoub 1996, 255–256, fig. 158 which suggests a sixth-century dating. A Near Eastern example of the lion motif is St Stephen’s Church in Umm ar-Rasas in Jordan: Piccirillo 1991, 83–85, fig. 44. Facing lions are also commonly used in synagogues in the Near East, for example, in Beth Alpha and Na’aran: Bowersock 2006, figs. 4.1, 4.4. For a comparative analysis of Balearic mosaic motifs and North

Mythological Subject Matter

Figure 2.25a Hergla, mosaic of the basilica, panel G of the left aisle. Ghalia (1998), 259, pl. XVII b-c. © Taher Ghalia.

Figure 2.25b Drawing of panel G of the basilica of Hergla. Ghalia (1998), 58, fig. 21. © Taher Ghalia.

East, were also frequent in late antique North Africa, including compositions of animals chased by horsemen on a white ground dotted with plants.150 Representations of ostriches, today extinct in North Africa, can also be found in North Africa and the Near East and must have added to the exotic flavour of Myrtilis’s hunting scene.151

150 151

African connections, see Orfila and Tuset 2003; Schlunk and Hauschild 1978, 57–58 and 75–85; Palol 1967, esp. 140. Yacoub 1996, 257–258, fig. 188. Lopes identifies the large bird in Myrtilis’s mosaic as a crane: Lopes 2003, 110. However, its pronounced legs ending in two toes permit only an identification of the bird as an ostrich, which is the only bird that has two toes. An exceptional motif is that of a lioness breastfeeding a snake. See Lopes 2003, 105. The ostriches in the north church of Cuicul and the early representation in the mosaic of the maison de l’âne in a private frigidarium in Cuicul in Numidia can be found in Pachtère 1911, 68–70. A hunting scene including an ostrich from a baptismal context can be seen in the Diaconicon baptistery at Mount Nebo in Jordan. See for instance Balty 2003, 165, fig. 10.

121

122

The Use of Non-Christian Imagery

Figure 2.26 Hergla, mosaic of the basilica, panel C of the right aisle. Ghalia (1998), 255, pl. XIII b. © Taher Ghalia.

North Africa also seems to be the point of origin for the iconographies found around the font. The motif of rosebuds inserted into scales (Figure 2.19) has direct parallels with a mosaic panel in the right aisle of the church at Hergla as well as to profane examples in Jebel Oust and Thuburbo Maius, all located in Africa Proconsularis.152 Moreover, representations of Bellerophon and the chimaera continued to be produced until a comparatively late period in North Africa. While the Iberian examples, to which we will return, originated sometime between the second and fourth centuries, scenes which draw on the myth of Bellerophon are preserved from sixth-century North Africa. A scene from Belalis Major in the region of Beja in Africa Proconsularis, today in the Musée Bardo in Tunis, shows Achilles riding the centaur Chiron and killing a stag with a bow and arrow (Figure 2.28). The scene is accompanied by the chimaera spitting fire from all three snouts as in the scene in Myrtilis. It is attributed to the late fifth or early sixth century.153 Another mosaic floor from Henchir Errich near Sufetula (Sbeitla) in Byzacena shows Bellerophon among a group of

152

153

On panel E, see Ghalia 1998, pl XV, figs. a,b ; on the apsidal mosaic in the ‘salle du jeu’ in Jebel Oust, see Fendri 1965, 548, fig. 12; on the fragment of portico VIII in the peristyle of the house of Nicentius in Thuburbo Maius, see Ben Abed-Ben Khader et al. 1985, plate XVIII, fig. 38b. The mosaic’s context is unclear. Yacoub 1996, 145, fig. 114; Yacoub 1995, 365, fig. 180.

Mythological Subject Matter

Figure 2.27 Mosaic of a pair of lions facing a stone pine from Carthage, fourth century, Musée National du Bardo, Tunis. Photography by the author.

Figure 2.28 Mosaic of Achilles riding the centaur Chiron from Belalis Maior (Henchir el Faouar), end of fifth to sixth century, Musée National du Bardo, Tunis. Photography by the author.

123

124

The Use of Non-Christian Imagery

mythological heroes. The scene is dated to after the Byzantine reconquest (533–534) for stylistic reasons.154 All things considered, it is most likely that a North African workshop produced the mosaics on the platform of Myrtilis in the fifth or sixth century.155

2.2.1.3

The Mosaic Decoration of the Platform as Evidence of Baptismal Use?

As the recent discovery of Font II complicates the identification of Font I’s function, we may ask whether the mosaic decoration would serve as evidence of a baptismal use of the space. Hypothetically, the mosaics could have been added when the building on the platform was transformed into a baptistery. Alternatively, the mosaics could have pre-dated the baptismal phase but still been considered appropriate for baptismal use. But is the iconography in accordance with an ecclesiastical use of the complex? The space is adorned with mosaics of three identifiable thematic groups: ‘rose pattern’ and ‘hunting scene’ on the portico and ‘Bellerophon killing the chimaera’ in the presumed baptismal room. None of these mosaics relates to the little that is preserved of Iberian baptismal decorations. However, given that the mosaicists were likely of North African origin or worked with North African models, we should also consider this tradition. The rose pattern is the most directly related to baptism. The motif of roses was commonly used for baptismal as well as funerary decorations in North Africa.156 The hunting scene has a single precedent in North African church decoration in the church at Hergla (Figures 2.25a and 2.25b). A large hunting scene in a baptismal context has only been preserved further in the Roman East, in the Diaconicon baptistery at Mount Nebo in Jordan (Figure 2.29).157 In the western tradition, however, the theme of the human hunter is (apart from Myrtilis) absent, and only one animal hunt is attested in a baptismal space. In the baptistery of Valentia, today

154 155 156

157

Bejaoui 2001, figs. 17, 21. On the sporadic African presence in Myrtilis, see Arce 2005, 358. Mosaics displaying rose motifs decorate the baptismal fonts of Basilica I at Dermech: Ristow 1998, n. 723, of the Basilica of St Vitalis and the apse of the baptistery of the church of Servus in Sufetula (Ristow 1998, 740–741), of Oued Zit (Ristow 1998, 736), of Henchir Sokrine (Ristow 1998, 717) and of Bît-el-Assa (Ristow 1998, 697). Raeck 1992. On the Diaconicon baptistery, see Piccirillo and Balty 1986, 63–64, pl. 5, fig. 44.

Mythological Subject Matter

Figure 2.29 Mount Nebo, Diaconicon baptistery, mosaic of a hunting scene, c. 530. Photography: Sean Leatherbury. © Manar al Athar.

Valence in south-eastern France, the octagonal font is adorned with a mosaic divided into two zones, an inner zone in direct contact with the font and an outer zone.158 The inner zone is designed with placid animals like lambs and deer, but also carnivorous animals such as a wolf or a bear drinking from water. In contrast, in the outer zone, a scene has been preserved of a bird of prey catching a running hare while another bird picks the hare’s eye (Figure 2.30). Due to a complicated conservation history, the dating of the mosaic has long been disputed, but a recent reexamination has shown that it was likely produced in the sixth or seventh century.159 In the decoration of Valentia, the paradisiac influence of the baptismal water is contrasted with the cruel reality of those yet to be saved. The loss of most of the mosaic decoration of the presumed baptismal space of Myrtilis prevents rash judgement about it having a similar configuration. Importantly, the mosaic of the basilica in Hergla, which comes closest to the portico’s pictorial programme, abstains from making clear distinctions

158 159

See the drawing by Laurent from 1866 in Blanc 1957, 110, fig. 19. Tardieu and Parron-Kontis 2006, 139–141. The dating was first problematized in Blanc 1957, 114–115. An eleventh-century fabrication was first suggested in Février 1957.

125

126

The Use of Non-Christian Imagery

Figure 2.30 Detail of the mosaic floor of the baptistery of Valentia (Valence), sixth to seventh century, Musée d’Art et d’Archéologie de Valence. Photography by the author.

between good and bad. In the mosaic pavement (Figures 2.25a and 2.25b and 2.26), panels displaying hunting or marine imagery alternate with ornamental panels and panels picturing spiral tendrils or medallions filled with animals.160 The combination of hunting scenes with imagery related to plenty and well-being can also be found in a range of eastern Mediterranean churches, suggesting that the hunting motif, like its counterparts, was associated with felicity and the good life.161 The basilica in Hergla does not offer any immanent narrative order but suggests that a free play of associations was acceptable.162 The human hunter (Figures 2.25a and 2.25b), a dog attacking a stag, or a maenad on a dolphin holding a whip (Figure 2.26), are interspersed in panels which

160 161

162

Ghalia 1998, fig. 9 and pl. XVII a. These include the monastery of Lady Mary in Skythopolis (Beth Shean), the church of St. Christopher in Qabr-Hiram, the aforesaid Diaconicon baptistery at Mount Nebo, the eastern church of Cyrene, and the basilica of Demetrios of Nicopolis. Raeck 1992, 42–48, esp. 47–48. Ghalia comes to a different conclusion, applying a sacramental model of interpretation to the decorative programme of Hergla: Ghalia 1998, 58–66, 75–90, 198.

Mythological Subject Matter

display predominantly peaceful interaction between various kinds of animals, including marine and forest animals.163 Thus, the withdrawal from visual strategies which were straightforwardly applicable to Christian allegory could have been just as much of a North African import as the inclusion of exotic motifs like ostriches and pairs of lions.164 A loosely structured narrative frame would also have suited the incorporation of the Bellerophon scene, which is not securely attested for any ecclesiastical space. Where it was used in sixth-century North Africa, it has no clear provenance or the architectural context could not be determined.165 When regarded with ‘North African’ eyes, the decorative concept of Myrtilis’s portico might have looked unconventional for a Christian space but was not out of the question. The rose pattern, the lion motif, and the hunting scene link the mosaic with North African church decorations. In consideration of the pictorial programme as a whole, an ecclesiastical function of the portico, at least, can neither be ruled out nor confirmed, which in itself is a signum of North African ecclesiastical art.166

2.2.1.4

The Christian and Graeco-Roman Bellerophon Scene

As discussed above, it is indeed possible to bring the thematic groups of ‘rose pattern’ and ‘hunting scene’ in line with a baptismal use of the font on the platform. A question which is more difficult to answer, however, is why the scene of Bellerophon killing the chimaera – a scene from a story handed down through Homer and Hesiod – would have been chosen, or reused, in a fifth- or sixth-century baptistery.167 To examine this question, it seems helpful to think again about the cultural affiliations the motif would have had for contemporary viewers. On the textual level, there is no evidence in Iberia or elsewhere of fifth- and sixth-century Christian interpretations of the myth of Bellerophon.168 The few Christian references

163 164

165

166 167

168

See Ghalia 1998, panel C on pl. XIII, panel G on pl. XVII, panel C’ on pl. XIX. An interesting parallel is offered by the ‘North African’ narrative strategies of Balearic church mosaics. For an example of the current state of research, see Orfila and Tuset 2003. On the lack of provenance of the Beja mosaic, see Yacoub 1996, 145. On the finds at Henchir Errich, see Bejaoui 2001. Cf. Chapter 1, 25–27. Homer, Iliad, 6,152–204; Hesiod, Theog., 319–325. On Bellerophon in Graeco-Roman mythology Hiller 1970, 9–13. On late antique Christian and pagan commentators of Bellerophon, see Brandenburg 1968, 60–61.

127

128

The Use of Non-Christian Imagery

which are preserved do not offer any reason for a Christian reinterpretation of the hero; instead, they indicate that he was seen critically.169 Amongst all late antique textual and visual sources related to Bellerophon, floor mosaics make up the largest group.170 The theme of Bellerophon fighting with the chimaera is nowhere as often documented on mosaics as in Iberia.171 All in all, Portugal (2) and Spain (6) can claim eight depictions, followed by England (5) and France (3), whereas Byzantine examples are much rarer.172 In Iberia, the preserved examples in Loma del Regadío (Teruel),173 Ucero (Soria),174 Torre de Bell-Lloch (Gerona),175 Puerta Oscura (Malaga),176 Sahelices el Chico (Salamanca),177 Augusta Emerita (today Mérida),178 Conimbriga,179 and Myrtilis point to a wide transmission on the peninsula. Except for Myrtilis and Augusta Emerita, all have been attributed with certainty to a domestic context, either a rural villa or a luxurious urban residence.180 Furthermore, all examples except for Myrtilis have been dated to between the second and fourth centuries. No other Iberian depictions of Bellerophon from the fifth or sixth century have come to my attention. What can be concluded from the extant Hispanic examples is that the Bellerophon scene was somewhat popular in wealthy households. When trying to place the image culturally, one association which we can relatively safely presume the image would have held for at least some viewers was that of traditional Roman elite culture. Scholars have debated for a long time over the Christian significance of Bellerophon. The question has been under discussion since the discovery of the fourth-century floor mosaic of a villa in Hinton St Mary (Dorset) in 1963.181 The mosaic became famous for the contemporaneous depiction of the bust of a man crowned with a chi-rho, usually interpreted as Christ,

169 170

171

172

173 174 175 177 180

181

Justin Martyr, Apol. 1, 21; Gregory of Nazianzus, Or. 43, 21. For textual evidence see the poems of the Anthologia Latina on Bellerophon 97 and on the Chimaera 98: Riese 1894, 124. On material culture beyond mosaics, see Raeck 1992, 118–121. Alarcão et al. 1992, 42–44. On Bellerophon in Iberia, see Blázquez Martínez 1994; Blázquez 1981, 77. A slightly outdated list is provided in Blázquez Martínez 1994, 43. France: Autun, Reims, Nîmes; England: Lullingstone, Hinton St. Mary, Frampton as well as the more recent finds at Croughton (discovered in 1991) and Boxford (discovered in 2017). For Croughton, see Cosh and Neal 2002; Dawson and Neal 2010. For Boxford, see Beeson, Nichol, and Appleton 2019. Azuara Galve and Villagordo Ros 2007. Blázquez and Ortego y Frías 1983, 50–51, 93, fig. 3. 176 Balil 1960b; Blázquez Martínez 1994, 282. Blázquez 1981, 77–80. 178 179 Chamoso 2007. Álvarez Martínez 1990, 98–101. Alarcão et al. 1992, 41–44. A domestic context is also a likely option in Augusta Emerita. See Álvarez Martínez 1990, 98–191. Toynbee 1964.

Mythological Subject Matter

and the scene of Bellerophon killing the chimaera.182 Two more midfourth-century mosaics in villas in Frampton (Dorset) and Lullingstone (Kent) attest to the scene being used by villa owners who chose Christian imagery as domestic decoration.183 Apart from the mosaic of Myrtilis, these are the only known examples of the theme’s connection to Christian patrons. The potential Christian significance of the image of Bellerophon killing the chimaera has received so much attention, however, that the depiction is still discussed as a possible marker of Christian identity, at least in the context of fourth-century England.184 Several suggestions have been put forward about what constituted the Christian message of Bellerophon, but the only hypothesis which has found wide acceptance is that the image of Bellerophon killing the chimaera was used to represent the fight of good against evil.185 The same interpretation has been proposed for Myrtilis.186 This basic idea has been developed in different directions: Bellerophon’s battle with the Chimaera has been taken to represent the poles of light and darkness,187 for instance, or life and death,188 or even Christianity (Bellerophon) and paganism (chimaera).189 Uniting all these readings is an exclusive concern with the allegorical message of the Bellerophon scene in a Christian context. The question that has been asked is: what is this image about? This is, of course, crucial for understanding an image’s significance. However, focusing on Christian meaning alone is not sufficient. All suggestions fail to explain why Christians in fourth-century England or sixth-century Portugal might have been interested in representing the fight of good against evil in the form of Bellerophon spearing the chimaera. We should grant for the possibility that Christians privileged the cultural affiliation of the image over its Christian interpretability. In other words, Bellerophon might have been of interest because he was a particularly well-liked mythological hero associated with Graeco-Roman elite culture, whereas the specific narrative was secondary as long as it allowed for a Christian interpretation. 182

183 184

185

186 189

Pearce’s argument against the traditional interpretation as Christ includes a useful overview of the debate: Pearce 2008. For Frampton and Lullingstone, see Henig 2012, 138–141. Cosh and Neal 2002, 236. See also Henig 2012. The idea of Bellerophon as a Christian symbol was, however, introduced much earlier. See Simon 1966; Huskinson 1974, 73–78. A critical perspective is offered in Brandenburg 1968 and Raeck 1992, 99–121, esp. 104–112. Huskinson dismisses the idea of Bellerophon as a pagan type of Christ: Huskinson 1974, 78. Hiller speculates that Bellerophon may be an image of imperial triumph, but could also be a ‘slayer of evil’: Hiller 1970, 82. 187 188 Lopes 2003, 117–118. Hiller 1970, 88. Huskinson 1974, 74. Simon 1966, 901.

129

130

The Use of Non-Christian Imagery

As Wulf Raeck has shown, late antique domestic representations of Bellerophon’s fight with the chimaera tend to integrate the scene into the larger subject of the hunt without highlighting the mythological content.190 Wealthy patrons seem to have associated the hero first and foremost with the prestige pastime of the hunt and the connected aspects of a virtuous and pleasant life. The mosaicists of Myrtilis’s platform possibly aimed to express a similar idea. Due to the considerable loss of the mosaic pavement in Myrtilis, this of course remains speculation. In any case, it is telling that researchers find it difficult to decide whether the building ensemble on the platform fits better with a palatial or a baptismal structure.191 Also, Myrtilis’s Bellerophon is the sort of image one would expect to find in a domestic Roman structure. If the assumed baptistery was not a transformed residential building, it looks as if it was meant to be reminiscent of one. In conclusion, the Bellerophon scene does not serve as an indicator of a Christian use of Myrtilis’s complex. However, it can certainly be interpreted in Christian terms. There is some evidence to suggest that the complex may have served as a baptistery at some point in its history, although none of it is definitive. The prevalence of secular and even mythological subject matters aligns well with the possibility of an initial or secondary Christian function of the space, especially considering the likely North African origin of the mosaicists. The mosaic design on the platform resembles the decorative concept seen in the basilica in Hergla in Africa Proconsularis. Both are characterized by the selection of themes prevalent in Roman domestic culture. If Myrtilis’s complex was indeed a baptistery or was transformed into one, I propose that the precise specificities of the Bellerophon motif were of secondary importance. The inclusion of the Bellerophon scene in the mosaic floor arguably occurred because it harmonized with the dominant theme of the hunt and was thereby connected to Roman elite culture. Similar to the basilica of Hergla or the villas of Hinton St Mary, Frampton, and Lullingstone in England, it is reasonable to assume that Christian interpretations were found for scenes with secular or mythological roots when needed. However, what stands out to me as more significant is that these Christian communities and families allowed themselves to engage with and be surrounded by mythological scenes. They 190 191

Raeck 1992, 104–112. Against a baptismal use, see Torres 1982, 5; Real 2012, 98. For a baptismal use Macías 1996, 54, n. 141; Lopes 2014, 196–197.

Mythological Subject Matter

demonstrated a commitment to classical Graeco-Roman culture and asserted independence from the normative Christian ethos to refrain from mythology. By showcasing paideia in the baptistery, Myrtilis Iulia’s Christian community would have allowed themselves to substantially redefine what correct Christian behaviour meant to the city.

2.2.2 Milreu: The Archaeological Evidence The last section of this chapter examines the reuse of formerly nonChristian mosaic friezes in an uncontested baptismal setting in the Lusitanian village of Milreu/Estói. The subject matter of this particular mosaic is the marine thiasus, the retinue of Poseidon and other marine divinities. The section focuses on the reconstruction of the decoration of the baptismal chamber and its significance for Milreu’s Christian community. The existing site predetermined a range of possibilities for the community to interact with the building. I will show that the Christian occupants chose depictions of hybrid sea creatures to adorn the interior and exterior of the baptismal area.

2.2.2.1

The Pre-Christian Archaeological Evidence

The baptistery was attached to an antique domed building of unclear original function, which scholars have variously interpreted as being a sanctuary, a mausoleum, or a dining hall (Figure 2.31).192 It was built opposite a Roman villa in the village of Milreu/Estói, about 8 km north of Faro, the ancient municipium of Ossonoba in Lusitania.193 The monument was reused as a church from the second half of the fifth or the sixth century at the latest. At this point the baptistery was added to the complex. There is archaeological evidence that a later settlement existed within the building between the Arabic conquest of Lusitania and the early tenth century.194 192

193

194

Hauschild 1964. For an updated assessment of the religious complex, see Oepen 2012, 96–109. On the function of the building as a temple dedicated to a water deity or as a nymphaeum, see Hauschild 1984, 2002; Teichner 2006, 83. On the building’s function as a mausoleum, see Graen 2004, 2005a, 2005b; Griesbach 2009. On the building’s function as an aula, see Teichner 2008b, 2011. An analysis of the religious authority of Hispanic rural elites on their estates, outdoing urbanoriented bishops, is offered in Bowes 2008, 179–187. Teichner 1994; Sidarus and Teichner 1996; Teichner 2006; Hauschild and Teichner 2002, 57–58.

131

132

The Use of Non-Christian Imagery

Figure 2.31 Milreu/Estói, late Roman monument, first half of the fourth century. Photography by the author.

The villa itself dates back to the first century AD. Both villa and monument have been excavated twice: in 1877 by Estácio da Veiga (Figure 2.32) and from 1971 to 1998 by Theodor Hauschild (Figure 2.33).195 The complex is located above the course of the Rio Seco close to one of the roads coming from Ossonoba.196 The villa is divided into a residential area with a peristyle and a large bath, and a pars rustica where wine and olive oil were produced.197 A fundamental refurbishment

195

196

Apart from the plan of the villa (Figure 2.32), Estácio da Veiga did not publish on the excavation. Excavation records are kept at the archive of the Museu Nacional de Arqueologia in Lisbon. I thank the MNA for giving me the opportunity to consult the archive as well as the stored archaeological material from Milreu. On the excavation history, see Oepen 2012, 97–109; Teichner 2008a, 98–102. On the archaeological interventions of the nineteenth and early twentieth centuries, see Teichner 2008b, 732–733. 197 Hauschild and Teichner 2002, 14. Teichner 2008a, 95–270.

Mythological Subject Matter

133

Figure 2.32 Detail of the plan of the excavation of Milreu under Estácio da Veiga (1877). The hexagonal basin is located in the centre of the monument (D). Da Veiga (1880). © DAI Madrid, D-DAI-MAD-Repro-03-2023.

134

The Use of Non-Christian Imagery

Figure 2.33 Plan of the villa and the monument of Milreu. Drawing: Theodor Hauschild. © DAI Madrid, D-DAI-MAD-A-53-Z-0033.

took place during the fourth century.198 The peristyle and a frigidarium were decorated with mosaic floors, several of which depict a variety of sea animals, mainly fish. The mosaic, originally covering the full 18 m length of the peristyle, has been preserved to a length of 6 m to this day. In addition, a monumental construction of opus testaceum (Figure 2.31) was built on the other side of the road, at a distance of 10 m from the villa, at some point in the first half of the fourth century.199 An almost quadrangular cella with an apse oriented to the south was covered by a cupola at a 198

199

This is the case of many other Lusitanian villas at the time. Besides the general tendency of high social classes to move to the countryside, an earthquake potentially triggered the wave of renovations. See Hauschild and Teichner 2002, 12; Graen 2004, 68. The suggestions for a dating of the monument differ and have changed over time; however, the first half of the fourth century is now generally agreed upon as the rough date of creation. Cf. Hauschild 1984, 287, who opts for a dating from the end of the third to the beginning of the fourth century. More recently, Hauschild suggested a dating in the middle of the fourth

Mythological Subject Matter

Figure 2.34 Milreu, mosaic frieze of the podium on the north-western side. Photography by the author.

height of more than 10 m. It was raised on a porticoed podium which was surrounded by a walled court. Fragments of opus sectile and a large number of coloured glass tesserae, many of them gilded, indicate that the inner walls and the cross vault had been richly decorated.200 Another monumental mosaic vault in Iberia has been preserved in the cupola of Centcelles near Tarragona.201 The podium supporting the monument is 1.2 m in height and is covered by a fragmented mosaic frieze 67 m long (Figure 2.34).202 The frieze’s mosaic decoration clearly corresponds with the maritime motif of the villa mosaics. The frieze mosaics on the northern façade are the best preserved and feature diverse types of fish as well as

200 201

202

century. See Hauschild 2002, 216. A dating in the first quarter of the fourth century is preferred by Graen 2004, 69. Teichner 2008a, 267; for a reconstruction of the extant opus sectile 263, fig. 136. For a recent discussion of the cupola of Centcelles, see Arbeiter and Korol 2016. See also Arce 2002; Schlunk 1988. Hauschild 2008, 21.

135

136

The Use of Non-Christian Imagery

dolphins. Many smaller fragments on the other sides of the frieze depict bodies of fish and other not clearly identifiable water creatures.203 Several functions have been suggested for the monument. In the past it has been regarded as a sanctuary, a nymphaeum, or a temple dedicated to a water deity, but more recently it has been argued that it was either a mausoleum or a dining hall. Hauschild has demonstrated that the form of the monument derives from Gallo-Roman ambulatory temples and argues that its decoration is best explained in the context of a cult of water: the frieze displays marine iconography, the middle of the cella accommodates a hexagonal basin (lost today but recorded by Estácio da Veiga; Figure 2.32), and what may be a ditch around the monument might have been used as a water basin.204 If the building did initially serve as a place of pagan worship, it would be one of only two known Hispanic examples of the Christian use of a previous temple structure.205 More recent scholarship has problematized the fact that no pre-Christian devotional objects or inscriptions have been found on the site. Furthermore, the conduits, which fed the basin at Milreu with water, destroyed parts of the mosaic decoration of the frieze and were likely installed after the initial construction.206 At the same time, a series of typologically similar Iberian buildings at the villa of São Cucufate in the Alentejo, the villa of Quinta de Marim in the Algarve, the villa of Maternus at Carranque in Castilla y León, and the villa at Los Castillejos in Extremadura have been brought into the discussion. This group of monuments shares many features with a series of monumental mausolea in Rome.207 The maritime iconography on the podium could also have had a funerary purpose.208 The group of Iberian monuments has further been interpreted as being representative banquet halls.209 Marine life is indeed a recurrent theme in the mosaic decoration of Iberian triclinia, but marine thiasus scenes are largely absent from

203

204

205

206 208

209

For a detailed description and hypothetical reconstruction of the frieze, see Hauschild 2008, 21–24, fig. 8. The arguments were first summarized in Hauschild 1964. Unfortunately, no information survives about the basin in the cella, which might have been added as late as in the Islamic period. Hauschild and Teichner 2002, 57–58; Teichner 2006. In Tarragona, church reuse of temples is attested from the eighth century onwards: Arce 2011b, 202. A doubtful case is Santa Eulalia de Bóveda near Lugo. There is much uncertainty as to the chronology of the building, claimed to have first been a pagan sanctuary. See BlancoRotea et al. 2009. 207 Graen 2004, 70; Graen 2005b, 378. Graen 2005b, 2005c. See also Griesbach 2009. On the theme of the marine thiasus on Roman sarcophagi, see Zanker and Ewald 2004, 117–134. Teichner 2008a, 516; Teichner 2011, 296–298.

Mythological Subject Matter

Figure 2.35 Milreu, view of the monument from the north-west. Photography by the author.

them.210 In light of these arguments, the structure’s proposed temple function is doubtful. As neither the mausoleum thesis nor the aula thesis have been conclusively proven thus far, I therefore simply refer to the preChristian building as a monument.

2.2.2.2

The Christian Phase of the Monument and the Baptismal Space

The phase of Christian use of the site is difficult to interpret archaeologically.211 The existence of a baptismal font situated in the north-west of the monument between the platform and the outer wall of the court, not far from the entrance (Figures 2.33 and 2.35), is the most striking argument for 210

211

Iberian triclinia decorated with fish and other sea animals have been preserved in the Casa del Anfiteatro in Augusta Emerita, in Vega del Ciego in Asturias, in Vega Baja de Toledo, and in the Casa dos Esqueletos in Conímbriga. See Guardia 1992, 309. For an exhaustive overview of marine mythological Roman mosaics in Portugal, see Mourão 2008, 2010, 2011. On the Christian reuse, see Hauschild 1980.

137

138

The Use of Non-Christian Imagery

Figure 2.36 Milreu, baptismal font above a grave, second half of the fifth or early sixth century. Photography: DAI Madrid, 1971. © DAI Madrid, D-DAI-MAD-FP-K0738–5270112/71.

the transformation of the monument into a church. The font (Figure 2.36) has been preserved: it is at least 92 cm deep and has an almost quadrangular floor and two sets of three narrow steps that extend the font to a length of 1.63 m.212 The font corresponds to the description of the ideal shape of a baptismal font given by the seventh-century Spanish bishops Isidore of Seville and Ildefonsus of Toledo (607–677); they state that the font should have three steps down and three up, with the floor counted as the seventh step and signifying Christ.213 The font of Milreu belongs to a common Iberian type of baptismal font of rectangular, oblong shape and approximately quadrangular floor, with two sets of stairs which were most frequently built in the second half of the fifth and the sixth century. There is a particular concentration of this type of font in Lusitania.214 Astonishingly, the northern stairs of the font are placed on the intact grave of a young adult (Figure 2.36). Remarkably unharmed by the font’s construction, the grave did not contain grave goods and cannot be dated with any certainty.215 It has been suggested that the grave was part of a Christian graveyard established after the monument had lost its original function.216 Most of the graves face eastwards and are located within the walled court of the monument, with the densest concentration in the northern entrance area.217 In the north-east of the courtyard, an almost 212 214

215 217

213 Hauschild 1980, 190. Isidore, De eccl. off., 2, 25, 4; Ildefonsus, De cogn. bapt., 110. Hauschild 1980, 200–201; Ulbert 1978, 156; Lück 2018, 79–83. Lück left the baptismal font of Milreu unconsidered. 216 Cf. Hauschild 1980, 201. On the additional burial sites, see Hauschild 1980, 204–207. Teichner 2008a, 255, fig. 29a.

Mythological Subject Matter

quadrangular construction with a niched apse in the east which contained three burials sticks out (Figure 2.33). It has been compared to similarly constructed Iberian mausolea and has been tentatively dated to the fifth century.218 This would make the mausoleum the earliest attestable Christian construction at Milreu. Undocumented finds of grave goods from the late nineteenth and early twentieth century testify to the site’s use as a Christian graveyard up to the late sixth or early seventh century.219 An excavator of this early period described the findings of ceramics and a buckle as ‘Visigothic’.220 The complexity of ethnic attributions to archaeological findings has been fully acknowledged only more recently, however.221 Due to a lack of archaeological records of the buckle and of information about the dress culture of Romans and Visigoths more generally, the attribution as ‘Visigothic’ can only be taken as an indication of the historical period in which the goods originated, not as evidence for the ethnic affiliation of the deceased.222 Whether the Christian occupants of Milreu’s monument were HispanoRomans, Barbarians, or both remains as unverifiable as the terms they would have used to describe themselves.223 Other material culture which may date back to the site’s Christian occupation is difficult to assess due to the early excavation date. Estácio da Veiga’s excavation brought to light various marble furnishings in the interior of the monument, including a marble plaque depicting two birds drinking from a kantharos.224 Furthermore, a group of bricks decorated with scratched figures like a pair of peacocks, a dolphin, a kantharos with ivy leaf, leaf ornaments, and, most interestingly for our question, a fragment of the frontal part of what looks like a horse (Figure 2.37), may also

218 219

220 221

222

223

224

Oepen 2012, 107, n. 530. In the Islamic period, the site was also used for burials. Teichner 2008a, 256. See also Hauschild 1980, 196; Teichner 1994, 94–95. Lyster Franco 1942, no pagination. The question of whether Visigothic metal work might signal ethnic affiliation is summarized in Ripoll 1999. On expressions of identity and the lack thereof of Hispanic Romans and Visigoths, see Arce 2018. For a summary on Visigothic dress, see Arce 2011a, 178–181. On the Barbarian occupation of Iberian villas more generally, see Chavarría Arnau 2004, 74–75. On Visigothic identity, see also Wood 2012. Estácio da Veiga calls the find spot of the plaque the ‘sala grande’ – likely the interior of the monument. In the same space he records many more marble fragments, among which are two Corinthian capitals: Estácio da Veiga 1885, no pagination. The plaque is preserved today in the collection of the Museu Nacional de Arqueologia in Lisbon (inv. no. E650).

139

140

The Use of Non-Christian Imagery

Figure 2.37 Brick fragment from Milreu, fifth to sixth century, Museu Nacional de Arqueologia, Lisbon, inv. no. 997.17.1. Photography by the author. © Museu Nacional de Arqueologia.

originate from there.225 The stretched front legs of the animal are reminiscent of conventional depictions of hippocamps, mythological creatures with the upper body of a horse and the lower body of a fish.226 Single horses are rarely shown in a running posture.227 The brick might therefore have mirrored the mythological repertoire of the mosaic decoration of the monument, as we will see shortly. Returning to the baptismal font, the ground plan made in 1880 after the first excavation by Estácio da Veiga (Figure 2.32) shows that the font had been integrated into a room formed out of the wall of the podium in the east, the outer wall which confines the court in the west, and two newly built walls in the south and north.228 This structure was about 4 to 5 m in

225

226 227

The bricks are preserved at the Museu Nacional de Arqueologia in Lisbon (inv. nos.: 15262, 2001.15.12, 997.17.1, 2006.372.14, 2006.372.12, 2001.15.14), which attributes them to the excavation of Milreu. Estácio da Veiga’s record entry, however, while listing thirty-one ornamented bricks alongside finds from Milreu, does not specifically mention Milreu as their find spot, but gives only the generic location of ‘Faro’. This also concerns the sixty-seven bricks stamped ‘VER FRONTINIANI’. See Estácio da Veiga 1878, 44. A selection of bricks is reproduced in Teichner 2008a, 266, fig. 139. I thank Gunnar Brands for bringing this to my attention. 228 Compare 88-89 in this Chapter. Hauschild 1980, 201; Teichner 2008a, 267.

Mythological Subject Matter

Figure 2.38a Milreu, wall of the podium at the height of the former baptismal chamber. Photography by the author.

width, hence, of a size of around 24 m2 in total. Today the walls to the north and south of the font are only partially traceable, while substantial parts of the podium wall in the east (Figures 2.38a and 2.38b) and, to a lesser extent, the outer court wall in the west still exist.229 A breach through the wall of the podium likely indicates that the entrance into the baptistery was via the church (Figure 2.36).230 The construction of a special baptismal room is not surprising, as a separate space was usually provided for the baptismal rite even in modest Iberian churches.231 What is perhaps more surprising is that the Christians who built the baptistery used the wall of the podium for its construction. This choice made the monument’s mosaic frieze part of the interior decoration of the baptismal chamber. Unfortunately, the mosaic frieze in the area of the baptistery is almost entirely destroyed. However, a drawing and two 229 231

Hauschild 1980, 192 and 201–202. Schlunk and Hauschild 1978, 48.

230

Hauschild 1980, 202.

141

142

The Use of Non-Christian Imagery

Figure 2.38b Milreu, fragment of the mosaic frieze depicting a curled tail at the height of the former baptismal chamber. Photography by the author.

Figure 2.39 Drawing of a mosaic fragment of a triton and a sea lion at Milreu, made during the excavation of Estácio da Veiga in 1877, Archive of the Museu Nacional de Arqueologia, Lisbon, inv. no. 25b. © Museu Nacional de Arqueologia.

Mythological Subject Matter

143

Figure 2.40 Photography by A.M. Xavier de Meireilles (1890s) of the mosaic fragment on a wall west of the monument of Milreu. Archive of the Museu Nacional de Arqueologia, Lisbon, inv. no. Foto25B. © Museu Nacional de Arqueologia.

photographs, made between 1877 and the 1890s, record parts of the decoration of the frieze in this area (Figures 2.39 and 2.40).232 They show the fragment of a triton with raised hands, holding a shell or a bowl in his right, and another fantastic sea creature with the torso of a lion and a curled, long fishtail facing the triton with an open and snarling snout. At the time the photographs were taken, the upper parts of the podium had been already removed, resulting in the loss of the faces of both creatures. Today the frieze has been destroyed. According to the caption in the photograph from the 1890s (Figure 2.40), the scene was situated on a wall

232

Drawing 25B, executed in 1877, is today in the archive of the Museu Nacional de Arqueologia in Lisbon. The drawing is not entirely accurate because horse legs were added to the figure of the triton. See Graen 2005b, 374, fig. 14. A first photograph was published in 1882 in the journal O Occidente, 138. http://hemerotecadigital.cm-lisboa.pt/OBRAS/Ocidente/Ocidente .htm (last retrieved 23 May 2023). A second photograph was taken in the 1890s. See Graen 2005b, 376.

144

The Use of Non-Christian Imagery

Figure 2.41 Milreu, detail of the mosaic frieze on the eastern side of the podium. Photography by the author.

in the west of the monument, the same site as the baptistery’s location: Concelho de Faro – Freguezia de Estoy. MILREU. Monstros mythologicos figurados em mosaico no muro de oeste de um edificio antigo. Other fragments on the podium confirm that the mosaic decoration included, besides fish, elements of a marine thiasus and possibly of a fishing scene. Still extant are a long, curled tail with spikes on the eastern side of the monument, similar to the tails of the triton and the sea lion (Figure 2.41), as well as a little human foot on the front side of the frieze (Figure 2.42). A fragment of the forepart of a boat is stored in the Museu Nacional de Arqueologia in Lisbon (Figure 2.43).233 Moreover, in the immediate surroundings of the font, at the right hand of the breach through the podium, is a small mosaic fragment of yet another curled tail (Figures 2.38a and 2.38b). The shape of this fragment matches the shape of the triton’s tail recorded in the photographs (Figures 2.39 and 2.40). The fragment’s curly tail consists of a dark upper 233

Graen identifies the boat as the Vegeiia or Placida type: Graen 2005b, 375, fig. 22. For a detailed catalogue of all archived fragments, including an additional fragment of a hoof, possibly from another sea creature and possibly from Milreu, today in the Museu Santa Rocha of Figueira da Foz, see Graen 2005b, 373–377.

