Public Baths and Bathing Habits in Late Antiquity: A Study of the Evidence from Italy, North Africa and Palestine A.D. 285-700 9004418725, 9789004418721

Citation preview

PUBLIC BATHS AND BATHING HABITS IN LATE ANTIQUITY

LATE ANTIQUE ARCHAEOLOGY (SUPPLEMENTARY SERIES) MANAGING EDITOR

LUKE LAVAN volume 6

The titles published in this series are listed at brill.com/laax

PUBLIC BATHS AND BATHING HABITS IN LATE ANTIQUITY A STUDY OF THE EVIDENCE FROM ITALY, NORTH AFRICA AND PALESTINE A.D. 285–700

BY

SADI MARÉCHAL

LEIDEN | BOSTON

Cover illustration: The Schola Baths in Leptis Magna (Libya). ©Photo by Sadi Maréchal, June 2012. Library of Congress Cataloging-in-Publication Data Names: Maréchal, Sadi, author. Title: Public baths and bathing habits in late antiquity : a study of the  evidence from Italy, North Africa and Palestine A.D. 285–700 / by Sadi  Maréchal. Other titles: study of the evidence from Italy, North Africa and Palestine  A.D. 285–700 Description: Leiden ; Boston : Brill, 2020. | Series: Late antique  archaeology (supplementary series), 2352–5177 ; vol.6 | Includes  bibliographical references and index. Identifiers: LCCN 2019048147 | ISBN 9789004418721 (paperback) |  ISBN 9789004419421 (ebook) Subjects: LCSH: Baths, Roman—Italy. | Baths, Roman—Africa, North. |  Baths, Roman—Palestine. | Public baths—Italy. | Public baths—Africa,  North. | Public baths—Palestine. | Bathing customs—History. |  Rome—Antiquities. Classification: LCC DG97 .M165 2020 | DDC 613/.410937—dc23 LC record available at https://lccn.loc.gov/2019048147

Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface. ISSN 2352-5177 ISBN 978-90-04-41872-1 (paperback) ISBN 978-90-04-41942-1 (e-book) Copyright 2020 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi, Brill Sense, Hotei Publishing, mentis Verlag, Verlag Ferdinand Schöningh and Wilhelm Fink Verlag. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. Brill has made all reasonable efforts to trace all rights holders to any copyrighted material used in this work. In cases where these efforts have not been successful the publisher welcomes communications from copyright holders, so that the appropriate acknowledgements can be made in future editions, and to settle other permission matters. This book is printed on acid-free paper and produced in a sustainable manner.

In loving memory of my grandfather, Prof. Dr. Raymond Bogaert, who was even more passionate about Antiquity than I am

…

To Nathalie

∵

Contents Foreword xi List of Illustrations xiii Abbreviations xx Key to the General Plans xxii Introduction 1 Why Baths Again? 1 General Aims and Outline of the Research 1 Geographical Framework 2 Chronological Framework 3 Methodology and Sources 4 Baths and Bathing Approached from Different Angles 4 Evidence of Baths and Bathing 4 The Structure of the Book 6 Notes on Terminology 6 Geographical Terminology and Place Names 6 Chronological Terminology 8 Cultural and Religious Terminology 8 Personal Names 8 Bath-Related Terminology 8 Notes on Abbreviations 9 1 General Introduction to Roman Baths and Earlier Research on Late Antique Baths 10 General Introduction to Roman Baths and Bathing Habits 10 Pre-Roman Origins of Public Bathing 10 Italic Bathing Habits 11 Early Roman Bathhouse 12 Imperial Period Bathhouse 13 The Bathing Itinerary 19 Modern Typologies of the Plans 20 Technology of the Baths 20 Hypocaust 21 Wall Heating 23 The Functioning of the Heating System 24 Water Management of Baths 26 Water Supply 26 Waste Water Disposal 27 Decoration 27 Popularity of Bathhouses 28 Earlier Research on Late Antique Baths 30 Late Antique Baths in General Works on Roman Baths and Bathing 30 Late Antique Baths in Regional Studies 31 Specific Studies on Late Antique Baths 32 2 Written Evidence of Baths 34 Late Antique Literature 34 Methodology and Approach 34 4th c. Latin Authors 34 4th c. Greek Authors 42 5th c. Latin Authors 49 5th c. Greek Authors 54 5th c. Syriac Literature 57

viii







Contents

6th c. Latin Authors 58 6th c. Greek Authors 61 6th c. Syriac Authors 66 7th c. Authors 67 Summarizing the Literary Attestations of Baths and Bathing Habits 68 Hammams in Early Medieval Arabic Literature 70 Legal Documents 71 Codex Theodosianus 71 Corpus Iuris Civilis 73 Ecumenical Councils and Synods 75 Summarizing the Legal Documents 75 Epigraphic Evidence 76 Methodological Problems and Approaches 76 Inscriptions of the Italian Peninsula 77 Inscriptions of North Africa 80 Inscriptions of Cyrenaica 85 Inscriptions of Palestina 85 Summarizing the Epigraphic Evidence 85 Papyrological Evidence 86 Introduction 86 Papyri 87 Summarizing the Papyrological Evidence 89

3 Archaeological Evidence 91 Introduction 91 Rome 91 Concise History of the Urban Fabric 91 Newly Built Late Antique Baths 95 Continued Use and Abandonment of Existing Baths 95 Private Baths 101 Baths on the Forma Urbis Roma 102 The Evidence for Baths in the Written Sources 103 Baths in the Late Antique Urban Context 105 Ostia 110 Concise History of the Urban Fabric 110 Newly Built Late Antique Baths 112 Continued Use and Abandonment of Existing Baths 113 Private Baths 128 Baths on the Isola Sacra 129 Baths in the Late Antique Urban Context 129 Cuicul 138 Concise History of the Urban Fabric 138 Newly Built Late Antique Baths 140 Continued Use and Abandonment of Existing Baths 140 Baths in the Late Antique Urban Context 142 Thamugadi 143 Concise History of the Urban Fabric 143 Newly Built Late Antique Baths 145 Continued Use and Abandonment of Existing Baths 145 Baths in the Late Antique Urban Context 148 Carthage 150 Concise History of the Urban Fabric 150 Newly Built Late Antique Baths 152

Contents

Continued Use or Abandonment of Existing Baths 152 Baths in the Late Antique Urban Context 155 Sufetula 157 Concise History of the Urban Fabric 157 Newly Built Late Antique Baths 158 Continued Use and Abandonment of Existing Baths 160 Baths in the Late Antique Urban Context 160 Ptolemais 163 Concise History of the Urban Fabric 163 Newly Built Late Antique Baths 165 Continued Use and Abandonment of Existing Baths 167 Private Baths 167 Baths in the Late Antique Urban Context 167 Scythopolis 168 Concise History of the Urban Fabric 168 Newly Built Late Antique Baths 170 Continued Use and Abandonment of Existing Baths 171 Baths in the Late Antique Urban Context 171 Archaeological Evidence of Early Hammams 175 North Africa and Cyrenaica 175 Egypt 175 Palestina 175 The Larger Middle Eastern Context 177 Summarizing the Archaeological Evidence of Early Hammams 180 4 Continuity and Change in Late Antique Public Baths and Bathing Habits 183 Architecture 183 Plans of the Baths 183 Size of the Baths 185 Rooms of the Baths 186 Construction Techniques and Building Materials 195 Decoration 196 Bath Technology 198 Hypocaust 198 Wall Heating 199 Furnaces 199 Water Supply 200 Waste Water Disposal 201 Public Baths and Their Contexts 202 Urban Contexts 202 Non-Urban Contexts 204 Secondary Use of Baths 204 Privatisation of the Bathing Habit? 206 The Church and the Baths 208 Decay or Continuity of Baths and Bathing Habits 210 Continuity in the 4th c. 210 Strong Regional Trends in the 5th c. 216 Difficulties in the 6th c. 218 Towards Medieval and Islamic Baths in the 7th c. 219 A Habit in Decline? 221 Epilogue—The Transition to the Hammams 223 General Concluding Remarks and Prospects for Future Research 225

ix

x

Contents

Appendix 1: Late Antique Inscriptions Mentioning Baths 227 Inscriptions from the Italian Peninsula 227 Inscriptions from North Africa 234 Inscriptions from Palestina 243 Appendix 2: Late Antique Papyri Mentioning Baths 244 Appendix 3: Lists of Late Antique Baths, Possible Baths and Continued Use of Baths 250

Gazetteer Introduction 261 Baths in the Italian Peninsula, Sicily and Sardinia 263 New Construction 263 Possible Late Antique Baths 296 Villa Baths with a Possible Public Use 299 Continued Use and Abandonment of Imperial Period Baths during Late Antiquity 301 Baths in Roman North Africa 310 New Construction 310 Possible Late Antique Baths 345 Villa Baths with a Possible Public Use 345 Other Private Baths with Possible Public Access 348 Ecclesiastical Baths 349 Fortress Baths 354 Continued Use and Abandonment of Imperial Period Baths during Late Antiquity 356 Baths in Cyrenaica 382 New Construction 382 Possible Late Antique Baths 392 Fortress Baths 392 Continued Use and Abandonment of Imperial Period Baths during Late Antiquity 393 Baths in Egypt 394 New Construction 394 Possible Late Antique Baths 420 Other Private Baths with a Possible Public Use 421 Ecclesiastical Baths 422 Fortress Baths 422 Continued Use and Abandonment of Imperial Period Baths during Late Antiquity 426 Baths in Palestina 428 New Construction 428 Possible Late Antique Baths 446 Private Baths 449 Ecclesiastical Baths 450 Fortress Baths 451 Continued Use and Abandonment of Imperial Period Baths during Late Antiquity 452 Glossary 456 Bibliography 461 Primary Sources 461 Modern Sources 464 Indices 498 Ancient Sources 498 Themes 499 People 500 Places 501

Foreword This book is the result of the research conducted for my doctoral thesis between 2012 and 2016, supervised by Prof. Dr. Frank Vermeulen at Ghent University (Belgium) and funded by the Special Research Fund of Ghent University (BOF). Additional funding for research stays and field trips was provided by scholarships from the Academia Belgica in Rome (2011 and 2012), and the Belgian Historical Institute in Rome (2013). Throughout my academic career, several people have contributed to the successful completion of this book. First, I would like to thank my mentor Prof. Dr. Frank Vermeulen, who guided me through my archaeological studies, permitted me to attend the excavations of Ammaia (Portugal), and encouraged me to pursue an academic career. My profound gratitude also goes to Dr. Nathalie de Haan (Radboud University, Nijmegen), who, despite her own busy schedule, helped shape this project, answer questions and discuss Roman baths. Her devotion to baths, archaeology and ancient history has inspired me to find my own path in academia. Several other colleagues at Ghent University have been of tremendous help, such as my colleagues in the ‘Roman Mediterranean Archaeology’ unit in the Department of Archaeology. A particular word of thanks to Dr. Devi Taelman, Dr. Dimitri Van Limbergen, Dr. Jan Trachet, Prof. Dr. Wim De Clercq, Sophie Dralans, Debby Van den Bergh, Tim Clerbaut, Dr. Patrick Monsieur, Dr. Lieven Verdonck and Dr. Francesca Carboni. Also at Ghent, the meetings with colleagues of the Institute for Classical Studies have been most stimulating. From the History Department, I thank Prof. Dr. Koen Verboven, Prof. Dr. Arjan Zuiderhoek, Dr. Lindsey Vandevoorde and Prof. Dr. Peter Van Nuffelen. At the Department of Classical Languages, my thanks go to Prof. Dr. Wim Verbaal, who kindly helped me with several Latin to English translations. My research also led me to other Belgian universities, where I was always welcomed with enthusiasm and helpfulness. I wish to thank the Department of Archaeology and the library staff at KU-Leuven, the Université Libre de Bruxelles, the Université de Liège and the Université Catholique de Louvain-la-Neuve. A special word goes to Dr. Jonas Danckers (KU-Leuven) and Dr. Delphine Tonglet (ULB). Last, but not least, I thank the many colleagues at the Université de Liège, in particular Dr. Stéphanie Derwael (Department of History of Art and Archaeology) and the study unit of papyrology (CEDOPAL), including Prof. Dr. Marie-Hélène Marganne, Dr. Antonio Ricciardetto and Dr. Nathan Carlig, for introducing me to and helping me with the papyrological databases. Their help with several translations has been much appreciated. No research about the Mediterranean can be complete without several field trips to the area in question. I dare say that Rome and the Academia Belgica have become my ‘homes-away-from-home’. I thank the two directors of the Academia, Prof. Dr. Walter Geerts and his successor Prof. Dr. Wouter Bracke, as well as the entire staff. A special word of thanks for Charles Bossu and Pamela Anastasio for these years of collaboration and friendship. In Rome, I also had the privilege of working in some of the finest libraries of Roman archaeology that Europe has to offer. I wish to thank the library staff of the British School at Rome, the Deutsches Archäologisches Institut, the École française de Rome and the Koninklijk Nederlands Instituut in Rome. I extend my gratitude to the Parco Archeologico di Ostia Antica for permitting access to the site and use of the library, the Archivio disegni and the Archivio fotografico. A word of gratitude is also due to the French archaeological mission in Libya, for introducing me to Roman baths and giving me the opportunity to work on the wonderful site at Leptis Magna. I am indebted to Prof. André Laronde (†), Prof. Dr. Vincent Michel, Prof. Dr. Michel Bonifay, Prof. Michel Paulin and Guillaume Dagnas. May we find ourselves together again in a peaceful Libya.

xii

Foreword

This monograph has benefitted from the useful insights and revisions of my PhD jury. I thank Prof. Dr. Neil Christie, Dr. Nathalie de Haan, Dr. Francesca Carboni, Dr. Luke Lavan and Prof. Dr. Roald Docter for their much appreciated remarks. I am also indebted to two anonymous peer-reviewers, whose stimulating remarks and corrections made this into a better book. Furthermore, I had the pleasure of meeting several scholars, who, in one way or another, contributed to my research. In no specific order, I would like to thank Dr. Paul Johnson, Dr. Béatrice Meyer, Dr. Hubertus Manderscheid, Dr. Michel Blonski, Prof. Dr. Lynne Lancaster, Prof. Dr. Patrizio Pensabene, Dr. Duncan Keenan-Jones, Prof. Dr. Eric Moormann, Dr. Vicky Manolopoulou, Dr. Stefanie Hoss and Dr. Thibaud Fournet. Also sincere thanks to all the authors who permitted the reproduction of illustrations and figures. To conclude, I owe gratitude to Dr. Michael Mulryan and his successor Dr. Peter Crawford for a fantastic editing job. All remaining errors are my own. Last, but not least, I thank my family, family-in-law and friends. No words can express my gratitude towards my parents and grandparents for their eternal love and support. A special word of thanks also to the family Baeyens-De Beule, who welcomed me into their home as one of their own. I certainly cannot forget my friends, especially those with (faint) moustaches, for dragging me out of the academic world from time to time. Grazie Jan, Arne, Thomas and Dimi! To the countless friends I met in the Academia Belgica in Rome: know that these moments spent in the Eternal City are priceless and unforgettable. To conclude, I thank my life-partner Nathalie, who had to endure my bathing frenzy, but nevertheless always remained my foremost supporter. Sadi Maréchal

Ghent, January 2020

Illustrations Figures 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Reconstruction of a tholos in a Greek-style bathhouse 11 Plan of the baths of Fregellae (Italy) with its early hypocaust 14 Plan of the Stabian Baths in Pompeii (Italy), with the alleged Greek hipbaths to the north 15 Plan of the Baths of Nero in Rome (Italy) 16 Plan of the Baths of Trajan in Rome (Italy) with the standing remains highlighted in black 17 Plan of the Small Baths in Hadrian’s Villa at Tibur (Italy) 18 Plan of the Baths of Caracalla in Rome (Italy) 19 Flow of the hot gasses inside the tubuli according to the experiment by H.-C. Grassmann 21 Reconstruction of a hypocaust system with testudo above the furnace, in connection to the alveus 22 Schematic representation of the location of chimney flues in connection to the hypocaust 23 Different types of wall heating 24 Plan of the baths under the Piazza dei Cinquecento in Rome (Italy), and the excavation during the construction of modern Termini train station 98 Drawing of the caldarium of the baths under the Piazza dei Cinquecento in Rome 99 Plan of the Baths under the Lateran Baptistery in Rome and plan of the Lateran Baptistery, with small pool to its north 100 Plan of the Baths of the Invidious in Ostia (Italy) 114 Baths of the Invidious in Ostia (Italy), small pool in the frigidarium seen from the west 114 Plan of the Baths of the Swimmer in Ostia (Italy) 115 Baths of the Swimmer in Ostia, small pool of the frigidarium seen from the east 116 Plan of the Baths of the Six Columns in Ostia (Italy) with the small heat lock 116 Baths of the Six Columns in Ostia (Italy), heat lock seen from the north 117 Plan of the Baths of Porta Marina in Ostia (Italy) 118 Plan of the Baths of the Coachmen in Ostia (Italy) 119 Baths of the Coachmen in Ostia (Italy), small basin in the caldarium seen from the south 120 Plan of the Baths of Neptune in Ostia (Italy), with the 4th c. walls highlighted in red 121 Drawings of the late antique mosaics in the Baths of Neptune in Ostia (Italy) 122 Plan of the Baths of Mithras in Ostia (Italy) 123 Baths of Mithras in Ostia (Italy), alleged church in the northern part of the baths seen from the south. Stele with ChiRho sign on foreground 124 Baths of the Trinacria in Ostia (Italy), small pool in the frigidarium seen from the north. Opus listatum blocking a pre-existing niche or door 124 Plan of the Forum Baths in Ostia (Italy) 125 Forum Baths in Ostia (Italy), late antique Corinthian columns from the fourth heated room seen from the south 126 Forum Baths in Ostia (Italy), recess with tubuli in the south wall of the exittepidarium 126