Mythological Subject Matter

Figure 2.42 Milreu, detail of the mosaic frieze on the north-eastern front side of the monument. Photography by the author.

side and a white lower side and exits from the lower abdomen, which is coloured yellow. This colour scheme is the same as that preserved in a drawing made during Estacio da Veiga’s excavation (Figure 2.39). It has been suggested that the small fragment next to the font may be the last remains of the triton on the photograph from 1890. The photograph also documents a stone structure located below the triton (Figure 2.40). This structure was identified as part of the now-lost walls of the baptistery.234 However, closer inspection of the shape of the simple guilloche below the tail of the preserved fragment in situ and the mythological scene recorded in the photograph makes it evident that the two fragments cannot be part of the same scene, as the pattern significantly differs in steepness between the two. The guilloche covering the podium is steeper than the one in the photograph in all the parts which are still in place (Figures 2.38b and 2.40). It is also questionable whether the guilloche on the photograph ever contained as many different colours as the one composed of different shades of rose, yellow, brown, black, and white on the podium. Although 234

Graen 2005b, 376.

145

146

The Use of Non-Christian Imagery

Figure 2.43 Mosaic fragment of the forepart of a boat from Milreu, Museu Nacional de Arqueologia, Lisbon, inv. no. 18701. Photography by the author. © Museu Nacional de Arqueologia.

the reduced colour scheme of the guilloche on the photograph could arguably be due to the quality of the shot or to an uncleaned surface, the photograph does represent other shades of grey in the stonework below the mosaic or in the sea lion with some sharpness and nuance. Two possibilities for an alternative location come to mind if we accept that the caption from 1890 is correct in its identification of the mosaic’s location on a wall in the west of the monument: either the guilloche pattern changed considerably in form and arguably in colour in the now-lost part at the south-western corner of the podium (Figure 2.35) – which seems unlikely as it has been preserved on most parts of the podium and is consistent in its shape everywhere else – or the triton and sea lion pair was, in fact, part of a mosaic decoration on another wall west of the monument. The logical alternative is that the caption of the photograph has been misunderstood. The mythological scene was attached to the outer wall which encircled the court around the monument (Figure 2.35).235 235

The description on drawing 25B (Figure 2.39) does not give any further insights as to the location of the scene: ‘Desenho do mosaico que revestia o muro G do lado G’ do edifício marcado na planta W25 com W11 e se presentado na estampa No 25A’. The official ground plan (Figure 2.32) marks the area between the podium and the outer wall with the letter ‘G’, but

Mythological Subject Matter

Because the photograph shows the brick wall of the monument rising up behind the wall to which the mythological scene is attached (Figure 2.40), the scene must have faced westwards, out of the inner precinct of the monument. The decorative programme of the baptismal chamber of Milreu can be reconstructed as follows: a triton facing a sea lion was located on the outer wall of the church precinct, on the side where the baptistery was located (Figure 2.20). Another sea creature, probably another triton, was placed on the podium wall, that is, the eastern interior wall of the baptistery (Figures 2.38a and 2.38b). There is no way to determine to what extent the mosaic decoration in the interior of the baptismal chamber remained untouched by the Christian community. We do know, however, that the mythological scene caught in the photographs was preserved to a degree that left its subject matter fully recognizable.

2.2.2.3

Reusing Roman Material Heritage: Pragmatics or Intention?

Nereids, tritons, and hybrid animals like hippocamps, centaurs, or cetea form the cortège of deities like Neptune, Oceanus, and Venus Marina whose stories are related to the sea.236 Members of the thiasus like the triton are not thought to embody mythological content by themselves; their role is to escort the gods.237 Marine mythology is omnipresent in the late antique visual culture of the Iberian Peninsula, similar to the situation in North Africa discussed in Chapter 1.238 In the extant churches from the late Roman and Visigothic period, however, such hybrid sea creatures are largely absent. This absence nourishes the suspicion that the field of associations triggered by the marine thiasus was too problematic for their inclusion in newly built churches. This appears all the more likely as targeted artistic receptions of Roman depictions of sea life, which omit mythological scenery, do exist in Visigothic Lusitania (Figure 2.44).239 Given the reluctance to include elements from the marine thiasus in newly built churches, should we assume that the construction of a

236 237 238

239

reference ‘G’ has unfortunately not been indicated on the plan. A search for additional plan material might bring clarity in the future. On myths related to the marine world circulating in Iberia, see Blázquez 2008, 11–19. LIMC, vol. 8.1, Tritones; Zanker and Ewald 2004, 396–397. A good account of the thematic spectrum and the geographical dissemination of marine mythological scenes in Hispanic mosaics is provided in Durán 1993, esp. 225–237. For Spain and Portugal, see also Guardia 1992, 307–309; Neira Jiménez 1997; Lancha 2003; Mourão 2008, 2010, 2011; López Monteagudo 2011. For a comparative discussion which takes into account the eastern Mediterranean evidence, see Neira Jiménez 2011. Cruz Villalón 1985, 102–103, cat. 209.

147

148

The Use of Non-Christian Imagery

Figure 2.44 Water conduit with carvings depicting sea life, sixth to seventh century, Museo del Arte y la Cultura Visigoda, Mérida, inv. no. 28. Photography by the author.

baptistery within pre-existing walls covered with depictions of mythological sea creatures was influenced by pragmatic motivations alone? In all likelihood, economic considerations came into play when Milreu’s magnificent fourth-century monument was reused as a church. In relation to the choice to locate the baptistery along the north-western walls within the complex, however, pure pragmatism is harder to defend. On the one hand, the location chosen for it is idiosyncratic; on the other, it would have been possible to build an independent baptismal room if the Christian community had thought one necessary. There is not a single comparable case on the Iberian mainland where a late antique baptismal font was erected on top of a grave (Figure 2.36).240 Furthermore, the north-western location of the font in proximity to the main entrance in the west runs counter to common Lusitanian practice. While in Iberia as a whole, and in particular in the north-east, there was a tendency to position the font in the west of the church, Lusitanian baptisteries were commonly situated northwards or southwards in close proximity to the apse.241 Most similar to the location of Milreu’s font in the north-west next to the entrance is the placement of the font in the martyrial church of

240

241

In Hispania, only in the basilica of Son Peretó near Manacor (Mallorca) were burials found next to the baptismal font. See Riera Rullan et al. 2006 with an updated bibliography. A burial practice in baptisteries certainly did exist in the West, however. A prohibition against such burials was issued at the Council of Auxerre of 578. See Krautheimer 1942, 28–30. Examples are (north) Casa Herrera, Torre de Palma, Alconétar, Valdecebadar Olivenza, and (south) San Pedro de Mérida, El Gatillo. On the locations of Lusitanian baptisteries, see Ulbert 1978, 140–156. See also Hauschild 1980, 202. For an overview of baptisteries’ locations in Hispania, see Godoy Fernández 2004, 481–485.

Mythological Subject Matter

Marialba (León) in northern Spain.242 However, in Marialba the font was placed in a complex annexed to the church.243 A similar solution would also have been feasible for Milreu, which would have made it possible to avoid using any of the pre-existent mosaic decoration in the interior or exterior of the baptistery. While independent baptisteries were less customary in Iberia than in northern Italy, for instance, there were no defining rules for the baptismal settings of Lusitania and Baetica other than that they were typically set apart from the open church spaces.244 As in Milreu, many of these fonts were placed in a baptismal chamber attached to the basilica, yet independently functioning annexes like the one at Torre de Palma in the Alto Alentejo, or free-standing rotundas as in Algezares in Murcia, existed as well.245 Given that the location of Milreu’s baptistery did not follow regional custom and that alternative building concepts and space would have been available, a purely pragmatic attitude seems unlikely. Since convention does not seem to fully explain the baptistery’s integration of the mosaic friezes, let us return to their subject matter. If one does not engage in detail with the figurative scenery, the overall theme is reducible to a depiction of life in water. It is not implausible that the patrons of the baptismal chamber were satisfied with the generic theme of aquatic life as a baptismal decoration.246 It could further be argued that the pre-existence of the mosaic frieze allowed Milreu’s Christians to choose from a thematic spectrum unavailable to communities building from scratch. The reuse of pre-existent, at times even mythological mosaics for ecclesiastical spaces is, in any case, not a unique incident on the Iberian Peninsula, which suggests that they were appealing to Christians. A potential case in point is the Church of San Miguel in Barcelona (Figure 2.45). The Church of San Miguel is attested in 1046, but the ground plan of the church does not conform to those typical of high medieval churches, and hence it has been speculated that the church could be older.247 San Miguel’s floor mosaic likely dates to the second half of the second century AD and probably decorated the frigidarium of a Roman

242 244

245 246 247

243 Hauschild 1980, 202. Oepen 2012, 398–407, esp. 404, pl. 49. Godoy Fernández 2004, 481–482 with further literature on North Italian baptisteries; Schlunk and Hauschild 1978, 48. For Torre de Palma Ristow 1998, nos. 528–530, for Algezares no. 568. Compare 60–65 in Chapter 1. See the conflicting datings of Barral i Altet 1978, 39: ‘probablement en place depuis le Moyen Âge (XIIe-XIIIe siècles)’, and Balil 1960a: ‘Debió existir ya en el s. X, con cementerio propio, pues en 1046 pasó a depender de los canónigos de la catedral de Barcelona. . . . un estudio de la planta permite observar que ésta no corresponde a una iglesia románica’.

149

150

The Use of Non-Christian Imagery

Figure 2.45 Floor mosaic of the destroyed church of San Miguel, Barcelona, second half of the second century, Museu d’Arqueologia de Catalunya, Barcelona, inv. no. 19042. © Museu d’Arqueologia de Catalunya (MAC)

bath.248 The fragmentary mosaic depicts two pairs of facing male tritons accompanied by dolphins on both longitudinal sides.249 The two figures of one of the pairs hold a fish between them. There is also a seahorse on one of the vertical sides, again accompanied by dolphins. This mosaic covered the nave of San Miguel before the destruction of the church.250 More definitively late antique is another Roman villa in Lusitania, a part of which was transformed into a church.251 A mosaic from the villa of Montinho das Laranjeiras (Figure 2.46) which depicts a fish belonging to the family of garfish as well as the fragments of a curled tail arguably formed part of what was recorded during an excavation by Estácio da Veiga in 1877 as being an ‘open font with a mosaic pavement ornamented with

248 249 250

251

Balil 1960a, esp. 22–26. It has been relocated to the Museu d’Arqueologia de Catalunya. For a ground plan of San Miguel, see Barral i Altet 1978, plate XIII. Another second-century mosaic depicting ‘a sea filled with algae and fish’ in the southern part of the crossing of the cathedral of León was likely only integrated in the early tenth century, which is the assumed building time of the cathedral. See Gómez-Moreno 1925, 25–26; Garcia y Bellido 1968, 19–21, fig. 16; Mañanes 1980, 25–28; Jiménez Sánchez and Sales Carbonell 2004, 197. On the Christianization of villas in Hispania, see Oepen 2012.

Mythological Subject Matter

Figure 2.46 Fragment from the mosaic pavement of the baptismal font of the villa of Montinho das Laranjeiras, first half of the fourth century, Museu Nacional de Arqueologia, Lisbon, inv. no. 18754. Photography by the author. © Museu Nacional de Arqueologia.

fish’.252 The rectangular basin with two sets of steps (length 1 m
width 60 cm
depth 60 cm) can be identified as a baptismal font; remains of a water conduit are further confirmation of this.253 Destroyed today, it was located in the north-western arm of a cruciform church which was integrated into the villa in the late sixth or early seventh century.254 The mosaic pavement has been dated to the first half of the fourth century.255 The emphasis on a rich palette of species, notably in combination with what

252

253 254 255

Maciel 1996, 91–92. The mosaic fragment is preserved at the Museu Nacional de Arqueologia, Lisbon, inv. no. 18754. A second mosaic fragment from Montinho das Laranjeiras shows a kantharos (inv. no. 18697). See also Santos Affonso dos 1972, 373–377; Saavedra Machado 1970, 282–283; Oepen 2012, 110–116. Maciel 1996, 94. Maciel 1996, 91–100; also Oepen 2012, 111–116; Hauschild 1980, 202–203. Oepen 2012, 114. Maciel suggests a dating of the mosaic to the end of the sixth or the seventh century, that is, contemporaneously with the construction of the church, with reference to comparable North African baptismal decorations in the fifth and sixth centuries. Maciel 1996, 94, 97. Maciel, however, cites the North African tradition only very selectively. The great variety of species still attested in the baptistery of Cuicul in the late fourth or at the turn of the fifth century is clearly diminished in North African baptismal mosaics of subsequent centuries. Maciel’s other comparative example, the baptismal mosaic of Mariana on Corsica, likely dates to the late fourth century. See n. 614 in Chapter 3. Furthermore, Maciel does not take the Lusitanian tradition of marine mosaics into account.

151

152

The Use of Non-Christian Imagery

was possibly a sort of sea monster with a curled tail, corresponds well with the detailedness and diversity of species of fourth-century marine mosaics like those of Milreu.256 In conclusion, in the villas of both Montinho das Laranjeiras and Milreu (around 60 km away from each other), a pre-existent mosaic became a focal point of the baptismal decoration when parts of the villas were reused as churches. The Christian construction phases can tentatively be dated to the fifth to sixth century in Milreu, and to the sixth to seventh century in Montinho das Laranjeiras. The terminus ante quem of the repurposing of the mosaic floor in the church of San Miguel in Barcelona is as late as 1046. In all three cases, Christians accepted the representations of water and the various (fantastic) species depicted therein for the decoration of their baptisteries and churches. While the thematic focus is reminiscent of newly created ecclesiastical mosaics like the one in Cuicul’s baptistery, Milreu’s mythological creatures distract from the assumed allegorical message of such aquatic scenes in baptisteries – namely that water is the element leading to baptismal salvation.257 This did not stop Milreu’s community from reusing the mosaic frieze. What we can learn from the case of Milreu is that debating whether the inclusion of Roman material culture was a pragmatic or intentional act is to miss the point. Milreu’s Christian community, led by the estate holders and those responsible for the building, had no alternative but to address the existing architectural circumstances. They did not have the luxury of developing an attitude towards the mythology in the church space independently from the material constraints imposed by the monument itself. It was within the spectrum of their agency, however, to embrace more or less proactively what they were faced with. The patrons of the baptistery decided against destroying the mosaic façade of the church and baptistery and against following local custom in their selection of the site. Instead, mythological sea creatures were made part of the visual stimuli which accompanied the baptismal ceremony, assumedly both inside and outside the baptismal chamber. Milreu’s Christians were demonstrably keen to integrate their preChristian heritage into the self-image of the community.

256

257

While there is a tradition of sea life mosaics in fourth century Lusitania, the mosaicists might have been from elsewhere. That workshops of North African descent travelled around Lusitania is argued in Lancha 2004, esp. 412–414. The arguments are made in response to Balil 1975. See 60 in Chapter 1.

Conclusion

2.3

Conclusion

In the baptisteries of Henchir el Koucha, Milreu and the assumed baptistery of Myrtilis Iulia (Mértola), the mosaic decorations affirm an inclusive vision of Christianness by acknowledging Roman culture as a part of Christian identity. In Henchir el Koucha, an iconography associable with the Roman circus was integrated within an ostensibly Christian framework. The motif of the facing horses held several connotations concurrently, one of which had the potential to put into question the very Christian codes of behaviour confirmed by the baptismal rite. In Myrtilis, we encountered a building with characteristics of both baptismal and palatial structures. Pre-Christian domestic and Christian visual languages may have coalesced into a mythological scene. The Christians of Milreu inserted a baptismal space within a pre-Christian Roman one and let the mosaic frieze and its rendition of water – infused with pre-Christian mythology – stand for the salvific water of baptism. They presented themselves as carrying on Roman heritage by building on it. If we compare the baptismal imagery from Cuicul in the first half of the fifth century with the three case studies dating to about the sixth century, at times with continuous effects in the seventh century, we cannot help but notice the strikingly different visual strategies: Cuicul’s baptismal imagery does not exhibit specifically Christian subject matters, and there is nothing particularly objectionable to the Christian eye. However, in this chapter, we encounter rural and urban Christian communities actively seeking to include non-Christian imagery which had the potential to evoke Christian contempt. The proceeding Christianization and the increasing temporal gap between the performance of non-Christian practices and the Christian usage of imagery referring to such practices did not diminish Christians’ interest in them. If anything, as the limited material evidence suggests, interest increased over time. Whether such imagery is a reuse of the decoration of earlier buildings, as in Milreu and possibly in Myrtilis, or a newly created image, as in Henchir el Koucha, is of little importance. Associating baptism with unorthodox imagery from a pre-existing site seems intentional, just as planning the creation of such imagery rooted in a century-old tradition. The chapter demonstrates that the reuse of older mosaics with mythological scenery was relatively popular on the Iberian Peninsula. The availability of ancient decoration seemed to facilitate the decision to incorporate mythological subject matter instead of causing resistance.

153

154

The Use of Non-Christian Imagery

When images are reused under changed social or religious circumstances, they lose some or all aspects of their original significance and acquire new ones. A reused image which has accumulated several layers of meaning can be expected to carry a less specific overall meaning than those created for a specific purpose. However, even the pair of facing horses in Henchir el Koucha’s baptistery reflect an iconographic development spanning at least four centuries (unless we consider its possible roots in Punic coinage, which would add another five hundred years of development). What I have argued in the case of the image of Bellerophon in Myrtilis holds true, I believe, more generally. Exploring Christian interpretations of non-Christian imagery can be a fascinating and fruitful endeavour. However, the fascination with the interpretatio Christiana should not overshadow the fact that images rarely have a single meaning. More importantly, we should consider the possibility that Christian communities may have chosen an image not primarily for its Christian interpretability but rather for the cultural affiliations they associated with the image. A unifying factor among the three case studies is that the decorations in question owe their origins to institutions and practices connotated as profoundly Roman in late antiquity. The imagery of Myrtilis and Milreu allowed Christian communities to showcase paideia, while in Henchir el Koucha Christians preserved the memory of the Roman circus, which remained active in the sixth century, within the Christian realm. All three baptisteries provided avenues for expressing Christian identity while still allowing an association with traditional Roman culture. They facilitated what identity theory calls the simultaneous activation of different identities which are experienced as congruent enough to be acted out together. Through baptismal art, the Christian communities handed down a flexible and inclusive understanding of Christian identity to future generations of Christians. All of the mosaics discussed in this chapter were part of the religious lives of Christians in post-Roman societies of the sixth century. The ethnic affiliation of the Christian communities cannot be determined for any of the examples. The extent to which these communities were affected by the arrival of the Vandals or Byzantines in Africa Proconsularis or by the Visigoths in Lusitania is likewise unclear. In any case, fundamental shifts in the political landscape did not hinder the three Christian communities from feeling connected to Roman culture. Indeed, it is interesting to ask whether the political transitions themselves might have motivated baptisms into a Christianity which seemed connected to Roman culture. The evidence from the Arian baptistery of Ravenna, discussed in the third chapter, points in this direction.

3

|

The Conversion of a Personification The River Jordan in Ravenna

In contrast to the previous chapters, this chapter does not primarily investigate the viewing habits of common Christians to come to an understanding of how they perceived the imagery of a baptismal site. Instead, we will rely on extant textual sources from the western Roman capital of Ravenna to examine how an ecclesiastical authority might have understood his creation of a baptismal iconography. Thus, by studying the cupola mosaic of the so-called Orthodox or Neonian Baptistery (c. 451–473 AD) of Ravenna, we can add the perspective of the ecclesiastical elite to our discussion, which has so far been about nonelite viewpoints. The Lived Religion approach has made us sensitive to the possibility of local, and at times non-orthodox, ways of making Christianity matter at the event of baptism. In the previous case studies, we concentrated our attention on individual Christian communities, both rural and urban, and on how they might have interpreted the baptismal imagery at hand. It was implied that the creation of the baptismal decoration or the incorporation of pre-existent art into a baptistery must have been approved by some church authority at some point. In the case of Ravenna, we might get a better sense of how a bishop aligned his wish to allude to Graeco-Roman culture with the authoritative voice of Christianity. The prominent position of the episcopal see of Ravenna has led to a comparatively good knowledge about the deeds and writings of the bishops of the capital. But there is more. To my knowledge, Ravenna offers the only opportunity to reconstruct the genesis of a new baptismal iconography from a deep exegetical tradition. In fact, it is likely that the iconography discussed here – the baptism of Christ witnessed by the River Jordan, as it appears in the Orthodox baptistery and was later copied in Ravenna’s Arian baptistery – was proactively developed for the Orthodox baptistery under the aegis of a bishop. This chapter seeks to demonstrate that Bishop Neon’s (sed. c. 450–473) translation of Sermon 160 by his predecessor Peter Chrysologus into a work of art created a visualization 155

156

The Conversion of a Personification

which casts the River Jordan as a representative of pre-Christian GraecoRoman antiquity.1 Neon and the mosaicists used the cultural connotation of ‘erstwhile Graeco-Roman deities’, arguably inherent to river personifications in late antiquity, to visualize the river’s transformation from a pagan entity into the Christian River Jordan. I argue that, unlike the imagery based on Roman art seen in other baptisteries, Neon’s iconography is careful to confine Graeco-Roman tradition to the past as long as it was not explicitly Christianized. Neon’s vision of antiquity is as positive as it is polemical: only in its adoption of Christianity does the Graeco-Roman past become the present. At the end of the chapter, I will delve into the Arian perspective on the same subject, as seen in the cupola mosaic of the Arian baptistery (c. 500 AD). The most notable contrast between the two mosaics resides in the depiction of the river Jordan, hinting at the possibility that the Arian commissioners intended to express their differing views on the relationship between Christianity and pre-Christian antiquity.

3.1

The Two Personifications

At the centre of the mosaiced dome of the Arian baptistery of Ravenna (c. 500 AD) (Figure 3.1a), the imposing figure of an aged, muscular man with a beard and well-kempt long white hair gestures towards his left (Figure 3.1b). His appearance is virile and dignified: he is bare-chested and clad up to his waist in a brilliant green garment with a red border; his face is marked by oversized, piercing eyes and rosy cheeks, his head by a pair of red crab claws; and he carries a shimmering green and blue reed. The physique and physiognomy of this personification of the River Jordan follows the conventions of representing the fluvial or marine deity Oceanus (Figure 3.3).2 The personification of Jordan in the centre of the cupola mosaic of the Orthodox or Neonian baptistery of Ravenna (c. 451–473 AD) (Figures 3.2a–c), which served the Arian baptistery as a model, follows a closely 1

2

PL 52, 620–622. The importance of this sermon in respect to the dating of the cupola mosaic in the Orthodox baptistery has been emphasized in Nordström 1953, 33–34. An exception is the attribute of a water jug that is typical for Jordan but untypical for Oceanus. See Dinkler-von Schubert 1972, 75. See also Deichmann 1969–1989, vol. 2, 254. On the iconographic tradition of Oceanus, see Cahn 1994; Klementa 1993, 72–102; Cavalieri 2002. Note Oceanus’s oscillating status between fluvial and marine deity: Durst 2012, 518–519.

The Two Personifications

Figure 3.1a Ravenna, cupola mosaic of the Arian Baptistery, c. 500. Photography by the author.

Figure 3.1b Ravenna, detail of the cupola mosaic of the Arian baptistery. Photography by the author.

157

158

The Conversion of a Personification

Figure 3.2a Ravena, cupola mosaic of the Neonian baptistery San Giovanni in Fonte, c. 451–473. Photography: Paola Cravino. © Getty Images, 491814591.

Figure 3.2b Ravenna, detail of the cupola mosaic of the Neonian baptistery. Photography: Flavio Vallenari. © Getty Images, 623199808.

The Two Personifications

Figure 3.2c Ravenna, Neonian baptistery, detail of the personification of Jordan. Photography by the author.

Figure 3.3 Floor mosaic depicting the mask of Oceanus from Aïn Temouchent, end of the fourth century, Musée National des Antiquités et des Arts Islamiques, Algiers. Photography by the author.

159

160

The Conversion of a Personification

related tradition, namely that of the representation of river gods.3 However, the athletic, bearded middle-aged man with stringy long grey hair interspersed with blue, green, and purple streaks witnesses the baptism from a less privileged position on the lower right side of the medallion. He is also considerably smaller than his later variation in the Arian baptistery. The rich scholarship on the mosaics of the Orthodox and Arian baptisteries of Ravenna has pointed out the classical inheritance of the two personifications as well as the differences between them long ago.4 It has problematized the fact that the use of personifications – anthropomorphic representations of a thing or a concept – seems to perpetuate the GraecoRoman tradition of attributing superhuman agency to non-human entities by giving them anthropomorphic shape.5 Representing a river in this classical manner means alluding to a type of representation commonly used in antiquity to emphasize the divine aspect of natural and artificial entities.6 Personifications were nevertheless widely used by Christians in late antiquity, be it in the visual arts, poetry, or theological writing.7 They increasingly came to represent abstract concepts or were reduced to topographical markers while mythological and divine aspects progressively diminished. This does not mean, however, that church officials were completely relaxed about the use of personifications. Some Christian writers animated natural phenomena only cautiously or refrained from them entirely. Augustine of Hippo (354–430), for instance, carefully unfolds in a sermon the nature of the ‘genius of Carthage’.8 Far from being a god, he says, the distinction that people identify as a genius is merely a stone. Then he

3 4

5 6

7

8

On the tradition of representing river gods, see 62–63 in this chapter. Deichmann 1969–1989, vol. 1, 210; Deichmann 1969–1989, vol.2; Jensen 1993, 2011, 118–123; Weinryb 2002, 52–53; Ward-Perkins 2007, 271; Sturaro 2013. Jensen 1993. On the antique tradition of personification and its afterlife, see Gerber 1883, esp. 269–284; Joosen and Waszink 1950; LCI, vol. 3, 394–407, Personifikation; Gombrich 1971, esp. 248–253; Ohly 1979; Paxson 1994; Hartmann 2003; Mertens Fleury 2015, 83–125. For river personifications and gods in art Klementa 1993; Dinkler-von Schubert 1972; Weiss 1981; Toutain 1926; Lehnerdt 1884; Leclercq 1907–1953. On the personification of Jordan in particular, see Dinkler-von Schubert 1972; 84–96; Waser 1919; Ristow 1957. See also Vollkommer 1981- with a catalogue of depictions of the personified Jordan. Piper 1847–1851, 7–10; Weitzmann 1960; Ohly 1977; Paxson 1994; esp. ch. 3 on Prudentius’s Psychomachia; Straubach 2002; Stafford and Herrin 2005. For personifications of natural phenomena Opelt 1962, esp. 1174–1178; Dinkler-von Schubert 1972, Alföldi-Rosenbaum and Ward-Perkins 1980, 37–40 and 121–133. The late antique play between anthropomorphic and topographical understandings of river personifications is discussed in regard to Homer and Philostratus in Elsner and Squire 2018, esp. 74–75. For city personifications Bühl 1995; Poulsen 2014. Augustine, serm., 62.10.

The Two Personifications

suggests a much more abstract interpretation which disentangles the genius from an individual personification: if the Christians of Carthage live well, they will be the genius of the city. Theodoret of Cyrus (c. 393–457), and even John of Damascus (675/676–749), felt they had to deny that natural phenomena personified in scripture were gifted with reason or with souls.9 A similarly cautious attitude can be found in the arts. Take, for example, the city gate sarcophagi of the late fourth century, which repeatedly depict a personified Jordan reclining below the ascending chariot of Elijah (Figure 3.9).10 The sarcophagus of Sant’Ambrogio in Milan stands out (Figure 3.10) because it depicts Adam and Eve instead of Jordan. The associative narration of the city gate sequence is rendered ineffective by this change: while Elijah parted the River Jordan before his ascension to heaven (2 Kings, 2:1–18), no obvious link can be drawn between Elijah and Adam and Eve. The substitution is best understood as a deliberate attempt to avoid depicting a personified River Jordan.11 Christian sensibilities towards personifications apparently originated in a concern that the idea of natural phenomena as supernatural agents would be widely promulgated. This chapter is not primarily concerned with the theological problems Christians encountered as they inherited preChristian classical traditions of allegorical speech and visualization. Nevertheless, these problems should be kept in mind when we examine the context in which the cupola mosaics of Ravenna’s baptisteries were designed. River personifications had previously been used in the visual arts to illustrate scenes from the Old Testament (like in Elijah’s ascension to heaven) but, as far as we can tell, not in the context of the New Testament.12 What was the purpose of depicting the River Jordan in the form of a personification? Why were the two personifications represented in the way they were? Why did they play such an essential role in the design of the Orthodox and the Arian baptisteries? The chapter builds on previous scholarship to reconstruct how the personification of the River Jordan became integrated into the cupola mosaics of Ravenna’s baptisteries. The story begins with an innovation which Ravenna’s Bishop Peter Chrysologus (sed. c. 431–450) introduced into the exegesis of the psalms, continues with his predecessor’s – Bishop Neon (sed. c. 450–473) – decoration of the Orthodox baptistery, and ends

9

10 12

Theodoret, Jerem., 2,12; John of Damascus; De fid. Orthod., 2,6 with a reference to the River Jordan. 11 Schoenebeck 1935; Dinkler-von Schubert 1972, 86. Schoenebeck 1935, 44. Dinkler-von Schubert 1972, 89.

161

162

The Conversion of a Personification

with the reinterpretation of the personification in the Arian baptistery, by patrons unnamed to us, perhaps around 500.13 The Orthodox and Arian baptisteries of Ravenna lie at the centre of a long-standing debate about whether ‘Arian’ art is different from ‘Catholic’ art. The umbrella term ‘Arian’ was used by Christians adhering to the Nicene Creed (originally adopted at the Council of Nicaea in 325) to refer to varying groups of Christians who did not believe in the absolute equality of the three persons of the Trinity. The ultimately triumphant (and therefore named ‘Catholic’) homoousios camp held that Father and Son were of the same substance. All of the deviating groups saw a difference in nature between God the Father and God the Son. They also believed the Holy Ghost to have been created by either one or both of the other two. One particular sub-group, the Homoians, believed that Father and Son were ‘like’ each other, but not the same in substance. Most Germanic peoples in the Roman West, amongst them the Ostrogoths, adhered to this belief.14 Despite efforts to determine a distinctly Arian visual language, this search has largely been unsuccessful.15 The Arian and Orthodox baptisteries in Ravenna in particular have been studied for iconographic traits which could speak to differences in the Christological positioning of Arians and Catholics in Ravenna.16 Ravenna’s baptisteries are obvious objects of study, as they each respond to the biblical call to spread the Trinitarian truth through baptism. The three persons of the Trinity are mentioned together for the first time in a Gospel passage related to Christ’s baptism: ‘Go therefore and make disciples of all the nations, baptizing them in the name of the Father and the Son and the Holy Spirit’ (Matthew 28:19). This baptismal formula emphasizes the equality of the three persons. The Arian baptistery is assumed to have been built not long after Theoderic made Ravenna his main residence and the capital of Ostrogothic Italy in

13

14

15

16

On the history of the Orthodox and Arian baptisteries, see Deichmann 1969–1989, vol.1, 130–151 and 209–212; Deichmann 1969–1989, vol.2, 15–47 and 245–258. For a synthesis of episcopal building in late antique Ravenna, see Deliyannis 2010; Jäggi 2016. The degree to which Arianism mattered for the identity construction of the Goths, is, however, a contested issue. For the extreme position that Arianism was irrelevant for ethnic-cultural identity, see Amory 1997b, 236–276. In response, see the more nuanced account, with particular emphasis on Ravenna, in Brown 2007. See further Gheller 2017 with a special regard for the Visigothic evidence. Ward-Perkins 2007; Mîrşanu 2009; Bockmann 2014. The traditional position is summarized in Sörries 1983. In favour of a specifically Arian message in the Arian baptistery are Sörries 1983, 97–100; Rizzardi 1989, 373–376; Rizzardi 2001, 927–930; Rizzardi et al. 2011, 83–86; Penni Iacco 2011, 63–68; Sturaro 2013, 12–16; Gardini 2017, 42.

The Orthodox and Arian Baptisteries

493.17 It stands to reason that the Arian congregation tried to communicate its stance on the Trinity when adapting the Orthodox dome mosaic for their baptistery. However, clear evidence for this is as scarce as it is for specifically Arian art and architecture overall.18 The visually most apparent difference between the two cupola mosaics – the depictions of the river personifications – should encourage us to reconsider what other considerations besides diverging Christological concepts might have determined the design of the two mosaics. I will start with an account of the history of the two baptisteries, the imagery of the cupola mosaics, and their restoration histories. I will then assess the iconographic and scriptural traditions which laid the foundations for the pictorial scheme of the mosaic medallion in the Orthodox cupola. The focus will be on how Chrysologus innovated a pre-existent eastern exegetical tradition. Finally, we will explore the possibility that the patrons of the two baptisteries sought to demonstrate diverging attitudes towards the preChristian classical heritage and Christianity’s relation to it.

3.2

The Orthodox and Arian Baptisteries of Ravenna and Their Dome Mosaics

The decoration of the Orthodox baptistery occurred at a time of fervent episcopal building activity in Ravenna.19 After the city became the capital of the western Roman Empire in 402, the emperors took care of the rebuilding and monumentalizing of the half-ruined city for half a century but increasingly preferred Rome as the imperial residence from 450 onwards. The bishops quickly filled the vacuum which the frequent absence of the emperors created.20 Since Ravenna’s bishop had been raised to the rank of a metropolitan, possibly in Peter Chrysologus’s time (sed. c. 431–450), episcopal building projects were aimed at maintaining the importance of Ravenna’s see, despite Ravenna’s decreasing political significance in the later fifth century. Bishop Neon’s (sed. c. 450–473) building activity in particular sought to underline the claim of equality with the See 17 18

19

20

Deichmann 1969–1989, vol. 2,1; Deichmann 1969–1989, vol. 1, 245. Ward-Perkins 2007, 268–271; Jäggi 2016, 199–200; Barber 2018, 152–153. Against an Arian theological message but in favour of visual markers of Arian ethnicity are Wharton 1995, 131–136; Deliyannis 2010, 184–187. On Catholic episcopal building projects in the fifth and sixth centuries, see Deliyannis 2016; Jäggi 2016; Brown 2007, 420–422. Deliyannis 2010, 84–85.

163

164

The Conversion of a Personification

of Rome.21 Under his episcopate, the largest of all of Ravenna’s churches – the no longer extant Basilica Petriana in Classe, founded by Chrysologus, – was finished and decorated. The episcopal palace received a representative dining hall, and the Orthodox baptistery was also completed.22 Most of what is known about episcopal building in late antique Ravenna derives from the Liber Pontificalis Ecclesiae Ravennatis, composed by Andreas Agnellus of Ravenna (c. 805–after 846).23 Agnellus reports that Neon extensively rebuilt the baptistery and decorated it.24 Much of the decoration created under Neon’s patronage has survived to this date, which makes the Orthodox baptistery such an important source for the art and rite of baptism, and the best-studied of all late antique baptisteries.25 However, the construction of the baptistery lies in the dark. It is commonly assumed that it was constructed in connection with the Catholic Cathedral under Bishop Ursus (sed. c. 399– c. 426).26 Yet, among all of Ursus’s building projects, Agnellus discusses only the construction of the cathedral and mentions unspecified others, which reduces the attribution to Ursus to a mere question of probability.27 Spiro Kostof has therefore rightly concluded that the baptistery, though arguably intended since the beginning of the cathedral, might not have even begun in Ursus’s episcopate, but could also have been constructed by Peter Chrysologus in the period between Ursus’s and Neon’s episcopates.28 The topmost section of the cupola mosaic, constructed under Neon, depicts the baptism of Christ (Figure 3.2b).29 Against a golden background, Christ’s slender but athletic body is shown naked, covered to the waist by the light blue, translucent water of the Jordan. John, taller than Christ, stands on a rocky ledge. He is dressed in an exomis made of animal skin

21 22 23 24

25

26 27

28 29

Deliyannis 2016, 43–51; Jäggi 2016. On fifth-century building activity, see Jäggi 2016, 116–147. Agnellus and Deliyannis 2006. See also Nauerth 1974; Benericetti 1994; Martínez Pizarro 1995. Agnellus and Deliyannis 2006, 175–184. On Neon’s patronage of the dome mosaic, see Jäggi 2013, 118–120. Nordström 1953; Mazzotti 1961; Kostof 1965; Deichmann 1969–1989, vol.1, 130–151; Deichmann 1969–1989, vol.2,1, 18–47; Wharton 1987, 1995, 105–136. For a selection of more recent literature Russo 2001; Rizzardi 2001; Foletti 2009; Deliyannis 2010, 88–100; Fabbi 2011; Brandt 2012, 191–241; Jäggi 2016, 118–129; Dresken-Weiland 2016, 63–100. On the genesis of this idea, see Kostof 1965, 12. ‘Post haec vero omnia consummata et aedificia pleniter constructa . . .’. Agnellus and Deliyannis 2006, De Sancto Urso, 169–170, esp. 170. Kostof suggests that, alternatively to the baptistery, the mention of Ursus’s other building projects might have referred to the construction of the episcopal palace. Kostof 1965, 12, n. 10. Kostof 1965, 12. On the mid-fifth-century dating of the dome and its mosaics, see Deichmann 1969–1989, vol.2,1, 18–19; Russo 2001.