xiv 32 Plan of the Baths of the Lighthouse in Ostia (Italy) 127 33 Plan of the Baths of the Philosopher in Ostia (Italy) 127 34 Baths of the Philosopher in Ostia (Italy), small pool at the beginning of the heated section seen from the east 128 35 Plan of the baths on the Isola Sacra (‘Terme di Matidia’) near Ostia 129 36 Plan of the Baths of the Capitolium in Cuicul (Algeria) 142 37 Plan of the Baths of the Filadelfes in Thamugadi (Algeria) 146 38 Plan of the Small East Baths in Thamugadi (Algeria) 147 39 Plan of the Baths of the Capitolium in Thamugadi (Algeria) 147 40 Plan of the Baths of the Donatist Cathedral in Thamugadi (Algeria) 150 41 Plan of the Baths of the Byzantine Fortress in Thamugadi (Algeria) 152 42 Plan of the Antonine Baths in Carthage (Tunisia) 153 43 Plan of the Large Baths in Sufetula (Tunisia) 161 44 Plan of Baths nr. 15 in Sufetula (Tunisia) 162 45 Plan of the Baths of House T in Ptolemais (Libya) 167 46 Plan of the Extra-muros Baths in Volubilis (Morocco) 175 47 Plan of the palace at Khirbet al-Mafjar (Palestine) with location of the baths 177 48 Baths of Khirbet al-Mafjar (Palestine), plan of the hypocaust system, furnace with reservoir 177 49 Plan of the Baths of al-Bara (Syria) in the Roman period 178 50 Plan of the Baths of al-Bara (Syria) in the Islamic period 178 51 Baths of al-Bara (Syria), cross-section of the furnace with superimposed boiler during the Islamic phase 179 52 Plan of the Baths of Halabiyya-Zenobia (Syria) 179 53 Plan of the Baths of Qasr al-Hayr East (Syria) 180 54 Plan of the Umayyad baths of Qusair Amra (Jordan) 181 55 Plan of the Umayyad baths attached to the Palace of the Caliph in Amman (Jordan) 181 56 Plan of the baths in Brad (Syria) 187 57 Plan of the baths in Serdjilla (Syria) 188 58 Plan of the North Baths in Volubilis (Morocco) 191 59 Plan of the thermal baths at Jebel Oust (Tunisia) with pools of different temperatures 193 60 Central Baths in Pompeii (Italy), detail of the small basin in the caldarium seen from the northwest 194 61 Baths along the Via Marciana in Ostia (Italy), seen from the west 270 62 Satellite imagery of Ostia (Italy), showing a street heading towards the Baths of Perseus. Also depicted are the Baths of Musiciolus, the Baths of Edificio b, the Baths of Porta Marina, the Small Baths along the Via Severiana and the Baths along the Via Marciana 271 63 Byzantine Baths in Ostia (Italy), detail of the mosaic of the courtyard 273 64 Baths on the via della Foce in Ostia (Italy), oval alveus of the caldarium seen from the southeast 274 65 Baths on the via della Foce, Ostia (Italy), detail of large tubulus found in the caldarium 276 66 Archaeological site of the Villa del Casale at Piazza Armerina (Sicily), with the Villa Baths and the Small South Baths 277 67 Small pool in a heated room of the Baths of the Lantern in Portus (Italy) 278 68 Reconstruction drawing in perspective view from the northeast of the Baths of Diocletian in Rome (Italy) 284 69 Frigidarium of the Baths of Diocletian in Rome (Italy), now turned into the Church of Santa Maria degli Angelli 285

Illustrations

Illustrations

70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108

Baths of Elagabalus in Rome (Italy), small east alveus of the caldarium seen from the west 288 Theatre of Volaterrae (Italy) with the late antique Vallebuona Baths in the peristyle square 295 Plan of the so-called ‘Temple of Minerva Medica’ in Rome (Italy), perhaps a bathhouse or, more likely, a heated reception hall 298 Plan of the alleged Baths of San Tommaso in Ticinum (Italy) 299 Archaeological site of the Villa of Theodoric in Galeata (Italy) 300 Plan of the Baths of Theodoric in Galeata (Italy) 300 Plan of the Villa Baths in Piazza Armerina (Sicily) 302 Plan of the Baths of Curinga (Italy) 304 Plan of the Baths of Herdonia (Italy) 305 Plan of the Baths of Misterbianco (Sicily) 306 Plan of the Baths of San Gaetano di Vada (Italy) 308 Plan of the House of the Coiedii in Suasa (Italy) 308 Plan of the Baths at Castel Porziano (Italy) 309 Plan of the House of Europe in Cuicul (Algeria) with semi-public bathhouse 317 Plan of the House of the Donkey in Cuicul (Algeria) with semi-public bathhouse 319 Plan of the House of Castorius in Cuicul (Algeria), with adjacent semi-public baths and private baths 320 Plan of phase 1 of the balnea privata in Carthage (Tunisia) 321 Sanctuary of Asclepius in Lambaesis (Algeria). The Small South Baths and the Sanctuary Baths can be identified 328 Plan of the Baths of Pompeianus in Oued Athmenia (Algeria) depicting the mosaic floors 330 Plan of the House North of the Capitolium in Thamugadi (Algeria) with adjacent bathhouse 338 Baths of the Laberii in Uthina (Tunisia), drawing of the mosaic of the frigidarium depicting Orpheus enchanting wild animals 344 Plan of the Baths of Sidi Ghrib (Tunisia) 346 Baths of Sidi Ghrib (Tunisia), different phases of use 347 Plan of the Baths of Baptistery in Cuicul (Algeria) 350 Plan of the ‘Church site’ in Tebessa Khalia (Algeria) with small baths (in box) 351 Plan of the Baths of Tebessa Khalia (Algeria) 352 Plan of the Basilica of Tipasa (Algeria) with adjacent baptistery and baths 353 Plan of the Baths of the Basilica in Tipasa (Algeria) 354 Plan of the Byzantine fortress in Thamugadi (Algeria) with location of the baths 355 Plan of the North Baths in Banasa (Morocco) 357 Plan of the Small West Baths in Banasa (Morocco) 357 Plan of the Baths of the Frescos in Banasa (Morocco) 358 Plan of Hadrian’s Baths in Leptis Magna (Libya) 361 Plan of the Hunting Baths in Leptis Magna (Libya) 362 Plan of the so-called ‘Unfinished Baths’ in Leptis Magna (Libya) with hypothetical reconstruction of the intended layout by Yvon Thébert 363 Plan of the Baths of the Schola in Leptis Magna (Libya) 364 Baths of the Schola in Leptis Magna (Libya), southern small apsidal pool in the frigidarium seen from the southwest 364 Plan of the church in the frigidarium of the West Baths in Mactaris (Tunisia) 365 Plan of the Large Baths in Madauros (Algeria) 366

xv

xvi

Illustrations

109 Plan of the church in the frigidarium of the Small Baths of Madauros (Algeria) 367 110 Plan of the Baths of En-Nagazza (Libya) 368 111 Plan of the Baths of the Theatre in Sabratha (Libya) 371 112 Plan of the Large Baths in Thenae (Tunisia) 372 113 Plan of the Summer Baths in Thuburbo Maius (Tunisia) with late antique reorganisation 374 114 Plan of the Winter Baths in Thuburbo Maius (Tunisia) 375 115 Plan of the Baths of the Labyrinth in Thuburbo Maius (Tunisia) 376 116 Plan of the Baths of the Capitolium in Thuburbo Maius (Tunisia), with its late antique reorganisation 376 117 Plan of the baths in the House of Venus in Volubilis (Morocco) 379 118 Plan of the Baths of the House of the Works of Hercules in Volubilis (Morocco) 379 119 Plan of the Baths of the Western Quarter in Volubilis (Morocco) 380 120 Plan of the Baths of the House of the Sundial in Volubilis (Morocco) 380 121 Plan of the Late Antique and Byzantine Baths in Apollonia (Libya) 382 122 Plan of the Baths of Trajan in Cyrene (Libya), with the insertion of the Byzantine Baths 384 123 Plan of the Fortress Baths in Taucheira (Libya) 393 124 Kom al-Dikka Baths in Alexandria (Egypt), cross-section and plan of the ‘pit-hypocaust’ 399 125 Plan of the Kom al-Dikka Baths in Alexandria (Egypt), with pools of different temperatures 400 126 Kom al-Dikka Baths in Alexandria (Egypt), cross-section of a possible type of testudo in the caldarium 401 127 North Baths of Karanis (Egypt), cross-section (west-east) of the caldarium and praefurnium 408 128 Drawing of the wall heating in the North Baths of Karanis (Egypt) 408 129 Byzantine Baths of Marea (Egypt), drawings of the fragments of claustra found inside the baths 413 130 Plan of the chapel and baths in the monastery of Bawit (Egypt) 423 131 Plan of the late antique fortress of Abu Sha’ar (Egypt) with a bathhouse outside its walls 424 132 Plan of the Fortress Baths of Abu Sha’ar (Egypt) 425 133 Plan of the late antique fortress at Nag el-Hagar (Egypt) 425 134 Plan of the Fortress Baths of Nag el-Hagar (Egypt) 426 135 Plan of the Baths and cistern (right) at Mampsis (Israel) 438 136 Baths in Rehovot (Israel), drawing of the roof of the baths with the system of a cistern, gullies, windows and orifices 442 137 Plan of the possible baths in Nahf (Israel) 448 138 Plan of the possible baths in Qalandia (Palestine) 449 139 Plan of the Mount Scopus monastery in Jerusalem (Israel / Palestine) with the location of the bathhouse 450 140 Plan of the Fortress Bath of Yotvata (Israel) 453 Maps 1 2 3

Late Antique inscriptions mentioning baths and bathing in the Italian Peninsula 7 4th c. inscriptions mentioning baths or bathing found in North Africa 81 5th and 6th c. inscriptions mentioning baths or bathing found in North Africa 84

Illustrations

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

Location of the most important buildings and public areas in Rome (Imperial Period and Late Antiquity) 93 Location of the public baths in Rome 96 The number of balnea and thermae in each region of Rome according to the Curiosum / Notitia 105 General plan of Ostia with the location of baths and main roads 111 Public baths in Ostia in relation to the main aqueducts, public spaces and new Christian buildings at the end of the 4th c. 130 Public baths in Ostia in relation to the main aqueducts, public spaces and Christian buildings at the end of the 5th c. 131 Regio I in Ostia during the 4th c., with location of baths and other buildings that were newly built or still in use 132 Regio I in Ostia during the 5th c., with location of baths and other buildings that were newly built or still in use 133 Regio II in Ostia during the 4th c., with location of baths and other buildings that were newly built or still in use 133 Regio II in Ostia during the 5th c., with location of baths and other buildings that were newly built or still in use 134 Regio III in Ostia during the 4th c., with location of baths and other buildings that were newly built or still in use 134 Regio III in Ostia during the 5th c., with location of baths and other buildings that were newly built or still in use 135 Regio IV in Ostia during the 4th c., with location of baths and other buildings that were newly built or still in use 136 Regio IV in Ostia during the 5th c., with location of baths and other buildings that were newly built or still in use 137 Regio V in Ostia during the 4th c., with location of baths and other buildings that were newly built or still in use 138 Regio V in Ostia during the 5th c., with location of baths and other buildings that were newly built or still in use 139 Cuicul (Algeria) in the 4th c., with the location of public baths and the most important buildings 141 Cuicul (Algeria) in the 5th c., with the location of public baths and the most important buildings 144 Thamugadi (Algeria) at the end of the 4th c., with the location of public baths and the most important buildings 149 Thamugadi (Algeria) at the end of the 6th c., with the location of public baths and the most important buildings 151 Carthage (Tunisia) at the end of the 4th c., with the location of public baths and the most important buildings and public areas 153 Carthage (Tunisia) at the end of the 5th c., with the location of public baths and the most important buildings and public areas 156 Carthage (Tunisia) at the end of the 6th c., with the location of public baths and the most important buildings and public areas 157 Sufetula (Tunisia) at the end of the 4th c., with the location of public baths and the most important buildings 159 Sufetula (Tunisia) at the end of the 5th c., with the location of public baths and the most important buildings 163 Sufetula (Tunisia) at the end of the 6th c., with the location of public baths and the most important buildings 164 Ptolemais (Libya), plan of the 4th c. city with the location of public baths and the most important buildings 166

xvii

xviii

Illustrations

26 Ptolemais (Libya), plan of the 6th c. city with the location of public baths and the most important buildings 169 27 Scythopolis (Israel), plan of the 4th c. city with the location of public baths and the most important buildings 172 28 Scythopolis (Israel), plan of the 5th c. city with the location of public baths and the most important buildings 173 29 Scythopolis (Israel), plan of the 6th c. city with the location of public baths and the most important buildings 174 30 Umayyad baths in the Middle and Near East 176 31 Public baths in late antique Italy 211 32 Public baths in late antique North Africa 212 33 Late antique public baths in Cyrenaica and Egypt 214 34 Late antique public baths in Palestina 215 35 Ravenna in the 6th c., with location of baths and the most important public areas and buildings 281 36 Bulla Regia (Tunisia), location of public baths and the most important buildings 313 37 Sabratha (Libya), plan of the city with the location of public baths and the most important buildings 370 38 Thuburbo Maius (Tunisia), plan of the city with the location of public baths and the most important buildings 373 39 Volubilis (Morocco), plan of the city with location of public baths 378 40 Alexandria (Egypt), hypothetical reconstruction of the late antique city grid with the location of public baths 397 Graphs 1 2 3 4 5 6 7 8 9 10 11 12

Inscriptions mentioning baths or bathing from Italy and North Africa between the 4th and 7th c. AD 80 Late antique papyri and ostraca mentioning baths and bathing found in Egypt 87 Public baths in the different regions during the 4th c., based on the data from Appendix 3 (possible baths not included) 210 Public baths still in use in late antique Italy, based on the data from Appendix 3A 212 Public baths still in use in late antique North Africa, based on the data from Appendix 3B 213 Public baths in late antique Cyrenaica, based on the data from Appendix 3C 213 Public baths in late antique Egypt, based on the data from Appendix 3D 214 Public baths still in use in late antique Palestina, based on the data from Appendix 3E 216 Public baths in the different regions during the 5th c., based on the data from Appendix 3 (possible baths not included) 218 Public baths in the different regions during the 6th c., based on the data from Appendix 3 (possible baths not included) 219 Public baths in the different regions during the 7th c., based on the data from Appendix 3 (possible baths not included) 220 Total number of all late antique baths in use (new and continued use) per region and per century (possible baths and baths with an unknown construction date not included) 221

Illustrations

13

Newly built public baths in Egypt during the Greek-Hellenistic period, High Empire and Late Antiquity. The numbers derive from the most recent catalogue of public baths in Egypt (Redon (2017a)) 222

Table 1 Buildings in each of the regiones of Rome as presented by the Curiosum and the Notitia Regionum 102

xix

Abbreviations Abbreviations are those used by the American Journal of Archaeology for periodicals, and the Oxford Classical Dictionary (3rd edn. 1999), Jones A. H. M. Later Roman Empire (Oxford 1964), Lampe G. W. H. A Patristic Greek Lexicon (Oxford 1961) for ancient sources. Other abbreviations used are listed below: AE

L’année épigraphique: revue des publications épigraphiques relatives à l’antiquité romaine AL Anthologia Latina BGU Aegyptische Urkunden aus den Königlichen (later Staatlichen) Museen zu Berlin, Griechische Urkunden BL Berichtigungsliste der Griechischen Papyrusurkunden aus Ägypten BSR British School at Rome Corpus Papyrorum Judaicarum C. Pap. Jud. CIIP Corpus Inscriptionum Iudaeae / Palaestinae CIL Corpus Inscriptionum Latinarum Carmina Latina Epigraphica CLE CPG Clavis Patrum Graecorum, ed. M. Geerard et al. (Turnhout 1939-1988) Corpus Papyrorum Raineri CPR Enquète sur l’épigraphie chrétienne d’Afrique EchrAfr Solin, H. (ed.) 1993. Le iscrizioni antiche di Trebula, Caiatia e Cubulteria IATrebula (Caserta 1993) Inscriptiones christianae urbis Romae ICUR Inscriptiones Graecae IG Inscriptiones Graecae Urbis Romae IGUR Inscriptions latines d’Afrique ILAfr Inscriptions latines d’Algérie ILAlg Inscriptiones Latinae Christianae veteres ILCV Inscriptiones Latinae Selectae ILS Inscriptions Latines de la Tunisie ILTun Inscriptiones Italiae InscrIt Inscriptions of Roman Tripolitania IRT Les ostraca grecs de Douch O. Douch O. Mich. Greek Ostraca in the University of Michigan Collection The Amherst Papyri, Being an Account of the Greek Papyri in the P. Amh. Collection of the Right Hon. Lord Amherst of Hackney, F.S.A. at Didlington Hall, Norfolk P. Apoll. Papyrus grecs d’Apollônos Anô Papyrological Texts in Honor of Roger S. Bagnall P. Bagnall P. Brem. Die Bremer Papyri P. Cair. Masp. Papyrus grecs d’époque byzantine, Catalogue général des antiquités égyptiennes du Musée du Caire P. Coll. Youtie Collectanea Papyrologica: Texts Published in Honor of H. C. Youtie P. Flor Papiri greco-egizii, Papiri Fiorentini Papiri dell’Università di Genova P. Genova Briefe des Apollonios-Archives aus der Sammlung Papyri Gissenses P. Giss. P. Iand. Papyri Iandanae P. Jena I, Jenäer Papyrus-Urkunden; II. late antique Greek Papyri in the Collection of the Friedrich-Schiller-Universität Jena P. Köln Kölner Papyri P. Lond. Greek Papyri in the British Museum

xxi

Abbreviations

P. Münch. P. Oslo P. Oxy. P. Ross. Georg. P. Ryl. P. Wisc. P. Würzb. Pap. Agon. PG PL PLRE PSI SB SEG Sel. Pap. Stud. Pal. SupIt W. Chr. ZPE

Die Papyri der Bayerischen Staatsbibliothek München Papyri Osloenses Oxyrhynchus Papyri Papyri russischer und georgischer Sammlungen Catalogue of the Greek and Latin Papyri in the John Rylands Library Wisconsin Papyri Mitteilungen aus der Würzburger Papyrussammlung Zehn agonistische Papyri Patrologia Graeca, Cursus Completus, ed. J.-P. Migne (Paris 1857–1866) Patrologia Latina, ed. J. P. Migne (Paris 1844–1855) Prosopography of the Later Roman Empire Papiri greci e latini Sammelbuch griechischer Urkunden aus Aegypten Supplementum Epigraphicum Graecum Select Papyri (The Loeb Classical Library) Studien zur Palaeographie und Papyruskunde Supplementa Italica Grundzüge und Chrestomathie der Papyruskunde Zeitschrift für Papyrologie und Epigraphik

Key to to the theGeneral GeneralPlans Plans Location of Inscriptions Regional capital 4th c. inscriptions 5th c. inscriptions 6th c. inscriptions

Location of Cities / Sites Regional capital

6th c. bath (new)

4th c. bath (new)

Existing bath still in use in the 6th c.

Existing bath still in use in the 4th c.

Roman site with abandoned bath(s)

5th c. bath (new)

Monastery baths

Existing bath still in use in the 5th c.