The Orthodox and Arian Baptisteries

and carries a crux gemmata in his left. He is about to pour water from a shallow bowl over Christ’s head with his right hand. The Holy Ghost in the guise of a dove descends directly above him. On the lower right side, the diminished personification of Jordan looks up at Christ (Figure 3.2c). An inscription above his head identifies him clearly as IORDANN.30 Jordan is of a darker complexion than Christ and John the Baptist; he carries a reed and a water jug. A green cloth embellished by shimmering reflections of light covers his right hand and forearm, which are stretched out as if he is trying to touch Christ. The scene is situated directly above the baptismal font.31 Although taken from the life of Christ, it is placed in the heavenly spheres, in the centre of the dome, in the space commonly designated for the depiction of celestial scenes. The gemmed cross in the centre, referring to Christ’s death and resurrection, situates his baptism as a stage in the history of salvation.32 It might also allude to Paul’s often repeated interpretation of the rite of baptism as the dying of the old man and resurrecting into Christ.33 The cupola medallion shows both the ideal prototype and the historical initiatory moment of the sacrament which the baptizand re-enacts in the here and now, making ‘time collapse’.34 The dome mosaics underwent drastic conservation work in 1791 and in the decade from 1854 to 1864, with the second efforts effectively effacing the first.35 Most of the present iconography of the medallion, however, can be traced back to the year 1690. Several features assumed to be additions by

30

31

32 33

34 35

CIL 11, 1, n. 256. Scholarship has not satisfactorily explained the second N in the inscription. Garrucci proposes that the last three letters have been added at some point. Garrucci is followed by Deichmann, with no obvious arguments. Garrucci 1872–1881, vol. 4, 34, n. 1; Deichmann 1969–1989, vol. 2,1, 33. The font in the baptistery today uses the ancient one in its foundations. It was internally circular in plan and the position of today’s font indicates that the original one was centrally located and approximately 2.7 m in diameter. See Kostof 1965, 36–37, figs. 20–21. Kostof’s discussion is based on the report Lanciani 1871. The original font was replaced by today’s octagonal font either in the twelfth century according to Kostof 1965, 146, appendix 2, doc. 5, or in the second half of the sixteenth century according to Deichmann 1969–1989, vol. 2,1, 19. Wisskirchen 1993, 168–170. Romans 6:1–11. The idea of baptism as death and rebirth was disseminated throughout late antiquity. See for instance Maximus of Turin, serm., 35; John Chrysostom, Catecheses ad illuminandos, 10.7; Gregory of Nyssa, Oratio Catechetica Magna, 35, 1–15. Wharton 1987, 370. The nineteenth-century Roman conservator Felice Kibel in particular has been criticized for his work, which was re-evaluated only two decades later. A brief historical overview of the phases of conservation is given in Verhoeven 2011, 252–256. More recently, on the restorations, see Antonellini 2012; Di Francesco et al. 2011. See further Gerola 1917; Ricci 1930–1937; Bovini 1966, 1969; Deichmann 1969–1989, vol. 2,1, 33; Iannucci 1984, 1985; Iannucci, Santopuoli, and Seccia 2001.

165

166

The Conversion of a Personification

Figure 3.4 Detail of the drawing of the dome mosaic of the Neonian baptistery, published in Giovanni Ciampini’s Vetera Monimenta (1699), 235. Photography by the author.

the nineteenth-century conservator Felice Kybel already appear in a drawing of the dome mosaic by Giovanni Ciampini in his Vetera Monimenta from 1690 (Figure 3.4).36 Restorations on the medallion previous to 1690 cannot be securely verified, but there is evidence that the lowest zone of the mosaic decoration, situated at eye level, may have been altered shortly after the Neonian construction.37 The first recorded restoration was envisaged in 1566, when an archepiscopal ordinance was issued ‘to restore the mosaics of the capelle and whatever else is devastated’.38 It is not clear exactly where mosaics were restored or replaced, since the ordinance concerns the capelle; this could refer to either the cupola or the baptistery 36

37

38

Ciampini 1699, 235. Iconographic peculiarities are the paten and the gemmed cross. See Rizzardi 2001, 918–922; Wisskirchen 1993. Two monograms supposedly meant to indicate Peter II (494–518) and Maximian (546–556) located in the lower zone of the mosaic walls of the baptistery suggest late antique restorations; however, the entire zone was re-restored in Kybel’s campaign, the accuracy of which is uncertain. See Deichmann 1969–1989, vol. 1, 19; Kostof 1965, 18–21. Further restorations initiated by Giulio della Rovere (1566–1578) that did not include the mosaics are recorded from 1573 onwards and must have been finished by 1591 at the latest. For more details, see Kostof 1965, 21.

The Orthodox and Arian Baptisteries

itself.39 A further document from 1591 indicates that little work was done on the mosaics in the decreed restoration.40 Kybel’s conservation efforts can be seen with the naked eye (Figure 3.2b) and are also recorded in Corrado Ricci’s drawing of the cupola mosaic (Figure 3.5). The upper central part of the medallion, distinguishable by the brighter colour of its golden background, has been totally redone. This area includes the upper part of the body of John the Baptist, the upper half of the gemmed cross, the paten, the neck and head of Christ, and the dove. A comparison with Ciampini’s drawing, which is not entirely correct, shows that Kybel added the beard of Christ and Saint John’s halo and neglected to include the water being poured from the paten; Kybel also seems to have slightly altered the position of the cross and the dove. The original fifth-century state of the restored area remains untraceable. The only figure in the scene which has been preserved in full is the personification of Jordan, together with the inscription identifying him. The roughly fifty-year-younger medallion of the Arian baptistery (Figure 3.1b) might be a point of reference for the reconstruction of the other figures.41 The central medallion of the Orthodox baptistery is encircled by two concentric bands, of which the inner one depicts a procession of the twelve apostles, labelled with their names, carrying golden wreaths (Figure 3.2a). The apostles’ procession has no obvious destination within the register of the apostles itself. Peter and Paul are in the lead and meet at a point located directly underneath Jesus and John the Baptist. In the outer band, empty thrones alternate with the four gospels lying open on altars. Various potential objects of the apostles’ veneration procession have been proposed. They include Christ himself (either as Christ triumphant or as Christ in his Theophany), the theophanic act of Christ’s baptism, and the four empty thrones in the outer band.42 Perhaps the most characteristic feature of the dome mosaic is that it allows viewers to interpret the programme in various ways, depending on how its different elements are evaluated and combined across the three registers. 39 40

41 42

Kostof 1965, 144, appendix 2, doc. 1 and 21, n. 48. See Kostof 1965, 21, n. 48 and n. 145, appendix 2, doc. 3: ‘. . .Ecclesia ipsa stat decenter, tantum remanserunt nonulle reliquie, que adhuc indigent reparatione modo prout actum fuit in locis, et figuris jam resarcitis ac restauratis’. ‘. . . the church is without any damage. There are only a few remains left which are still in need of repair, in the same way as was necessary in places [of the church] which have already been repaired and remade.’ Transl. by the author. Kostof 1965, 86–87; Wharton 1987, 370. See amongst others Nordström 1953, 36–46; Kostof 1965, 82–93; Deichmann 1969–1989, vol. 2,1, 38–42; Wharton 1987, 372–375; Engemann 1989 with an overview of the debate, esp. 481–483; Jäggi 2016, 127–128.

167

168

The Conversion of a Personification

Figure 3.5 Drawing of the dome mosaic of the Neonian baptistery with indications of the restoration work by Felice Kybel. Ricci (1930–1937), pl. 10–13. Photography by the author.

The narrative structure of the seemingly very similar cupola mosaic of the Arian baptistery operates rather differently (Figure 3.1a). The baptistery belongs to an Arian church only a few hundred metres from the Basilica Ursiana. The Arian church was rededicated to S. Spirito in the fifteenth century.43 S. Spirito, due to its baptistery and a nearby episcopium, has often 43

It is controversial whether the Arian cathedral, potentially S. Spirito, was originally dedicated to Sta. Anastasia or to the Anastasis. See Deliyannis 2010, 174–175. Cf. Schäferdiek 2009, 216–217.

The Orthodox and Arian Baptisteries

been assumed to have been the Arian cathedral, which is presumed to have been constructed by Theoderic.44 The common dating of the Arian baptistery to c. 500 AD, shortly after Theoderic took residence in Ravenna in 493, relies on this assumption. However, Agnellus gives testimony that the Basilica Salvatoris (today S. Apollinare Nuovo) also had a baptistery and an episcopium, which allows for the alternative that the Basilica Salvatoris was the Arian cathedral.45 Therefore, neither the cathedral status of S. Spirito nor Theoderic’s patronage at around 500 AD is certain. S. Spirito and its baptistery could also have been constructed by the Arian bishopric. Unfortunately, next to nothing is known about Ravenna’s Arian bishops.46 The cupola mosaic of the Arian baptistery is, apart from small patches restored in the seventeenth and mid-nineteenth centuries, mostly intact.47 Stylistic differences between the central medallion and the area around the throne (including Peter, Paul, and the apostle behind Paul on the one hand, and the rest of the mosaic on the other) testify to the work of two workshops, although the sequence of their involvement is not clear.48 The building likely functioned as an Arian baptistery for a relatively short period. Andreas Agnellus reports that all Arian churches were recatholicized during the episcopate of Archbishop Agnellus (556/7–569/70) following the capitulation of Ostrogothic Ravenna in 540 during the Justinianic reconquest of Italy.49 Though ostensibly inspired by the architecture and decoration of the Orthodox baptistery, the Arian baptistery suppresses the interrelation of

44 45

46

47 48

49

The Arian baptistery has mostly been discussed in relation to the Ostrogothic or Arian presence in Ravenna, and also in comparison to the Orthodox baptistery. For a recent reconsideration of its mosaic programme, see Barber 2018. See further Mazzotti 1957; Breschi 1965; Mazzotti 1970; Deichmann 1969–1989, vol. 1, 209–212; Deichmann 1969–1989, vol. 2,1, 251–258; Johnson 1988, esp. 79–80; Rizzardi 1989. I have not been able to consult Cummins 1994. See more recently Weinryb 2002; Wood 2007, 251–252; Deliyannis 2010, 177–187. Deichmann 1969–1989, vol. 2,1, 245; Johnson 1988, 79; Weinryb 2002, 42. Agnellus, Liber Pontificalis, chs. 86–89. On the baptistery, see Deliyannis 2010, 146–147. On the episcopium, see Jäggi 2016, 169–170. The only attested Arian bishop of Ravenna is one Unimundus, who built a church named S. Eusebii in 518. Agnellus, Liber Pontificalis, chs. 85–86. See also Deichmann 1969–1989, vol. 2,1, 243; Deliyannis 2010, 143–145. An important document about Arian clerical representation in Justinianic Ravenna is the papyrus Tjäder 34. Tjäder 1972, 1954–1982, vol. 2, papyri 29–59; Schäferdiek 2009. Gerola 1923, 122–127; Breschi 1965, 41–47; Deichmann 1969–1989, vol. 2,1, 252–253. The central part is consistently identified as the older one. There is no evidence against a quick succession of the two phases. See Gerola 1923, 125–127; Bovini 1957; Breschi 1965, 81–85; Deichmann 1969–1989, vol. 2,1, 255. Agnellus, Liber Pontificalis, chs. 86–89.

169

170

The Conversion of a Personification

the registers.50 The dome mosaic has only half the diameter of its model and displays only two registers. However, excavations have brought to light 170 kg of mosaic tesserae which could have formed part of a more extensive decoration of the cupola, or alternatively, of the four conches.51 The central medallion still depicts the baptism of Christ. However, by eradicating the gemmed cross from the scene, the medallion lost the very element which referred most clearly to the time-transcending significance of baptism. Instead, its historical character is put into focus.52 The medallion is surrounded by an apostolic procession, which in this case has a clear destination. Between Peter and Paul, a gemmed cross covered by a purple cloth hovers in front of a richly decorated empty throne. This throne is often interpreted as the Hetoimasia, which symbolizes the current absence and second coming of Christ.53 The empty throne, however, cannot always be interpreted in eschatological terms only; often it is already occupied by Christ in the form of symbols representing him.54 This also applies here, as the jewelled cross can be interpreted as symbolizing the resurrected Christ.55 The contemporaneous contemplation of the two narratives (the baptism of Christ and the apostles’ veneration of the resurrected Christ) is made difficult by the position of the outer band, which is rotated by 180 in comparison to the arrangement in the Orthodox baptistery (Figure 3.1a). Like in the Orthodox baptistery, the baptismal scene and the meeting point of the procession at the throne are still in a direct axis. This creates a link between baptism and salvation, but as the elements are placed upside down, this is not immediately evident. It has been suggested that a dogmatic decision motivated the reorientation of the mosaic. According to this view, the detachment of the two spheres prevents the apostles from being shown as venerating the human Christ directly, but only the non-anthropomorphic representation of Christ in the form of a cross.56 However, against this view, it has been argued that 50

51 52

53

54 55

56

Jäggi 2016. The architecture also shows significant differences. The most important ones are the addition of an apse, providing for a clear eastern orientation, and the addition of a now-lost ambulatory. See Gerola 1923; Mazzotti 1957, 48–50. Bovini 1957, 8. An author who differentiates clearly between time-transcending and historical scenes is Engemann 1989, 484. For a comparison of representations of empty thrones in the two baptisteries, see Nordström 1953, 46–57; Wharton 1995, 134–146. Nordström 1953, 47. For an overview of Christological interpretations of the gemmed cross, see Deichmann 1969–1989, vol. 2,1, 254. Opinions deviate on what the throne and cross might represent exactly. See Sörries 1983, 99–100; Rizzardi 2001, 929.

The Orthodox and Arian Baptisteries

the wish to formally separate the historical scene of Christ’s baptism and the time-transcending apostolic procession could have been motivated by reasons other than dogmatic difference. Alternative readings have stressed intrinsically artistic reasons as well as the wish to demonstrate the authority of the Arian bishops or the presumed patron Theoderic.57 Moreover, it is the depiction of Christ’s baptism which would qualify first and foremost as the Trinitarian bone of contention.58 However, the baptismal scene is, iconographically speaking, almost identical to the medallion in the Orthodox baptistery in respect to the representation of the naked Christ, the (visually absent) father, and the Holy Ghost. Both depictions are based on Mark 1:9–11: ‘Jesus . . . was baptized by John in the Jordan. And when he came up out of the water, immediately he saw the heavens being torn open and the Spirit descending on him like a dove. And a voice came from heaven, “You are my beloved Son; with you I am well pleased.”’59 The repositioning of John the Baptist from left to right is a relatively minor change.60 In fact, the by far most pertinent difference in the iconography is the upgrading of the river personification, now significantly enlarged and placed in the foreground on Christ’s right side and opposite John. As Bryan WardPerkins has put it: ‘If we were to take seriously the iconographical differences that there are between these two representations, I think our conclusion would be the absurd one that Arians attributed greater importance to rivergods than did Catholics . . .’.61 When reformulated slightly, would this conclusion be really that absurd? Could it be that the representation of the personified Jordan occupied Orthodox and Arian patrons as much as, or more than, promulgating a particular view on the Trinity? The next section reassesses the iconographical tradition of river personifications in order to demonstrate that the depiction of the River Jordan in the Orthodox baptistery was indeed novel in its time. This will prepare us

57

58 60

61

On art and theological clarification, see Jäggi 2016, 199; on the authority of the Arian bishop Barber 2018, 154–155; on Theoderic Weinryb 2002, 50–54. See also Ward-Perkins 2007. Evidence for liturgical differences between Catholic and Arian baptismal ceremonies are rare, and non-existent for Ostrogothic Italy, which makes speculation about their relevance for the artistic choices pointless. The discussion concerning Ravenna is summarized in Deliyannis 2010, 178–179. 59 See 162–163 in this chapter. See also Matthew 3:13 and, less explicitly, Luke 3:21–22. John’s baptizing gesture with a paten in the Orthodox baptistery is likely a nineteenth-century addition. See Deichmann 1969–1989, vol. 2,1, 33; Rizzardi 2001, 918–922. On the iconography of the dove emitting water, potentially once present also in the Orthodox baptistery, see Deichmann 1969–1989, vol. 2,1, 255–258. Ward-Perkins 2007, 271.

171

172

The Conversion of a Personification

for the discussion on why Bishop Neon may have wanted to put so much emphasis on the figure of Jordan.

3.3

The Creation of a New Iconography: The Personification of Jordan

The label IORDANN, written in large, black letters on the dome of the Orthodox baptistery, leaves no doubt about the identity of the river personification below it (Figure 3.2c). Based on surviving examples, however, it seems that later artists no longer felt the need to label the personification of Jordan.62 The practice of labelling mythological figures was not new to late antique art and was, amongst other things, a means of preserving and communicating classical knowledge in a Christianizing world.63 The labelling of Jordan in Ravenna is arguably an example of such a necessary reminder. Besides the novelty of the subject matter, the specific iconography of the personified Jordan has neither precursors nor successors: it is unique. The way Jordan covers his right hand with a green cloth (which partially dips in the water) while reaching out to Christ is especially noticeable. Because of this stance, Jordan has been compared to the angels holding towels for Christ which begin to appear in eastern depictions of Christ’s baptism in the sixth century (Figures 3.6 and 3.7).64 It would, however, be more accurate to say that Jordan is covering his hands in veneration of Christ. He holds the green cloth over his left forearm; the thumb and index finger which hold the reed are not covered. In this way he resembles John, whose hand holding the gemmed cross is likewise covered in a gesture of reverence, while the hand holding the bowl is not; the apostles, too, carry their crowns with mostly covered hands. Since covering the hands was a gesture of reverence common in both court ceremonial and religious imagery, Jordan is clearly depicted as being in adoration of Christ.65 The reverential manner in which Jordan covers his hands, looking attentively up to the baptized, stands in marked contrast to both earlier and later Christian personifications of bodies of water. The earliest of such 62

63 65

Late antique parallels are the labelled personifications of the rivers of Paradise at the Martyr Theodore chapel in the cathedral of Madaba, at the Church of St Paul at Umm ar-Rasas, both in Jordan and at the Jabaliyah baptistery in the Gaza strip. See Piccirillo 1994; Talgam 2014; Humbert 1999. 64 Raeck 1992, 160–166. See also Cameron 2011, 705. Waser 1919, 194. Deichmann 1969–1989, vol.1, 37; Dresken-Weiland 2016, 81.

Figure 3.6 Throne of Maximian, panel depicting the baptism of Christ, sixth century, Museo Arcivescovile, Ravenna. Photography: Judith McKenzie. © Manar al Athar.

Figure 3.7 Column drum depicting the baptism of Christ from Constantinople (?), attended by two angels and the River Jordan, sixth century, Archaeological Museum Istanbul. © Guntram Koch.

174

The Conversion of a Personification

Figure 3.8 Rome, Santa Maria Antiqua, sarcophagus depicting a personification of Neptune, c. 270. © DAI Rome, D-DAI-ROM-59.423.

images date from the third century. According to the extant material, at this early date personifications only appear in the more personal context of sarcophagi.66 The first preserved depiction is the reclined, beardless figure of Neptune holding a trident located on the narrow left-hand side of the tub-shaped sarcophagus of Santa Maria Antiqua (c. 270) (Figure 3.8).67 The water which spills from his jug flows into a sea on the front of the sarcophagus, the same sea into which Jonah is thrown from the boat. The earliest personifications of Jordan also appear on sarcophagi in the second half of the fourth century. They are all connected to scenes of Elijah’s ascension to heaven (2 Kings, 2:1–18) (Figures 3.9 and 3.10).68 These early representations stand in the iconographic tradition of river personifications

66

67

68

Beyond Ravenna’s baptisteries, only one other decoration in a church depicts a personified Jordan: the church of Hosios David (c. 500 AD) in Thessaloniki. Grabar 1959, 291–297; Ihm 1960, 44–45 and 182–184; Matthiae 1962; Spieser 2001. The identification as Jordan goes back to Severus of Antioch (d. 538): Ihm 1960, 46, n. 4. Bovini et al. 1967–2003, vol. 1, 306–307, n. 747. See further Jensen 2011b, 46–51; Dinkler-von Schubert 1972, 84–85. For an overview of all city gate sarcophagi depicting the personified Jordan, see Schrenk and Rexin 2007, 189, n. 52.

The Creation of a New Iconography

175

Figure 3.9 City gate sarcophagus, left-hand panel depicting the ascension of Elijah, late fourth century, Musée du Louvre, inv. no. Ma 2980.1. Photography: Hervé Lewandowski. https://collections.louvre.fr/ark:/53355/cl010303464

of the Roman imperial period. A good example of this is the only surviving monumental depiction of Jordan, preserved in a relief on the Arch of Titus (c. 70 AD) in Rome (Figure 3.11). Jordan, depicted as a middle-aged and athletic figure, is dressed to his waist in a garment and lies on a ferculum, or bier, while holding a reed and leaning on an amphora.69 River personifications from fifth- and sixth-century ecclesiastical contexts also follow this tradition.70 In the Justinian re-foundation of

69

70

Klementa 1993, 109. Jordan’s position on a ferculum suggests that the personification is not chiefly designed to represent a river but that it is a depiction of a personification of Jordan meant to be carried as an item of booty in a triumphal procession, symbolizing the victory over a country identified with the river. The carrying of depictions of river personifications in triumphal processions following the victorious return of the emperor was living praxis in Imperial Rome. See Dinkler-von Schubert 1972, 75. A late antique exception to the commonly figurative depictions of Jordan in Christian spaces is the representation of Jordan in the apse of SS. Cosmas and Damian in Rome. Pope Felix IV (sed.

176

The Conversion of a Personification

Figure 3.10 City gate sarcophagus, left-hand panel depicting the ascension of Elijah, late fourth century, Sant’Ambrogio, Milan. Photography: Giovanni Dall’Orto. © Wikimedia Commons.

Cyrenaica (Qasr Lybia) in modern Libya, personifications of the rivers of Paradise adorn four of the fifty panels of a floor mosaic (6 m  10 m) in one of Cyrenaica’s churches, dedicated in 539.71 They are represented as resting male nudes of different ages and with different attributes (Figure 3.12). All have pieces of cloth draped around them, but their genitals are mostly exposed. Although largely destroyed, the personifications of the rivers of Paradise in the medallions of the Martyr Theodore chapel in the cathedral of Madaba and the Church of St Paul in Umm arRasas in Jordan seem to have been depicted in a similar way. Whether they were also ostensibly naked cannot be reconstructed.72

71

72

526–530) chose a mere inscription of Jordan’s name for the decoration of the apse; the four rivers of Paradise are likewise indicated by name. See Ihm 1960, 137. For the relation between Jordan and the rivers of Paradise and the iconography of SS. Cosmas and Damian Spieser 2015, 275–278 and 281–283. Chick 2014. On further personifications of natural phenomena, see Alföldi-Rosenbaum and Ward-Perkins 1980, 37–40 and 121–133. Hachlili 2009, 182–183, fig. VIII-2; Talgam 2014, 216.

The Creation of a New Iconography

Figure 3.11 Rome, Arch of Titus, detail of the personification of Jordan, c. 70. © DAI Rome, D-DAI-ROM-79.2324.

Most late antique baptismal decorations depicting river personifications, however, tend to show only the heads or upper bodies of the rivers of Paradise. The floor mosaic of the small baptistery of Mariana in Corsica (Figure 3.13), arguably from the late fourth century, as well as that of the fifth-century baptistery of Ohrid in North Macedonia (Figure 3.14), both display Pishon, Gihon, Euphrates, and Tigris as male heads which frame diverse animal scenes. The mosaic floor of the Jabaliyah baptistery in the Gaza Strip shows the river personifications, both male and female, up to the chest. A female personification is similarly known from the Martyr Theodore chapel in Madaba.73 This brief survey leaves us with the 73

For Jabaliyah, see Humbert 1999; Hachlili 2009, 181, fig. VIII-I. See on female personifications Hachlili 2009, 182–183; Talgam 2014, 216. For Ohrid, see Bitrakova-Grozdanova 1975. For Mariana Moracchini-Mazel 1962a, 1962b; Moracchini-Mazel et al. 1971.

177

178

The Conversion of a Personification

Figure 3.12 Cyrenaica (Qasr Libya), mosaic floor panel depicting the personification of the river Gehon. © Jane Chick.

Figure 3.13 Mosaic pavement of the baptistery of Mariana (Corsica), late fourth century, Musée de Mariana, inv. no. 2018.2.4. © Patrick Ageneau/Atelier de restauration de mosaïques – Saint-Romainen-Gal.

The Creation of a New Iconography

Figure 3.14 Ohrid, floor mosaic of the baptistery, fifth century. Photography: Diego Delso. © Wikimedia Commons.

observation that, where personifications of rivers occur in Christian public space, they tend to continue or abbreviate traditional Graeco-Roman ways of representing rivers, at times accentuating the figures’ nudity or varying their sex. The emphasis is not on reformulating such imagery in an explicitly Christian way; instead, there is a tendency to either accentuate or abbreviate the existing iconographic tradition. Against this background, the depiction of Jordan’s active and positive attitude towards the baptism of Christ (Figure 3.2c) is astonishing. It also stands out amongst the group of about a dozen late antique examples of personifications of Jordan witnessing Christ’s baptism collected by Sabine Schrenk and Gerhard Rexin.74 Jordan continued to be frequently depicted at Christ’s baptism throughout the Middle Ages, particularly in the eastern Mediterranean, but also in the Frankish-Carolingian world.75 The earliest

74

Schrenk and Rexin 2007.

75

Strzygowski 1885; Ristow 1967, 151–156.

179

180

The Conversion of a Personification

Figure 3.15 Panel of the Werden Casket depicting the baptism of Christ, fifth century or Carolingian copy of a late antique model, Victoria and Albert Museum, inv. no. 2009CA7202. © Victoria and Albert Museum.

example of a relatively secure date, in the third quarter of the fifth century, is the depiction of Jordan in the Orthodox baptistery. Together with the small damaged figure of Jordan depicted below the baptized Christ and two angels on a column drum in the Archaeological Museum of Istanbul (Figure 3.7), dated to the sixth century for stylistic reasons, it forms the small group of figures of Jordan which are standing up to their waist in water.76 The personification on the marble segment, attributed to a workshop in Constantinople, is much smaller than the other figures and appears to be directed towards the baptismal scene. The figure’s right arm, and with it any action it might have performed, has been erased. Among the earliest anthropomorphic representations of Jordan (besides the Orthodox baptistery) are the Jordan of the Arian baptistery (Figure 3.1b) and potentially the ivory Werden Casket (Figure 3.15), attributed to northern Italy. The casket shows the male personification of Jordan in a reclined position, with a water jug and wavy plants protruding from the space between his torso and arms. Suggested datings for the Werden Casket range from as early as the beginning of the fifth century to the Carolingian period, in which case it may have been copied from a late antique model.77 Both the Jordan from the Arian baptistery and the Werden casket have in common the traditional mode of representation. 76 77

Mendel 1914, 440–442, n. 659; Firatli 1990, 103, n. 191. On the various positions, see Williamson 1999 versus Weigel 1999. Cf. Schrenk and Rexin 2007, 138.

The Creation of a New Iconography

A further group of six examples, dating from the sixth to eighth centuries, show Jordan with his body facing away from the baptism, usually with his hands raised and head turned over his shoulders to look back at the scene.78 One specimen of this iconography, sometimes called the ‘fleeing’ Jordan, was commissioned for the ivory throne of Maximian (Figure 3.6), bishop of Ravenna from 546 to 556, which is of disputed but likely eastern origin.79 The specific rendition of the figure of Jordan has mostly been taken as a side issue.80 Only Schrenk and Rexin have taken the posture of the ‘fleeing’ Jordan as the starting point for their investigation.81 They demonstrate the rootedness of this posture in pre-Christian iconography (Figure 3.16). Antique river personifications, in particular, were endowed with the function of expressing astonishment in the face of wonderful events. Their position and stance – turning away from the central scene, at times with raised hands – usually corresponds to a standard formula of transmitting the idea of an epiphany or a godly revelation. In analogy, ‘fleeing’ personifications of Jordan at the baptism of Christ should not be interpreted as expressing an aversion to the baptism. Rather, such personifications of Jordan are marvelling at the epiphanic character of the baptism of Christ, where the threefold nature of God is revealed for the first time.82 The raised hand of Jordan of the Arian baptistery (Figure 3.1b) is reminiscent of this type of representation. For the River Jordan’s role in Ravenna’s Orthodox baptistery, however, Schrenk’s and Rexin’s acute interpretation of the ‘fleeing’ Jordan is of limited value. Confronted with an epiphany, Jordan’s gaze is observing, turned upwards, his covered hand stretched out towards Christ.

78 80

81 82

79 See Schrenk and Rexin 2007. Volbach 1952, 68–69. Scholars have focused instead on providing a theological explanation for the integration of the personified river and the baptism of Christ. Robin Jensen suggests that the personification became a symbol for the sanctification of the River Jordan at Jesus’s baptism. Jensen 1993, 36–39; Jensen 2011b, 118–123. On Jordan’s sanctification, see also n. 661 in this chapter. According to Annabel Jane Wharton, the personification refers to an Old Testament prefiguration of Jesus’s baptism (Joshua 3:16–17). Wharton 1987, 94. See also 183–186 in this chapter. The attempt to find an Arian sub-text in the inclusion of Jordan in the baptism of Christ is misplaced. Sturaro 2013. Schrenk and Rexin 2007 argue against Ristow 1957; Deichmann 1969–1989, vol. 1. Schrenk and Rexin 2007, 182–188. On the baptism of Christ interpreted as a theophany in Christian art, see Deichmann 1969–1989, vol. 1, 210; Deichmann 1969–1989, vol. 2,1, 33–34.

181

182

The Conversion of a Personification

Figure 3.16 Patera depicting the triumph of Attis and Cybele with two river personifications at the bottom from Parabiago, second half of the fourth century, Civico Museo Archeologico di Milano, inv. no. A 0.9.14264 – St. 5986. © Comune di Milano – Civico Museo Archeologico.

3.4

Chrysologus’s Reinterpretation of the River Jordan: East Meets West

A plausible explanation for the iconographic anomaly in the Orthodox baptistery was provided long ago but has received little attention in modern scholarship. Carl Nordström was the first to note the similarities between the representation of the Orthodox baptistery and a passage on the subject of Jordan’s behaviour at Christ’s baptism in Sermon 160 by Peter Chrysologus.83 The bishop wrote: Why is it that the Jordan, who fled at the presence of the Ark of the Covenant [Joshua 3:16–17], did not run away when the whole Trinity is present? Why is it? It is because the one who shows homage and 83

Nordström 1953, 33–34.

Chrysologus’s Reinterpretation of the River Jordan

deference begins to be fearful [Ps. 114:3–6] no longer. Here the Trinity puts all its grace into operation and expresses all its love for the world; on that prior occasion the Trinity took hold of the elements in order to instruct its little servants to fear it.84

Nordström cited this passage to strengthen his case that Chrysologus could have been the commissioner of the mosaic cupola of the Orthodox baptistery. Scholarship has long since agreed that the mosaic was instead created under Chrysologus’s episcopal successor Neon, although Chrysologus cannot be ruled out entirely as the patron of the baptistery.85 In any case, Chrysologus’s words must be considered carefully, because Sermon 160 likely had an impact on the cupola mosaic’s design, as I will argue in the next section. Sermon 160 is a sermon on the Epiphany – one of the three feast days, besides Easter and Pentecost, on which the annual celebration of baptism traditionally took place in the West.86 Under the episcopate of Chrysologus, the baptism of Christ featured as one of three events celebrated on Epiphany in Ravenna, the other two being the adoration of the magi and the miracle at Cana.87 Chrysologus’s sermon was tied, at least indirectly, to the baptismal rite. There is no evidence to indicate the precise day on which the baptismal liturgy was celebrated in Ravenna, but Epiphany is one of several plausible options. Chrysologus’s commentary juxtaposes Joshua 3:16–17, which describes the Hebrews’ crossing of the River Jordan, with the event of Jesus’s baptism. Under the guidance of Joshua, who placed the Ark of the Covenant at the shore of the Jordan, the river split and allowed the Hebrews free passage. This passage stands in a long exegetical tradition of Old Testament prefigurations of the baptism of Christ. Two other crossings of large bodies of water, the crossing of the Red Sea (Exod. 14:15–15:22) and the division of the Jordan by Elisha (2 Kings 2:12–14),

84

85

86 87

Chrysologus, serm., 160 ‘. . . quid est quod Jordanis qui fugit ad praesentiam legalis arcae, ad totius Trinitatis praesentiam non refugit. Quid est? Quia qui pietati obsequitur, incipit non esse timori. Hic Trinitas exercet gratiam totam, totam secum loquitur charitatem: ibi elementa corrupit, ut servulos instituat ad timorem . . ..’ The translation follows Palardy (ed.) 2005, 280 with slight modifications by the author. For Neon’s patronage Deichmann 1969–1989, vol. 2, 17–19 and 34–35. See also 164 in this chapter. Ferguson 2009, 333 and 345. On Epiphany and Pentecost, see also 115–117 and 166–170. This can be concluded for instance from sermons 157 and 160, both of which relate to Epiphany. See Palardy (ed.) 2005, 270, nos. 20 and 281.

183

184

The Conversion of a Personification

were also understood as prefigurations pointing to baptism.88 Both crossing and baptism represent overcoming a boundary and an arrival, and both are connected to water.89 The connection between them can also be found in the Bible itself, namely in Paul’s interpretation of the crossing of the Red Sea as a baptism.90 Significantly, in Joshua 3:16–17, the river is not described as a personified deity who is frightened or fleeing but as a body of water responding to the plan of God: . . . the waters coming down from above stood and rose up in a heap very far away, at Adam, the city that is beside Zarethan, and those flowing down toward the Sea of the Arabah, the Salt Sea, were completely cut off. And the people passed over opposite Jericho.

Chrysologus, however, describes Jordan’s reaction towards the presence of the Ark of the Covenant as a flight: ‘Why is it that the Jordan, who fled at the presence of the Ark of the Covenant, did not run away when the whole Trinity is present?’91 Chrysologus alludes here to a second biblical source, namely several psalms frequently referenced in connection with the baptism of Christ since the fourth century. These psalms praise the power of the Lord over the waters (Ps. 74) and describe the tormented waters (Ps. 77) and the flight of Jordan (Ps. 114) at the arrival of the Lord.92 The theme of flight occurs explicitly in Psalm 114:3–6: . . . The sea looked and fled; Jordan turned back. The mountains skipped like rams, the hills like lambs. What ails you, O sea, that you flee? O Jordan, that you turn back? O mountains, that you skip like rams? O hills, like lambs?

Psalm 114 gave rise to a widespread exegetical tradition which interpreted Jordan as frightened and fleeing from God.93 Exegetes usually set this scene in the context of the baptism of Christ. The interpretation apparently derives from eastern sources and is mostly recorded in Syrian and

88 90

91 92

93

89 Jensen 2012, 20–23. Jensen 2012, 184–186. 1 Cor 10:1–2 ‘For I do not want you to be unaware, brothers, that our fathers were all under the cloud, and all passed through the sea, and all were baptized into Moses in the cloud and in the sea’. Paul is referring here to Exod. 13:21 and Exod. 14. Chrysologus, serm., 160. Ps. 74:15: ‘You split open springs and brooks; you dried up ever-flowing streams.’, Ps. 77:17: ‘The clouds poured out water; the skies gave forth thunder; your arrows flashed on every side.’ Artistic reactions to this tradition are difficult to verify. See Schrenk and Rexin 2007.

Chrysologus’s Reinterpretation of the River Jordan

Greek.94 Daniel Vigne differentiates between three major types of interpretation of Jordan’s movements: (1) the abrupt cessation of the flow of the waters of Jordan, which is an exclusively eastern tradition; (2) the retreat of the waters in fear of Christ, which is for the most part an eastern tradition and also the most popular version; and (3) the return of the waters, often after they have been called back by Christ, which was also transmitted in the West.95 The oldest interpretation of Jordan’s movements is found in the homily on the Theophany by Pseudo Hippolytus, which dates to approximately the third century, and is a good example of Jordan’s retreat out of fear (Vigne’s type 2): The prophet who had seen this miracle a long time in advance, interrogated them: ‘Why was it, sea, that you fled? Why, Jordan, did you turn back?’ And the waters replied: ‘We saw the creator of the world as a slave; and since we ignored the mystery of the divine frugality, we drew back out of horror.’96

The imperial court at Ravenna, a rich interface between East and West, was probably one of the best places to learn about this exegetical tradition. However, Chrysologus’s own interpretation does not fall into any of the three categories. He was evidently acquainted with the eastern tradition of a retreat out of fear.97 But he does not situate Jordan’s fear and flight at the moment of Christ’s baptism, but earlier, when the Hebrews cross the Jordan (Joshua 3:16–17). Why is it that the Jordan, who fled at the presence of the Ark of the Covenant [Joshua 3:16–17], did not run away when the whole Trinity is present?98 94

95

96 97

98

A comprehensive overview is provided in Jacoby 1902 and, more recently, Vigne 1992, 279–285; Schrenk and Rexin 2007, 191–196. See also Deichmann 1969–1989, vol. 1, 34–35; Nordström 1953, 60; Vollkommer 1981-, 658; Ristow 1957, 120. Vigne 1992, 279–285, on type 1 esp. 279–281. Vigne claims that type 1 does not occur with direct reference to Christ’s baptism, which is disproven by the Chronicon Paschale 42,1/5 (c. 630): ‘John baptized in the River Jordan on the 11th tybi at the tenth hour, and the Jordan drew back [Ps. 113:3]. The Lord said to John: tell the Jordan to stand still, the Lord has come to us. And immediately the waters stood still.’ Transl. by the author. Cf. also the report by the Piacenza Pilgrim (late sixth century) of the Jordan standing still on the Feast of Epiphany during the baptism of the neophytes in the Jordan: Piacenza Pilgrim, Itin. 11. Pseudo Hippolytus, Hom. Theoph., 2. The earliest Latin reception of type 2 is recorded in Sedulius’s (d. c. 450) Carmen paschale, 2,162/5: ‘The elements sensed the God, the sea fled and even/ Jordan turned his course in waves flowing backwards./ Since also the prophet says when he sings: ‘Why was it, sea, that you fled?/ Why, Jordan, did you turn back?’ See Schrenk and Rexin 2007, 192. Chrysologus, serm., 160 ‘. . . quid est quod Jordanis qui fugit ad praesentiam legalis arcae, ad totius Trinitatis praesentiam non refugit’.