Possible late antique baths

Plans of Cities Public baths

Private / imperial building (Rome)

Public building

Private bath

Reservoir / cistern / latrine

Abandoned building

Temple / sanctuary

Forum

Church

Possibly abandoned building

Area of economic activity

Evidence of late antique occupation

Houses / streets

Well

City walls

Christian burials

Aqueduct

Key to the General Plans

Catalogue Catalogue Wall Hypothetical wall Door blocked at a later date Pre-existing wall Wall added at a later date Water supply (piping) Water disposal (sewage channel) Hypocaust with wall heating Hypocaust with possible wall heating Possible hypocaust with possible wall heating Hypocaust without wall heating Possible hypocaust without wall heating Main entrance Furnace Possible furnace Water boiler Possible water boiler Stairs Pool Natural rock Well Reservoir Bench

xxiii

Introduction Kein Volk des Altertums oder der Neuzeit badete mit solcher Leidenschaft wie die Römer; kein Volk hat so Grosses geschaffen und gebaut, um diese Leidenschaft zu befriedigen. No other people, ancient or modern, bathed with such passion as the Romans; no other people has invented and built to that extent to satisfy this passion.1

∵ Why Baths Again? By the start of the 20th c., scholars had already labelled the public bathhouse as one of the most distinctive markers of Roman culture. The countless references to baths and bathing in ancient literature as well as archaeological remains of sumptuous bathhouses in the 4 corners of the empire were the undeniable proof of a Roman ‘culture of the body’. The romantic picture of Roman ancestors with a neat appearance and impeccable hygiene, that would not fail contemporary standards, was promoted in art and reconstructions.2 In the eyes of the Renaissance and Early Modern educated elite, Roman culture was indeed the worthy predecessor of modern Western European culture. In this line of thought, the fall of the Western Roman empire and the decline of Roman culture into the so-called Dark Ages necessarily entailed the decline of Roman bathing habits and the disappearance of the public bathhouse.3 The period after the High Empire was only a footnote in the history of Roman baths and bathing habits.4 However, archaeological excavations and a renewed interest in ‘late antique’ literature in the last three decades have transformed our knowledge and appreciation of the ‘later’ Roman world. On an archaeological level, the number of ‘late’ sites not only increased, but a better knowledge of the material culture also enabled a re-assigning of some buildings and features to this ‘late’ period. The topic of water management has also come to renewed attention. In the 21st c. with an ever-growing 1  Marcuse (1903) 19; translation by the author. 2  Di Capua (1941); Pazzini (1941) 11. 3  Already criticised by Thorndike (1928). 4  For a detailed overview of earlier research into Roman baths and the attention that was paid to late antique baths, see chapter 1.

© koninklijke brill nv, leiden, ��20 | doi:��.��63/9789004419421_002

global population and concerns about climate change, water is, more than ever, the planet’s most valuable resource. Issues of water supply, waste water disposal and water treatment for reuse are high on the agenda of governing bodies. Engineers are turning to history and archaeology to understand how past civilizations coped with crises in water management.5 A study of Roman public baths and bathing habits in Late Antiquity could provide us with an example of how an empire and its baths, “the most conspicuous expression of Roman hydraulic engineering,”6 responded to a period of major change. Furthermore, questions of public hygiene are also of current interest in a rapidly ‘urbanizing’ world.7 Even if the Romans, despite their bathing habits, did not have the same conception of public hygiene as we have today (see below, chapter 1), the continued use of public bathhouses in a period hit by severe plagues (the Antonine and Justinianic plague) may have prevented even worse catastrophes. It is remarkable that a structure that has been considered as ‘the most Roman’ of all public buildings has not been re-examined in the light of all these current developments. General Aims and Outline of the Research The general aim of this book is to get a better understanding of Roman-style public baths and bathing habits during the period that saw the transformation of classical Roman society into that of the Early Medieval period. However, this book does not aim to give an exhaustive overview of all late antique baths so far discovered. The main aim is rather to assemble a representative sample of the literary, epigraphic and archaeological evidence of late antique baths, in order to write a cultural history of bathing habits between AD 285 and AD 700. Furthermore, this study not only targets an audience that is specialized in public baths; it also aims to reach out to students, scholars and field archaeologists who have an interest in late antique baths and bathing habits. To detect possible changes, it is clear that we should approach the baths and bathing habits of Late Antiquity in the longue durée of Mediterranean bathing habits. This ‘long-term approach’ has also been the hallmark of the Balnéorient project, focusing on the development of baths and bathing habits in the Near and Middle 5  Most recently Sedlak (2014). 6  Aryamontri (2009) 330–31. 7  Boëtsch and Tamarozzi (2014).

2 East from the earliest settlements to modern times.8 Furthermore, the subject of public baths and bathing in Late Antiquity is so vast that some restriction concerning the dataset, geographical and chronological framework had to be imposed. A study that does not only aim to investigate the architectural and technological evolution of the bathhouse as a building, but also possible changes in bathing habits, should include all possible sources—literature, inscriptions and archaeology (see below ‘Methodology and sources’). The focus of this study lies on public baths, i.e. baths that were accessible to everyone, whether they were owned by the city / state or by a private individual.9 Private baths on the other hand, which were only accessible to the owner of a house or villa,10 will only be touched upon when they provide us with important parallels or relevant information about public bathing. Thermal baths—which relied on natural hot water and gasses and were mainly used for curative purposes11— will similarly be involved if relevant for the discussion about public bathing. As their location depended on the presence of natural hot springs and their design, architecture and technology were adapted to enable the best possible use of the natural spring,12 thermal baths were very different from ‘regular’ bathhouses. Several in-depth studies of this special category of baths have already shown the continued popularity of spas in Late Antiquity.13 Baths in fortresses and ecclesiastical structures will be discussed if there is reason to believe that they were publicly accessible. The choice to study several Roman regions instead of concentrating upon a single province might seem overconfident. However, by studying different regions, each with its own history and relations with Rome, we can obtain a better picture of the ‘connectivity’ that existed in the Mediterranean during this period, rather than resorting to ‘cultural-container thinking’ as so eloquently phrased by Versluys,14 even if this means that such an overview might not be as exhaustive as region-specific studies. Each region studied in this book represents a certain ‘tradition’ within the Roman empire: the Italian Peninsula as the Roman ‘heartland’, North Africa as prosperous provinces in the western Mediterranean

8  Charpentier and Denoix (2011). 9  Following Nielsen (1993a) 119. 10  Following de Haan (2010) 5. 11  Köhler (2006); Dvorjetski (2007) 120; Broise (2015). 12  Dvorjetski (2007) 120–23; Broise (2015). 13  Jouffroy (1992); Pettenò (1996); Dvorjetski (2007); Kreiner and Letzner (2012); Guérin-Beauvois (2015). 14  Versluys (2014) 11.

Introduction

tradition, and Cyrenaica, Egypt and Palestine as rich provinces in the eastern Mediterranean tradition. The choice for Late Antiquity has been inspired by the recent mass of new (archaeological) data that has come to light in recent years, as well as the possibilities for reinterpreting old data (for the specific chronological framework, see below). If we consider that buildings indeed form a sort of non-verbal language,15 we can try to discern if the changes in the buildings can tell us something about late antique society in general, especially as Roman-style bathhouses are identified as ‘markers’ for a Mediterranean connectivity. Without returning to the debate about Romanisation,16 we could indeed ask ourselves if the changes in baths and bathing habits, so deeply rooted in and connected to the antique world,17 can inform us on the transition to a new, different Mediterranean world. Geographical Framework This study will focus on the Roman-style public baths in the Italian Peninsula, North Africa in the modern geographical sense and the Palestine regions. As explained above, the choice for these three regions is inspired by the general aim of this book, namely to compare the evolution of Mediterranean bathing habits in the heartland, western and eastern provinces. The exact delimitation of each region and the choice for these specific regions merit some explanation. Italy, which includes the mainland peninsula and the islands of Sicily and Sardinia, was chosen for its central role within the western Mediterranean, even after the imperial administration was moved to Constantinople under Constantine. Rome still exerted an important cultural influence and enjoyed a somewhat mythical status among the cities of the empire. It is therefore interesting to investigate how Rome and the Roman ‘heartland’ evolved under the new socio-political realities of Late Antiquity. With Mediolanum (Milan) and Ravenna, the peninsula also hosted three administrative capitals of Late Antiquity. In contrast to the northern provinces and the Iberian Peninsula, Italy remained one administrative unit under imperial control until the late 5th c. The impact of the subsequent Gothic, ‘Byzantine’ and Lombard reigns in different parts of the peninsula allows 15  See among others Rapoport (1974); Rapoport (1976); Gieryn (2002); Smith and Bugni (2006). 16  For the most recent debate, see Archaeological dialogues 21.1, esp. Versluys (2014) and Woolf (2014) with anterior bibliography. 17  Blonski (2014a).

3

Introduction

us to assess how the new rulers dealt with baths and bathing habits. Furthermore, Italy was also a core region for the development of Christian doctrine, enabling an assessment of bathing habits in a new Christian context. The large cities are also well represented in the literary evidence about baths and bathing. As Anette Haug concluded about the Italian heartland: “Diese neue politische, aber auch wirtschaftliche, geistliche und geistige “Zentralität” in der Spätantike prädestiniert die Region für eine urbanistiche Analyse.” (“This new political, but also economic, religious and intellectual “centrality” in Late Antiquity predestines this region for an urbanistic analysis.”)18 North Africa was a region of unusual interest for Rome. The area comprises the northern part of modern-day Morocco (until the imaginary line Rabat-Meknes-Fez), the coastline of Algeria and its immediate hinterland, Tunisia and the coastline of modern Tripolitania and its immediate hinterland. The agrarian production of Africa fed the people of the Italian Peninsula and made the land-based elite extremely wealthy. During the 4th c., North Africa also enjoyed a continuous period of prosperity, with investment in public buildings just as in the Italian Peninsula. The superb state of preservation of its Roman remains gives archaeologists an exceptional insight into its built heritage and of its evolution over a long period of time. The impact of a ‘barbarian’ invasion on the built heritage, such as the Vandals in the early 5th c., can hence be investigated through the archaeological record instead of relying only on literature. Furthermore, the ‘restoration’ of this heritage under Byzantine rule can also be assessed. The history of the bathing habits in North Africa faced a third hurdle in the Islamic conquest. As the Islamic conquerors later settled on different lands, the archaeological remains of bathhouses in North Africa offer us a unique look onto the last phase of the building and possible transformations under Early Islamic rule. Even if Cyrenaica and Egypt have always been treated somewhat apart in the archaeology of the Roman empire, these regions form an important link between the western and eastern part of the (southern) Mediterranean. Cyrenaica consisted of the coastline of modern Cyrenaica and the Gebel-Ahkdar mountains around Cyrene, while Roman Aegyptus corresponded more or less to modern day Egypt, until the first cataract of the Nile at Aswan. Both regions offer us some exceptionally well preserved sites, which had their roots in the Greek-Hellenistic building tradition. Their historical trajectory further diverged after they were spared ‘barbarian’ incursion. Cyrenaica was not conquered by the 18  Haug (2003) 1; translation by the author.

Vandals and remained under imperial control in the 5th c., while Egypt was heavily protected from eastern threats because of its crucial role in the food supply of first Rome and later Constantinople. In Egypt, papyrological evidence also gives us an exceptional look at ordinary problems and daily concerns regarding baths and bathing that would never have made it to books or inscriptions. Its long bathing tradition (Hellenistic public baths) and the strong Christian presence in Egypt, not least in the form of ascetism and pilgrimage sites, gives the evolution of baths and bathing habits in this region additional dimensions. The Islamic conquest started slightly earlier than in Roman Africa and was of a more permanent character, offering an interesting comparison between the baths and bathing traditions in both the western and eastern part of North Africa under Early Islamic rule. The Palestine regions form the definite transition to the eastern part of the empire. In this book, the area under investigation corresponds to late antique Palestina Prima, Palestina Secunda and Palestina Tertia. This roughly falls within the borders of the modern states of Israel, Palestine and the western part of Jordan. The Palestine region prospered between the early 4th and late 6th c., benefitting from trade with Egypt and the other eastern provinces. The region’s transitional role between the Mediterranean world and the East, as well as the extraordinary quantity and quality of the archaeological information, make Palestina an appealing comparison for other regions. Furthermore, it is interesting to assess the evolution of baths and bathing in a region that is known as the cradle of Judaism and Christianity. The preservation of Roman sites in the Italian, North African and Palestine regions, combined with their more or less stable political and economic context between the 4th and late 6th c., enable an interesting comparison of baths and bathing habits between the Italian heartland, western and eastern parts of the Mediterranean. Chronological Framework The period that will be covered in this study lies between AD 285 and AD 700. The choice of the start date seems straightforward. It is the year in which the so-called ‘crisis of the 3rd c.’ ended and Diocletian became the new emperor of the Roman empire. The introduction of the Tetrarchy (the ‘rule of four’) and the subsequent reorganisation of imperial administration and the economy had an empire-wide impact. By including the reign of Diocletian and the Tetrarchy, rather than starting with ‘the first Christian emperor Constantine’, we thus

4 include the first baths that were constructed under these changed socio-political and economic circumstances.19 The end date seems more random, as AD 700 does not mark a specific historical event. The date was chosen with the Islamic conquest of North Africa in mind. In AD 698, Carthage fell into Muslim hands and by AD 708, Tangier had been taken. With the end of Roman rule, a new historical chapter commenced. By including the second half of the 7th c., we could (theoretically) observe how the Roman-style baths in North Africa and the Middle East ‘survived’ the first decades of Muslim rule. The further development of bathing habits under Early Islam will only be presented as an ‘epilogue’, as an exhaustive study of transition to the hammams would merit a separate publication. In the Italian Peninsula, the Roman societal model had already disappeared by the late 6th c. The end of Theoderic’s rule, the last of some sort of administrative unity, is often chosen as end date of late antique Italy. By pushing back the chronological limit to the end of the 7th c., we can include the tumultuous period of the Gothic-Byzantine Wars, Lombard-Byzantine Wars and their effects on public baths and bathing habits. In this way, we can include an outline for baths and bathing in the Early Medieval period. The geographic and chronological framework of this study enables us to investigate the evolution of the baths and bathing habits in response to political changes (administrative reforms under Diocletian and Constantine, Vandal, Gothic, Byzantine, Lombard and Islamic rule), economic changes (decline of certain cities, administrative abandonment of certain regions, shifting trade routes) and cultural changes (rise of Christianity, Christian doctrinal disputes, dwindling classical cults, Jewish conservatism, rise of the Islam). Methodology and Sources Baths and Bathing Approached from Different Angles As the title of this book implies, this research will not only focus on the bathhouse as a building, but also on bathing habits. Evidently, the actual architectural structure will be a very important part of the evidence. Just as in most of the seminal works on Roman baths, the architecture, technology and decoration of the baths 19  Several baths are indeed dated ‘at the turn of the 3rd and 4th centuries. By including the period between AD 285 and AD 305, we can include these baths. However, for the inscriptions and the papyri, the year AD 300 has been chosen as a start date to simplify search queries in the databases.

Introduction

will be investigated.20 The use of different rooms in the bathhouse, which is closely linked to the three aspects of baths mentioned above, has also been discussed in most of the general works on Roman baths.21 The perception of baths and bathing by the people that built and used them is more difficult to investigate. For this aspect, the written evidence by users of these baths takes a suitably prominent position.22 If we want to understand the evolution of baths and bathing habits in Late Antiquity, it is not sufficient to restrict the research to newly built baths. Several establishments of the High Empire were still in use during Late Antiquity, sometimes with important changes to their original plan. This study will therefore try to detect which pre-existing baths were still in use, if these baths were restored and if so, if these restorations brought about changes to the original design. The restorations and possible transformations can then be compared with the newly constructed baths in order to detect similar trends in the 5 aspects of a bathhouse discussed above. As the public bathhouses are often described as social hubs within Roman communities, this research will also focus on the position of the bath within its larger environment. For public baths in an urban context, the location of the baths within the city fabric is important.23 The relation of the baths to the city’s water supply network and sewer system, other social hubs (e.g. fora, basilicae, theatres, amphitheatres, nymphaea, public latrines, churches) and road networks will be examined where possible. For the baths in a non-urban context, the relation to water sources, possible traffic arteries and surrounding structures (villas or even small settlements) can turn out to be important factors for the location of the baths. Evidence of Baths and Bathing For the best possible insight into these three different aspects of the research into baths and bathing—the bathhouse itself, its use and perception, and its position within the larger environment—all types of evidence should be included. It is important to stress that this book does not present an exhaustive list of all available texts, inscriptions and archaeological sites of late antique baths, but only provides a representative sample 20  Krencker et al. (1929); Brödner (1983); Heinz (1983); Yegül (1992); Nielsen (1993a). 21  See above n20, to which should be added Fagan (1999a) and Blonski (2014a). 22  Fagan (1999a). 23  Wilkes (1999) 23.

Introduction

that permits study of the evolution of baths and bathing habits between AD 285 and AD 700. The abundant literary output of late antique writers informs us on different aspects of baths and bathing. The writings of early Christian authors are especially important to understand the position of the Church towards the Romans’ favourite pastime. Furthermore, the works of Greek-speaking and Syriac-speaking writers will be included in this study, as these authors often travelled and wrote about the different parts of the empire (see ‘Introduction’ of chapter 2). Besides the passages by the ‘usual suspects’ such as Jerome, Eusebius and Augustine, other bath-related texts by less well-known authors such as Vegetius, Commodianus, John Cassian and John of Nikiu prove to be very informative and will be included. There will also be special attention for medical authors such as Theodorus Priscianus, Cassius Felix and Alexander of Tralles, who frequently mention the use of baths in their prescriptions and cures. The second category of written sources, the juridical texts, is often quoted parsimoniously in modern research. A thorough examination of these laws, however, informs us about a wide variety of aspects of baths and bathing habits, especially about practical issues such as water supply, building permits and inheritance of privately owned baths. Canonical laws and Church rules give a more general counterbalance to the personal opinions of the Early Church fathers. Inscriptions regarding late antique baths and bathing merit an independent study. The corpus of inscriptions that was assembled for this work relies on earlier studies, supplemented by bath-specific queries on online databases and going through the last volumes of the most important epigraphic journals (for more detail, see ‘Introduction’ of chapter 2). The inscriptions that were included all refer directly to bathhouses or bathing. Inscriptions that were found inside baths but do not have a bath-related content were excluded. When the text of the inscription is entirely preserved, it can give us important information on construction or restoration dates, the patron(s) paying for the work and the type of work that was carried out. In addition, the epigraphic data can give us the name of the baths, the costs and duration of the work and other information. The number of inscriptions pertaining to public buildings has often been used to chart the evolution of building activities in the Roman empire. Even if the link between inscription and building activity is not as straightforward as statistics would have us believe, it is nevertheless interesting to compare the archaeological and literary data about late antique baths to the epigraphic corpus, especially on a region-by-region basis.