185

186

The Conversion of a Personification

Christ’s baptism is for Chrysologus, contrary to most other interpretations, a joyful event for the river. Chrysologus likely took inspiration from Jerome’s (c. 347–420) homily on the Epiphany. Jerome precedes Chrysologus’s interpretation of a Jordan delighted to accompany Christ at his baptism:99 The Jordan River that dried up when Joshua led the Israelites into the Land of Promise [Joshua 3:16–17], now longed to gather together all its waters into one place, if it could, to bathe the body of the Lord.100

For his sermon on the Epiphany, Chrysologus combined the idea of Jerome’s homily of a joyful River Jordan witnessing Christ’s baptism with the exegetical tradition of Psalm 114, which saw Jordan as fleeing from God out of fear. This original and unique exegesis has not yet been adequately recognized as a valuable interpretative tool for the baptismal scene in the Orthodox baptistery. Let us examine Chrysologus’s words again step by step. The declared aim of the passage is to explain the presence of Jordan at the Epiphanic event of Christ’s baptism: Why is it that the Jordan, who fled at the presence of the Ark of the Covenant [Joshua 3:16–17], did not run away when the whole Trinity is present?101

Chrysologus sees fear as the root cause of Jordan’s flight at the crossing; in his interpretation, Jordan is afraid at the crossing but unafraid at the baptism: It is because the one who shows homage and deference begins to be fearful no longer [Ps. 114:3–6]. Here the Trinity puts all its grace into operation and expresses all its love for the world; on that prior occasion the Trinity took hold of the elements in order to instruct its little servants to fear it.102

99

100 101

102

It is only from the late fifth or early sixth century onwards that a few eastern commentators begin to interpret the Psalms as an expression of joy and excitement on the part of the watery elements present at the baptism. See Jacob of Serough, On the appearance of the Lord and his baptism in the Jordan, and Sophronius of Jerusalem, Rituale Graecorum in Schrenk and Rexin 2007, 192–193 in the original and with a German translation. Jerome, Hom. 89, transl. Ewald (ed.) 1964, vol. 2, 230. Chrysologus, serm., 160 ‘. . . quid est quod Jordanis qui fugit ad praesentiam legalis arcae, ad totius Trinitatis praesentiam non refugit’. Chrysologus, serm., 160 ‘Quia qui pietati obsequitur, incipit non esse timori. Hic Trinitas exercet gratiam totam, totam secum loquitur charitatem: ibi elementa corrupit, ut servulos instituat ad timorem.’

The Personifications of Jordan

In both passages, taken from the Old and New Testaments, respectively, it is the Trinity which provokes the emotional response in Jordan. In the presence of the Ark of the Covenant, the Trinity frightens the ‘little servants’ (servulos), while at Christ’s baptism the Trinity acts with grace and charity. Accordingly, in the Old Testament Jordan acts under compulsion (elementa corrupit), whereas at the baptism, the river, inspired by grace and charity, obeys piety (pietati obsequitur). Jordan’s change in behaviour marks a fundamental shift from the Jewish relation to God, as Chrysologus sees it, in which obedience is motivated by fear, to the Christian relation to God. The key Christian virtues of grace and charity produce a voluntary piety.103 In other words, Chrysologus describes the conversion of Jordan as a consequence of the Epiphanic event of Christ’s baptism. In his Old Testament encounter with God, the River Jordan witnessed the Trinity mediated by the presence of the ‘lawful Ark (of the Covenant)’ (ad praesentiam legalis arcae). At this point in history, of course, the Trinitarian nature of God had not yet been revealed, and so Jordan experienced it only in the form of the ark symbolizing the Old Law. At the baptism Jordan faces the whole Trinity for the first time, which, in its wholeness, is emblematic of the revelation of Christian as opposed to Jewish truth, and is thus compelling to the river, transforming it fundamentally.

3.5

The Personifications of Jordan in the Orthodox and Arian Baptisteries: Responses to Chrysologus and Representations of Antiquity

That Chrysologus’s exegesis inspired the unique representation of Jordan in the Orthodox baptistery (Figure 3.2b) seems likely. As I have shown above, it is the only extant river personification in late antiquity which reflects active support of the Christian faith. The artists went a step further by nuancing the relationship and hierarchy between the participants of the baptism scene through the careful rendition of their physical characteristics and postures. With his tanned body and scrubby strands of blue and green hair, Jordan carries himself in a manner distinct from the fairer and larger bodies of Jesus and John. Although athletic and fitted with attributes worthy of a river god, Jordan, venerating Christ in an appropriate

103

Anti-Jewish rhetoric is common in Chrysologus’s sermons. See Allen 2018, 148.

187

188

The Conversion of a Personification

Christian fashion, seems fully content with his new role as doting servant. The representation of Jordan stresses both his excellence and his inferiority; Jordan is characterized by his otherness, his marked difference from the superior bodies of Christ and his baptizer. In this carefully arranged image, Jordan is represented as a natural force that has been subdued and converted. But is it likely that Chrysologus’s Sermon 160 had a tangible impact on the design of the cupola mosaic? Neon, its patron, took up a central role in the united effort of Ravenna’s bishops to express the importance of the see through the episcopal architecture.104 He adorned and finished not only the Orthodox baptistery, begun under Ursus or possibly Chrysologus, but also Chrysologus’s monumental Basilica Petriana. While we do not know anything about the relationship between Peter Chrysologus and his successor Neon, it stands to reason that Ravenna’s most famous bishop left a mark on various of his successor’s endeavours.105 To Ravenna’s eighth- and ninth-century clerics, Chrysologus’s fame derived especially from his sermons. The first collections of a total of 179 preserved sermons are recorded in three codices from the second half of the sixth to the seventh century. Later, Bishop Felix of Ravenna (sed. 709–725) collected, indexed, and published Chrysologus’s sermons, and in the ninth century he was called ‘golden word’ by Agnellus.106 Chrysologus’s fame as a preacher is further indicated by at least a dozen homilies wrongly attributed to him.107 Besides falsifications, the broad transmission of sermons – leading to reuse, shortening, and adaptations – is well attested for late antique preachers, and concerns, in particular, the works of the most famous preachers of the period.108 When we keep the common practice of sharing and adapting sermons in mind, an old argument brought forward against Chrysologus’s influence on the mosaic loses its urgency. Deichmann objected to Nordström’s attribution of the design to Chrysologus because only two of the three persons of the Trinity are represented in the medallion. Chrysologus’s insistence on the Trinity’s influence on the Jordan in Sermon 160, Deichmann argued, would have made the depiction of the hand of God necessary.109 However, corresponding representations in scenes of Jesus’s baptism, in which the 104

105 106 108 109

On the almost ceaseless chain of episcopal building activity in the fifth and sixth centuries, see Deliyannis 2016; Jäggi 2016. On Chrysologus’s distinguished role amongst Ravenna’s bishops, see Deliyannis 2010, 84–85. 107 Bizzozero 2018, 403. Allen 2018, 149. See also Bizzozero 2018, 404–410. Rebillard 2018, 89–93. On Chrysologus especially, see Allen 2018, 149. Deichmann 1969–1989, vol. 2, 35.

The Personifications of Jordan

depiction of a hand signifies the voice of God the Father acknowledging his son (Mark 1:9–11), date back to the sixth century at the earliest.110 In the third quarter of the fifth century, this visual formula might not have been available to the mosaicists yet. More importantly, there is no need to assume a complete implementation of the sermon into a work of art, as Neon could well have been selective in his reception of Chrysologus’s sermon.111 Under Neon’s patronage of the Orthodox baptistery, one original choice was to personify the river instead of visualizing Jordan as a stream of water, as had been the norm up to this point.112 The labelling of Jordan indicates that the representation was unusual enough. This pictorial choice corresponds well with Chrysologus’s ambition to personify the river as much as possible in his sermon – making him an agent motivated by fear and faith. Chrysologus’s sermon had marked Jordan as once bound by ‘Jewish’ fear but now voluntarily pious thanks to the grace and love of the Trinity. Depicting a personification instead of a body of water had the advantage of being able to show Jordan as an active agent who underwent a conversion, much in line with Chrysologus’s sermon. None of the other late antique personifications of Jordan depicted him as tamed and reverent. Considering how anomalous the representation is, and how well this image fits the description of Jordan in the sermon, we should conclude that Sermon 160 served as a template for the cupola mosaic. However, Neon’s pictorial choice also added a new aspect to Jordan which has no correspondence in the sermon. Putting words into images always opens up the potential for additional layers of meaning, and this holds particularly true for personifications: the conventional type of river personifications available to the mosaicists showed rivers in the tradition of Greek and Roman river gods. When the River Jordan was depicted in the Orthodox baptistery in this ambivalent guise, Jordan’s role as a servile Christian was necessarily complemented by a second aspect, that of a partaker in the classical world of mythology. By embodying Jordan in this shape, the river suddenly outgrew Chrysologus’s interpretation of him as a ‘formerly Jewish’ convert to Christianity. Additionally, Jordan became a representative of the pre-Christian world of classical mythology.

110 111 112

Kötzsche 1986b, 439–440. On the interdependence of art and sermonic literature, see Dresken-Weiland 2018. LCI, vol. 4, 247–255, Taufe Jesu, esp. 248–259; Strzygowski 1885, esp. 3–9; Ristow 1967, 5–18. Cf. Fausone 1982.

189

190

The Conversion of a Personification

As such, the cupola mosaic of the Orthodox baptistery depicts, in a strong self-affirmatory gesture, the beginning of the Christian mission as the Christianization of antiquity. The respectful representation of Jordan underlines the praiseworthiness of the encounter between antiquity and Christianity. The two worlds visually overlap and become inseparably intertwined in the figure of Jordan – legendary Christian river and representative of Graeco-Roman antiquity. Does the same interpretation hold true for the Jordan of the Arian baptistery (Figure 3.1b)? Seated at the right side of Christ, as a counterpart to John standing at Christ’s left, Jordan gestures to Christ. Jordan’s enlargement in size as compared to the prototype, his fair skin colour (which matches the complexion of Christ and John), his more prominent position in the foreground next to the new-born Christ, and the respectful representation of an aged and venerable man, all imply a high appreciation of the River Jordan. The absence of a label is likely due to the public’s increased familiarity with the iconography in Ravenna. His classical appearance might be easily explained as a recurrence of the antique model of river personifications devoid of any additional meaning. However, Jordan features far too prominently in the medallion to be merely a contingent witness to the baptism or a symbol for the river. A common explanation is that Jordan was granted his relative prominence in order to guarantee a harmonized composition.113 Yet, the presence of a highly connotative and similar, but not identical, iconography of the same subject less than half a mile away raises doubts that artistic convention alone determined the choice. Ittai Weinryb has suggested that the increase in prominence and monumentality results from Theoderic’s wellknown admiration of and concern for Roman culture.114 The figure’s emphatically antique traits were to point to the fact that the Jordan of the Arian baptistery represents antiquity.115 Since Theoderic’s patronage of the baptistery is unproven, Weinryb’s argument remains speculative. Suppose Chrysologus’s exegesis was still circulating when the Arian baptistery was decorated. In that case, Jordan’s status as a convert to Christianity might have provoked an Arian response. The Arian baptistery’s mosaic no longer stresses Jordan’s belief in Christianity; rather, it shows him as a natural participant. By approximately 500, it seems, the

113 114

115

Wharton 1995, 122; Ward-Perkins 2007, 271 n. 12. Weinryb 2002, 52–53. On Theoderic’s conservation and building policy, see Arnold 2018, 218–228; Wiemer 2018, 433–443, 453–467. See also Hen 2018, 66–67. Weinryb 2002, 53.

The Personifications of Jordan

river was no longer thought to need conversion. Jordan had become fully Christian. Moreover, if we allow for a continuous understanding of the river as personifying antiquity itself, the message associated with the mosaic of the Orthodox baptistery would have been subtly nuanced: Christianity and antiquity are naturally aligned; Christianity’s triumph encompasses all times, but Christianity also honours its ancestral past. The Arian baptistery would thus surpass the Orthodox model in its praise of antiquity. A final iconographic detail needs to be mentioned in this context. The large, bright red crab claws on top of Jordan’s head (Figure 3.1b) are puzzling, to say the least. No other late antique personification of Jordan wears them – but they are a typical component of depictions of Oceanus (Figure 3.3).116 Could it be possible that the Arians’ wish to reinterpret the Orthodox cupola mosaic went further than previously imagined? As it seems highly unlikely that adding crab claws would have served as an aid to recognizing Jordan, the possibility that the Arians wanted to depict Oceanus instead would be worth exploring. The antique tradition sees Oceanus as the generator of all springs, rivers, and the sea. Best attested for late antiquity is the idea of Oceanus as the primordial stream which flows around the inhabited earth.117 In Christian cosmographic representations of the physical universe, the ocean is typically visualized as a body of water hosting a wide range of marine animals and plants. The most famous depiction of earth surrounded by the ocean is the mosaic floor in the north transept of the basilica of Dumetios in Nikopolis, dating to the second quarter of the sixth century.118 Sometimes, however, personifications visualize earth or ocean instead (Figure 3.17).119 Could the personification in the Arian baptistery signify all water? A popular theme in exegetical literature is how the River Jordan was consecrated or sanctified at Christ’s baptism by the influence of the

116

117

118 119

I thank Gunnar Brands for his feedback on the issue. Jordan wearing crab claws reappears in Carolingian times, for example, on an ivory casket from Metz from the third quarter of the ninth century, today at Herzog Anton Ulrich Museum in Braunschweig, inv. no. MA 59. See also Deichmann 1969–1989, vol. 1.2, 254; Kurth 1901, 196–197; Strzygowski 1885, 11. Durst 2012, 514–519. On the iconographic tradition, see Maguire 1987, 20–30 and 69–72; Dinkler-von Schubert 1972, 97. Maguire 1987, 21–24. On the personifications of the earth in the sixth-century church of the priest John, and of the sixth-century church St George, both in Khirbat Al-Makhayyat (Jordan), see Maguire 1987, 69–72. On the personification of Thalassa in the church of the Apostles in Madaba (Jordan), dated to 578–579, see Lux 1968, 13–14 and 19 with literature on personifications in churches; Noth 1968; 133–135; Piccirillo 1989, 96–106.

191

192

The Conversion of a Personification

Figure 3.17 Madaba, church of the Apostles, detail of the floor mosaic depicting the bust of Thalassa, 578–579. Photography: Steve Welsh. © Manar al Athar.

Holy Ghost or by physical contact with Christ.120 According to some, the sanctification had a purifying effect on the element of water as such.121 Replacing Jordan by Oceanus could have been a way of evoking all the waters of the terrestrial world, and of underlining the effect Jesus’s baptism had on all coming generations of baptizands. Regardless of whether the river personification represented Jordan or Oceanus (or both), the Arian church of Ravenna made sure to develop an original interpretation of this figure and to distinguish itself from Chrysologus’s and Neon’s view of the river Jordan. 120

121

The sanctification of the Jordan at the baptism of Christ is a popular theme. For North Italy, see Ambrose, De sacr., 1,15; Maximus of Turin, serm. de S. Epiphania, 100,3. See further Ephrem the Syrian, Hymnus de Epiphania, 6,1 who ascribes the sanctification to the Holy Ghost. For Ignatius of Antioch, Ad Ephesios, 18,2; Ambrose, De sacr., 1,18; Leo the Great, serm., 1, it is rooted in the physical contact with the body of Christ. According to Ignatius of Antioch, Ad Ephesios, 18,2 Jesus’s baptism had purified the water of the Jordan. This made all the water the source of purification for ordinary humanity. Compare Tertullian, De baptism, 3.2, 4.1, 5.1; Ambrose, De sacramentis, 1.15–18; John Chrysostom, Baptismal Instructions, 11.12. Cf. Gregory of Nyssa, Catechet. Mag., 35,15 who attests to the purifying effect of water. This interpretation privileges water as baptismal medium. See Jensen 2011b, 119–120, 134–136.

Conclusion

3.6

Conclusion

The emergence of a multitude of iconographies of Jordan in scenes of the baptism of Christ in the fifth and sixth centuries was not incidental. These new iconographies relate to exegetical literature in the Greek, Syrian, and Latin tradition which discussed the role of Jordan at the baptism from the fourth century onwards. Chrysologus’s Sermon 160 is a particularly original take on the exegetical tradition which emphasized that Jordan’s conversion turned the ‘Jewish’ river into a ‘Christian’ one. Notably, the mosaic of the cupola of the Orthodox baptistery of Ravenna responded to the stories’ latent theme of Jordan’s status within the Christian realm. However, by visualizing the story, the patron and the artists added additional layers of meaning to it. For the cupola mosaic, we can conclude that a plausible reading emphasizes both Jordan’s conversion to Christianity and his role as a representative of pre-Christian antiquity. The iconographic innovation celebrates an ultimately triumphant Christianity but also pays homage to the classical past. In contrast to the baptismal decorations of Cuicul, Henchir el Koucha, Myrtilis, and Milreu, Bishop Neon’s pictorial choice ostentatiously determines a hierarchical relationship between Christianity and Roman culture – the latter is confined to a noble past era which acknowledges the superiority of the present. The iconography reveals the great importance that was given to underlining this hierarchy at the moment of baptism. At the same time, it points to Christianity’s ambition towards inclusiveness. While Christological differences between the cupola mosaics of the Orthodox and Arian baptisteries are hard to determine, the Arian version of the baptism of Christ differed in respect to the personified river, depicted at eye level with Jesus and John the Baptist. Different interpretations of the Arian personification, as either Jordan or Oceanus, can be imagined. The rendition accentuates, in any case, its classical appearance in the guise of a river god and suppresses any indicators of servility. This nourishes the suspicion that Ravenna’s Arian commissioners were commenting on the relationship between Christianity and classical antiquity. The Arian take on the iconography drops all polemics and elevates the status of antiquity to that of an authoritative ally in Christianity’s baptismal mission.

193

Coda

‘The differences have become smaller and, especially among the young generation, they have actually completely vanished.’1 In 2015, on the twenty-fifth anniversary of German unification, then German president Joachim Gauck described the situation of East and West Germans with these words. The president believed that the younger generation, those born after the fall of the wall in 1989, were completely alike, whether they were born in former West Germany or in the area which used to be the GDR. As has subsequently become apparent, the opinion of many young East Germans today differs from that of President Gauck on this point. Let us return to my comparison at the beginning of this book. Nothing prevents those born ‘after the fact’ – after the replacement of Roman religion with Christianity as the state religion, the fall of the Roman Empire, and Rome’s uncontested cultural dominance – from respecting the lived experiences of their parents and ancestors who lived under very different circumstances, from paying attention to their heritage, or from continuing some traditions of past generations in their daily lives. When these subsequent generations feel aversion towards the past, they are likely to feel it more strongly than others who are not affected by the past in the same way. However, post-communist-transition children who still identify as East Germans, or perhaps second-generation migrants who do not feel completely at home where they were born, as well as many other groups born ‘after the fact’, can only provide us with limited glimpses of what it must have been like to be or aspire to become Roman in the post-Roman period. If even the short-lived GDR profoundly affects the self-image of the children of its former citizens, the impact of thousand-year-old ancient Rome on its descendants must have been of an entirely different quality.

1

194

‘Die Unterschiede sind kleiner geworden und besonders in der jungen Generation, da sind sie doch eigentlich gänzlich verschwunden.’ www.bundespraesident.de/SharedDocs/Reden/DE/ Joachim-Gauck/Reden/2015/10/151003-Festakt-Deutsche-Einheit.html?nn¼1891680, last retrieved 16 June 2023.

Coda

The narrative strategy with which Gauck negated East German identity for the greater goal of German unity has equivalents in late antiquity. In the hope of a complete reunification which influences not only politics and economics but also the hearts and minds of all Germans, the president postulated a post-GDR generation who no longer defines its identity through the past. A clear separation of past and present is also evident in the cupola mosaic of Ravenna’s Orthodox baptistery (Fig. 3.2b). Positioned at a distance from the baptism of Christ, the smallish figure of Jordan, in the guise of a Graeco-Roman river deity and with a gesture of reverence and submission, embodies the relationship Bishop Neon deemed fitting for an encounter between present and past. As we have seen, Jordan functions here as a representation of the past, whether understood as classical antiquity with pagan overtones, as suggested by the visualization, or as a representation of Jewish believers, as suggested by Chrysologus’s sermon 160. In his encounter with Christ, Jordan is shown relinquishing all ancient superstition and instead adoring Christ’s overwhelming superiority. The past denies itself in order to be acceptable to the present. This narrative pattern is well known to students of late antique Christianity. In this book, I have argued that we need to redefine our model of Christian identity in late antiquity and that visual and material evidence can help us do so. If we only follow the lead of elite Christian voices, we would miss a wealth of deviating conceptions of Christian identity. In order to do justice to the muted voices of Christians who were opposed to the equalizing tendencies of Christian apologetics, we need to broaden the spectrum of our sources as well as our methods. We should account for the great social breadth of Christian religious practices, for the variations which occurred across provincial and urban locales, as well as for the horizontal structure of personhood, which can accommodate spiritual concerns, cultural affinities, local rooting, familial affection, gender roles, social standing, and ethnic self-images, all on an equal footing. Identity formation in multi-ethnic societies has become a major concern in the study of (late) antiquity, as it has in a wide array of other disciplines, including anthropology, psychology, and religious studies.2 Historians and archaeologists of the migration period have made significant contributions to the deconstruction of monolithic models of identity and social belonging. As a result of such efforts, notions of the fluidity and 2

See for instance the historical studies Mattingly 2010; Pohl et al. 2018; Kaldellis 2019. Crossdisciplinary perspectives are offered in Josselson and Harway 2012; Benet-Martínez and Hong 2014.

195

196

Coda

permeability of ethnicities have become mainstream concepts in this field, serving to shift scholarly interest to how identities were constructed.3 Moreover, scholarship now widely acknowledges that the concepts of identity promoted by earlier generations of scholars contributed to the legitimization of understandings of fixed ethnicities and, deriving from this discourse, of modern nationhood.4 Conversely, the universalizing concept of religious identity which assigns Christian faith the most determinative social role in the life of a Christian at all times still tends to dominate historical writing on late antique Christian societies. This concept is certainly common in studies of Christian art. Moreover, studies of late antique Christianity not infrequently fail to differentiate between regional, cultural, and class distinctions with respect to how Christianity was actually enacted. In marked contrast to the study of ethnicities, they tend to operate with concepts which leave notions of a monolithic origin for modern Christianity untouched – with all the consequences such assumptions might entail for the construction of Christian identity today.5 Scholars of modern and contemporary religion, on the other hand, have embraced identity theory as indispensable for the study of religions: it is commonplace in religious studies to account for the specific spiritual and social needs of individuals or groups by complementing the study of the institutional and doctrinal core tenets of religious identity with examinations of individual responses to them as well as of popular non-normative practices.6 This non-centrist approach to religious identity should guide all scholarship on late antique and early medieval religiosity, but it is especially important for those religions which became mass movements in that period and are commonly called ‘world religions’ today.7 In this study, I have focused on how the cultural identity of fifth- to seventh-century Christian communities, and more precisely the affirmation of aspects of traditional Roman culture, intersected with the innovations of a Christian-focused identity. I have argued that Christian communities who used visual and material culture with both Roman and Christian cultural affiliations affirmed that Roman culture was an integral

3 5 6

7

4 Pohl and Reimitz (eds.) 1998; Geary 1999; Gillett 2002. Jones and Graves-Brown 1996, 18. See Smith 1990. An overview over this wide field of research is given in Coleman and Collins 2004; Voss Roberts 2010, 49–51; McGuire 2008. On the construction of ‘world religions’, see Masuzawa 2005.

Coda

part of their sense of Christianity. They did so in a key space of Christian identity construction: the baptistery. The case studies demonstrate that Roman culture itself meant different things to different communities, from traditional ritual practice and Roman ludi to mythology. In all cases, however, baptismal art encouraged the faithful to cultivate their Roman identity (or the aspiration to have a Roman identity) as part of their identity as Christians. Their Christianness did not stop these Christians from crafting complex bonds between various aspects of their worlds, both inherited and recent. Moreover, the majority of the case studies show that Roman aspects of their identity could influence how Christians defined their Christianness. For instance, when furnishing their baptisteries with mythological imagery and maintaining the decoration over centuries, the Christian communities of Myrtilis (Mértola) and Milreu violated the code of correct Christian ethical conduct as known from the church fathers. However, these Lusitanian Christians arguably did not perceive their decorative choices as un-Christian. Rather, their positive attitude towards Roman cultural heritage influenced the way in which they performed their Christian identity. In terms of identity theory, we could say that they saw the relationship between Roman culture and the Christian religion as lateral rather than hierarchical.8 Their Christianity did not invariably determine their behaviour; instead, other category memberships prevailed at times, even in religious contexts. As a result, a careful examination of baptismal art and architecture reveals that the lived religion of the communities presented in Chapters 1 and 2 differed from the aspirations formulated by clerical elites between the fifth and seventh centuries. This conclusion can presumably be applied both to lay members and to the clergy of these communities. Baptisteries were under the control of the clergy and served as the catalyst for both conversion to and collective affirmation of the Christian faith. When Christian visual and material culture merged with its Roman, preChristian counterpart in a baptistery, it can be assumed that both the clergy and the lay community supported this choice. The visual and archaeological evidence suggests that the positions of lay practitioners and their ordained leaders regarding the significance of Roman culture in Christian life were more similar than contemporary writing would suggest.

8

Handelman 1977, 191. See also 9–11 in the Introduction.

197

198

Coda

While Bishop Neon in Ravenna insisted on a clear divide between the Roman past and the Christian present, this division is blurred in both rural and urban areas which received less public attention than the western Roman capital. This book’s ambition extends beyond its impact on scholarship in late antiquity. It aims to open new avenues for historical research on identities in general. The preceding case studies demonstrate the benefits of analysing art and architecture for identity research. To this end, let me draw a conclusion about the kind of information we can access through art and architecture and what methodological steps need to be taken to make this a fruitful enterprise. In the current research landscape, historical identity research typically focuses on one of two approaches. It either examines textual records, often written from an individual point of view, and involving conscious ascriptions of certain qualities. These tend to be prescriptive rather than descriptive, at least in the study of late antique Christianity due to the available source material. Alternatively, material culture is often evaluated to gain insights into habitual ways of performing identity. In contrast to these approaches, the study of public art and architecture enables us to gain insights into the construction of communal identities which go beyond the realm of the habitual, and which may challenge the views of individual Christian writers. In the initial discussion regarding Berlin’s former socialist Palace of the Republic (Fig. 0.1) and its successor, the recently opened Humboldt Forum (Fig. 0.2), we were reminded of the power of architecture to serve as a platform for public negotiation of communal identity (Fig. 0.3; 0.4; 0.5). If a community uses a building to create and represent its group identity, as is the case for many late antique baptisteries, it is important to carefully observe the building’s design, decoration, but also later refurbishments, graffiti, and other interventions. By doing so, we gain insights into the lived reality and self-concept of the community, insights which may not be accessible through the examination of written sources or non-artistic material culture. For instance, when I collected source material in order to examine the impact of Roman culture on Christian identities, I encountered depictions of racehorses and hunting scenes, marine imagery suited to decorate baths or dining rooms, as well as sea monsters and legendary heroes. The baptisteries were furnished with elements of popular Roman mass culture. These findings starkly contrast with views of the classical past and its impact on Christian lives presented in writings by educated Christians,

Coda

who often had vested interests in church politics.9 By shifting our primary attention to the material world, we gain a more nuanced, complete, and sometimes unexpected view of those aspects of Graeco-Roman culture which subsequent generations deemed important enough to preserve and integrate into their religious lives. However, the study of public art and architecture presents a methodological challenge because it needs to carefully deduce the views held by the recipients of artworks on the basis of mostly material evidence. This leaves us often unable to distinguish between the perspectives of the recipients, commissioners, and artists. The information we obtain certainly requires additional evidence to make conclusive arguments. Nonetheless, investigating art and architecture broadens our access to more diverse historical perspectives. The case studies have expanded our knowledge about the interrelation of Roman and Christian identity in the post-Roman world in terms of geographical scope, the similarities and differences between rural and urban areas, and the diversity of social class and ethnicity. The ongoing debate in late antique scholarship about the identification of Barbarians with Rome could be enriched by considering proclamations in Arian art regarding Arians’ role in preserving Roman heritage which had not been previously taken into account.10 The analysis of the cupola mosaic in the Ostrogothic baptistery in Ravenna revealed that the commissioners emphasized the relevance of Graeco-Roman antiquity for Christianity more wholeheartedly than the Nicene rendition of the same subject. This finding is particularly relevant considering the inconclusive search for specifically Arian ecclesiastical art, which diverges explicitly in theological terms from the imagery used in Nicene contexts.11 There is little value in redirecting our focus to the visual if we are not prepared to give serious consideration to this type of evidence. The belief that visual culture ultimately depends on textual culture and has limited autonomy as a source of knowledge is deeply ingrained in the history of humanist scholarship.12 However, by allowing visual evidence to take the lead in this study, we have arrived at precisely the opposite conclusion:

9 10

11 12

Mango 1963; Liebeschütz 1995; Cameron 1996a. See for instance Stouraitis 2018, 131–132; Pohl 2018, 23–24; Conant 2015, 191–192; Pohl 2013, 22; Greatrex 2000, 277. Ward-Perkins 2007; Mîrşanu 2009; Bockmann 2014. Squire 2009, 16; Koerner 2004, 27–51. On the challenges of material culture studies in the field of history of religion, see Uehlinger 2015.

199

200

Coda

namely that our predominantly text-based understanding of late antique baptism requires modification. For example, examining the layout of Cuicul’s baptistery-cum-bath has led us to believe that the architecture facilitated pre-baptismal cleansings at the baptistery site, aligning with pre-ritual washings which were likely still practised in various non-Christian religious contexts during the same period. This observation challenges Augustine’s view that the custom of collective bathing before baptism was merely a social activity.13 In Henchir el Koucha’s baptistery, we encountered a mosaic floor detail featuring a pair of horses reminiscent of Roman circus imagery. The image survived despite the fact that late antique baptizands were typically required to renounce the devil by explicitly referring to the idolatrous nature of the Roman games, including the circus.14 Architecture and art are more than the subservient materialization of orthodox Christian liturgy. They may even have functioned as ‘hidden transcripts’, offering in a gesture of silent resistance a local counter-image to Christian doctrine.15 A balanced handling of all available sources, ideally through close cooperation between research disciplines and employing a broad methodological framework, is crucial for advancing late antique scholarship.

13

See 72–75 in Chapter 1.

14

See 99–101 in Chapter 2.

15

Scott 1990.

References

Aasgaard, Reidar. 2011. ‘Ambrose and Augustine. Two Bishops on Baptism and Christian Identity’, in David Hellholm, Tor Vegge,
yvind Norderval, and Christer Hellholm (eds.), Ablution, Initiation, and Baptism: Late Antiquity, Early Judaism, and Early Christianity (De Gruyter: Berlin): 1253–1282. Abulafia, David. 2011. The Great Sea: A Human History of the Mediterranean (Allen Lane: London). Acquaro, Enrico (ed.). 1991. Monete puniche nelle collezioni italiane. Vol. 3. Bolletino di numismatica (Istituto Poligrafico e Zecca dello Stato: Rome). Adrych, Philippa, Robert Bracey, Dominic Dalglish, Stefanie Lenk, and Rachel Wood. 2017. Images of Mithra (Oxford University Press: Oxford). Agnellus of Ravenna. 2006. Agnelli Ravennatis Liber pontificalis ecclesiae Ravennatis, ed. Deborah Mauskopf Deliyannis (Brepols: Turnhout). Ahbe, Thomas. 2013. ‘Die ostdeutsche Erinnerung als Eisberg’, in Elisa GoudinSteinmann and Carola Hähnel-Mesnard (eds.), Ostdeutsche Erinnerungsdiskurse nach 1989. Narrative kultureller Identität (Frank & Timme: Berlin): 27–58. Alarcão, Adíla Moutinho de, Vergílio Correia, João M. Bairrão Oleiro, and Rui Nunes Pedroso. 1992. Conimbriga. Casa dos Repuxos (Instituto Português de Museus/Museu Monográfico de Conímbriga: Coimbra). Albertini, Eugène. 1922. ‘Inscription chrétienne sur mosaïque provenant de Djémila’, Bulletin archéologique du Comité des travaux historiques et scientifiques. Fasc. B, Afrique du Nord: 26–32. 1934. ‘L’archéologie chrétienne en Algérie’, in Pontificio Istituto di Archeologia Cristiana (ed.), Atti del III Congresso internazionale di archeologia cristiana (Pontificio Istituto di archeologia cristiana: Rome): 411–428. Albiol López, Esther 2013. ‘Una pintura de sostre de l’antiguitat tardana al baptisteri de Barcelona’, Quarhis – Quaderns d’arqueología i història de la ciutat de Barcelona, 9: 165–183. Albrecht, Janico, Christopher Degelmann, Valentino Gasparini, Richard Gordon, Maik Patzelt, Georgia Petridou, Rubina Raja, Anna-Katharina Rieger, Jörg Rüpke, Benjamin Sippel, Emiliano Rubens Urciuoli, and Lara Weiss. 2018. ‘Religion in the Making: The Lived Ancient Religion Approach’, Religion, 48.4: 568–593.

201

202

References

Alföldi-Rosenbaum, Elisabeth, and John Bryan Ward-Perkins. 1980. Justinianic Mosaic Pavements in Cyrenaican Churches (L’Erma di Bretschneider: Rome). Allais, Yvonne. 1938. Djémila (Société d’édition Les belles lettres: Paris). 1967. ‘Une basilique cimétériale à Djemila (Cuicul)’, Bulletin d’Archéologie Algérienne, 1: 189–205. Allen, Pauline. 2018. ‘Impact, Influence, and Identity in Latin Preaching. The Cases of Maximus of Turin and Peter Chrysologus of Ravenna’, in Anthony Dupont, Shari Boodts, Gert Partoens, and Johan Leemans (eds.), Preaching in the Patristic Era. Sermons, Preachers and Audiences in the Latin West (Brill: Leiden): 135–157. Álvarez Martínez, José María. 1990. Mosaicos romanos de Mérida: nuevos hallazgos (Ministerio de Cultura. Dirección General de Bellas Artes y Archivos: Madrid). Amory, Patrick. 1997. People and Identity in Ostrogothic Italy, 489–554 (Cambridge University Press: Cambridge). Angenendt, Arnold. 2009. ‘Kirche als Träger der Kontinuität’, in Theo Kölzer and Rudolf Schieffer (eds.), Von der Spätantike zum frühen Mittelalter: Kontinuitäten und Brüche, Konzeptionen und Befunde (Thorbecke: Ostfildern): 101–141. Antonellini, Caterina. 2012. ‘Restauri ottocenteschi ai mosaici di Ravenna: gli interventi di Felice Kibel alla cupola del Battistero Neoniano’, Ravenna, studi e ricerche, 17.1: 175–195. Apostolos-Cappadona, Diane. 2011. ‘I Understand the Mystery, and I Recognize the Sacrament: On the Iconology of Ablution, Baptism and Initiation’, in David Hellholm, Christer Hellholm,
yvind Norderval, and Tor Vegge (eds.), Ablution, Initiation, and Baptism. Late Antiquity, Early Judaism, and Early Christianity (Walter de Gruyter: Berlin): 1721–1741. Appadurai, Arjun. 1986. ‘Commodities and the Politics of Value’, in Arjun Appadurai (ed.), The Social Life of Things: Commodities in Cultural Perspective (Cambridge University Press: Cambridge): 3–63. Appiah, Kwame Anthony. 2018. The Lies that Bind. Rethinking Identity (Liveright: New York). Arbeiter, Achim, and Dieter Korol. 2016. Der Kuppelbau von Centcelles: Neue Forschungen zu einem enigmatischen Denkmal von Weltrang (Philipp von Zabern: Mainz). Arbi Nsiri, Mohammed. 2015. ‘La survivance du Donatisme après la conférence de Carthage de 411’, in Paola Ruggeri (ed.), L’Africa romana: Momenti di continuità e rottura (Carocci: Rome): 1195–1204. Arce, Javier. 2002. Centcelles: el monumento tardorromano: iconografía y arquitectura (L’Erma di Bretschneider: Rome). 2005. ‘Spain and the African Provinces in Late Antiquity’, in Kimberly Diane Bowes and Michael Kulikowski (eds.), Hispania in Late Antiquity: Current Perspectives (Brill: Leiden, Boston, MA): 341–361.