5 The papyrological evidence is a special type of data. The papyri that were relevant for this study have been collected through bath-related queries in online databases and through earlier publications. The geographical spread of this corpus is limited to modern-day Egypt and pertains almost exclusively to the baths and bathing habits in this same region. As Roman Egypt has often been treated as a somewhat separate case in modern research, use of papyri for research on Roman baths was frequently marginal.24 However, the papyri are an invaluable source of information on the baths of Egypt, in addition to the archaeological and literary evidence. As the papyri contained personal or administrative messages, not intended for widespread publication or immortalization in stone, the information we gained from these excerpts shows us a different side of the baths: from the payment of 50 panes of glass for restoration work to the complaint of citizens over high entrance fees. It is also worth investigating if the more general information on baths and bathing habits provided by the papyri corresponds with the evidence from the other regions that are discussed in this study. Egypt might not have been such an exception as usually assumed. The present study is the result of a bibliographical survey, complemented with some on-site observations in a limited number of cases.25 No new archaeological fieldwork was conducted. The description of the bathhouses in this book is the result of a critical reading of primary excavation reports (if available) and of the detailed analysis of the plan provided by this primary publication. The secondary literature is only included if it adds new elements to the primary source. After the discussion, a gazetteer of late antique Roman-style baths is presented. For each region, not only the newly built late antique baths, but also the High Imperial baths with a late antique phase have been included. The newly built late antique bathhouses are ‘valued’ according to the type of evidence that was used by the excavators for dating. The baths with an A-rating have been dated by scientific analyses (radiocarbon dating, dendrochronology, etc.), contextual finds or inscriptions. These baths can be attributed with certainty to Late Antiquity. Baths with a B-rating only have a terminus post or ante quem, but can still be dated fairly securely to Late Antiquity. Baths with a C-rating could only be dated on the basis of general construction tech24  Yet see Nielsen (1993a) 207–208. 25  If additional observations were made on site by the author, these are indicated in the relevant parts of the texts as footnotes.

6 niques, decorative schemes, use of spolia or architectural styles. The D-rated baths can only be attributed to Late Antiquity on the basis of regional trends or site-specific elements. These last two categories most likely belong to the late antique period, although we often lack the hard archaeological evidence to confirm this. It is sometimes difficult to discern restorations and new construction, and often the construction date can only be narrowed down to a century. For a building to be interpreted as a bathhouse, the archaeological remains should at least reveal a hypocaust and a pool (with steps for access), or point incontestably to a bathing function (e.g. a mosaic with a text welcoming bathers). The archaeological remains that have been firmly dated to Late Antiquity, but that display only certain features of baths, such as basins (without steps for access) or a hypocaust, or for which only the presence of specific building materials such as tubuli, hydraulic mortar or pillar tiles can be ascertained, are categorized under ‘possible late antique baths’. Only future research may determine whether these buildings were actually baths or not. Besides the description of late antique public baths in the gazetteer, a number of urban case-studies will be discussed. These have been selected on the basis of the number of newly constructed late antique public baths, the state of preservation of the archaeological remains (both of the baths and the urban fabric) and of the available published archaeological data. This resulted in a list of 8 cities, spread across the different regions, which will be discussed in more detail: – Rome and Ostia for the Italian Peninsula – Cuicul, Thamugadi, Carthage and Sufetula for North Africa – Ptolemais for Cyrenaica – Scythopolis for Palestina For each city, the construction of late antique baths and the restoration of pre-existing baths will be framed in the historical and urban framework. Hence the relationship to the historical events and of the evolution of the urban fabric can be assessed, especially in relation to water management and other public buildings and spaces. The Structure of the Book The general structure of this book has been shaped by a logical deconstruction of the main research question, also allowing the reader to consult the body of evidence by source (written or archaeological) as well as by region. The detailed structure and research methodology of each part is explained at the start of each individual

Introduction

section. First, a short introduction to Roman baths and bathing habits will be presented, with a special focus on modern research into late antique baths (chapter 1). In the following two chapters, the evidence for late antique baths is discussed. The written evidence (chapter 2) includes ancient literary texts, the juridical corpora, inscriptions and papyri that can inform us not only about the actual buildings, but also about the people involved in their construction and use. A brief overview of the earliest literary attestations of hammams will also be presented. The archaeological evidence (chapter 3) from excavations and the standing remains of the bathhouses can inform us about the architecture, technology and decoration of the bath building and will form the main body of evidence of this study. The different sites, both new construction and continued use, are discussed in the Gazetteer. For 8 case studies, the baths will be discussed in their urban framework. To conclude, there is a concise overview of the archaeological evidence of early hammams. Finally, the evidence from the previous two chapters will be combined to answer general research questions about the evolution of late antique baths (chapter 4). In order to frame late antique baths in the longue durée of Mediterranean bathing habits, this evidence will also be compared to the evolution during High Empire and, for North Africa and the Levant, compared to early Islamic period. Notes on Terminology Geographical Terminology and Place Names In this study, the Latin terminology and delimitation of the different regions are preferred to modern delimitations. Hence North Africa corresponds to the coastal regions and the hinterland of modern Morocco, Algeria, Tunisia and Tripolitania in Libya, while Cyrenaica and Egypt are discussed separately. The term Palestina is preferred to modern Israel / Palestine because it encompasses also the western border region of Jordan. The collective term ‘the Middle East’ points to the regions now made up by Israel and the Palestine regions, Jordan, Lebanon, the Turkish region around Antakya and western Syria. The Near East, by contrast, points to eastern Turkey, eastern Syria, Iraq and Iran. For the smaller subdivision of each region, the antique provinces are followed. For the Italian Peninsula, this means a division according to the Augustan regiones (regio I to XI, see map 1). For North Africa, the provinces as shaped by the late antique reorganisation by Diocletian are used. The place names are referred to in their Latin spelling, always followed by the equivalent modern city and the

Introduction

map 1

Late Antique inscriptions mentioning baths and bathing in the Italian Peninsula. The numbers refer to the EI-numbers of Appendix 1, not shown: EI-24, Cornus on Sardinia. plan by the author

7

8 modern country (in brackets). As such, the reader will find Cuicul (Djemila, Algeria). The transcription of modern Arabic place names into western script follows Brill’s New Pauly when possible, or the transcription by the author of the primary source in other cases, e.g. Oued Athmenia. If the ancient name is unknown, only the modern place name will be mentioned, followed by the modern country, e.g. Ouled Hafouz (Tunisia). However, some well-known ancient cities have Anglicized versions of the Latin or Greek name, such as Rome, Carthage, Jerusalem, Constantinople or Antioch, which will be preferred. The same goes for modern cities for which the name is the same as the Roman version, such as Ostia, Ravenna or Alexandria. Chronological Terminology The term ‘late antique’ is preferred over ‘Late Roman’ or ‘Later Roman’ as it expresses a certain independence from the previous periods, yet also stresses the unmistakably ‘classical’ character of the period. Furthermore, ‘late antique’ shifts the attention from the cultural (‘Roman’) sphere to a more neutral ‘chronological’ (‘antique’) indication, embracing all the new sociocultural developments of this period, such as the rise of the non-Roman ‘successor kingdoms’ in Europe or the Muslim expansion in North Africa. For the Middle East, the term ‘late antique’ is maintained, even if modern scholars specializing in this region often tend to use the term ‘Byzantine’, stressing the cultural bond with Constantinople. For the sake of continuity in this study and because the collective term ‘late antique’ is perfectly capable of encompassing different cultural influences, the term ‘Byzantine’ will only be used in a political sense, i.e. as the rule of Constantinople extending to North Africa and the Italian Peninsula from the reign of Justinian onwards. It should be stressed that ‘late antique’, especially concerning baths and even architecture in general, is a mere chronological indication, i.e. pertaining to the period AD 285–700, and does not imply the presence of specific characteristics. Similarly, the term ‘Islamic’ is only a chronological marker to designate the start of Muslim rule. It does not imply special characteristics. Cultural and Religious Terminology The words ‘Romanisation’ and ‘Christianisation’ are avoided throughout this book, as these are somewhat loaded teleological terms, which can exclude notions of interaction and dialogue. The terms ‘Roman’ and ‘Christian’ are used as general collective terms including a heterogeneous mix of contrasting elements. The terms ‘Jewish’ and ‘Muslim’ should similarly be interpreted as collective terms that leave room for different internal

Introduction

fractions. As it is difficult to label the heterogeneous amalgam of polytheist beliefs of the classical world as ‘religion’ due to the lack of dogmatic framework and holy scripture, the definition ‘classical or traditional cults’ or the ‘classical pantheon’ is used throughout the book. The term ‘pagan’ is avoided as much as possible, as the word implies a belief system in opposition to Christianity and is hence defined ‘in relation to’ Christian faith.26 The term ‘Berber’, which finds its origin in the Greek βάρβαρος (barbaros) denoting all non-Greek speaking people, and which was later adopted by both Romans and Muslims, shall not be used in this study, as it still carries the pejorative connotation of ‘wild’ or ‘uneducated’. Instead, the word ‘Amazigh’—a term chosen by the modern descendants—will be used to denote the indigenous communities of North Africa. Personal Names The personal names of well-known ancient authors and emperors are given in their standard English form following OCD, such as Augustine instead of Augustinus or Justinian instead of Justinianus. Note that for some authors, the name has no Anglicized version, e.g. Ammianus Marcellinus. For Greek authors and rulers, the Latinized versions are used, such as Procopius instead of Προκόπιος or Prokopios. The names of average Romans mentioned in texts and inscriptions are given in their original Latin form, or for Greek, Syriac, Hebrew and Arab names in their standard Latinized forms (following OCD, The Oxford Dictionary of Byzantium and The Oxford Dictionary of Islam). Bath-Related Terminology Throughout this book, the English terms baths, bath, and bathhouse are used as synonyms. The Latin terms balnea and thermae are avoided, except when quoting an ancient text or inscription. The specification ‘private bath’ is used in the modern conception of the word, i.e. the bath of a private house, which was used only by the proprietor or his family. Regarding ownership, the specification ‘privately owned bath’ will avoid any confusion. For the separate parts of the bathhouse— for rooms as well as for architectural elements and building materials—a standardised Latin terminology will be used. There is a glossary of these terms at the end of this book.

26  Due to a lack of alternatives, the word ‘pagan’ in this study is only used in opposition to a Christian counterpart, e.g. Christian versus ‘pagan’ burial ground.