References

2011a. Esperando a los árabes: los visigodos en Hispania (507–711) (Marcial Pons Historia: Madrid). 2011b. ‘Fana, templa, delubra destrui praecipimus: The End of the Temples in Roman Spain’, in Luke Lavan and Michael Mulryan (eds.), The Archaeology of Late Antique Paganism (Brill: Leiden, Boston, MA): 195–208. 2018. ‘Goths and Romans in Visigothic Hispania’, in Walter Pohl, Clemens Gantner, Cinzia Grifoni, and Marianne Pollheimer-Mohaupt (eds.), Transformations of Romanness. Early Medieval Regions and Identities (De Gruyter: New York, Berlin): 371–378. Ardeleanu, Stefan. 2018. ‘Directing the Faithful, Structuring the Sacred Space: Funerary Epigraphy in the Archaeological Context in Late-Antique Tipasa’, Journal of Roman Archaeology, 31: 475–500. Arnold, Jonathan J. 2018. Theoderic and the Roman Imperial Restoration (Cambridge University Press: Cambridge). Assmann, Jan. 1992. Das kulturelle Gedächtnis. Schrift, Erinnerung und politische Identität in frühen Hochkulturen (C.H.Beck: Munich). Azuara Galve, Sara, and Carolina Villagordo Ros. 2007. La villa romana de La Loma del Regadío (Urrea de Gaén) (Prames: Zaragoza). Bady, Guillaume, and Laurence Foschia. 2014. ‘Chrétiens, rabbins, païens dans le même bain? Les bains dans l’Orient romain (IVe-VIIe s.) ou comment s’en accommoder’, in Marie-Françoise Boussac, Sylvie Denoix, Thibaud Fournet, Bérangère Redon, Institut français d’archéologie orientale du Caire and Institut français du Proche-Orient (eds.), 25 siècles de bain collectif en Orient: Proche-Orient, Égypte et péninsule Arabique: balaneia ¼ thermae ¼ hammāmāt: actes du 3e colloque international Balnéorient (Institut français d’archéologie orientale: Cairo): 985–1000. Balil, Alberto. 1960a. ‘El mosaico romano de la Iglesia de San Miguel’, Cuadernos de Arqueología e Historia de la Ciudad, 1: 21–74. 1960b. ‘Mosaico de Bellerofonte y la Quimera, de Torre de Bell-Lloch (Gerona)’, Archivo Español de Arqueología, 33: 98–112. 1975. ‘Sobre los mosaicos Romanos de Galicia: identificación de un taller musivario’, in Henri Stern and Marcel Le Glay (eds.), La mosaïque grécoromaine II: actes du IIe Colloque international pour l’étude de la mosaïque antique, Vienne, 30 août-4 septembre 1971 (A.&J. Picard: Paris): 259–263. Ballu, Albert. 1921. ‘Les ruines de Djemila (antique Cuicul)’, Revue Africaine, 62: 204–274. 1923. Rapport sur les travaux de fouilles et de consolidations effectués en 1922 aux monuments historiques de l’Algérie (Jules Carbonel: Algiers). Balty, Janine. 2003. ‘La place des mosaïques de Jordanie au sein de la production orientale’, in Noël Duval and Jean-Charles Balty (eds.), Les églises de Jordanie et leurs mosaïques: actes de la journée d’études sur les églises de Jordanie et leurs mosaïques organisée à la occasion de l’inauguration de l’exposition “Mosaïques Byzantine de Jordanie” au musée de la Civilisation

203

204

References

gallo-romaine à Lyon en avril 1989 (Institut français d’archéologie du Proche-Orient: Beirut): 153–188. Baratte, François. 2018. ‘Les villes du nord de l’Afrique entre Antiquité tardive et conquête arabe. Historiographie récente et nouvelles perspectives’, in Sabine Panzram and Laurent Callegarin (eds.), Entre civitas y Madina: el mundo de las ciudades en la Península Ibérica y en el norte de África, siglos IV-IX (Casa de Velázquez: Madrid): 191–201. Baratte, François, and Fathi Bejaoui. 2001. ‘Églises urbaines, églises rurales dans la Tunisie paléochrétienne: nouvelles récherches d’architecture et d’urbanisme’, Comptes rendus des séances de l’Académie des inscriptions et BellesLettres, 145.4: 1447–1498. Baratte, François, Fathi Bejaoui, Noël Duval, Sarah Berraho, Isabelle Gui, and Hélène Jacquest. 2014. Basiliques chrétiennes d’Afrique du Nord. II. Inventaire des monuments de la Tunisie (Ausonius: Bordeaux). Barber, Sam. 2018. ‘Defining Difference or Connecting Spaces? Similarity and Meaning in the Arian Baptistery, Ravenna’, in Meg Boulton, Jane Hawkes and Heidi Stoner (eds.), Place and Space in the Medieval World (Routledge: London): 149–158. Barney, Stephen Allen, W. J. Lewis, J. A. Beach, and Oliver Berghof (eds.). 2006. The Etymologies of Isidore of Seville (Cambridge University Press: Cambridge). Barnish, Sam J., and Federico Marazzi (eds.). 2007. The Ostrogoths. From the Migration Period to the Sixth Century (Boydell Press: Woodbridge). Barral i Altet, Xavier. 1978. Les mosaïques romaines et médiévales de la Regio Laietana (Universidad de Barcelona. Instituto de Arqueología y Prehistoria: Barcelona). Barth, Fredrik. 1969. Ethnic Groups and Boundaries: The Social Organization of Culture Difference (Universitetsforlaget: Oslo). Beeson, Anthony, Matt Nichol, and Joy Appleton. 2019. The Boxford Mosaic: A Unique Survivor from the Roman Age (Countryside Books: Newbury). Bejaoui, Fathi. 2001. ‘Deux mosaïques tardives de la région de Sbeïtla, l’antique Sufetula en Tunisie’, Comptes rendus des séances de l’Académie des inscriptions et Belles-Lettres, 45.1: 489–515. 2016. ‘Quelques nouvelles découvertes d’époque chrétienne en Tunisie’, in Vincenzo Nicolai Fiocchi, Gabriele Castiglia and Olof Brandt (eds.), Acta XVI congressus internationalis archaeologiae christianae (Romae 2228.9.2013): Costantino e i Costantinidi: l’innovazione costantiniana, le sue radici e i suoi sviluppi (Pontifico Istituto di Archeologia Cristiana: Vatican City): 1679–1690. Belayche, Nicole. 2007. ‘Des lieux pour le « profane » dans l’empire tardo-antique? Les fêtes entre koinônia sociale et espaces de rivalités religieuses’, Antiquité Tardive, 15: 36–46. Beltrán de Heredia Bercero, Julia. 2017. ‘Los nuevos hallazgos arqueológicos de la basílica dels sants màrtirs Just i Pastor y la dualidad de baptisterios en

References

Barcelona’, in Julia Beltrán de Heredia Bercero (ed.), La dualitat de baptisteris en les ciutats episcopals del cristanisme tardoantic. Actes del I simposi d’arqueologia cristiana (Ateneu Universitari Sant Pacià: Barcelona): 99–125. Beltrán de Heredia Bercero, Julia, and Cristina Godoy Fernández (eds.). 2017. La dualitat de baptisteris en les ciutats episcopals del cristianisme tardoantic: actes del I Simposi d’arqueologia cristiana, Barcelona, FHEAG-AUSP, 26–27 de maig de 2016 (Ateneu Universitari Sant Pacià: Barcelona). Beltrán de Heredia, Julia, and Charles Bonnet. 2007. ‘Nouvelles données sur le baptistère de Barcelona’, in Mario Marcenaro (ed.), Albenga Città Episcopale. Tempi e dinamiche della cristianizzazione tra Liguria di Ponente e Provenza (Istituto Internazionale di Studi Liguri: Genoa): 717–820. Ben Abed-Ben Khader, Aïcha, Mongi Ennaifer, Marie Spiro, and Margaret A. Alexander. 1985. Thuburbo Majus. Les mosaïques de la région des grands thermes (Institut National d’Archéologie et d’Art: Tunis). Ben Abed-Ben Khader, Aïcha, Michel Fixot, and Sylvestre Roucole. 2011. Sidi Jdidi II. Le groupe épiscopal (École française de Rome: Rome). Benericetti, Ruggero. 1994. II Pontificale di Ravenna. Studio critico (Seminario Vescovile Pio XII: Faenza). Benet-Martínez, Verónica, and Ying-yi Hong. 2014. The Oxford Handbook of Multicultural Identity (Oxford University Press: New York). Bierbaum, Lee Kirsten. 2014. Die Ausstattung des Lateranbaptisteriums unter Urban VIII. (Michael Imhof Verlag: Petersberg). Bisconti, Fabrizio. 2000. Temi di Iconografia Paleocristiana (Pontificio Istituto di Archeologia Cristiana: Vatican City). 2001. ‘L’iconografia dei battisteri paleocristiani in Italia’, in Pontificio Istituto di archeologia cristiana (ed.), L’edificio battesimale in Italia. Aspetti e problemi (atti dell’VIII congresso nazionale di archeologia cristiana: Genova, Sarzana, Albenga, Finale Ligure, Ventimiglia, 21-26 settembre 1998): 406–460. Bitrakova-Grozdanova, Vera. 1975. Monuments paléochrétiens de la région d’Ohrid: Starohristijanski spomenici vo Ohridsko (Inst. pour la protection des monuments de la culture et Musée national d’Ohrid: Ohrid). Bizzozero, Andrea. 2018. ‘Peter Chrysologus’, in Anthony Dupont, Shari Boodts, Gert Partoens and Johan Leemans (eds.), Preaching in the Patristic Era. Sermons, Preachers and Audiences in the Latin West (Brill: Leiden): 403–429. Blanc, André. 1957. ‘Le baptistère de Valence (Drôme)’, Gallia, 15: 87–116. Blanchard-Lemée, Michèle. 1975. Maisons à mosaïques du quartier central de Djemila, Cuicul (Ophrys: Paris). 1996. ‘The Sea: Fish, Ships and Gods’, in Michèle Blanchard-Lemée and Gilles Mermet (eds.), Mosaics of Roman Africa: Floor Mosaics from Tunisia (British Museum Press: London): 121–145. Blanco-Rotea, Rebeca, Rosa Benavides García, Jorge Sanjurjo Sánchez, and Daniel Fernández Mosquera. 2009. ‘Evolución constructiva de Santa Eulalia de Bóveda (Lugo, Galicia)’, Arqueología de la Arquitectura, 6: 149–198.

205

206

References

Blas de Roblès, Jean-Marie, and Claude Sintès. 2003. Sites et monuments antiques de l’Algérie (Édisud: Aix-en-Provence). Blázquez, José María. 1981. Mosaicos romanos de Córdoba, Jaén y Málaga (Instituto Español de Arqueología “Rodrigo Caro”: Madrid). 2008. Arte y religión en el Mediterráneo antiguo (Cátedra: Madrid). Blázquez, José María, and Teógenes Ortego y Frías. 1983. Mosaicos romanos de Soria (Instituto Español de Arqueología “Rodrigo Caro” del Consejo Superior de Investigaciones Científicas: Madrid). Blázquez Martínez, José María 1994. ‘Mosaicos hispanos de tema homérico’, in César María Batalla (ed.), VI Coloquio Internacional sobre Mosaico Antiguo (Palencia – Mérida 1990, Guadalajara 1994) (Asociación Española del Mosaico: Palencia): 279–292. Bockmann, Ralf. 2013. Capital Continuous: A Study of Vandal Carthage and Central North Africa from an Archaeological Perspective (Reichert: Wiesbaden). 2014. ‘The Non-Archaeology of Arianism – What Comparing Cases in Carthage, Haidra and Ravenna can tell us about “Arian” Churches’, in Guido Michael Berndt and Roland Steinacher (eds.), Arianism: Roman Heresy and Barbarian Creed (Ashgate: Farnham): 201–218. Bodel, John Putnam, and Saul Mitchell Olyan. 2008. Household and Family Religion in Antiquity (Blackwell: Oxford). Bouras, Laskarina. 1982. ‘Byzantine Lighting Devices’, Jahrbuch der Österreichischen Byzantinistik, 32.3: 479–491. Bourdieu, Pierre. 1991. ‘Identity and Representation: Elements for a Critical Reflection on the Idea of Region’, in John Bookshire Thompson (ed.), Language and Symbolic Power (Polity: Cambridge): 220–228. Bourdy, Franck. 1990. ‘Les chevaux des courses de chars à Rome’, in Véronique Fuentes, Véronique Kramérovskis and Christian Landes (eds.), Le Cirque et les courses de chars. Rome-Byzance: catalogue de l’exposition (Editions Imago, Musée archéologique Henri Prades: Lattes): 147–148. Bovini, Giuseppe. 1957. ‘Nota sulla successione delle antiche fasi di lavoro nella decorazione musiva del Battistero degli Ariani’, Felix Ravenna, 75: 5–24. 1966. ‘Principali restauri compiuti nel secolo scorso da Felice Kibel nei mosaici di S. Apollinare Nuovo di Ravenna’, Corso di Cultura sull’arte ravennate e bizantina, 13: 83. 1969. Edifici di culto di Ravenna d’età preteodoriciana (Pàtron: Bologna). Bovini, Giuseppe, Friedrich Wilhelm Deichmann, Hugo Brandenburg, Thilo Ulbert, and Jutta Dresken-Weiland. 1967–2003. Repertorium der christlichantiken Sarkophage (F. Steiner: Wiesbaden). Bowersock, Glenn Warren. 1990. Hellenism in Late Antiquity: Thomas Spencer Jerome Lectures (University of Michigan Press: Ann Arbor). 2006. Mosaics as History: The Near East from Late Antiquity to Islam (Belknap Press of Harvard University Press: Cambridge, MA, London).

References Bowes, Kim. 2005. ‘“Une coterie espagnole pieuse”: Christian Archaeology and Christian Communities in Fourth- and Fifth-Century Hispania’, in Kim Bowes and Michael Kulikowski (eds.), Hispania in Late Antiquity: Current Perspectives (Brill: Leiden, Boston, MA): 189–258 2008. Private Worship, Public Values, and Religious Change in Late Antiquity (Cambridge University Press: Cambridge). Bradshaw, Paul Frederick. 2002. The Search for the Origins of Christian Worship: Sources and Methods for the Study of Early Liturgy (Oxford University Press: London, New York). 2006. ‘The Profession of Faith in Early Christian Baptism’, Evangelical Quarterly, 78: 101–115. Bradshaw, Paul Frederick, Maxwell E. Johnson, L. Edward Phillips, and Harold William Attridge. 2002. The Apostolic Tradition: A Commentary (Fortress Press: Minneapolis). Brandenburg, Hugo. 1968. ‘Bellerophon christianus? Zur Deutung des Mosaiks von Hinton St. Mary und zum Problem der Mythendarstellungen in der kaiserzeitlichen dekorativen Kunst’, Römische Quartalschrift, 63.1/2: 49–86. 1983. ‘Die Darstellungen maritimen Lebens’, in Herbert Beck and Peter Cornelis Bol (eds.), Spätantike und frühes Christentum. Ausstellung im Liebighaus 16. Dezember 1983-11. März 1984 (Liebighaus: Frankfurt a.M.): 249–259. 1995. ‘Altar und Grab, zu einem Problem des Märtyrerkultes im 4. und 5. Jahrhundert’, in Mathijs Lamberigts and Peter van Deun (eds.), Martyrium in Multidisciplinary Perspective: Memorial Louis Reekmans (Leuven University Press: Leuven). Brandt, Olof. 2006a. ‘Deer, Lambs and Water in the Lateran Baptistery’, Rivista di archeologia cristiana, 81: 131–156. 2006b. ‘The Lateran Baptistery and the Diffusion of Octagonal Baptisteries from Rome to Constantinople’, in Acta Congressus Internationalis XIV Archaeologiae Christianae Vindobonae 19-26 settembre 1999 (Vatican City): 221–227. 2012. Battisteri oltre la pianta: gli alzati di nove battisteri paleocristiani in Italia (Pontificio Istituto di archeologia cristiana: Vatican City). 2016. ‘Constantinian Baptisteries’, in Olof Brandt and Gabriele Castiglia (eds.), Acta XVI Congressus Internationalis archaeologiae christianae (Ponteficio Istituto di Archeologia Cristiana: Vatican City). Bredekamp, Horst, and Stefan Trinks (eds.). 2017. Transformatio et Continuatio. Forms of Change and Constancy of Antiquity in the Iberian Peninsula 5001500 (De Gruyter: New York). Brent, Allen. 1995. Hippolytus and the Roman Church in the Third Century: Communities in Tension before the Emergence of a Monarch-Bishop (Brill: Leiden). Breschi, Maria Grazia. 1965. ‘La Cattedrale e il Battistero degli Ariani a Ravenna’, Collana di quaderni di antichità ravennati, cristiane e bizantine, 6.

207

208

References

Brons, Martin. 2017. Augustins Trinitätslehre praktisch: Katechese, Liturgie, Predigt. Ritual und Unterweisung auf dem Weg zur Taufe (Mohr Siebeck: Tübingen). Brown, Peter. 1997. ‘Christianization and Religious Conflict’, in Averil Cameron and Peter Garnsey (eds.), The Cambridge Ancient History: Volume 13: The Late Empire, AD 337–425 (Cambridge University Press: Cambridge): 632–664. Brown, Thomas S. 2007. ‘The Role of Arianism in Ostrogothic Italy: The Evidence from Ravenna’, in Federico Marazzi and Sam J. Barnish (eds.), The Ostrogoths from the Migration Period to the Sixth Century: An Ethnographic Perspective (Boydell Press: Woodbridge): 417–441. Brubaker, Rogers. 2004. Ethnicity Without Groups (Harvard University Press: Cambridge, MA). Brubaker, Rogers and Frederick Cooper. (2000). ‘Beyond “Identity”’, Theory and Society, 29.1: 1–47. Bruderer Eichberg, Barbara. 2003. ‘Die Erneuerung des Lateranbaptisteriums durch Sixtus III. (432–440) als Sinnbild päpstlicher Tauftheologie und Taufpolitik. Die Apsismosaiken des Vestibüls und das Taufgedicht Sixtus’ III’, Marburger Jahrbuch für Kunstwissenschaft, 30: 7–34. Buchberger, Erica. 2017. Shifting Ethnic Identities in Spain and Gaul, 500–700: From Romans to Goths and Franks (Amsterdam University Press: Amsterdam). Bühl, Gudrun. 1995. Constantinopolis und Roma. Stadtpersonifikationen der Spätantike (Akanthus: Kilchberg, Zurich). Buhler, Frédéric M. 1975. Les baptistères en France. De l’époque paléochrétienne à la période romane: Inventaire déscriptif (Centre de Culture Chrétienne: Mulhouse). Burke, Peter J. 2003a. Advances in Identity Theory and Research (Kluwer Academic/Plenum Publishers: New York). 2003b. ‘Relationships Among Multiple Identities’, in Peter J. Burke (ed.), Advances in Identity Theory and Research (Kluwer Academic/Plenum Publishers: New York): 195–216. Burke, Peter J., and Jan E. Stets. 2009. Identity Theory (Oxford University Press: New York, Oxford). Burns Jr., James Patout, and Robin Margaret Jensen. 2014. Christianity in Roman Africa. The development of its practices and beliefs (Eerdmans: Grand Rapids, MI, Cambridge). Burrus, Virginia (ed.). 2005. Late Ancient Christianity (Fortress: Minneapolis). Busine, Aude. 2015. Religious Practices and Christianization of the Late Antique City (4th–7th Cent.) (Brill: Leiden, Boston). Caballero Zoreda, Luis, and Juan Carlos Sánchez Santos. 1990. ‘Reutilización de material romano en edificios de culto cristiano’, Antigüedad y cristianismo, 7: 431–485.

References Cagnat, René. 1922. ‘Séance, Janvier 17’, Bulletin archéologique du comité des traveaux historiques et scientifiques: XXVI–XXXII. Cahn, Herbert Adolf. 1994. ‘Okeanos’, in Nikolaos Gialurēs (ed.), Lexicon Iconographicum Mythologiae Classicae (Artemis: Zurich, Munich): 30–33. Cameron, Alan. 2004. Greek Mythography in the Roman World (Oxford University Press: New York, Oxford). 2011. The Last Pagans of Rome (Oxford University Press: Oxford). Cameron, Averil. 1993. The Mediterranean World in Late Antiquity, AD 395-600 (Routledge: London). 1996a. Changing Cultures in Early Byzantium (Variorum: Aldershot). 1996b. ‘Byzantine Africa: The Literary Evidence’, in Averil Cameron (ed.), Changing Cultures in Early Byzantium (Variorum: Aldershot): 1–51. Carr, Karen Eva. 1999. ‘A Changing World – African Red Slip in Roman and Visigothic Baetica’, in Alberto Ferreiro (ed.), The Visigoths: Studies in Culture and Society (Brill: Leiden): 219–261. Cast, Alicia D., and Peter J. Burke. 2002. ‘A Theory of Self Esteem’, Social Forces, 80: 1041–1068. Castelfranchi, Marina Falla. 1995. ‘Battisteri e pellerinaggi’, in Ernst Dassmann and Josef Engemann (eds.), Akten des XII. Internationalen Kongresses für Christliche Archäologie (Aschendorffsche Verlagsbuchhandlung: Münster): 234–248. Cavalieri, Marco. 2002. ‘La maschera di Oceano: valore e simbologia di un’iconografia romana’, Aurea Parma, 1: 49–72. Cayrel, Pierre. 1934. ‘Une basilique donatiste de Numidie’, Mélanges d’archéologie et d’histoire, 51: 114–142. Chamoso, Martín 2007. ‘Villa Romana de Sahelices el Chico (Salamanca). El Mosaico de Belerofonte y la Quimera’, Revista de Arqueología 312: 24–33. Chastagnol, André, and Noël Duval. 1974. ‘La survivance du culte impérial dans l’Afrique du Nord à l’époque vandale’, in William Seston (ed.), Mélanges d’histoire ancienne offerts à William Seston (E. de Boccard: Paris): 87–118. Chavarría Arnau, Alexandra. 2004. ‘Interpreting the Transformation of Late Roman Villas: The Case of Hispania’, in Neil Christie (ed.), Landscapes of Change: Rural Evolutions in Late Antiquity and the Early Middle Ages (Ashgate: Aldershot): 67–102. Chick, Jane. 2014. ‘The Large Pavement at Qasr el-Lebia in Cyrenaica: Episodes, Narratives and Transformations (diss.)’, University of East Anglia, Norwich (unpubl.). Christern, Jürgen. 1976. Das frühchristliche Pilgerheiligtum von Tebessa (Franz Steiner: Wiesbaden). Ciampini, Giovanni Giustino. 1699. Vetera Monimenta; in Quibus Præcipuè Musiva Opera Sacrarum, Profanarumque Ædium Structura, Ac Nonnulli Antiqui Ritus, Dissertationibus, Iconibusque Illustrantur. Vol. 2. (Rome).

209

210

References Cintas, Jean, and George Louis Feuille. 1952. ‘Église et baptistère de Thibiuca’, Karthago, 3.4: 195–205. Clarke, John R. 2003. Art in the Lives of Ordinary Romans: Visual Representation and Non-Elite Viewers in Italy, 100 B.C.-A.D. 315 (University of California Press: Berkeley, London). Coleman, Kathleen Mary. 1990. ‘Fatal Charades: Roman Executions Staged as Mythological Enactments’, Journal of Roman Studies, 80: 44–73. Coleman, Simon, and Peter Collins. 2004. Religion, Identity and Change: Perspectives on Global Transformations (Ashgate: Aldershot). Conant, Jonathan. 2015. Staying Roman: Conquest and Identity in Africa and the Mediterranean (Cambridge University Press: Cambridge). 2016. ‘Donatism in the Fifth and Sixth Centuries’, in Richard Miles (ed.), The Donatist Schism. Controversy and Contexts (Liverpool University Press: Liverpool): 345–386. Cosh, Stephen Robert, and David S. Neal. 2002. ‘Croughton. Mosaic 86.1’, Roman Mosaics of Britain, 1: 234–236. Courtois, Christian. 1955a. Les Vandales et l’Afrique (Arts et métiers graphiques: Paris). 1955b. ‘Sur un baptistère découvert dans la région de Kélibia (Cap Bon)’, Karthago, 6: 98–123. Croci, Chiara. 2019. Tessere per un nuovo inizio. Il battistero paleocristiano di Napoli e i suoi mosaici (Arte’m: Naples). Cruz Villalón, María. 1985. Mérida Visigoda. La escultura arquitectónica y litúrgica (Colección Roso de Luna: Badajoz). Cummins, Sheila. 1994. ‘The Arian Baptistery of Ravenna (diss.)’, Indiana University, Bloomington (unpubl.). Dally, Ortwin. 2003. ‘“Pflege” und Umnutzung der spätrömischen Stadt im östlichen Mittelmeerraum’, in Gunnar Brands and Hans-Georg Severin (eds.), Die spätantike Stadt und ihre Christianisierung (Reichert: Wiesbaden). Danthine, Hélène. 1937. Le palmier-dattier et les arbres sacrés dans l’iconographie de l’Asie occidentale ancienne (P. Geuthner: Paris). Daugherty, Bradley John. 2015. ‘The Bishops of North Africa: Rethinking Practice and Belief in Late Antiquity (diss.)’, Vanderbuilt University, Nashville, TN (https://etd.library.vanderbilt.edu/etd-11232015-215043). Dawson, Mike, and David S. Neal. 2010. ‘New Fragments and Further Excavation of the Bellerophon Mosaic from Croughton’, Britannia, 41: 313–339. Day, Juliette. 2018. ‘Entering the Baptistery. Spatial, Identity and Salvific Transitions in Fourth- and Fifth-Century Baptismal Liturgies’, in Emilie M. van Opstall (ed.), Sacred Thresholds: The Door to the Sanctuary in Late Antiquity (Brill: Leiden, Boston): 66–90. de Blauuw, Sible. 2020. ‘Architektur und Textilien in den spätantiken Kirchen Roms’, in Sible de Blauuw, Elisabet Enss and Petra Linscheid

References

(eds.), Contextus. Festschrift für Sabine Schrenk (Aschendorff: Münster): 144–163. de Blauuw, Sible, and Klára Dolezalová. 2019. ‘Constructing Liminal Space? Curtains in Late Antiquity and Early Medieval Churches’, in Ivan Foletti and Klára Dolezalová (eds.), The Notion of Liminality and the Medieval Sacred Space (Brepols: Turnhout): 46–66. de Bruyne, Lucien. 1957. ‘La décoration des baptistères paleochrétiens’, in Congresso internazionale di archeologia cristiana (ed.), Actes du Ve Congrès International d’Archéologie Chrétienne (Pontificio Istituto di archeologia cristiana: Vatican City). de Waal, Anton. 1896. ‘Taufe Christi auf vorkonstantinischen Gemälden der Katakomben’, Römische Quartalschrift, 10: 335–349. Deichmann, Friedrich Wilhelm. 1969–1989. Ravenna: Hauptstadt des spätantiken Abendlandes (Franz Steiner: Wiesbaden). Dennis, Nathan. 2020. ‘A Tale of Two Inscriptions: Tipasa, Djemila, and the Role of Textual Icons in the North African Cult of Saints’, Mosaic, 47: 20–32. 2018. ‘Bodies in Motion: Visualizing Trinitarian Space in the Albenga Baptistery’, in Jelena Bogdanović (ed.), Perceptions of the Body and Sacred Space in Late Antiquity and Byzantium (Routledge: London, New York): 124–148. 2017. ‘Living Water, Living Presence: Animating Sacred Space in the Early Christian Baptistery’, in Alexei Lidov (ed.), Holy Water in the Hierotopy and Iconography of the Christian World (Russian Academy of Arts: Moscow): 89–117. Di Francesco, Carla, Claudia Tedeschi, Antonella Ranaldi, and Cetty Muscolino. 2011. Il Battistero Neoniano: uno sguardo attraverso il restauro (Longo: Ravenna). Dinkler-von Schubert, Erika. 1972. ‘Fluss 2 (ikonographisch)’, in Theodor Klauser, Franz Joseph Dölger, and Ernst Dassmann (eds.), Reallexikon für Antike und Christentum (Hiersemann: Stuttgart): 73–100. Domagalski, Bernhard. 1990. Der Hirsch in spätantiker Literatur und Kunst: Unter besonderer Berücksichtigung der frühchristlichen Zeugnisse (Aschendorff: Münster). Dresken-Weiland, Jutta. 2016. Die frühchristlichen Mosaiken von Ravenna: Bild und Bedeutung (Schnell & Steiner: Regensburg). 2018. ‘Visual Art and Iconography’, in Anthony Dupont, Shari Boodts, Gert Partoens and Johan Leemans (eds.), Preaching in the Patristic Era. Sermons, Preachers and Audiences in the Latin West (Brill: Leiden): 59–85. Drewer, Lois. 1981. ‘Fisherman and Fish Pond: From the Sea of Sin to the Living Waters’, Art Bulletin, 63.4: 533–547. Droogan, Julian. 2013. Religion, Material Culture and Archaeology (Bloomsbury Academic: London, New York).

211

212

References

Dujarier, Michel. 1979. A History of The Catechumenate: The First Six Centuries (Sadlier: New York). Dunbabin, Katherine. 1978. The Mosaics of Roman North Africa. Studies in Iconography and Patronage (Clarendon Press: Oxford). 2012. ‘Expresiones de cultura en un mundo cristiano: scenas mitológicas en mosaicos del Mediterráneo oriental durante los siglos V y VI D.C’, in Luz Neira (ed.), Civilización y barbarie: el mito como argumento en los mosaicos romanos (CVG: Madrid): 213–229. 2014. ‘Mythology and Theatre in the Mosaics of the Graeco-Roman East’, in Stine Birk, Troels Myrup Kristensen and Birte Poulsen (eds.), Using Images in Late Antiquity (Oxbow Books: Oxford): 227–252. Durán, Mercedes. 1993. Iconografía de los mosaicos romanos en la Hispania altoimperial (Universitat Rovira i Virgili: Barcelona). Durst, Michael, Rita Amedick, and Elisabet Enß. 2012. “Meer”, in Theodor Klauser, Franz Joseph Dölger, und Ernst Dassmann (eds.), Reallexikon für Antike und Christentum (Hiersemann: Stuttgart): 505–610. Duval, Noël. 1971. Les églises africaines à deux absides. Recherches archéologiques sur la liturgie chrétienne en Afrique du Nord. I: Les basiliques de Sbeitla à deux sanctuaires opposés (E. de Boccard: Paris). 1972. ‘Études d’architecture chrétienne nord-africaine’, Mélanges de l’Ecole française de Rome. Antiquité, 84.2: 1071–1172. 1974. ‘Le dossier de l’église d’El Mouassat (au sud-ouest de Sfax, Tunisie)’, Antiquités Africaines, 8: 157–173. 1980. ‘Les baptistères d’Acholla (Tunisie) et l’origine des baptistères polylobés en Afrique du Nord’, Études d’Archéologie Chrétienne Nord-Africaine. IX. Ant. Africains, 15: 329–342. 1989. ‘L’évêque et la cathédrale en Afrique du Nord’, Actes du XIe Congrès international d’archéologie chrétienne: 345–399. 1990. ‘Les prix du cirque dans l’antiquité tardive’, in Christian Landes (ed.), Le cirque et les courses de char. Rome-Byzance (Edition Imago: Lattes). 1992. ‘Recherches nouvelles sur les prix de concours représentés sur les mosaïques’, Bulletin archéologique du Comité des travaux historiques et scientifiques. Fasc. B, Afrique du Nord, 22: 177–209. 2011. ‘Les prix de concours de l’Antiquité tardive: nouveautés, hypothèses et doutes’, Bulletin de la Société nationale des antiquaires de France, année 2004–2005: 250–257. Duval, Noël, and Paul-Albert Février. 1991. ‘La basilique cimétériale de l’est à Djémila. Une église à crypte méconnue’, in Marc Van Uytfanghe and Roland Demeulenaere (eds.), Aevum inter utrumque: mélanges offerts à Gabriel Sanders (Abbatia S. Petri: Steenbrugis). Duval, Noël, and Nejib Ben Lazreg. 1995. ‘Le baptistère de Bekalta’, in ParisMusées (ed.), Carthage. L’histoire, sa trace, son écho: catalogue de l’exposition à Paris (Musée du Petit Palais: Paris): 304–307.

References Edwards, Mark. 2016. ‘The Donatism Schism and Theology’, in Richard Miles (ed.), The Donatist Schism: Controversy and Contexts (Liverpool University Press: Liverpool): 101–119. Elsner, Jaś. 1995. Art and the Roman Viewer (Cambridge University Press: Cambridge). 2007. ‘Genders of Viewing. Visualizing Women in the Casket of Projecta’, in Jaś Elsner (ed.), Roman Eyes: Visuality and Aubjectivity in Art and Text (Princeton University Press: Princeton, Oxford): 200–224. 2012. ‘Material Culture and Ritual: State of the Question,’ in Bonna Daix Wescoat and Robert G. Ousterhout (eds.), Architecture of the Sacred: Space, Ritual, and Experience from Classical Greece to Byzantium (Cambridge University Press: Cambridge): 1–26. Elsner, Jaś, and Michael Squire. 2018. ‘Homer and the Ekphrasists: Text and Picture in the Elder Philostratus’ “Scamander” (Imagines I.1)’, in John L. Bintliff and N. Keith Rutter (eds.), The Archaeology of Greece and Rome: Studies in Honour of Anthony Snodgrass (Edinburgh University Press: Edinburgh): 57–99. Engemann, Josef. 1989. ‘Die Huldigung der Apostel im Mosaik des ravennatischen Orthodoxenbaptisteriums’, in Hans-Ulrich Cain, Hanns Gabelmann, and Dieter Salzmann, (eds.), Festschrift für Nikolaus Himmelmann (Philipp von Zabern: Mainz): 481–489. Ennaifer, Mongi. 1983. ‘Le thème des chevaux vainqueurs à travers la série des mosaïques africaines’, Mélanges de l’Ecole française de Rome. Antiquité, 95.2: 817–858. 1994. ‘La mosaïque aux chevaux d’El-Maharune (près de Thuburbo Minus, l’actuel Thébourba)’, Mélanges de l’Ecole française de Rome. Antiquité, 106: 303–318. Estácio da Veiga, Sebastião Philippes Martins. 1880. A tabula de bronze de Aljustrel: lida, deduzida e commentada em 1876; memoria presentada á la Academia Real das Sciencias de Lisboa (Academia: Lisbon). Ewald, Marie Liguori (ed.). 1964. The Homilies of Saint Jerome (Catholic University of America Press: Washington, DC). Ewert, Christian. 1973. ‘Die Moschee von Mértola (Portugal)’, Madrider Mitteilungen, 14: 217–246. Fabbi, Fulvia. 2011. ‘Tipologie battisteriali nel litorale adriatico: il battistero degli ariani e il battistero degli ortodossi di Ravenna, tra premesse simboliche, archetipi architettonici e geometria’, in Maurizio Tagliaferri (ed.), Architetture del sacro nel bacino adriatico. Figure, forme e liturgie (Edizioni Dehoniane: Bologna): 165–198. Faus, Rainer, and Simon Storks. 2019. Im vereinten Deutschland geboren – in den Einstellungen gespalten? OBS-Studie zur ersten Nachwendegeneration (Otto Brenner Stiftung: Frankfurt a.M.). Fausone, Alfonso. 1982. Die Taufe in der frühchristlichen Sepulkralkunst (Pontificio Istituto di Archeologia Cristiana: Vatican City).

213

214

References Fendri, Mohamed. 1965. ‘Évolution chronologique et stylistique d’un ensemble de mosaïques dans une station thermale à Djebel Oust (Tunisie)’, in La Mosaïque Gréco-Romaine (Éditions du Centre national de la recherche scientifique: Paris): 157–173. Fentress, Elizabeth. 1988. ‘Sétif, les thermes du Ve siècle’, in Attilio Mastino (ed.), L’Africa Romana. Atti del VI Convegno di studio (Edizioni Gallizzi: Sassari): 321–337. Ferguson, Everett. 2009. Baptism in the Early Church: History, Theology, and Liturgy in the First Five Centuries (Eerdmans: Grand Rapids, MI). Fernandez Alonso, Justo. 1955. La cura pastoral en la España romanovisigoda (Iglesia Nacional Española: Rome). Ferrari, Anna. 2011. ‘La rilettura cristiana dei miti pagani’, Antiquité Tardive, 19.1: 209–222. Ferreiro, Alberto (ed.). 1999. The Visigoths: Studies in Culture and Society (Brill: Leiden, Boston). Ferri, Giovanna. 2013. I mosaici del battistero di San Giovanni in Fonte a Napoli (Tau Editrice: Todi). Feuille, George Louis. 1949. ‘Une mosaïque chrétienne de l’Henchir Messaouda’, Cahiers archéologiques, 4: 9–16. Février, Paul-Albert. 1957. ‘Les baptistères de Provence pendant le moyen âge’, Actes du Ve congrès international d’archéologie chrétienne, 5: 423–432. 1965. Fouilles de Sétif. Les basiliques chrétiennes du quartier Nord-Ouest (Éditions du Centre national de la recherche scientifique: Paris). 1966. ‘Remarques sur les mosaïques de basse époque à Djemila (Algérie)’, Bulletin de la Société nationale des antiquaires de France, année 1965: 85–92. 1967. ‘Inscriptions chrétiennes de Djémila (Cuicul)’, Bulletin d’Archéologie Algérienne, 1 (1962–1965): 207–226. 1984. ‘L’abeille et la seiche (à propos du baptistère de Kélibia)’, Rivista di Archeologia Cristiana, 60: 277–292. 1996. ‘Notes sur le développement urbain en Afrique du Nord. Les exemples comparés de Djemila et Sétif (1964)’, in Paul-Albert Février (ed.), La Méditerranée de Paul-Albert Février (reprint) (École française de Rome: Rome): 651–697. Février, Paul-Albert, Adolfo Gaspary, and Roger Guéry. 1970. Fouilles de Sétif: 1959-1966: quartier nord-ouest, rempart et cirque (République algérienne démocratique et populaire, Ministère de l’éducation nationale (antiquités): Algiers). Firatli, Nezih. 1990. La sculpture byzantine figurée au Musée archéologique d’Istanbul (Librairie d’Amérique et d’Orient: Paris). Foletti, Ivan. 2009. ‘Saint Ambroise et le baptistère des orthodoxes de Ravenne. Autour du lavement des pieds dans la liturgie baptismale’, in Ivan Foletti and Serena Romano (eds.), Fons Vitae. Baptême, baptistères et rites d’initiation (IIe-VIe siècle) (Viella: Rome): 121–155.

References

Foletti, Ivan, and Serena Romano (eds.). 2009. Fons Vitae: Baptême, Baptistères et Rites d’initiation (IIe-VIe siècle) (Viella: Rome). Fowden, Garth. 1998. ‘Polytheist Religion and Philosophy’, in Averil Cameron and Peter Garnsey (eds.), The Late Empire, A.D. 337–425 (Cambridge University Press: Cambridge): 538–560. Fowler, Chris. 2010. ‘From Identity and Material Culture to Personhood and Materiality’, in Dan Hicks and Mary Carolyn Beaudry (eds.), The Oxford Handbook of Material Culture Studies (Oxford University Press: Oxford): 352–385. Frakes, Robert, and Elizabeth DePalma Digeser (eds.). 2006. Religious Identity in Late Antiquity (Edgar Kent: Toronto). Frankfurter, David. 1998. Pilgrimage and Holy Space in Late Antique Egypt (Brill: Leiden). Frend, William Hugh Clifford. 1952. The Donatist Church: A Movement of Protest in Roman North Africa (Clarendon Press: Oxford). 1997. ‘Altare Subnixus: A Cult of Relics in the Romano-British Church?’, The Journal of Theological Studies, 48.1: 125–128. Gandolfi, Daniela (ed.). 2001. L’edificio battesimale in Italia: Aspetti e problemi: atti dell’VIII Congresso Nazionale di Archeologia Cristiana, Genova, Sarzana, Albenga, Finale Ligure, Ventimiglia; 21-26 settembre 1998 (Ist. Internazionale di Studi Liguri: Bordighera). Gantner, Clemens, Walter Pohl, and Richard Payne (eds.). 2012. Visions of Community in the Post-Roman World: The West, Byzantium and the Islamic World, 300–1100 (Ashgate: Farnham). Garcia González, Teodoro. 1979. ‘La iglesia desde la conversión de Recaredo hasta la invasion árabe’, in Ricardo Garcia Villoslada (ed.), Historia de la Iglesia en España (Biblioteca de Autores Cristianos: Madrid): 401–748. García Moreno, Luis Agustín. 2001. ‘El Cristianismo y el final de los ludi en las Españas’, in Luis Agustín García Moreno and Sebastián Rascón Marqués (eds.), Hispánia en la Antigüedad Tardía: ocio y espectáculos, Actas del 2 Encuentro Hispania en la Antigüedad Tardía, Alcalá de Henares, 15 a 17 de octubre de 1997 (Universidad del Alcalá: Alcalá Henares): 7–17. García y Bellido, Antonio. 1968. Nueve Studios sobre la Legio VII Gemina y su Campamento en León (Diputacion Provincal de León: León). Gardini, Giovanni. 2017. ‘Battisteri ravennati: architettura, iconografia, liturgia’, in Julia Beltrán de Heredia Bercero and Cristina Godoy Fernández (eds.), La dualitat de baptisteris en les ciutats episcopals del cristianisme tardoantic (Ateneu Universitari Sant Pacià, Facultat Antoni Gaudí d’Història, Arqueologia i Arts Cristianes: Facultat de Teologia de Catalunya: Barcelona): 31–46. Garipzanov, Ildar, Caroline Goodson, and Henry Maguire (eds.). 2017. Graphic Signs of Identity, Faith, and Power in Late Antiquity and the Early Middle Ages (Brepols: Turnhout).

215

216

References

Garrucci, Raffaele. 1872–1881. Storia della arte cristiana nei primi otto secoli della chiesa (G. Guasti: Prato). Gasparini, Valentino, Maik Patzelt, Rubina Raja, Anna-Katharina Rieger, Jörg Rüpke, and Emiliano Urciuoli (eds.). 2020. Lived Religion in the Ancient Mediterranean World: Approaching Religious Transformations from Archaeology, History and Classics (De Gruyter: Berlin, Boston). Gauckler, Paul. 1910. Inventaire des mosaïques de la Gaule et de l’Afrique. Afrique Proconsulaire (Tunisie). Vol. 2. (Ernest Leroux: Paris). 1914. Inventaire des mosaïques de la Gaule et de l’Afrique. Planches. Tunisie. Vol. 2 (Ernest Leroux: Paris). Gaumer, Matthew Alan. 2016. Augustine’s Cyprian: Authority in Roman Africa (Brill: Leiden). Geary, Patrick Joseph. 1999. ‘Barbarians and Ethnicity’, in Peter Brown, Glen Bowersock and André Grabar (eds.), Late Antiquity. A Guide to the Postclassical World (Harvard University Press: Cambridge, MA): 106–129. Gerber, Adolf. 1883. Naturpersonification in Poesie und Kunst der Alten (B. G. Teubner: Leipzig). Gerola, Giuseppe. 1917. ‘La tecnica dei restauri ai mosaici di Ravenna’, Estratto dagli Atti e memorie della R. Deputazione di Storia Patria per la Romagna (Atti per la Romagna), 7: 101–194. 1923. ‘Il restauro del Battistero Ariano di Ravenna’, in Heinrich Glück (ed.), Studien zur Kunst des Ostens: Josef Strzygowski zum sechzigsten Geburtstage von seinen Freunden und Schülern (Avalun-Verlag: Vienna, Hellerau): 112–129. Gessel, Wilhelm. 1981. ‘Monumentale Spuren des Christentums im Römischen Nordafrika’, Antike Welt, 12: 3–76. Geyer, Angelika. 1993. ‘“Ne ruinis urbs deformetur. . .” Ästhetische Kriterien in der spätantiken Baugesetzgebung’, Boreas. Münsterische Beiträge zur Archäologie, 16: 63–77. Ghalia, Taher. 1998. Hergla et les mosaïques de pavement des basiliques chrétiennes de Tunisie: plan, décor et liturgie (Institut national du patrimoine: Tunis). 2002. ‘L’architecture religieuse en Tunisie aux Ve et VIe siècles’, Antiquité Tardive, 10: 213–222. 2006. ‘Landscapes and Scenes from Daily Life in Pavement Mosaics from Ancient Tunisia’, in Aïcha Ben Abed Ben Khader (ed.), Stories in Stone: Conserving Mosaics of Roman Africa: Masterpieces from the National Museums of Tunisia (J. Paul Getty Museum: Getty Conservation Institute, Institut National du Patrimoine: Los Angeles): 31–46. 2009. ‘Mosaike aus der Vandalenzeit: Eine späte Blüte christlich-römischer Kunst’, in Badisches Landesmuseum Karlsruhe (ed.), Das Königreich der Vandalen (Philipp von Zabern: Darmstadt): 273–284. 2016. ‘Recherches sur le christianisme rural en Afrique Proconsulaire IVe – VIIe siècles’, Bulletin de la Société nationale des antiquaires de France, année 2016: 303–333.

References Gheller, Viola. 2017. ‘Identità’ e ‘arianesimo gotico’: genesi di un topos storiografico (Bononia University Press: Bologna). Giet, Stanislas. 1949. Basile de Césarée, Homélies sur l’Hexaéméron (Éditions du Cerf: Paris). Gillett, Andrew. 2002. On Barbarian Identity: Critical Approaches to Ethnicity in the Early Middle Ages (Brepols: Turnhout). Gilson, Etienne (ed.). 1999. Philosophie et incarnation selon saint Augustin, suivi de Saint Augustin: Lettre XVIII, Sermon contre les païens, Dolbeau 26 (Ad Solem: Genoa). Glover, Terrot Reaveley, and Gerald Henry Rendall (eds.). 1931. Tertullian. Apology, De Spectaculis; Marcus Minucius Felix (Harvard University Press: Cambridge, MA). Godoy Fernández, Cristina. 1986. ‘Reflexiones sobre la funcionalidad litúrgica de pequeñas pilas junto a piscinas mayores en los baptisterios cristianos hispánicos’, in Mário Barroca and Manuel Louis Real (eds.), Actas del I Congreso de Arqueología Medieval Española (Diputación General de Aragón, Departamento de Educación y Cultura: Zaragoza): 125–137. 1989. ‘Baptistérios Hispánicos (Siglos IV-VIII). Arqueología y Liturgia’, in Noël Duval and Philippe Pergola (eds.), Actes du XIe Congrès international d’archéologie chrétienne (École française de Rome: Rome): 607–635. 1995. Arqueología y liturgia. Iglesias hispánicas (siglos IV al VIII) (Universitat de Barcelona, Port de Tarragona: Barcelona). 2004. ‘A los pies del templo. Espacios litúrgicos en contraposición al altar: una revisión’, in José María Blásquez Martínez and Antonino González Blanco (eds.), Sacralidad y Arqueología (Universidad de Murcia: Murcia): 473–498. Gombrich, Ernst. 1971. ‘Personification’, in Robert Ralph Bolgar (ed.), Classical Influences on European Culture A.D. 500–1500 (Cambridge University Press: Cambridge): 247–258. Gómez-Moreno, Manuel. 1925. Catálogo Monumental de España. Provincia de León (Ministerio de Instruccíon Pública y Bellas Artes: Madrid). Goodman, Martin. 1994. Mission and Conversion: Proselytizing in the Religious History of the Roman Empire (Clarendon Press: Oxford). Grabar, André. 1946. Martyrium. Récherches sur le culte des reliques et l’art chrétien antique. Architecture (Collège de France: Paris). 1959. ‘À propos d’une icône byzantine du XIVe siècle au musée de Sofia’, Cahiers Archéologique, 10: 289–304. 1961. Christian Iconography. A Study of Its Origins (Princeton University Press: Princeton). Graen, Dennis. 2004. ‘“Sepultus in villa” – Bestattet in der Villa: Drei Zentralbauten in Portugal zeugen vom Grabprunk der Spätantike’, Antike Welt, 35.3: 65–74. 2005a. ‘Kaiserzeitliche Grabmonumente in der Lusitania zwischen Romanisierung und Afrikanisierung’, in Günther Schörner (ed.),

217

218

References Romanisierung – Romanisation. Theoretische Modelle und praktische Fallbeispiele (Archaeopress: Oxford): 135–147. 2005b. ‘Os mosaicos do “santuário” de Milreu, Estói (Algarve), no contexto de uma nova interpretação’, O Arqueólogo Português, 4.23: 367–415. 2005c. ‘The So-Called Temples of Milreu, Sao Cucufate and Quinta de Marim – A Suggestion of a New Interpretation of Their Function, Based on Actual Excavations and Iconographic Studies’, Xelb. Actas do II Encontro do Arqueologia do Algarve, Silves 17.-19.10.2003, 5: 70–80. Grahame, Mark. 1998. ‘Material Culture and Roman Identity: The Spatial Layout of Pompeian Houses and the Problem of Ethnicity’, in Ray Laurence and Joanne Berry (eds.), Cultural Identity in the Roman Empire (Routledge: London): 156–178. Graves-Brown, Paul, Clive Gamble, and Siân Jones. 1996. Cultural Identity and Archaeology: The Construction of European Communities (Routledge: London). Greatrex, Geoffrey. 2000. ‘Roman Identity in the Sixth Century’, in Stephen Mitchell and Geoffrey Greatrex (eds.), Ethnicity and Culture in Late Antiquity (Duckworth; Classical Press of Wales: London, Swansea): 267–292. Grégoire, Henri. 1938. ‘Les baptistères de Cuicul et de Doura’, Byzantion, 13: 589–593. Griesbach, Jochen. 2009. ‘Grabbauten und Villen auf der iberischen Halbinsel’, in Dietrich Boschung (ed.), Grabbauten des 2. und 3. Jahrhunderts in den gallischen und germanischen Provinzen. Akten des Internationalen Kolloquiums Köln, 22. bis 23 Februar 2007 (Reichert: Wiesbaden). Grig, Lucy. 2017. Popular Culture in the Ancient World (Cambridge University Press: Cambridge). Grossmann, Peter. 1986. Abu Mina: A Guide to the Ancient Pilgrimage Center (Fotiadis & Co. Press: Cairo). Guardia, Milagros. 1992. Los Mosaicos de la antigüedad tardía en Hispania: Estudios de iconografía (PPU: Barcelona). Gui, Isabelle, Noël Duval, and Jean-Pierre Caillet. 1992. Basiliques chrétiennes d’Afrique du Nord. Inventaire des Monuments de l’Algérie (Institut d’Études Augustiniennes: Paris). Günther, Linda-Marie. 2015. ‘Pferd und Quadriga auf sikulopunischen Münzen: griechische oder punische Symbolik’, in Linda-Marie Günther and Bärbel Morstadt (eds.), Phönizische, griechische und römische Gottheiten im historischen Wandel (Brepols: Turnhout): 137–152. Hachlili, Rachel. 2009. Ancient Mosaic Pavements. Themes, Issues, and Trends (Brill: Leiden). Hacker, Michael, Judith Enders, Stephanie Maiwald, Johannes Staemmler, Adriana Lettrari, Hagen Pietzcker, Mandy Schulze, and Henrik Schober. 2012. Dritte Generation Ost. Wer wir sind, was wir wollen (Christoph Links Verlag: Berlin).

References

Halsall, Guy. 2007. Barbarian Migrations and the Roman West, 376–568 (Cambridge University Press: Cambridge). Handelman, Don. 1977. ‘The Organization of Ethnicity’, Ethnic Groups, 1: 187–200. Hanfmann, George Maxim Anossov. 1980. ‘The Continuity of Classical Art: Culture, Myth and Faith’, in Kurt Weitzmann (ed.), Age of Spirituality: A Symposium (The Metropolitan Museum of Art: New York): 75–98. Harmless, William. 2014. Augustine and the Catechumenate (Liturgical Press: Collegeville, MN). Harrison, Carol, Caroline Humfress, Isabella Sandwell, and Gillian Clark. 2014. Being Christian in Late Antiquity: A Festschrift for Gillian Clark (Oxford University Press: Oxford). Hartmann, Volker. 2003. ‘Personifikation’, in Gert Ueding (ed.), Historisches Wörterbuch der Rhetorik (Max Niemeyer Verlag: Tübingen): 810–813. Hauschild, Theodor. 1964. ‘Der Kultbau neben dem römischen Ruinenkomplex bei Estoi in der Provincia Lusitania (diss.)’, Technische Universität Berlin (unpubl.). 1980. ‘Milreu/Estoi (Algarve): Untersuchungen neben der Taufpiscina und Sondagen in der Villa – Kampagnen 1971 und 1979’, Madrider Mitteilungen, 21: 189–219. 1984. ‘Die Mosaiken am Podium des Wasserheiligtums von Milreu, Estoi (Algarve)’, in Jean-Pierre Darmon and Alain Rebourg (eds.), La Mosaïque Gréco-Romaine IV (Supplément au Bulletin de l’A.I.E.M.A.: Paris): 285–291. 2002. ‘O “ninfeu” de Milreu’, in Museu Nacional de Arqueologia (ed.), Religiões da Lusitânia. Loquuntur saxa (Museu Nacional de Arqueologia: Lisbon): 241-244. 2008. ‘A arquitectura a os mosaicos do “edifício de culto” ou “aula” da villa romana de Milreu’, Revista de História da Arte, 6: 17–31. Hauschild, Theodor, and Felix Teichner. 2002. Milreu. Ruins (IPPAR – Departamento de Estudos: Lisbon). Heather, Peter John. 1999. The Visigoths from the Migration Period to the Seventh Century: An Ethnographic Perspective (Boydell: Woodbridge). 2007. The Fall of the Roman Empire: A New History of Rome and the Barbarians (Oxford University Press: New York; Oxford). Hellholm, David, Tor Vegge,
yvind Norderval, and Christer Hellholm (eds.). 2011. Ablution, Initiation, and Baptism: Late Antiquity, Early Judaism, and Early Christianity ¼ Waschungen, Initiation und Taufe: Spätantike, frühes Judentum und frühes Christentum (De Gruyter: Berlin, Boston). Hen, Yitzhak. 2007. Roman Barbarians: The Royal Court and Culture in the Early Medieval West (Palgrave Macmillan: Basingstoke). 2018. ‘Compelling and Intense: The Christian Transformation of Romanness’, in Walter Pohl, Clemens Gantner, Cinzia Grifoni and Marianne PollheimerMohaupt (eds.), Transformations of Romanness. Early Medieval Regions and Identities (De Gruyter: Berlin, Boston): 41–57.

219

220

References Henig, Martin. 2012. ‘From Romano-British Hero to Patron-Saint of England: The Transformations of Bellerophon and His Chimera’, in Giuseppina Carlotta Cianferoni, Mario Iozzo and Elisabetta Setari (eds.), Myth, Allegory, Emblem: The Many Lives of the Chimaera of Arezzo: Proceedings of the International Colloquium, Malibu, the J. Getty Museum (December 4–5, 2009) (Aracne: Rome). Herrin, Judith. 2021. Ravenna: Capital of Empire, Crucible of Europe (Princeton University Press: Princeton). Heucke, Clemens. 1994. Circus und Hippodrom als politischer Raum (OlmsWeidmann: Hildesheim). Hidri, Hava, Sali Hidri, and Renate Pillinger. 2011. Die frühchristliche Basilika in Arapaj/Durrës (Albanien) (Verlag der Österreichischen Akademie der Wissenschaften: Vienna). Hiller, Stefan. 1970. Bellerophon. Ein griechischer Mythos in der römischen Kunst (Wilhelm Fink: Munich). Hingley, Richard. 2005. Globalizing Roman Culture: Unity, Diversity and Empire (Routledge: London). Hofmeister, Philipp. 1948. Die heiligen Öle in der morgen- und abendländischen Kirche (Augustinus-Verlag: Würzburg). Holfelder, Moritz. 2008. ‘Der Zweifler. Interview mit Lars Ramberg, Installationskünstler *1864 Oslo (Norwegen)’, in Moritz Holfelder (ed.), Palast der Republik. Aufstieg und Fall eines symbolischen Gebäudes (Ch. Links Verlag: Berlin): 196–197. Horden, Peregrine, and Nicholas Purcell. 2000. The Corrupting Sea: A Study of Mediterranean History (Blackwell: Oxford). Houtman, Dick, and Birgit Meyer (eds.). 2012. Things. Religion and the Question of Materiality (Fordham University Press: New York). Howes, David. 1996. ‘Cultural Appropriation and Resistance in the American Southwest. Decommodifying “Indianness”’, in David Howes (ed.), CrossCultural Consumption: Global Markets, Local Realities (Routledge: London, New York): 138–160. Humbert, Jean-Baptiste. 1999. ‘The Rivers of Paradise in the Byzantine Church Near Jabaliyah – Gaza’, in Michele Piccirillo and Eugenio Alliata (eds.), The Madaba Map Centenary (Franciscan Printing Press: Jerusalem). Humphries, Mark, and David M. Gwynn. 2010. ‘The Sacred and the Secular: The Presence or Absence of Christian Religious Thought in Secular Writing in the Late Antique West’, in David M. Gwynn, Susanne Bangert and Luke Lavan (eds.), Religious Diversity in Late Antiquity (Brill: Leiden, Boston): 493–509. Huskinson, Janet. 1974. ‘Some Pagan Mythological Figures and their Significance in Early Christian Art’, Papers of the British School at Rome, 42: 68–97. Huss, Werner. 2004. Die Karthager 3rd ed. (C.H. Beck: Munich).

References Iannucci, Anna Maria. 1984. ‘Una ricognizione al Battistero Neoniano’, in Raffaella Farioli Campanati (ed.), 31. Corso di Cultura sull’Arte Ravennate e Bizantina (Del Girasole: Ravenna): 297–339. 1985. ‘Nuove ricerche al Battistero Neoniano’, in Raffaella Farioli Campanati (ed.), 32. Corso di Cultura sull’Arte Ravennate e Bizantina. Seminario Internazionale di Studi su “Cipro e il Mediterraneo orientale” (Del Girasole: Ravenna): 79–107. Iannucci, Anna Maria, Nicola Santopuoli, and Leonardo Seccia. 2001. ‘Il nuovo atlante dei mosaici del Battistero Neoniano: catalogazione fotografica e riflettografica’, in Raffaella Farioli Campanati (ed.), 44. Corso di Cultura sull’Arte Ravennate e Bizantina. Seminario internazionale di studi sul tema: Le grandi isole del Mediterraneo orientale tra tarda antichità e medioevo; Ravenna, 19–21 settembre 1998 (Del Girasole: Ravenna): 333–359. Ihm, Christa. 1960. Die Programme der christlichen Apsismalerei vom vierten Jahrhundert bis zur Mitte des achten Jahrhunderts (Franz Steiner: Wiesbaden). Insoll, Timothy. 2011. ‘Christianity’, in Timothy Insoll (ed.), The Oxford Handbook of the Archaeology of Ritual and Religion (Oxford University Press: Oxford): 948–966. Iturgaiz, Domingo. 1967. ‘Baptisterios paleocristianos de Hispania’, Analecta Sacra Tarraconensia, 40.2: 209–295. 1968. ‘Baptisterios paleocristianos de Hispania’, Analecta Sacra Tarraconensia, 41.2: 209–244. Ivanovici, Vladimir. 2014. ‘“Luce renobatus”. Speculations on the Placement and Importance of Lights in Ravenna’s Neonian Baptistery’, in Daniela Mondini and Vladimir Ivanovici (eds.), Manipolare la luce in epoca premoderna (Mendrisio Academy Press: Mendrisio): 19–29. 2016. Manipulating Theophany. Light and Ritual in North Adriatic Architecture (ca. 400-ca. 800) (De Gruyter: Berlin). Jacobs, Ine. 2013. Aesthetic Maintenance of Civic Space. The ‘Classical’ City from the Fourth to the Seventh c. AD (Peeters: Leuven, Paris, Walpole MA). 2015. ‘Holy Goals and Worldly Means. Urban Representation Elements in Church Complexes’, in Aude Busine (ed.), Religious Practices and Christianization of the Late Antique City (4th–7th Cent.) (Brill: Leiden, Boston): 71–114. 2020. ‘Old Statues, New Meanings. Literary, Epigraphic and Archaeological Evidence for Christian Reidentification of Statuary’, Byzantinische Zeitschrift, 113.3: 789–836. Jacoby, Adolf. 1902. Ein bisher unbeachteter apokrypher Bericht über die Taufe Jesu nebst Beiträgen zur Geschichte der Didaskalie der zwölf Apostel und Erläuterungen zu den Darstellungen der Taufe Jesu (Trübner: Strasbourg). Jäggi, Carola. 2013. Ravenna: Kunst und Kultur einer spätantiken Residenzstadt: die Bauten und Mosaiken des 5. und 6. Jahrhunderts (Schnell & Steiner: Regensburg).

221

222

References 2016. ‘Ravenna in the Sixth Century: The Archaeology of Change’, in Judith Herrin and Jinty Nelson (eds.), Ravenna. Its Role in Earlier Medieval Change and Exchange (University of London Press: London): 87–110. Jenkins, Richard. 1996. Social Identity (Routledge: London). Jensen, Robin Margaret. 1993. ‘What Are Pagan River Gods Doing in Scenes of Jesus’ Baptism?’, Bible Review Washington, 9.1: 35–41. 2000. Understanding Early Christian Art (Routledge: London). 2011a. ‘Baptismal Practices at North African Martyr’s Shrines’, in David Hellholm, Tor Vegge,
yvind Norderval and Christer Hellholm (eds.), Ablution, Initiation, and Baptism: Late Antiquity, Early Judaism, and Early Christianity ¼ Waschungen, Initiation und Taufe: Spätantike, frühes Judentum und frühes Christentum (De Gruyter: Berlin, Boston). 2011b. Living Water: Images, Symbols, and Settings of Early Christian Baptism (Brill: Leiden). 2012. Baptismal Imagery in Early Christianity: Ritual, Visual, and Theological Dimensions (Baker Academic: Grand Rapids, MI). Jeremias, Joachim. 1958. Die Kindertaufe in den ersten vier Jahrhunderten (Vandenhoeck & Ruprecht: Göttingen). Jiménez Sánchez, Juan Antonio. 2010. Los juegos paganos en la Roma cristiana (Fondazione Benetton studi ricerche, Viella: Treviso, Rome). Jiménez Sánchez, Juan Antonio, and Jordina Sales Carbonell. 2004. ‘Termas e Iglesias durante la antigüedad tardía: Reutilización arquitectónica o conflicto religioso? Algunos ejemplos hispanos’, in José María Blásquez Martínez and Antonino González Blanco (eds.), Sacralidad y Archeología (Universidad de Murcia: Murcia): 185–201. Johnson, Mark Joseph. 1988. ‘Toward a History of Theoderic’s Building Program’, Dumbarton Oaks Papers, 42: 73–96. Johnson, Maxwell E. 1999. The Rites of Christian Initiation: Their Evolution and Interpretation (Liturgical Press: Collegeville, MN). Jones, Lindsay. 2000. The Hermeneutics of Sacred Architecture: Experience, Interpretation, Comparison (Harvard University Press: Cambridge, MA). Jones, Siân, and Paul Graves-Brown. 1996. ‘Introduction. Archaeology and Cultural Identity in Europe’, in Paul Graves-Brown, Clive Gamble and Siân Jones (eds.), Cultural Identity and Archaeology: The Construction of European Communities (Routledge: London): 1–24. Joosen, Josephus C., and Jan Hendrik Waszink. 1950. ‘Allegorese’, in Theodor Klauser (ed.), Reallexikon für Antike und Christentum (Hiersemann: Stuttgart): 283–293. Jorge, Ana Maria. 1999. ‘Church and Culture in Lusitania in the V-VIII Centuries: A Late Roman Province at the Crossroads’, in Alberto Ferreiro (ed.), The Visigoths: Studies in Culture and Society (Brill: Leiden): 99–122. Josselson, Ruthellen, and Michele Harway. 2012. Navigating Multiple Identities: Race, Gender, Culture, Nationality, and Roles (Oxford University Press: Oxford, New York).

References

Jullien, François. 2021. There Is No Such Thing as Cultural Identity (Polity: Cambridge). Kahlos, Maijastina. 2007. Debate and Dialogue: Christian and Pagan Cultures c. 360-430 (Ashgate: Aldershot). Kaldellis, Anthony. 2007. Hellenism in Byzantium: The Transformations of Greek Identity and the Reception of the Classical Tradition (Cambridge University Press: Cambridge). 2019. Romanland: Ethnicity and Empire in Byzantium (Harvard University Press: Cambridge, MA). Khatchatrian, Armen. 1962. Les baptistères paléochrétiens: plans, notices et bibliographie (Imprimerie nationale: Paris). 1982. Origine et typologie des baptistères paléochrétiens (Centre de culture chrétienne: Mulhouse). Klementa, Sylvia. 1993. Gelagerte Flussgötter des Späthellenismus und der römischen Kaiserzeit (Böhlau Verlag: Cologne). Koerner, Joseph Leo. 2004. The Reformation of the Image (University of Chicago Press: Chicago). Kostof, Spiro Konstantine. 1965. The Orthodox Baptistery of Ravenna (Yale University Press: New Haven). Kötting, Bernhard. 1980. Peregrinatio Religiosa. Wallfahrten in der Antike und das Pilgerwesen in der alten Kirche (Th. Stenderhoff: Münster). Kötzsche, Lieselotte. 1986a. ‘Der Basilikaleuchter in Leningrad’, in Otto Feld and Urs Peschlow (eds.), Studien zur spätantiken und byzantinischen Kunst. Friedrich Wilhelm Deichmann gewidmet (Habelt: Bonn). 1986b. ‘Hand II (ikonographisch)’, in Theodor Klauser, Ernst Dassmann, Carsten Colpe, Albrecht Dihle, Josef Engemann, and Wolfgang Speyer (eds.), Reallexikon für Antike und Christentum (Hiersemann: Stuttgart): 402–482. Krautheimer, Richard. 1942. ‘Introduction to an “Iconography of Mediaeval Architecture”’, Journal of the Warburg and Courtauld Institutes, 5: 1–33. Kretschmar, Georg. 1970. Die Geschichte des Taufgottesdienstes in der Alten Kirche (Stauda: Kassel). Kubiak, Daniel. 2011. ‘Ostdeutsche Identität im Wandel der Zeiten. 30 Jahre und noch kein Ende’, in Sören Becker and Matthias Naumann (eds.), Regionalentwicklung in Ostdeutschland. Dynamiken, Perspektiven und der Beitrag der Humangeographie (Springer: Berlin, Heidelberg). Kubiak, Daniel. 2018. ‘Socialization, Downgrading and Othering: The Formation of Identity of Young “East Germans”’, in Stephan Ehrig, Marcel Thomas and David Zell (eds.), The GDR Today. New Interdisciplinary Approaches to East German History, Memory and Culture (Peter Lang: Bern): 195–214. Kurth, Julius. 1901. Die Mosaiken der christlichen Ära. I. Die Wandmosaiken von Ravenna (Deutsche Bibelgesellschaft: Leipzig, Berlin).

223

224

References

Ladner, Gerhart Burian. 1992. Handbuch der frühchristlichen Symbolik: Gott, Kosmos, Mensch (Belser: Stuttgart). Lafli, Ergün, and Hadrien Bru. 2013. ‘Inscriptions gréco-romaines d’Anatolie II.’, Chronique d’Orient. Dialogues d’histoire ancienne, 39.2: 287–303. Lancel, Serge (ed.). 1974. Gesta conlationis Carthaginiensis, anno 411: Accedit Sancti Augustini breviculus conlationis cum Donatistis (Brepols: Turnhout). Lancha, Janine. 2003. ‘Mythologie classique et mosaïques tardives d’Hispanie (IVe s.)’, Antiquité Tardive, 11: 197–214. 2004. ‘De nouvelles données sur les mosaïstes itinérants en Lusitanie’, in JeanGérard Gorges, Enrique Cerrillo and Trinidad Nogales Basarrate (eds.), V Mesa Redonda Internacional sobre Lusitania Romana: Las comunicaciones (Ministerio de Cultura: Madrid): 409–426. Lanciani, Filippo. 1871. Cenni intorno ai monumenti e alle cose piu’ notabili di Ravenna (Tip. Della Ditta G. Angeletti: Ravenna). Lander, Shira. 2016. Ritual Sites and Religious Rivalries in Late Roman North Africa (Cambridge University Press: Cambridge). Laporte, Jean-Pierre, and Henri Lavagne. 2006. ‘Sousse: la domus de Sorothus et ses mosaïques’, Comptes rendus des séances de l’Academie des inscriptions et Belles-Lettres, 150.2: 1327–1392. Lassus, Jean. 1965. ‘Les edifices du culte autour de la basilique’, in Congresso internazionale di archeologia cristiana (ed.), Atti del VI Congresso internazionale di archeologia cristiana (Pontificio Istituto di Archeologia Cristiana: Vatican City): 581–610. 1970. ‘Les baptistères africains’, Corsi di cultura sull’arte ravennate e bizantina, 17: 235–252. Laurence, Ray, and Joanne Berry. 1998. Cultural Identity in the Roman Empire (Routledge: London). Leader-Newby, Ruth E. 2004. Silver and Society in Late Antiquity: Functions and Meanings of Silver Plate in the Fourth to Seventh Centuries (Ashgate: Aldershot). Leatherbury, Sean. 2020. Inscribing Faith in Late Antiquity: Between Reading and Seeing (Routledge: London). Leclercq, Henri. 1907–1953. ‘Fleuves’, in Fernand Cabrol and Henri Leclercq (eds.), Dictionnaire d’archéologie chrétienne et de liturgie (Letouzey et Ané: Paris): 1760–1761. Leglay, Marcel. 1957. ‘Note sur quelques baptistères d’Algérie’, in Congresso internazionale di archeologia cristiana (ed.), Actes du Ve congrès international d’archéologie chrétienne (Pontificio Istituto di Archeologia Cristiana: Vatican City): 401–406. Lehmann, Tomas. 2010. ‘Die frühchristlichen Mosaiken im Dombereich von Aquileia’, in Giuseppe Cuscito and Tomas Lehmann (eds.), La Basilica di Aquileia: storia, archeologia ed arte ¼ Der Dom von Aquileia: Geschichte, Archäologie und Kunst (Editreg: Trieste).

References Lehnerdt, Max. 1884. ‘Flussgötter’, in Wilhelm Heinrich Roscher (ed.), Ausführliches Lexikon der griechischen und römischen Mythologie (Teubner: Leipzig): 1487–1496. Lenk, Stefanie. 2017. ‘Iberian Christians and the Classical Past – The Baptistery of Milreu/Estói (Algarve) at the End of Late Antiquity’, in Horst Bredekamp and Stefan Trinks (eds.), Transformatio et Continuatio. Forms of Change and Constancy of Antiquity in the Iberian Peninsula 500–1500 (De Gruyter: New York): 63–91. 2021. ‘Response to C. Uehlinger “Material Religion in Comparative Perspective: How Different is BCE from CE?”’, in Jaś Elsner and Rachel Wood (eds.), Imagining the Divine (British Museum Press: London): 170–175. Leone, Anna. 2013. The End of the Pagan City. Religion, Economy, and Urbanism in Late Antique North Africa (Oxford University Press: Oxford). 2016. ‘Tracing the Donatist Presence in North Africa: An Archaeological Perspective’, in Richard Miles (ed.), The Donatist Schism. Controversy and Contexts (Liverpool University Press: Liverpool): 317–344. Lepelley, Claude. 1979–1981. Les cités de l’Afrique Romaine au Bas-Empire (Études Augustiniennes: Paris). 1994. ‘Le musée des statues divines. La volonté de sauvegarder le patrimoine artistique païen à l’époque théodosienne’, Cahiers Archéologiques – Fin de l’Antiquité et Moyen Âge, 42: 5–15. 1998. ‘L’aristocratie lettrée païenne: une menace aux yeux d’Augustin (à propos du sermon Dolbeau 26 – Mayence 62)’, in Goulven Madec (ed.), Augustin prédicateur (Institut d’Études Augustiniennes: Paris): 397–413. 2002. ‘Le lieu des valeurs communes. La cité terrain neutre entre païens et chrétiens dans l’Afrique romaine tardive’, in Hervé Inglebert (ed.), Idéologies et valeurs civiques dans le Monde Romain (Picard: Paris): 271–285. 2010. ‘Sacred and Secular. The Use of Secularised Latin Pagan Culture by Christians’, in David M. Gwynn and Susanne Bangert (eds.), Religious Diversity in Late Antiquity (Brill: Leiden, Boston, MA): 475–492. Leschi, Louis. 1950. Djemila. Antique Cuicul (Imprimerie Officielle: Algiers). 1953. Djemila. Antique Cuicul. Algiers, Imprimerie officielle. Lettrari, Adriana, Christian Nestler, and Nadja Troi-Boeck. 2016. Die Generation der Wendekinder. Elaboration eines Forschungsfeldes (Springer: Wiesbaden). Liebeschütz, Wolfgang. 1995. ‘Pagan Mythology in the Christian Empire’, International Journal of the Classical Tradition, 2.2: 193–208. Lopes, Virgílio. 2003. Mértola na Antiguidade Tardia (Campo Arqueológico de Mértola: Mértola). 2005. ‘Les mosaïques de Mértola et le baptistère. Bilan des fouilles recentes’, in Hélène Morlier (ed.), La mosaïque gréco-romaine. Actes du IXème Colloque international pour l’étude de la mosaïque antique et médiévale organisé à Rome, 5-10 novembre 2001 (École française de Rome: Rome): 489–501.

225

226

References 2011. ‘Os mosaicos do baptistério de Mértola’, in Comissao Executiva do X Col AIEMA (ed.), O mosaico romano nos centros e nas periferias (Museu Monográfico de Conímbriga: Coimbra): 915–925. 2014. ‘Mértola e o seu território na antiguidade tardia (séculos IV-VIII) (diss.)’, Universidad de Huelva, Huelva (http://hdl.handle.net/10272/8053) 2015. ‘Mértola in Late Antiquity’, in Susana Gómez-Martínez, Santiago Macias and Virgílio Lopes (eds.), O sudoeste peninsular entre Roma e o Islão ¼ The Southwestern Iberian Peninsula between Rome and Islam (Campo arqueológico de Mértola: Mértola): 381–387. 2017a. ‘Mértola na antiguidade tardia. A topografia histórica’, in José Morais Arnaud and Andrea Martins (eds.), Arqueologia em Portugal (Associação dos Arqueólogos Portugueses: Lisbon): 1379–1390. 2017b. ‘Mértola na antiguidade tardia. O complexo religioso e os batistérios’, in Julia Beltrán de Heredia Bercero and Cristina Godoy Fernández (eds.), La dualitat de baptisteris en les ciutats episcopals del cristanisme tardoantic. Actes del i simposi d’arqueologia cristiana (Ateneu Universitari Sant Pacià: Barcelona): 127–145. Lopes, Virgílio, and Susana Gómez-Martínez. 2006. ‘Trabalhos arqueológicos de Estácio da Veiga em Mértola’, Xelb. Actas do IV Encontro do Arqueologia do Algarve, 7: 269–282. Lopez Jr., Donald Sewell. 1998. ‘Belief’, in Mark C. Taylor (ed.), Critical Terms for Religious Studies (University of Chicago Press: Chicago, London): 21–35 López Monteagudo, Guadalupe. 1994. ‘Mosaicos hispanos de circo y anfiteatro’, in César María Batalla (ed.), VI Coloquio Internacional sobre Mosaico Antiguo: Palencia-Mérida, octubre 1990 (Asociación Española del Mosaico: Valladolid): 343–358. 2011. ‘La iconografía del dios Oceanos en los mosaicos hispano-romanos’, in Museu Monográfico de Conímbriga (ed.), O mosaico romano nos centros e nas periferias: originalidades, influências e identidades: actas do X Coloquio Internacional da Associação Internacional para o Estudo do Mosaico Antigo (AIEMA) (Instituto dos Museus e da Conservação: Conímbriga): 287–302. Lück, Stefanie. 2018. Taufanlagen Hispaniens: 4. bis 10. Jahrhundert (Cuvillier: Göttingen). Lugaresi, Leonardo. 2007. ‘Regio Aliena. L’attegiamento della chiesa verso i luoghi di spettacolo nella città tardoantica’, Antiquité Tardive, 15: 21–34. Lux, Ute. 1968. ‘Die Apostel-Kirche in Madeba’, Zeitschrift des Deutschen Palästina-Vereins, 84: 104–129. Lyster Franco, Mário. 1942. As ruínas romanas do Milreu: e os últimos trabalhos nelas realizados por intermédio da Dirrecção Geral dos Edifícios e Monumentos Nacionais (Boletim da Junta de Província do Algarve: Faro) (no pagination). Mac Mullen, Ramsay. 1986. ‘What Difference Did Christianity Make?’, Historia: Zeitschrift für Alte Geschichte, 35.3: 322–343.

References 1989. ‘The Preacher’s Audience. AD 350-400’, The Journal of Theological Studies, 40: 503–511. Macías, Santiago. 1996. Mértola Islâmica. Estudo histórico-arqueológico do bairro da Alcáçova: séculos XII-XIII (Campo arquéologico de Mértola: Mértola). Maciel, Justino. 1996. Antiguidade Tardia e Paleocristianismo em Portugal (Justino Maciel: Lisbon). Madoz y Moleres, José. 1945. ‘Una nueva recensión del ‘De correctione rusticorum’, de Martín de Braga (Ms. Sant Cugat, n. 22)’, Estudios eclesiásticos, 19.74: 335–353. Maguire, Henry. 1987. Earth and Ocean. The Terrestrial World in Early Byzantine Art (Pennsylvania State University Press: University Park, London). 1999. ‘The Good Life’, in Glen Warren Bowersock, Peter Brown and Oleg Grabar (eds.), Late Antiquity. A Guide to the Postclassical World (Harvard University Press: Cambridge): 238–259. Mañanes, Tomas. 1980. ‘El mosaico romano de las termas de la Catedral de León’, in Tomas Mañanes and Alberto Balil (eds.), Estudios sobre Mosaicos Romanos (Universidad de Valladolid: Valladolid): 25–28. Mango, Cyril. 1963. ‘Antique Statuary and the Byzantine Beholder’, Dumbarton Oaks Papers, 17: 53–75. Marcenaro, Mario (ed.). 2007. Albenga Città Episcopale. Tempi e dinamiche della cristianizzazione tra Liguria di Ponente e Provenza (Istituto Internazionale di Studi Liguri: Bodhigera). Marcone, Arnaldo. 2011. ‘Persistenze classiche e innovazioni cristiane: iconografia ed epigrafia nelle Venezie’, in Peter Brown and Rita Lizzi Testa (eds.), Pagans and Christians in the Roman Empire: The Breaking of a Dialogue (IVth-VIth Century A.D.) (LIT Verlag: Zurich, Berlin): 291–308. 2014. ‘Alla ricerca di un’identità. Tradizioni classiche nella prima iconografia cristiana’, in Stine Birk, Troels Myrup Kristensen and Birte Poulsen (eds.), Using Images in Late Antiquity (Oxbow Books: Oxford): 253–267. Maréchal, Sadi. 2020. Public Baths and Bathing Habits in Late Antiquity: A Study of the Evidence from Italy, North Africa and Palestine A.D. 285-700 (Brill: Leiden, Boston). 2016. ‘Public and Private Bathing in Late Antique North Africa. Changing Habits in a Changing Society?’ in Matthew J. Mandich, Thomas J. Derrick, Sergio Gonzalez Sanchez, Giaccomo Savani and Eleonora Zampieri (eds.), TRAC 2015: Proceedings of the Twenty-Fifth Annual Theoretical Roman Archaeology Conference (Oxbow Books: Oxford): 125–140. Markus, Robert Austin. 1990. The End of Ancient Christianity (Cambridge University Press: Cambridge). Martínez Pizarro, Joaquín. 1995. Writing Ravenna. The Liber Pontificalis of Andreas Agnellus (University of Michigan Press: Ann Arbor). Masaro, Gabriele. 2014. ‘Vario formata decore. Reminiscenze classiche e autori cristiani nelle dediche metriche delle basiliche tardo-antiche’, in Antonio

227

228

References

Pistellato (ed.), Memoria poetica e poesia della memoria. La versificazione epigrafica dall’antichità all’umanesimo (Ca’Foscari: Venice): 177–205. Masuzawa, Tomoko. 2005. The Invention of World Religions, or, How European Universalism was Preserved in the Language of Pluralism (University of Chicago Press: Chicago, London). Mathisen, Ralph Whitney. 2014. ‘Barbarian “Arian” Clergy, Church Organization, and Church Practices’, in Guido Michael Berndt and Roland Steinacher (eds.), Arianism: Roman Heresy and Barbarian Creed (Ashgate: Farnham): 145–192. Matthiae, Guglielmo. 1962. ‘La cultura figurativa di Salonicco nei secoli V e VI’, Rivista di Archeologia Cristiana, 83: 163–213. Mattingly, David J. 2010. Imperialism, Power, and Identity – Experiencing the Roman Empire (Princeton University Press: Princeton). Mauskopf Deliyannis, Deborah. 2010. Ravenna in Late Antiquity (Cambridge University Press: Cambridge). 2016. ‘Episcopal Commemoration in Late Fifth-Century Ravenna’, in Judith Herrin and Jinty Nelson (eds.), Ravenna: Its Role in Earlier Medieval Change and Exchange (University of London Press: London): 39–51. Mazzotti, Mario. 1957. ‘La “Anastasis Ghotorum” di Ravenna ed il suo battistero’, Felix Ravenna, 75.1: 25–62. 1961. ‘Il battistero della cattedrale di Ravenna’, Corsi di Cultura sull’Arte ravennate e bizantina, 8: 255–278. 1970. ‘Scavi recenti al battistero degli ariani in Ravenna’, Felix Ravenna, 101.1: 115–123. McGuire, Meredith. 2008. Lived Religion: Faith and Practice in Everyday Life (Oxford University Press: New York, Oxford). McLynn, Neil B., Arietta Papaconstantinou, and Daniel Louis Schwartz. 2015. Conversion in Late Antiquity: Christianity, Islam, and Beyond (Ashgate: Farnham). Mead, George Herbert, and Charles William Morris (eds.). 1934 (2015). Mind, Self and Society (University of Chicago Press: Chicago, London). Meer, Frederik van der, Brian Battershaw, and G. R. Lamb. 1961. Augustine the Bishop: The Life and Work of a Father of the Church (Sheed and Ward: London). Mendel, Gustave. 1914. Catalogue des sculptures grecques, romaines et byzantines (Musée Imperial: Constantinople). Mercado Hernández, César, and Esther Sánchez Medina. 2001. ‘Visión isidoriana de los espectáculos públicos’, in Luis Agustín García Moreno and Sebastián Rascón Marqués (eds.), Hispánia en la Antigüedad Tardía: ocio y espectáculos, Actas del 2 Encuentro Hispania en la Antigüedad Tardía, Alcalá de Henares, 15 a 17 de octubre de 1997 (Universidad del Alcalá: Alcalá Henares): 221–229. Merrills, Andy, and Richard Miles. 2010. The Vandals (Wiley-Blackwell: Oxford).

References Merritt Bassett, Paul. 1976. ‘The Use of History in the Chronicon of Isidore of Seville’, History and Theory, 15.3: 278–292. Mertens Fleury, Katharina. 2015. Zeigen und Bezeichnen. Zugänge zu allegorischem Erzählen im Mittelalter (Königshausen u. Neumann: Würzburg). Meyer, Birgit. 2015. ‘Picturing the Invisible. Visual Culture and the Study of Religion’, Method and Theory in the Study of Religion, 27: 333–360. Mikat, Paul. 1984. Doppelbesetzung oder Ehrentitulatur: zur Stellung des westgotisch-arianischen Episkopates nach der Konversion von 587/89 (Westdeutscher Verlag: Opladen). Miles, Richard (ed.). 2016. The Donatist Schism. Controversy and Context (Liverpool University Press: Liverpool). 2005. ‘The Anthologia Latina and the Creation of a Secular Space in Vandal Carthage’, Antiquité Tardive, 13: 305–320. Mîrşanu, Dragoş 2009. ‘The Aesthetic ‘Shadow’of Gothic Arianism: Archaeology, Architecture and Art in the Age of Heresies’, in Mihail Neamţu and Bogdan Tǎtaru-Cazaban (eds.), Memory, Humanity, and Meaning (Zeta Books: Bucharest): 411–431. Monceaux, Paul. 1908. ‘Enquête sur l’épigraphie chrétienne d’Afrique’, Mémoires présentés par divers savants à l’Académie des inscriptions et belles-lettres de l’Institut de France, 12.1: 161–339. 1922. ‘Découverte d’un groupe d’édifices chrétiens à Djemila’, Comptes rendus des séances de l’Académie des inscriptions et Belles-Lettres, 66.5: 380–707. 1923. ‘Cuicul chrétien (Numidie)’, Miscellanea Giovanni Battista de Rossi. Atti della pontificia accademia romana di archeologia (serie III), 1.1: 89–112. Moracchini-Mazel, Geneviève. 1962a. ‘Une basilique et un baptistère paléochrétiens découverts à Mariana (Corse)’, Comptes-rendus des séances de l’Académie des Inscriptions et Belles-Lettres, 106.2: 99–102. 1962b. ‘Le pavement en mosaïque de la basilique paléo-chrétienne et du baptistère de Mariana (Corse)’, Cahiers Archéologiques, 13: 137–160. Moracchini-Mazel, Geneviève, André Jodin, Gabrielle Démians d’Archimbaud, and Claude Brenot. 1971. Les Fouilles de Mariana (Corse) (Fédération d’associations et groupements pour les études corses: no location). Morfino, Debora. 2011. ‘Le strutture battesimali nell’Africa Romana IV-VI secolo d. C’, in Rosa Maria Bonacasa Carra (ed.), Il primo christianesimo nell’Africa romana e in Sicilia (Carlo Saladino: Palermo): 9–93. Morgan, David. 2010a. ‘Introduction. The Matter of Belief’, in David Morgan (ed.), Religion and Material Culture: The Matter of Belief (Routledge: London): 1–17. 2010b. Religion and Material Culture: The Matter of Belief (Routledge: London). 2016. ‘Material Analysis and the Study of Religion’, in Tim Hutchings and Joanne McKenzie (eds.), Materiality and the Study of Religion: The Stuff of the Sacred (Taylor & Francis: London): 14–32.

229

230

References

Morgan, Teresa. 2015. Roman Faith and Christian Faith: Pistis and Fides in the Early Roman Empire and Early Churches (Oxford University Press: Oxford). Morvillez, M. Éric. 2012. ‘Les mosaïques des bains d’Oued Athmenia (Algérie): les calques conservés à la Médiathèque de l’Architecture et du Patrimoine’. Bulletin de la Société nationale des Antiquaires de France, année 2006: 304–322. Mourão, Cátia. 2011. ‘Mosaicos romanos com motivos aquáticos em Portugal’, in Museu Monográfico de Conímbriga (ed.), O mosaico romano nos centros e nas periferias: originalidades, influências e identidades: actas do X Coloquio Internacional da Associação Internacional para o Estudo do Mosaico Antigo (AIEMA) (Instituto dos Museus e da Conservação: Conímbriga): 343–351. 2010. ‘AVTEM NON SVNT RERVM NATVRA. Figurações heteromórficas em mosaicos hispano-romanos (diss.)’, Universidade Nova de Lisboa: Lisbon (http://hdl.handle.net/10362/6268). 2008. ‘Motivos aquáticos em mosaicos antigos de Portugal – Decorativismo e Simbolismo’, Revista de História da Arte, 6: 115–131. Mugnai, Niccolo, Julia Nikolaus, and Nick Ray. 2016. De Africa Romaque: Merging Cultures Across North Africa: Proceedings of the International Conference Held at the University of Leicester (26–27 October 2013) (The Society for Libyan Studies: London). Muhlberger, Steve. 1992. ‘Looking Back from Mid Century: The Gallic Chronicler of 452 and the Crisis of Honorius’ Reign’, in John Drinkwater and Hugh Elton (eds.), Fifth-Century Gaul: A Crisis of Identity? (Cambridge University Press: Cambridge): 28–37. Müller-Wiener, Wolfgang. 1989. ‘Bischofsresidenzen des 4.-7. Jhs. im östlichen Mittelmeerraum’, in Françoise Baritel, Noël Duval and Philippe Pergola (eds.), Actes du XIe Congrès international d’archéologie chrétienne : Lyon, Vienne, Grenoble, Genève et Aoste (21-28 Septembre 1986) (Ecole française de Rome: Rome): 651–709. Munier, Charles. 1974. Concilia Africae: a. 345-a. 525 (Brepols: Turnhout). Mürbe, Stefanie. 2007. ‘Der Palast der Republik – Ein Ort geteilter Erinnerungen?’ in Alexander Schug (ed.), Palast der Republik: politischer Diskurs und private Erinnerung (BWV: Berlin): 108–121. Muth, Susanne. 2001. ‘Eine Kultur zwischen Veränderung und Stagnation. Zum Umgang mit den Mythenbildern im spätantiken Haus’, in Franz Alto Bauer and Norbert Zimmermann (eds.), Epochenwandel? Kunst und Kultur zwischen Antike und Mittelalter (Philipp von Zabern: Mainz): 95–116. Nagel, Eduard. 1980. Kindertaufe und Taufaufschub. Die Praxis vom 3. - 5. Jahrhundert in Nordafrika und ihre theologische Einordnung bei Tertullian, Cyprian und Augustinus [Peter D. Lang: Frankfurt (Main)]. Nauerth, Claudia. 1974. Agnellus von Ravenna. Untersuchungen zur archäologischen Methode des ravennatischen Chronisten (Weidmann: Hildesheim).

References Neira Jiménez, Luz. 1997. ‘Algunas consideraciones sobre mosaicos romanos con nereidas y tritones en ambientes termales de Hispania’, in María Jesús Peréx Agorretea (ed.), Termalismo Antiguo; actas (Universidad Nacional de Educación a Distancia: Madrid): 481–496. 2011. ‘The Sea Thiasos of Nereids and Tritons in the Roman Mosaics of Turkey’, in Mustafa Şahin (ed.), 11th International Colloquium on Ancient Mosaics, October 16th–20th, 2009, Bursa, Turkey: Mosaics of Turkey and Parallel Developments in the Rest of the Acient and Medieval World (Uludağ Üniversitesi, Mozaik Araştırmaları Uygulama ve Araştırma Merkezi: Istanbul): 631–654. Neumann, Elke (ed.). 2019. Palast der Republik: Utopie, Inspiration, Politikum (Exhibition catalogue, Kunsthalle Rostock, 1 June–13 October 2019) [Mitteldeutscher Verlag: Halle (Saale)]. Neuner, Stefan, Markus Klammer, and Beate Fricke. 2011. ‘Pikturale Horizonte der Gemeinschaftsproduktion’, in Beate Fricke, Markus Klammer and Stefan Neuner (eds.), Bilder und Gemeinschaften. Studien zur Konvergenz von Politik und Ästhetik in Kunst, Literatur und Theorie (Fink: Paderborn): 10–39. Nola, Paulinus of, and Matthias Skeb. 1998. Epistulae. Fontes Christiani 25.2 (Herder: Freiburg im Br.). Nora, Pierre. 1989. ‘Between Memory and History: Les lieux de mémoire’, Representations, 26: 7–24. Nordström, Carl Otto. 1953. Ravennastudien; ideengeschichtliche und ikonographische Untersuchungen über die Mosaiken von Ravenna (Almquist and Wiksell: Stockholm). North, John A., and Simon Rowland Francis Price (eds.). 2011. The Religious History of the Roman Empire: Pagans, Jews, and Christians (Oxford University Press: Oxford). Noth, Martin. 1968. ‘Die Mosaikinschriften der Apostel-Kirche in Madeba’, Zeitschrift des Deutschen Palästina-Vereins, 84: 130–142. O’Donnell, James Joseph. 1979. ‘The Demise of Paganism’, Traditio, 35: 45–88. Oepen, Alexis. 2012. Villa und christlicher Kult auf der Iberischen Halbinsel in Spätantike und Westgotenzeit (Reichert Verlag: Wiesbaden). Öğüş-Uzun, Esen. 2010. ‘Fruit Trees on Late Antique Mosaics: From the Worcester Hunt to Jordanian Churches’, OLBA, 18: 395–420. Ohly, Friedrich. 1977. ‘Vom geistigen Sinn des Wortes im Mittelalter’, in Friedrich Ohly (ed.), Schriften zur mittelalterlichen Bedeutungsforschung (Wissenschaftliche Buchgesellschaft: Darmstadt). 1979. ‘Typologische Figuren aus Natur und Mythus’, in Walter Haug (ed.), Formen und Funktionen der Allegorie (Metzlersche Verlagsbuchhandlung: Stuttgart). Opelt, Ilona. 1962. ‘Erde’, in Theodor Klauser, Franz Joseph Dölger, and Ernst Dassmann (eds.), Reallexikon für Antike und Christentum (Hiersemann: Stuttgart): 1113–1179.

231

232

References Orfila, Margarita, and Francisco Tuset. 2003. ‘Descripción, paralelos y análisis de los mosaicos de la iglesia de Son Fadrinet (Campos, Mallorca)’, Mayurqa, 29: 189–207. Orsi, Robert Anthony. 2005. Between Heaven and Earth: The Religious Worlds People make and the Scholars who study them (Princeton University Press: Princeton, NJ, Oxford). Pachtère, Félix-Georges de. 1911. Inventaire des mosaïques de la Gaule et de l’Afrique (Ernest Leroux: Paris). Palardy, William B. (ed.). 2005. St. Peter Chrysologus: Selected Sermons (Catholic University of America Press: Washington, DC). Palazzo, Eric. 1992. ‘Iconographie et liturgie: la mosaïque du baptistère de Kélibia (Tunisie)’, Archiv für Liturgiewissenschaft, 34: 102–120. Palol, Pedro de. 1967. ‘En torno a la iconografía de los mosaicos cristianos de las islas Baleares’, in Pedro de Palol (ed.), Actas de la primera Reunión Nacional de Arqueología Paleocristiana: Vitoria, 29-31 octubre 1966 (Sancho el Sabio: Valladolid): 131–150. 1989. ‘La arqueología cristiana en la Hispania romana y visigoda. Descubrimientos recientes y nuevos puntos de vista’, in Françoise Baritel, Noël Duval and Philippe Pergola (eds.), Actes du XIe Congrès international d’archéologie chrétienne (Ecole française de Rome: Rome): 1975–2022. Parsons, Wilfrid (ed.). 1951. Augustine. Letters. Volume 1, (1–82) (Catholic University of America Press: Washington, DC). Patout Burns, James. 2011. ‘Augustine’s Theology of Baptism’, in David Hellholm, Tor Vegge,
yvind Norderval, and Christer Hellholm (eds.), Ablution, Initiation, and Baptism: Late Antiquity, Early Judaism, and Early Christianity (De Gruyter: Berlin, New York): 1290–1303. Paxson, James J. 1994. The Poetics of Personification (Cambridge University Press: Cambridge). Pearce, Susan. 2008. ‘The Hinton St Mary Mosaic Pavement: Christ or Emperor?’, Britannia, 39: 193–218. Pelikan, Jaroslav. 1993. Christianity and Classical Culture: The Metamorphosis of Natural Theology in the Christian Encounter with Hellenism (Yale University Press: New Haven, London). Penni Iacco, Emanuela. 2011. L’arianesimo nei mosaici di Ravenna (Longo Editore: Ravenna). Perler, Othmar. 1964. ‘Die Taufsymbolik der vier Jahreszeiten im Baptisterium bei Kélibia’, Jahrbuch für Antike und Christentum, 1: 282–290. Pflaum, Hans-Georg, and Xavier Dupuis. 2003. Inscriptions Latines de l’Algérie (E. de Boccard: Paris). Pharr, Clyde. 2001. The Theodosian Code and Novels and the Sirmondian Constitutions (The Lawbook Exchange: Clark, NJ). Piccirillo, Michele. 1989. Chiese e mosaici di Madaba (Franciscan Printing Press: Jerusalem).

References

1991. I mosaici di Giordania [catalogue of an exhibition], Castello di Malpaga, Cavernago-Bergamo, marzo-giugno 1991 (Studium Biblicum Franciscanum: Jerusalem). 1994. ‘The Byzantine Mosaics of Jordan as an Historical Source – II (1980–1984)’, in Jean-Pierre Darmon and Alain Rebourg (eds.), La Mosaïque Gréco-Romaine IV: IVe Colloque International pour l’Étude de la Mosaïque antique (Supplément au Bulletin de l’A.I.E.M.A.: Paris): 219–225. Piccirillo, Michele, and Janine Balty. 1986. I mosaici di Giordania (Quasar: Rome). Piper, Ferdinand. 1847–1851. Mythologie der christlichen Kunst von der ältesten Zeit bis in’s sechzehnte Jahrhundert (Verlag des Landes-IndustrieComptoirs: Weimar). Pitts, Martin. 2007. ‘The Emperor’s New Clothes? The Utility of Identity in Roman Archaeology’, American Journal of Archaeology, 111.4: 693–713. Pohl, Walter. 1997. ‘The Barbarian Successor States’, in Leslie Webster and Michelle Brown (eds.), The Transformation of the Roman World, AD 400900 (British Museum Press: London): 33–47. 2013. ‘Christian and Barbarian Identities in the Early Medieval West: Introduction’, in Walter Pohl and Gerda Heydemann (eds.), Post-Roman Transitions: Christian and Barbarian Identities in the Early Medieval West (Brepols: Turnhout). 2018. ‘Introduction: Early Medieval Romanness – a Multiple Identity’, in Walter Pohl, Clemens Gantner, Cinzia Grifoni and Marianne Pollheimer-Mohaupt (eds.), Transformations of Romanness. Early Medieval Regions and Identities (De Gruyter: Berlin, Boston): 3–39. Pohl, Walter, Clemens Gantner, Cinzia Grifoni, and Marianne PollheimerMohaupt (eds.). 2018. Transformations of Romanness. Early Medieval Regions and Identities (De Gruyter: Berlin). Pohl, Walter, and Gerda Heydemann (eds.). 2013. Post-Roman Transitions: Christian and Barbarian Identities in the Early Medieval West (Brepols: Turnhout). Pohl, Walter, and Helmut Reimitz (eds.). 1998. Strategies of Distinction: The Construction of the Ethnic Communities, 300-800 (Brill: Leiden). Poulsen, Birte. 2014. ‘City Personifications in Late Antiquity’, in Stine Birk, Troels Myrup Kristensen and Birte Poulsen (eds.), Using Images in Late Antiquity (Oxbow Books: Oxford): 209–226. Puech, Henri-Charles. 1949. ‘Le cerf et le serpent, note sur le symbolisme de la mosaïque découverte au baptistère de l’Henchir Messaouda’, Cahiers archéologiques de l’antiquité et moyen age, 4: 17–60. Quaresma, José Carlos. 2010. ‘Les importations de céramique culinaire africaine à Chaos Salgados, Santiago de Cacém (Mirobriga?), en la villa de Toralla (Vigo, Galicia, Espana)’, in Simonetta Menchelli, Sara Santoro, Marinella Pasquinucci and Gabriella Guiducci (eds.), Late Roman Coarse Wares, Cooking Wares and Amphorae in the Mediterranean: Archaeology and

233

234

References

Archaeometry: Comparison Between Western and Eastern Mediterranean (Archaeopress: Oxford): 157–166. Raeck, Wulff. 1992. Modernisierte Mythen. Zum Umgang der Spätantike mit klassischen Bildthemen (Steiner: Stuttgart). Raja, Rubina. 2012. Urban Development and Regional Identity in the Eastern Roman Provinces, 50 BC-AD 250: Aphrodisias, Ephesos, Athens, Gerasa (Museum Tusculanum Press: Copenhagen). Raja, Rubina, and Jörg Rüpke. 2015a. ‘Archaeology of Religion, Material Religion, and the Ancient World’, in Rubina Raja and Jörg Rüpke (eds.), A Companion to the Archaeology of Religion in the Ancient World (Wiley Blackwell: Chichester): 1–25. Raja, Rubina, and Lara Weiss. 2015. ‘The Role of Objects: Meanings, Situations and Interaction’, Religion in the Roman Empire, 1.2: 137–147. Ravoisié, Amable. 1846. Exploration scientifique de l’Algérie pendant les années 1840, 1841, 1842. Vol. 1 (Librairie de Firmin Didot Frères: Paris). Real, Manuel Luís. 2012. ‘Materiais de construção utilizados na arquitectura cristã da alta Idade Média, em Portugal’, in Arnaldo Sousa Melo and Maria do Carmo Franco Ribeiro (eds.), História da Construção. Os Materiais (CITCEM: Braga): 89–126. Real, Ulrich. 2003. ‘Die Bischofsresidenz in der spätantiken Stadt’, in Gunnar Brands and Hans-Georg Severin (eds.), Die spätantike Stadt und ihre Christianisierung (Reichert: Wiesbaden): 219–237. Rebillard, Éric. 2012. Christians and their Many Identities in Late Antiquity, North Africa, 200-450 CE (Cornell University Press: Ithaca, NY). 2015. ‘Material Culture and Religious Identity in Late Antiquity’, in Rubina Raja and Jörg Rüpke (eds.), A Companion to the Archaeology of Religion in the Ancient World (Wiley Blackwell: Chichester): 427–436. 2018. ‘Sermons, Audience, Preacher’, in Anthony Dupont, Shari Boodts, Gert Partoens and Johan Leemans (eds.), Preaching in the Patristic Era. Sermons, Preachers and Audiences in the Latin West (Brill: Leiden): 87–102. Rebillard, Éric, and Jörg Rüpke. 2015. Group Identity and Religious Individuality in Late Antiquity (The Catholic University of America Press: Washington, DC). Redjel, Tahar, and S. Tewfik. 2010. Guide Djémila (Araja éditions: Constantine). Reinbold, Fabian and Mirjam Novak. 2007. ‘Leidenschaftliche Kämpfer. Die Abrissdebatte und ihre Akteure’, in Alexander Schug (ed.), Palast der Republik: politischer Diskurs und private Erinnerung (BWV: Berlin): 67–81. Remotti, Francesco. 2007. Contro l’identità (Laterza: Rome, Bari). Ricci, Corrado. 1930–1937. Tavole storiche dei mosaici di Ravenna: Battistero della Cattedrale (Instituto Poligrafico dello Stato: Rome). Riera Rullan, Mateu, Magdalena Salas Bruguera, Margalida Munar Grimalt, and Miguel Angel Cau Ontiveros. 2006. ‘El proyecto de revisión y adecuación del

References

yacimiento de Son Peretó (Manacor, Mallorca)’, Association pour l’Antiquité Tardive – Bulletin, 15: 69–74. Riesco Terrero, Luis. 1975. Epistolario de San Braulio (Editorial Católica Española: Seville). Riese, Alexander. 1894. Anthologia Latina : sive Poesis Latinae supplementum. Pars prior, Carmina in codicibus scripta (B.G. Teubner: Leipzig). Riggs, David. 2001. ‘The Continuity of Paganism between the Cities and Countryside of Late Roman Africa’, in Thomas Samuel Burns and John William Eadie (eds.), Urban Centers and Rural Contexts in Late Antiquity (Michigan State University Press: East Lansing, MI): 285–300. Ripoll, Gisela. 1999. ‘Symbolic Life and Signs of Identity in Visigothic Times’, in Peter Heather (ed.), The Visigoths. From the Migration Period to the Seventh Century. An Ethnographic Perspective (Boydell: Woodbridge): 403–431. Ripoll, Gisela, and Isabel Velásquez. 1999. ‘Origen y desarrollo de las parrochiae en la Hispania de la Antigüedad Tardía’, in Philippe Pergola (ed.), Atti delle giornata tematica dei Seminari di Archeologia cristiana (École française de Rome – 19 marzo 1998) (Pontificio Istituto di Archeologia Cristiana: Rome): 101–165. Ristow, Günter. 1957. ‘Zur Personifikation des Jordan in Taufdarstellungen der frühen christlichen Kunst’, in Johannes Irmscher (ed.), Aus der byzantinischen Arbeit der Deutschen Demokratischen Republik (Akademie Verlag: Berlin): 120–126. 1967. Die Taufe Christi (Aurel Bongers Verlag: Recklinghausen). Ristow, Sebastian. 1998. Frühchristliche Baptisterien (Aschendorff: Münster). Rizzardi, Clementina. 1989. ‘L’arte dei Goti a Ravenna: motivi ideologici, aspetti iconografici e formali nella decorazione musiva’, in Corso di cultura sull’arte ravennate e bizantina (ed.), XXXVI Corso di cultura sull’arte ravennate e bizantina: Seminario di studi sul tema: “Ravenna e l’Italia fra Goti e Longobardi (Ravenna, 14-22 aprile 1989) (Del Girasole: Ravenna): 373–375. 2001. ‘La decorazione musiva del Battistero degli Ortodossi e degli Ariani a Ravenna: alcune considerazioni’, in Daniela Gandolfi (ed.), L’edificio battesimale in Italia (Ist. Internazionale di Studi Liguri: Bordighera): 915–930. Rizzardi, Clementina, Linda Kniffitz, Letizia Sotira, and Barbara Vernia. 2011. Il mosaico a Ravenna: ideologia e arte (Ante Quem: Bologna). Roth, Kersten Sven. 2008, ‘Der Westen als “Normal Null”. Zur Diskurssemantik von “ostdeutsch” und “westdeutsch”’, in Kersten Sven Roth and Markus Wienen (eds.), Diskursmauern. Aktuelle Aspekte der sprachlichen Verhältnisse zwischen Ost und West (Hempen: Bremen): 69–89. Rouwhorst, Gerard. 2019. ‘Salbung’, in Christian Hornung, Sible de Blaauw, Therese Fuhrer, Hartmut Leppin, Winrich Löhr, Heinz-Günther Nesselrath, Maren Niehoff, Georg Schöllgen, and Ilinca Tanaseanu-Döbler (eds.), Reallexikon für Antike und Christentum (Hiersemann: Stuttgart): 339–370.

235

236

References

Rúa, Enrique Jorge Montenegro. 2008. Santa Eulalia de Bóveda (Xunta de Galicia, Consellería de Cultura e Deporte). Rujivacharakul, Vimalin. 2011. Collecting China: The World, China, and a History of Collecting (University of Delaware Press: Newark, DE). Rummel, Philipp von. 2013. ‘Unrömische Römer und römische Barbaren. Die Fluidität vermeintlich präziser Leitbegriffe der Forschung zum spätantiken Gallien’, in Steffen Diefenbach and Gernot Michael Müller (eds.), Gallien in Spätantike und Frühmittelalter. Kulturgeschichte einer Region (De Gruyter: Berlin, Boston): 277–296. 2016. ‘The Transformation of Ancient Land- and Cityscapes in Early Medieval North Africa’, in Susan T. Stevens and Jonathan Conant (eds.), North Africa Under Byzantium and Early Islam (Dumbarton Oaks Research Library and Collection: Washington, DC): 105–117. Rüpke, Jörg. 2016. On Roman Religion: Lived Religion and the Individual in Ancient Rome (Cornell University Press: Ithaca, London). Rüpke, Jörg, and Wolfgang Spickermann. 2012. Reflections on Religious Individuality: Greco-Roman and Judaeo-Christian Texts and Practices (De Gruyter: Berlin). Russo, Eugenio. 2001. ‘Sulla sopraelevazione neoniana del battistero della cattedrale di Ravenna’, in Daniela Gandolfi (ed.), L’edificio battesimale in Italia (Bordighera: Florence): 891–914. Saavedra Machado, João 1970. ‘Documentos de Estácio da Veiga para o estudo da arqueologia do Algarve. Catálogo de plantas, desenhos e mosaicos’, in Actas das I Jornadas Arqueológicas (Associação dos Arqueólogos Portugueses: Lisbon): 335–385. Sabrow, Martin. 2010. ‘Die DDR erinnern’, in Martin Sabrow (ed.), Erinnerungsorte der DDR (Bundeszentrale für Politische Bildung: Bonn): 11–29. Salzman, Michele Renee. 2002. The Making of a Christian Aristocracy: Social and Religious Change in the Western Roman Empire (Harvard University Press: Cambridge, MA, London). Sandwell, Isabella. 2007. Religious Identity in Late Antiquity: Greeks, Jews, and Christians in Antioch (Cambridge University Press: Cambridge). Santos, Affonso dos, and Maria Luísa Estácio da Veiga. 1972. Arqueologia Romana do Algarve (Associação dos Arqueólogos Portugueses: Lisbon). Saxer, Victor. 1988. Les rites de l’initiation chrétienne du IIe au VIe siècle (Centro italiano di studi sull’alto medioevo: Spoleto). Schäferdiek, Knut. 1967. Die Kirche in den Reichen der Westgoten und Suewen bis zur Errichtung der westgotischen katholischen Staatskirche (De Gruyter: Berlin). 2009. ‘Die Ravennater Papyrusurkunde Tjäder 34, der Codex argenteus und die ostgotische arianische Kirche’, Zeitschrift für Kirchengeschichte, 120.2: 215–231.

References

Schlunk, Helmut. 1988. Die Mosaikkuppel von Centcelles (Philipp von Zabern: Mainz). Schlunk, Helmut, and Theodor Hauschild. 1978. Die Denkmäler der frühchristlichen und westgotischen Zeit (Philipp von Zabern: Mainz). Schneider, Hannah. 2011. ‘Die Entwicklung der Taufbecken in der Spätantike’, in David Hellholm, Tor Vegge,
yvind Norderval, and Christer Hellholm (eds.), Ablution, Initiation, and Baptism: Late Antiquity, Early Judaism, and Early Christianity (De Gruyter: New York, Berlin): 1697–1719. Schoenebeck, Hans Ulrich von. 1935. Der Mailänder Sarkophag und seine Nachfolge (Pontificio Istituto di Archeologia Cristiana: Vatican City). Schrenk, Sabine, and Gerhard Rexin. 2007. ‘Erstaunen oder Flucht? Zur Darstellung des Jordan in den spätantiken Bildern der Taufe Jesu’, Jahrbuch für Antike und Christentum, 50: 180–198. Scott, James C. 1990. Domination and the Arts of Resistance: Hidden Transcripts (Yale University Press: New Haven). Sears, Gareth. 2007. Late Roman African Urbanism: Continuity and Transformation in the City (Archaeopress: Oxford). 2016. ‘A Catholic Stronghold? Religious and Ethnic Identites at Tipasa’, in Niccolo Mugnai, Julia Nikolaus and Nick Ray (eds.), De Africa Romaque: Merging Cultures Across North Africa: Proceedings of the International Conference Held at the University of Leicester (26–27 October 2013) (The Society for Libyan Studies: London): 131–141. Seligman, Adam B., Robert Paul Weller, Michael James Puett, and Benett Simon. 2008. ‘Ritual and the Subjunctive’, in Adam B. Seligman, Robert Paul Weller, Michael James Puett and Benett Simon (eds.), Ritual and its Consequences: An Essay on the Limits of Sincerity (Oxford University Press: New York, Oxford): 17–42. Settles, Isis H., and Nicole T. Buchanan. 2014. ‘Multiple Groups, Multiple Identites, and Intersectionality’, in Verónica Benet-Martínez and Ying-yi Hong (eds.), The Oxford Handbook of Multicultural Identity (Oxford University Press: New York): 160–180. Sfameni, Carla. 2014. Residenze e culti in età tardoantica (Scienze e Lettere: Rome). Shackleton Bailey, David Roy. 1982. Anthologia Latina. I, Carmina in codicibus scripta. Fasc. 1, Libri Salmasiani aliorumque carmina (B.G. Teubner: Stuttgart). Shaw, Brent. 1992. ‘African Christianity: Disputes, Definitions, and “Donatists”’, in Malcolm R. Greenshields and Thomas Arthur Robinson (eds.), Orthodoxy and Heresy in Religious Movements. Discipline and Dissent (Edwin Mellen Press: Lewiston, NY): 5–34. 2004. ‘Who were the Circumcellions?’ in Andy Merrills (ed.), Vandals, Romans and Berbers: New Perspectives on Late Antique North Africa (Ashgate: Aldershot): 227–258. 2011. Sacred Violence. African Christians and Sectarian Hatred in the Age of Augustin (Cambridge University Press: New York).

237

238

References Sidarus, Adel, and Felix Teichner. 1996. ‘As Termas romanas no Gharb AlÂndaluz. As inscrições árabes de Milreu (Estói)’, Arqueologia Medieval, 5: 177–189. Simon, Marcel. 1966. ‘Bellérophon chrétien’, in Jacques Heurgon, William Seston and Gilbert Charles-Picard (eds.), Mélanges d’archéologie, d’épigraphie et d’histoire offerts à Jérôme Carcopino (Hachette: Paris): 889–904. Sinding-Larsen, Staale. 1984. Iconography and Ritual: A Study of Analytical Perspectives (Universitetsforlaget AS: Oslo). Sintès, Claude, and Ymouna Rebahi. 2003. Algérie antique [catalogue de l’exposition 26 avril au 17 août 2003, Musée de l’Arles et de la Provence antique] (Éditions du Musée de l’Arles et de la Provence antiques: Arles). Smati, Naima. 2001. ‘Les mosaïques figurées à thème marin en Afrique du Nord (Tunisie-Algérie-Maroc): étude descriptive et analytique (Tunisie, Algérie, Maroc) (diss.)’, Univ. Paris IV, Paris (unpubl.). Smith, Jonathan Zittell. 1987. To Take Place: Toward Theory in Ritual (University of Chicago Press: Chicago, London). 1990. Drudgery Divine: On the Comparison of Early Christianities and the Religions of Late Antiquity (University of Chicago Press: Chicago). Smith, Julia Mary Howard. 2005. Europe After Rome: A New Cultural History 500–1000 (Oxford University Press: Oxford). Sölch, Brigitte. 2019. ‘Das Rathaus, der Stadtplatz und die Revitalisierung der Forumsidee – mit einem Ausblick auf das “Humboldt-Forum” in Berlin’, in Bardo Maria Gauly, Gernot Michael Müller, and Michael Rathmann (eds.), Dialoge mit dem Altertum. Sinnstiftungen aus der Vergangenheit in Antike, Früher Neuzeit und Moderne (Universitätsverlag Winter: Heidelberg): 271–317. Sörries, Reiner. 1983. Die Bilder der Orthodoxen im Kampf gegen den Arianismus (Lang: Frankfurt a.M.). Spieser, Jean-Michel. 2001. ‘Further Remarks on the Mosaic of Hosios David’, in Jean-Michel Spieser (ed.), Urban and Religious Spaces in Late Antiquity and Early Byzantium (Ashgate: Aldershot): 1–12. 2015. Images du Christ (Droz: Geneva). Spinks, Bryan Douglas. 2006. Early and Medieval Rituals and Theologies of Baptism. From the New Testament to the Council of Trent (Ashgate: Aldershot). Squire, Michael. 2009. Image and Text in Graeco-Roman Antiquity (Cambridge University Press: Cambridge). Stafford, Emma, and Judith Herrin. 2005. Personification in the Greek World: From Antiquity to Byzantium (Ashgate: Aldershot, Burlington, VT). Stasolla, Francesca Romana. 2002a. ‘Balnea ed edifici di culto: relazioni e trasformazioni tra tarda antichità e alto medioevo’, in Alessandra Guidobaldi Guiglia and Federico Guidobaldi (eds.), Ecclesia Urbis (Pontificio Istituto di Archeologia Cristiana: Vatican City): 143–151.

References

2002b. Pro labandis curis: il balneum tra tarda antichità e Medioevo (Fratelli Palombi editori: Rome). Stern, Henri. 1957. ‘Le décor des pavements et des cuves dans les baptistères paléochrétiens’, in Congresso internazionale di archeologia cristiana (ed.), Actes du Ve Congrès International d’Archéologie Chrétienne (Pontificio Istituto di Archeologia Cristiana: Vatican City): 381–390. Stevens, Susan T. 1988. ‘The Circus Poems in the Latin Anthology’, in John Herbert Humphrey (ed.), The Circus and a Byzantine Cemetery at Carthage (University of Michigan Press: Ann Arbor): 153–178. Stevens, Susan T., Angela V. Kalinovski, and Hans van der Leest. 2005. Bir Ftouha: A Pilgrimage Church Complex at Carthage, Journal of Roman Studies Supplementary Series, 59 (Cambridge University Press: Cambridge). Stewart, Alistair. 2015. On the Apostolic Tradition, Hippolytus; An English Version with Introduction and Commentary (St Vladimir’s Seminary Press: New York). Stewart, Michael, David Parnell, and Conor Whately (eds). 2022. The Routledge Handbook on Identity in Byzantium (Routledge: London, New York). Stiegemann, Christoph (ed.). 2001. Byzanz – Das Licht aus dem Osten. Kult und Alltag im Byzantinischen Reich vom 4. bis 15. Jahrhundert. [Katalog der Ausstellung im Erzbischöflichen Diözesanmuseum Paderborn] (Philip von Zabern: Mainz). Stiftung Humboldt Forum im Berliner Schloss and Bernhard Wolter (eds.). 2020. Das Humboldt Forum im Berliner Schloss (Prestel: Munich). Stiftung Humboldt Forum im Berliner Schloss, Judith Prokasky and Mareen Maaß (eds.). 2017. Palast der Republik: Ein Erinnerungsort neu diskutiert (Humboldt Forum im Berliner Schloss: Berlin). Stirling, Lea Margaret. 2005. The Learned Collector: Mythological Statuettes and Classical Taste in Late Antique Gaul (University of Michigan Press: Ann Arbor). Stotz, Peter. 1998. Handbuch zur lateinischen Sprache des Mittelalters: Formenlehre, Syntax und Stilistik (Beck: Munich). Stouraitis, Ioannis. 2018. ‘Byzantine Romanness: From Geopolitical to Ethnic Conceptions’, in Walter Pohl, Clemens Gantner, Cinzia Grifoni and Marianne Pollheimer-Mohaupt (eds.), Transformations of Romanness: Early Medieval Regions and Identities (De Gruyter: Berlin): 123–139. Straubach, Nikolaus. 2002. ‘Signa utilia – signa inutilia. Zur Theorie gesellschaftlicher und religiöser Symbolik bei Augustinus und im Mittelalter’, in Gerd Althoff and Christel Meier (eds.), Frühmittelaterliche Studien. Jahrbuch des Instituts für Frühmittelalterforschung der Universität Münster (De Gruyter: Berlin, New York): 19–49. Strube, Christine. 1996. ‘Zur Datierung der Baudekoration von Tebessa’, in Beat Brenk (ed.), Innovation in der Spätantike (Reichert: Wiesbaden): 423–455. Stryker, Sheldon. 1968. ‘Identity Salience and Role Performance: The Relevance of Symbolic Interaction Theory for Family Research’, Journal of Marriage and Family, 30.4: 558–564.

239

240

References 2000. ‘Identity Competition: Key to Differential Social Movement Involvement’, in Robert White (ed.), Identity: Self and Social Movements (University of Minnesota Press: Minneapolis): 21–40. Stryker, Sheldon, and Peter J. Burke. 2000. ‘The Past, Present, and Future of an Identity Theory’, Social Psychology Quarterly, 63.4: 284–297. Strzygowski, Josef. 1885. Ikonographie der Taufe Christi: Ein Beitrag zur Entwicklungsgeschichte der christlichen Kunst (Theodor Riedel: Munich). Sturaro, Chiara. 2013. ‘Il battesimo di Cristo nei mosaici delle cupole dei battisteri di Ravenna: una interpretazione della personificazione del Giordano’, Iconographica, 12: 9–21. Talgam, Rina. 2014. Mosaics of Faith: Floors of Pagans, Jews, Samaritans, Christians, and Muslims in the Holy Land (Pennsylvania State University Press: University Park). Tannous, Jack. 2018. The Making of the Medieval Middle East: Religion, Society, and Simple Believers (Princeton University Press: Princeton). Tardieu, Joëlle, and Isabelle Parron-Kontis. 2006. ‘Les mosaïques dites du “baptistère de Valence”’, in Pascale Soleil (ed.), De mémoires de palais. Archéologie et histoire du groupe cathédral de Valence (Musée de Valence: Valence): 131–141. Teichner, Felix. 1994. ‘Acerca da villa romana de Milreu/Estói. Continuidade de ocupação na época árabe’, Arqueologia Medieval, 3: 89–100. 2006. ‘“De lo Romano a lo Árabe”. La Transición del sur de la provincia de Lusitania a Al-Gharb Al-Andalus. Nuevas investigaciones en los yacimientos de Milreu y Cerro de Vila’, in Alexandra Arnau Chavarría, Javier Arce Martínez and Gian Pietro Brogiolo (eds.), Villas tardoantiguas en el Mediterráneo occidental (Editorial CSIC Consejo Superior de Investigaciones Científicas: Madrid): 207–220. 2008a. Entre tierra y mar. Zwischen Land und Meer. Architektur und Wirtschaftsweise ländlicher Siedlungsplätze im Süden der römischen Provinz Lusitanien (Portugal) (Museo Nacional de Arte Romano: Mérida). 2008b. ‘La pars urbana tardorromana de la villa de Milreu (Estói, Portugal): nuevos descubrimientos y antiguos documentos’, in Carmen Fernández Ochoa, Virginia García-Entero and Fernando Gil Sendino (eds.), Las villae tardorromanas en el occidente del Imperio: arquitectura y función: IV Coloquio internacional de Arqueología en Gijón (Trea: Gijón): 732–740. 2011. ‘“Nam primum tibi mater Hispania est, terris omnibus terra felicior”. Spätantike Großvillen und Residenzen auf der Iberischen Halbinsel’, in Gerda von Bülow and Heinrich Zabehlicky (eds.), Bruckneudorf und Gamzigrad: Spätantike Paläste und Großvillen im Donau-Balkan-Raum: Akten des Internationalen Kolloquiums in Bruckneudorf vom 15. bis 18. Oktober 2008 (Habelt Verlag: Bonn): 293–308. Thébert, Yvon. 1983. ‘L’évolution urbaine dans les provinces orientales de l’Afrique Romaine Tardive’, OPUS, 2: 99–130.

References

2003. Thermes romains d’Afrique du Nord et leur contexte méditerranéen (École française de Rome: Rome). Thomas, Thelma K. 2016. ‘Material Meaning in Late Antiquity’, in Thelma K. Thomas, Jennifer Ball, and Edward Bleiberg (eds.), Designing Identity: The Power of Textiles in Late Antiquity (Princeton University Press: Princeton). Tilley, Maureen Antonio. 1997. The Bible in Christian North Africa: The Donatist World (Fortress Press: Minneapolis). Tjäder, Jan-Olof. 1954–1982. Die nichtliterarischen lateinischen Papyri Italiens aus der Zeit 445-700 (Gleerup: Lund). 1972. ‘Der Codex argenteus in Uppsala und der Buchmeister Viliaric in Ravenna’, in Ulf Erik Hagberg (ed.), Studia Gotica. Die eisenzeitlichen Verbindungen zwischen Schweden und Südosteuropa (Almqvist & Wiksell: Stockholm): 144–164. Tomlinson, John. 1991. Cultural Imperialism: A Critical Introduction (Pinter: London). Toner, Jerry P. 2009. Popular Culture in Ancient Rome (Polity: Cambridge). Torres, Cláudio. 1982. ‘Alcáçova de Mértola. História e arqueologia urbana’, Arqueologia, 6: 86–95. 1991. Museu de Mértola. Núcleo do Castelo (Campo arqueológico de Mértola: Mértola). 1998. ‘O Guadiana – Envolvimento histórico’, Odiana, CCR Algarve: 8–12. 2015. ‘From Early Christianity to Islam’, in Susana Gómez-Martínez, Santiago Macias and Virgilio Lopes (eds.), O sudoeste peninsular entre Roma e o Islão ¼ Southwestern Iberian Peninsula Between Rome and Islam (Campo arqueológico de Mértola: Mértola): 361–363. Toutain, Jules. 1926. ‘Les cultes des fleuves, sa forme primitive et ses principaux rites, chez les peuples de l’antiquité classique’, L’ethnographie, 13–14: 1–7. Toynbee, Jocelyn Mary Catherine. 1964. ‘A New Roman Mosaic Pavement found in Dorset’, The Journal of Roman Studies, 54.1–2: 7–14. Trombley, Frank R. 1993–1994. Hellenic Religion and Christianization c. 370–529 (Brill: Leiden). Tsushima, Teresa, and Peter J. Burke. 1999. ‘Levels, Agency, and Control in the Parent Identity’, Social Psychology Quarterly, 62: 173–189. Turner, Ralph Herbert. 1978. ‘The Role and the Person’, American Journal of Archaeology, 84.1: 1–23. Uehlinger, Christoph. 2021. ‘Material Religion in Comparative Perspective: How Different is BCE from CE?’ in Jaś Elsner and Rachel Wood (eds.), Imagining the Divine (British Museum Press: London): 155–169. 2015. ‘Approaches to Visual Culture and Religion. Disciplinary Trajectories, Interdisciplinary Connections, and some Suggestions for further Progress’, Method and Theory in the Study of Religion, 27: 384–422.

241

242

References

Ulbert, Thilo. 1978. Frühchristliche Basiliken mit Doppelapsiden auf der Iberischen Halbinsel: Studien zur Architektur- und Liturgiegeschichte (Mann: Berlin). Uytterhoeven, Inge. 2009. ‘Know Your Classics! Manifestations of Classical Culture in Late Antique Elite Houses’, in Peter van Nuffelen (ed.), Faces of Hellenism: Studies in the History of the Eastern Mediterranean (4th Century B.C.-5th Century A.D.) (Peeters: Leuven): 321–342. Vásquez, Manuel A. 2011. More than Belief: A Materialist Theory of Religion (Oxford University Press: Oxford). Verhoeven, Mariette 2011. The Early Christian Monuments of Ravenna: Transformations and Memory (Brepols: Turnhout). Versluys, Miguel John. 2015. ‘Roman Visual Material Culture as Globalising Koine’, in Martin Pitts and Miguel John Versluys (eds.), Globalisation and the Roman World: World History, Connectivity and Material Culture (Cambridge University Press: Cambridge, New York): 141–174. 2008. ‘Exploring Identities in the Phoenician, Hellenistic and Roman East’, Bibliotheca Orientalis, 65.3: 342–356. Veyne, Paul. 2005. L’Empire gréco-romain (Le Seuil: Paris). Viciano i Vives, Albert. 2017. ‘Teologia baptismal en les controvèrsies amb el donatisme i l’arrianisme (segles IV-VII)’, in Julia Beltrán de Heredia Bercero and Cristina Godoy Fernández (eds.), La dualitat de baptisteris en les ciutats episcopals del cristianisme tardoantic (Ateneu Universitari Sant Pacià: Barcelona): 199–211. Vigne, Daniel. 1992. Christ au Jourdain: le baptême de Jésus dans la tradition judéochrétienne (Gabalda: Paris). Vives, José, Tomás Marín Martínez, and Gonzalo Martínez Díez. 1963. Concilios visigóticos e hispano-romanos (Consejo Superior de Investigaciones Científicas, Instituto Enrique Flórez: Barcelona, Madrid). Volbach, Wolfgang Fritz. 1952. Elfenbeinarbeiten der Spätantike und des frühen Mittelalters (Verlag des Römisch-Germanischen Zentralmuseums: Mainz). Vollkommer, Rainer. 1981–2009. ‘Iordanes’, in Hans Christoph Ackermann and Jean-Robert Gisler (eds.), Lexicon Iconographicum Mythologiae Classicae (Artemis: Zurich): 658–661. Voss Roberts, Michelle. 2010. ‘Religious Belonging and the Multiple’, Journal of Feminist Studies in Religion, 26: 43–62. Ward-Perkins, Bryan. 2005. The Fall of Rome and the End of Civilization (Oxford University Press: Oxford). 2007. ‘Where Is the Archaeology and Iconography of Germanic Arianism?’ in David M. Gwynn and Susanne Bangert (eds.), Religious Diversity in Late Antiquity (Brill: Leiden, Boston): 265–289. Waser, Otto. 1919. ‘Vom Flussgott Jordan und anderen Personifikationen’, in Eduard Hermann (ed.), Festgabe Adolf Kaegi (Huber: Frauenfeld): 191–217. Waszink, Jan Hendrik. 1947. ‘Pompa diaboli’, Vigiliae Christianae, 1.1: 13–41.

References Webster, Jane. 2001. ‘Creolizing the Roman Provinces’, American Journal of Archaeology, 105.2: 209–225. Wefing, Heinrich. 2010. ‘Der Palast der Republik’, in Martin Sabrow (ed.), Erinnerungsorte der DDR (Bundeszentrale für Politische Bildung: Bonn): 134–141. Weigel, Thomas. 1999. ‘53’, in Jan Gerchow (ed.), Klosterwelt Werden 799-1803: Das Jahrtausend der Mönche (Ruhrlandmuseum Essen: Cologne): 357–359. Weinryb, Ittai. 2002. ‘A Take of Two Baptisteries: Royal and Ecclesiastical Patronage in Ravenna’, Asaph: Studies in Art History, 7: 41–58. Weiss, Corinna. 1981. ‘Fluvii’, in Hans Christoph Ackermann and Jean-Robert Gisler (eds.), Lexicon Iconographicum Mythologiae Classicae (Artemis: Zurich): 139–148. Weitzmann, Kurt. 1960. ‘The Survival of Mythological Representations in Early Christian and Byzantine Art and their Impact on Christian Iconography’, Dumbarton Oaks Papers, 14: 43–68. Whalin, Douglas. 2021. Roman Identity from the Arab Conquests to the Triumph of Orthodoxy (Palgrave Macmillan: Basingstoke). Wharton, Annabel Jane. 1987. ‘Ritual and Reconstructed Meaning: The Neonian Baptistery in Ravenna’, Art Bulletin, 69.3: 358–375. 1995. Refiguring the Post Classical City: Dura Europos, Jerash, Jerusalem and Ravenna (Cambridge University Press: Cambridge). Whelan, Robin. 2018. Being Christian in Vandal Africa: The Politics of Orthodoxy in the Post-Imperial West (University of California Press: Oakland). Wickham, Chris. 2005. Framing the Early Middle Ages: Europe and the Mediterranean 400-800 (Oxford University Press: Oxford). 2009. The Inheritance of Rome: A History of Europe from 400 to 1000 (Allen Lane: London). Wiemer, Hans-Ulrich. 2018. Theoderich der Große. König der Goten – Herrscher der Römer (Beck: Munich). Williamson, Paul. 1999. ‘VII.37’, in Christoph Stiegemann and Matthias Wemhoff (eds.), 799 – Kunst und Kultur der Karolingerzeit: Karl der Große und Papst Leo III. in Paderborn [Eine gemeinsame Ausstellung der Stadt Paderborn, des Erzbistums Paderborn und des Landschaftsverbandes Westfalen-Lippe vom 23. Juli-1. November 1999] (Philipp von Zabern: Mainz). Wisskirchen, Rotraut. 1993. ‘Zum Medaillon im Kuppelmosaik des Orthodoxenbaptisteriums’, Jahrbuch für Antike und Christentum, 36: 164–170. Wood, Ian. 2007. ‘Theoderic’s Monuments in Ravenna’, in Sam J. Barnish and Federico Marazzi (eds.), The Ostrogoths from the Migration Period to the Sixth Century: An Ethnographic Perspective (Boydell Press: Woodbridge): 249–262. Wood, Jamie. 2012. The Politics of Identity in Visigothic Spain: Religion and Power in the Histories of Isidore of Seville (Brill: Leiden).

243

244

References

Woolf, Greg. 1998. Becoming Roman: The Origins of Provincial Civilization in Gaul (Cambridge University Press: Cambridge). 2015. ‘Ritual Traditions of Non-Mediterranean Europe’, in Rubina Raja and Jörg Rüpke (eds.), A Companion to the Archaeology of Religion in the Ancient World (Wiley Blackwell: Chichester): 465–477. Yacoub, Mohamed. 1994. ‘La mosaïque de Lahmimine et le thème africain des chevaux de cirque affrontés’, in Jean-Pierre Darmon and Alain Rebourg (eds.), La Mosaïque Gréco-Romaine IV: IVe Colloque International pour l’Étude de la Mosaïque antique (Supplément au Bulletin de l’A.I.E.M.A.: Paris): 249–257. 1995. Splendeurs des Mosaïques de Tunisie (Agence nationale du patrimoine: Tunis). 1996. Le musée du Bardo (Éditions de l’agence nationale du patrimoine: Tunis). Yarnold, Edward, and Cyril. 2000. Cyril of Jerusalem (Routledge: London). Yasin, Ann Marie. 2009. Saints and Church Spaces in the Late Antique Mediterranean: Architecture, Cult, and Community (Cambridge University Press: Cambridge, New York). 2016. ‘Beyond Spolia. Architectural Memory and Adaptation in the Churches of Late Antique North Africa’, in Susan T. Stevens and Jonathan Conant (eds.), North Africa under Byzantium and Early Islam (Dumbarton Oaks Research Library and Collection: Washington, DC): 215–236. Zaleski, John. 2014. ‘Religion and Roman Spectacle’, in Paul Christesen and Donald G. Kyle (eds.), A Companion to Sport and Spectacle in Greek and Roman Antiquity (Wiley Blackwell: Chichester): 590–602. Zanker, Paul, and Björn Christian Ewald. 2004. Mit Mythen leben: die Bilderwelt der Römischen Sarkophage (Hirmer: Munich). Zimmermann, Norbert, and Markus Öhler (eds.). 2017. ‘Sacra Privata: Domestic Religion in Greco-Roman Antiquity and Early Christianity’, Archiv Für Religionsgeschichte, 18–19.1: 3–263.

Archival Material Estácio da Veiga, Sebastião Philippes Martins. 1878. ’Catálogo dos monumentos e objectos de arte antiga, descobertos e obtidos no reconhecimento das antiguidades do distrito de Faro, feito desde março de 1877 até outubro de 1878, para o levantamento da Carta Archeologica do Algarve em virtude da portaria de 15 de janeiro de 1877 por S. P. M. Estácio da Veiga’, Archive of the Museu Nacional de Arqueologia, Lisbon. 1885. ‘Inventário do Museu Archeologico do Algarve fundado em 1880 na Academia Real de Bellas Artes de Lisboa por Sebastião Philippes Martins Estácio da Veiga’, Archive of the Museu Nacional de Arqueologia, Lisbon.

Index

Adam and Eve, 161 Africa Proconsularis, 23, 40, 66–67, 79–80, 82, 89–90, 96, 119–122, 130, 154 Albenga, 6n8, 117 Alentejo, 23, 79, 105, 136, 149 Alexander, bishop of Tipasa, 47, 49, 58 Algarve, 23, 79, 136 Algeria, 5, 11, 22, 25, 28, 70, 98 Algezares, 149 Ambrose of Milan, 12n34, 60n114, 61, 192 Ammaedara, 68n149, 69 Andreas Agnellus of Ravenna, 164, 169, 188 angel, 85–86, 99–100, 172, 180 Anthologia Latina, 96, 128n170 apostle, 167, 169–172 Apostolic Tradition, 14–15 apse, 32n27, 35, 38, 43–44, 69–70, 81–83, 86–87, 108, 112, 117, 124n156, 134, 139, 148, 170n50, 176n70 counter-apse, 83, 87–88 Arch of Titus, 175 architecture, xvii–xix, xxiv, xxvi, xxix, 4, 6, 12, 16–17, 19, 20n84, 24–25, 28, 52, 67, 77, 152, 169, 188, 197–200 Arian art, 23–24, 156, 162–163, 171n57, 190–193 bishop, 116, 117n135, 169, 171 church, 2, 8, 59, 104, 116–117, 169 Augusta Emerita (Mérida), 104, 118n142, 128n180, 128, 137n210 Augustine of Hippo, 9, 13–15, 50n69, 57n98, 58, 60–61, 65–66, 69, 72–75, 78, 85, 99, 160, 200 Baetica, 79n2, 115, 149 Balearic Islands, 24, 120, 120n149, 127n164 baptism infant baptism, 13–14, 113–117 of Jesus, 23, 171, 181, 183, 185n95, 188, 191–192, 192n121 baptismal abjuration, 99–101

font, 25, 37, 39, 69, 72, 80, 82–83, 108–110, 112–118, 124, 137–138, 140, 148–151, 165 liturgy, see liturgy water, 7, 13, 31, 36–37, 60–61, 75n169, 85, 88, 99, 109, 112, 125, 149, 151–153, 164–189, 191–192 baptistery Diaconicon, 24, 88n35, 121n151, 124 double baptistery, 116–117 free-standing, 31, 52, 149 Barbarian, 2, 18–19, 103–104, 139, 199 Barcino (Barcelona), 6n8, 112, 116–117, 149 Basil the Great, 61 bath of the churches of Cuicul, 34–35, 41, 46, 70, 78, 200 of the Great Basilica of Tipasa, 70 Great bath of Cuicul, 63 of Capsa (Gafsa), 97 of the church of Thamugadi (Timgad), 70–72 private, 98, 132 bathing, 21, 72–75, 200 Bellerophon, 110–112, 118, 122–124, 127–131, 154 Berlin, xvii, xxiv, 198 Braulio of Zaragoza, 103 Bulla Regia, 6n8, 50n67, 72, 88n33 burial, 12, 139, 148n240 Byzacena, 40, 66, 69, 83, 89, 96–97, 119, 122 Byzantine art and architecture, 4, 52–53, 61, 66n140, 83, 95n57 reconquest of North Africa, 66 Byzantium, 2n1, 4, 59, 154, 169 candle, 85–86 Capsa (Gafsa), 97 Carthage, 5n8, 58–59, 66–67, 81, 83n17, 91–92, 96–97, 99, 120n149, 160 centaur, 94n55, 122, 147 Centcelles, 135

245

246

Index

chi-rho, 26–27, 40, 128 chrism, 44–45, 50–52, 70 Christian doctrine, 3, 7, 12–13, 15, 57, 101, 196 ethos, 57, 80, 104, 131, 197 identity, xix, xxiii, 3, 7–13, 15, 18, 22–23, 58–59, 69, 78, 96, 101, 129, 153–154, 195–199 non-Christian, 79, 200 Christological, 162–163, 170n55, 193 church of Alexander, Tipasa, 49, 59 of Cuicul (Djemila), 29–30, 46, 52–60 double church, 25, 31, 44, 50, 52, 54–55, 59 of Henchir el Koucha, 81, 83, 87–88 of Hergla, 120, 122, 124, 127, 130 of San Miguel, Barcelona, 149–150, 152 Ciampini, Giovanni, 166–167 ciborium, 31, 36, 43, 82 circus (games), 21, 79–81, 89, 92–102, 104, 153–154, 200 commemoration, xxiii, 43, 47–49, 55–57, 58n104, 59, 72, 77 East German, see German/y; East; commemoration Conference of Carthage (411), 52, 57n98, 58, 77 consignatorium, 35n30, 70 Constantinople, 4, 24, 52–54, 180 creed, 13, 59, 72, 104, 162 Cresconius, bishop of Cuicul, 44, 49, 52, 58 inscription, see inscription; Cresconius inscription cross, 57n97, 83, 88, 89n37, 117, 165, 166n36, 167, 170, 172 crypt, 43–44, 49 Cuicul (Djémila), 22, 25, 29 cultural affiliation, xix, 4, 20–22, 98, 127, 129, 154, 196 culture dress-culture, 19, 139 literary, 19, 103n84, 199 Roman, xviii, xxix, 2–4, 11, 19, 21–24, 28, 63–65, 97n63, 101–102, 104n87, 131, 153–155, 190, 193, 196–199 visual and material, 3–7, 11, 15–23, 28, 65, 78, 81, 101–102, 139, 147, 152, 196–200 Cyprian of Carthage, 60 Cyrenaica (Qasr Lybia), 175–176 Cyril of Jerusalem, 100 De correctione rusticorum, 103 De Spectaculis, 99–100 deer, 27, 85–88, 118, 125

deity, 22, 62, 66, 99, 102–103, 131n192, 136, 147, 156, 184 Diaconicon baptistery, see baptistery; Diaconicon Djémila, see Cuicul (Djémila) donation, 10, 55–57, 67, 169, 190 Donatist bishop, 58, 77 dissidents, 8, 27–28, 52, 57–59, 70, 78 Easter, 72, 85, 183 El-Erg, 6n8, 83 Elijah, 161, 174 Emona (Ljubiliana), 117 Epiphany, 181–183, 186–187, 192n120 episcopal palace, 46n57, 164n27, 164, 168–169 patronage, 49, 58, 161, 163–164, 181–183, 188–189, 193 tomb, 44, 48–49, 58–59 Eutychianism, 116 Felix of Ravenna, 188 fidelis, 14–15, 65 fish, 7, 25, 37–40, 60–65, 89, 134–136, 140, 144, 150–151 flamen perpetuus, 56, 68–69 fountain, 27, 38, 62, 65, 87, 117–118 Frampton (Dorset), 128n172, 129–130 France, 5, 125, 128 fresco, 113 frigidarium, 35, 118n143, 121n151, 134, 149 German/y Democratic Republic, xvii, xx–xxii, xxiv–xxvii, xxix, 194–195 East, xix–xxiv, xxvi, 194–195 commemoration, xxi–xxii identity, xix–xxiii, xxix, 195 Federal Republic, xx–xxiii, xxv reunification, xvii, xix–xx, xxii, 195 West, xx–xxii, xxvi, 194 Good Shepherd, 60n114, 89 Greek, 4–5, 24, 51n71, 60, 90, 92, 100, 104n87, 116, 185, 189, 193 guilloche, 145–146 Hadjeb-el-Ayoun, 89, 98 Hadrumetum (Sousse), 63n128, 85n22, 91, 119 Henchir el Koucha, see church; of Henchir el Koucha Hergla, see church; of Hergla

Index

heritage East German, xvii, xix, xxvii Roman, 3–4, 20–21, 66–67, 96, 152–153, 160–161, 163, 194, 197, 199 Hetoimasia, 170 Hexameron, 61 Hinton St Mary (Dorset), 128, 130 Hippo Regius, 6n8, 26, 66, 69, 72 hippocamp, 140, 147 hippodrome, see circus (games) Holy Ghost, 162, 165, 171, 192 homoousios, 162 horse, 87–96, 98, 101–102, 121, 139–140, 153–154 Humboldt Forum, xxv, xxvii–xxix, 198 hunt, 88n35, 97, 101, 118–127, 130, 198 Iberian Peninsula, 5, 13, 18, 79, 83n19, 90, 103–104, 112n112, 113n120, 115–117, 118n142, 122–124, 136–139, 139n223, 141, 147–149, 153 identity Christian, see Christian; identity concept of, xx, xxiii–xxiv, 9–10 East German, see German/y; East; identity ethnic, 9, 18–19, 104, 195–196 group, xix, xxix, 12, 16–17, 27–28, 198–199 religious, 16–18, 104, 196 Roman, xix, xxiii, xxix, 2–4, 4, 11, 18–22, 196–199 theory, 10–11, 154, 196–197 idolatry, 98–100, 104, 200 individuality, 3, 8–9, 11–13, 16–18, 27, 74, 93, 103, 155, 196, 198 inscription Alexander inscription, 47–51, 78 Cresconius inscription, 44, 46–52, 57–59, 70, 77 inscriptions of the baptistery of Cuicul, 37–39, 41–43 inscriptions of the church of Nola, 47n59 Isidore of Seville, 99, 103, 104n87, 138 Italy, 5, 6n11, 149, 162, 169, 171n57, 180, 192n120 Jabaliyah, 172n62, 177 Jebel Oust, 122 Jerome, 60n114, 186 Jesus Christ, 60, 167, 193 Jewish, 187, 189–190, 193, 195 John of Damascus, 161 John the Baptist, 165, 167, 171, 193 Joshua, 183, 186 Justinian, 4, 53, 169n46, 175

kantharos, 25, 36–37, 88, 90, 94, 95n57, 139–140, 151n252 Kelibia, 6n8, 27, 40n42, 86 Kybel, Felice, 166–167 Lahmimine, 95–96 lamb, 40, 125, 184 Latin, 13, 193 Lent, 15, 72 Liber Pontificalis Ecclesiae Ravennatis, 164 lion, 111, 118–120, 121n151, 127 sea lion, 143–147 liturgy, 4, 6–7, 13, 20n84, 41, 51, 57, 65, 69–70, 80, 88, 99, 101–103, 112n112, 118n142, 171n57, 183, 200 lived religion, 3, 16–18, 20, 25, 27, 78, 155, 197 locura, 99, 104 Los Castillejos, 136 luck, 25–26, 94–95, 98, 102 ludi, 80, 96, 99–101, 197 Lullingstone (Kent), 128n172, 129–130 Lusitania, 19n76, 23, 79, 103–105, 115–116, 118n142, 119n147, 131, 134n198, 138, 147–150, 151n255, 152n256, 154, 197 Madaba, 172n62, 176–177, 191n119 magic, 65, 75 Marialba, 149 Mariana, 6n8, 151n255, 177 Martin of Braga, 103 martyr, 46, 49, 58n104, 148, 172n62, 176–177 Maundy Thursday, 72–75 Mauretania Caesariensis, 29, 47 Mauretania Sitifensis, 29, 55n90, 68n149 mausoleum, 105, 131, 136–137, 139 Mérida, see Emerita Augusta Mértola, see Myrtilis Iulia Messmer, Arwed, xxvi Milreu/ Estói, 6n8, 79, 131, 140n225, 152 modius, 92–94 Moknine, 94n55, 98n65, 98 Montinho das Laranjeiras, 6n8, 150–152 mosaic mosaicist, 24, 27, 49, 88n35, 119–120, 124, 130, 152n256, 189 tomb-mosaic, 25–26, 89 Myrtilis Iulia (Mértola), 5n8, 79–80, 104–108, 116, 124, 127–131, 197 Mystagogic Catecheses, 100 mystery religion, 66, 75 mythology, xix, 21, 79–80, 96n59, 102–104, 124, 127n167, 129–131, 137n210, 140, 144–149, 152–154, 160, 172, 189–190, 197

247

248

Index

Neon, bishop of Ravenna, 23, 155–156, 163–164, 172, 183, 188–189, 192, 195, 198 Neptune, 147, 174 New Testament, 5, 14, 161, 187 Nicene church, 2, 8, 13, 23, 28, 44, 52, 57–60, 65, 77–78, 116–117, 162, 199 nudity, 176–179 Numidia, 22, 24–26, 28, 52, 55, 57, 66, 68n149, 78 Oceanus, 63n128, 156, 191–193 Ohrid, 85n22, 177 Old Testament, 14, 161, 181n80, 183, 187 Ossonoba, 131–132 ostrich, 118, 121, 127 Ostrogoths, 18, 162–163, 169n43, 169, 171n57 Osuna, 90n44, 98 Oued Athmenia, 98 paganism, 2, 3n3, 8–9, 22, 65–69, 79n1, 98–99, 102n82, 103n84, 127n168, 129, 136, 156, 195 paideia, 102, 104, 131, 154 Palace of the Republic (Palast der Republik), xvii, xxiv–xxv, xxvii–xxix, 198 palm tree, 25, 87, 91–92 Paulinus of Nola, 47n59 peacock, 87–88, 139 personification, 22–23, 156, 160–162, 164–193 Peter Chrysologus, 182–190, 192–193, 195 pilgrimage, 24, 28, 46, 49, 51–52, 58, 72n163, 77 poetry, 96–97, 160 pompa, 99–101 Portugal, 5, 79, 118, 128–129, 137n210, 147n238 psalm, 5, 27, 43, 85, 88, 161, 184–186 Pseudo Hippolytus, 185 Punic, 91–92, 154 purification, 28, 66, 69, 72–75, 192 Quinta de Marim, 136 Quodvultdeus, 66 Ramberg, Lars Ø, xvii, xxvii Ravenna, 5n8, 23, 116n132, 154–156, 162–164, 169, 181, 183, 185, 188, 192–193, 195, 198–199 Reccared I, 104, 116 Red Sea, 183–184 river god, 23–24, 156, 160, 187, 189, 193, 195 River Jordan, 23, 155–156, 161, 171–195, 181n80 rivers of Paradise, 85, 172n62, 175–179 Roman Empire, 4, 18, 24–25, 29, 94, 163, 194 Rome, 14, 19–21, 68, 136, 163–164, 175, 194, 199 Rusguniae, 72

sacerdos provinciae, 55, 68 Salvian of Marseille, 99 São Cucufate, 136 sarcophagus, 136n208, 161, 174 sermon, 9, 15, 23, 61, 65, 99, 155, 160, 182–183, 186, 187n103, 188–189, 193, 195 sheep, 88–89 Sidi Abdallah, 98, 98n65 Sidi Jdidi, 6n8, 86n25, 86, 88n35, 89n39 Sitifis (Sétif), 26–28, 57 Sousse, see Hadrumetum Spain, 4, 90n44, 103n84, 115n125, 128, 131n193, 136, 138, 147, 147n238, 148n240, 149, 150n251 Sufetula (Sbeitla), 6n8, 40n42, 50n67, 72, 122, 124n156 swastika, 40–41 temple, 56, 66–67, 75, 117, 131n192, 136–137, 152 Tertullian, 60, 99–100, 192n121 Thabraca, 88n33, 89, 95, 98 Thamugadi (Timgad), 6n8, 26, 70–74 Theoderic, 162, 169, 171, 190 Theodoret of Cyrus, 161 Theophany, 167, 185 Theveste (Tebessa), 26, 52 thiasus, 79, 131, 136, 144, 147–148 Thuburbo Maius, 6n8, 37n37, 122n152, 122 Thysdrus (El Djem), 96 Tipasa, 6n8, 46–49, 51, 58–59, 60n111, 70–72, 74, 78 Torre de Palma, 115n126, 148n241, 149 Tractatus contra paganos, 65 Trinity, 12–13, 162–163, 171, 182–189 triton, 143–147, 150 Tunisia, 5, 23, 25, 79, 120 Umm ar-Rasas, 120n149, 172n62, 176 Ursus, bishop of Ravenna, 164, 188 Vandals, 18, 23, 29, 59, 68n149, 77n171, 82–83, 96–97, 99, 154 Veiga, Estácio da, 105, 132, 136, 139–140, 145, 150 Venus Marina, 147 villa, 62, 98, 102, 128–129, 134n198, 139n223, 147n238 Visigoths, 18, 23, 103–104, 139, 147, 154, 162n14 Werden Casket, 180