Introduction

Notes on Abbreviations Throughout this study, some late antique baths will be followed by a catalogue number (e.g. C23), referring to the plan in the Gazetteer. The late antique inscriptions pertaining to bathhouses will be followed by a code referring to the list of inscriptions found in Appendix 1. The inscriptions are grouped according to the region in which they have been found, namely Italy (EI-numbers), North Africa (EA-numbers) and Palestina (EP-numbers). The late antique papyri pertaining to baths are followed by code referring to Appendix 2 (e.g. P-1). The late antique ostraca (pottery sherds) with inscriptions concerning baths have a code referring to the same Appendix 2 (e.g. O-3). The abbreviations for the different epigraphic corpora can be found above p. xx. The names and titles of the works of late antique authors follow

9 the abbreviations found in OCD (3rd ed., (1996)). The names and works of early Christian authors which are not included in the dictionary, follow A. H. M. Jones The Later Roman Empire (Oxford 1964) vol. 2, 1462–76, then G. W. H. Lampe A Patristic Greek Lexicon (Oxford 1961). Years pertaining to antiquity and to the Early Medieval or Islamic periods are given in the standard western BC and AD format. For measurements, the following abbreviations are used: km for kilometre(s), m for metre(s), cm for centimetre(s), and m2 for square metre(s). The abbreviations of the following geometric terms occur in the book: ‘diam.’ for diameter, ‘r’ for radius and ‘s’ for side. The dimensions of surfaces are given in the following format: ‘length’ x ‘width’, with the number of decimals after the decimal point an indication of the precision of the measurement.

chapter 1

General Introduction to Roman Baths and Earlier Research on Late Antique Baths General Introduction to Roman Baths and Bathing Habits Pre-Roman Origins of Public Bathing The relationship between man and water has always been somewhat particular. Water is the essential prerequisite for life itself and so control of water is of primordial importance for any society.1 The life-giving power of water and its ability to remove dirt gave special meaning to this element of nature, especially in combination with the other elements, earth and fire.2 It should not come as a surprise that the act of washing was intimately linked with a society’s world views and what Georges Vigarello called “l’imaginaire du corps.”3 The Romans were not the first to develop a (communal) bathing habit. Already in the Bronze Age Central Asia, nomadic groups called ‘Scythes’ by Herodotus practiced a form of bodily hygiene in sweat chambers.4 In ancient Mesopotamian cultures, no actual bathhouses have been discovered, but sewer systems, canalizations—sometimes with terracotta pipes—and cisterns do survive.5 In pharaonic Egypt, basic bath rooms with drainage systems were found in the private residences of high dignitaries from the Middle Dynasty onwards (1938–ca. 1630 BC). There is no evidence of public baths in Egypt until Greek-style bathhouses were introduced under the Ptolemies.6 The Aegean region seems to have played an important role in introducing eastern water technology with what we would now call ‘personal hygiene’. This was especially evident in the Minoan civilization on Crete (first half of the 2nd millennium BC), where we find some of the earliest examples of both terracotta piping and bath tubs in the Mediterranean. Similar tubs were found in Mycenaean palaces on the Greek mainland, for example at Mycenae and Pylos.7 1  Sedlak (2014). 2  Munn (2009). 3  Vigarello (1985) 11. 4  Hdt. 4.73–75; Brödner (1977) 45. Interestingly, a similar sweat bath on the Russian steppes is still described by the 10th c. Persian explorer Ibn Rustah (Kitab al-ala’iq an-nafisah). 5  Tamburino (2010); Muller et al. (2014). 6  Gräzer (2009) 37; For the Greek-style baths, see Gallo (2009). 7  Brödner (1978) 48–50; Ginouvès (1962) 32; see also Shaw (2012) for Mycenaean bathing habits.

© koninklijke brill nv, leiden, ��20 | doi:��.��63/9789004419421_003

The Greeks were the first to introduce a form of communal bathing, besides the most basic form of washing in a river or public fountain—the latter represented on some Greek vases.8 From around 500 BC, public facilities start to emerge.9 The Greek-style public baths were composed of rectangular or round rooms (θόλοι, tholoi) with a series of individual hip-baths against their walls.10 These bath tubs were usually in terracotta, cut out of stone or occasionally constructed with stones and hydraulic mortar. Sitting up straight, the bather was not fully immersed but probably poured hot water over himself, or had this done by a servant (fig. 1).11 These bathing facilities, called a βαλανεῖον / balaneion (also used for Greek private baths), were built during the 4th c. BC and mainly during the Hellenistic period (3rd to 1st c. BC) in most parts of the Mediterranean, including southern France, southern Italy, Sicily, the Middle East and North Africa.12 The innovative character of the bathhouses in ‘Hieronian’ Sicily (3rd c. BC) is particularly stressed in recent research.13 These baths had a bottle-shaped furnace channel heating a water boiler for the nearby hip-baths of the tholos, while a subterranean heating duct placed under a full immersion bathtub created what Monica Trümper called the first “relaxing bathing forms.”14 A second type of bathing facility in Greece was linked to an even more popular building, namely the gymnasium. Especially in the late Classical and Hellenistic periods, this large public building reserved for men combined physical exercises with intellectual education within its walls.15 Besides the actual exercise yard, this building

8  Ginouvès (1962) 21–28. 9  For a catalogue of all known Greek-style public bathhouses, see Fournet et al. (2013). 10  Ginouvès (1962) 187; Hoffmann (1999) 58–71. 11  Villard (1994) 43–45. Evidence for bathing personnel pouring warm water over bathers is found in Egyptian papyri, see Fournet and Redon (2013) 241. 12  See the map by Fournet (2013) 264. 13  Especially the Greek baths of Gela (4th–3rd c. BC), Syracuse (3rd c. BC), Morgantina (3rd c. BC) and Megara Hyblaea (3rd c. BC.) are good examples of Greek style public baths in Sicily. For further reading, see also Broise (1994) 19–25; Lucore (2013). 14  Trümper (2009) 142. 15  Delorme (1960) 462–69; Hoffmann (1999) 58–61; Scholz (2004).

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

11

figure 1 Reconstruction of a tholos in a Greek-style bathhouse in Fournet and Redon (2009) 134, fig. 10; with permission

comprised dressing rooms, a bathroom (λουτρόν)16 with cold water basins and sometimes an open air swimming pool, a massage room (ἀλειπτήριον) and from the 2nd c. BC onwards a round sweat chamber (πυριατήριον).17 One can remark, however, that the act of washing was only a secondary activity, to remove the oil, sweat and sand. One visited the gymnasium to exercise, not to bathe.18 The combination of sports and bathing persisted into Roman times, especially in the Hellenized east. A third widespread form of bathing in the Greek-Hellenistic period took place in private houses. This form probably originated from free-standing tubs in the palaces of the east and found its way to Greece through Minoan and Mycenaean culture.19 In most of the cases, a private bathroom consisted only of a small room with waterproof flooring. In some cases, however, the private baths had street entrances, enabling non-residents to make

16  The term loutron was also used to describe independent structures (Trümper (2013) 35–36), see glossary. 17  Ginouvès (1962) 125–50; Yegül (1992) 17–21; Nielsen (1993a) 9–11; Trümper (2014) 218. 18  Blonski (2014a) 179. 19  Ginouvès (1962) 29–30.

use of the baths. The water was brought in from a reservoir or cistern by means of portable basins and jugs.20 Greek-style bathing habits in public or private contexts thus established themselves in most parts of the Mediterranean world between the 5th and 1st c. BC. Yet, bathhouses never seem to have been an indispensable part of a town’s fabric and thus were never considered by the local authorities to be an indispensable amenity which had to be offered to its citizens.21 Italic Bathing Habits Little is known about the bathing habits of the Etruscans. They were familiar with piped water and sewers, but conclusive evidence for bathing practices is lacking. It is also very difficult to distinguish original Etruscan elements from adopted Greek practices, as cultural exchange was common in the Hellenistic period.22 Southern Etruria is nevertheless an area with lots of natural hot springs, so there is the possibility that the Etruscans knew and used

20  de Haan (2010) 12; Trümper (2014) 210. 21  Trümper (2013) 62; Trümper (2014) 219–25. 22  Broise and Jolivet (1991) 81.

12 these thermal sites. However, they did not build bathhouses around them.23 Before Rome ‘unified’ large parts of the Italian Peninsula, some of the Italic people had already developed bathing habits in specialized facilities.24 A small bathroom (lavatrina) with some kind of waterproof flooring and portable water basins was usually built next to the kitchen, making use of the oven to heat the water.25 Although similar facilities existed in the Hellenistic world, there is no reason to doubt an independent evolution of this simple type of bathroom. The same goes for the larger bathrooms comprising two or even three rooms, usually having a separate dressing room (apodyterium) with niches to store clothes, a cold room ( frigidarium) with a wash basin (labrum) and a hot room (caldarium) heated by a brazier and with a pool.26 These basic bathrooms exhibit an interesting combination of Hellenistic, Punic and local influences, including the Greek type of rectangular immersion bath tub, Punic geometrical black and white mosaic patterns imbedded in opus signinum and a simple drainage system of the bathing water directly spilling on the floor and being evacuated through drain holes in the walls.

chapter 1

Early Roman Bathhouse As (public) bathhouses are seen as a typically Roman building, the origins of the technique behind these facilities have been the focus of fierce debates.27 Mid20th c. Italian scholars were convinced that the principles of Roman baths were invented in southern Italy, and more specifically in Campania, where the seismically active area of the Flegrean Fields with its numerous natural hot springs and steam vents was the perfect breeding ground for conceiving artificially heated bathing facilities.28 In Latin literature, these scholars found a valuable ally in a certain C. Sergius Orata.29 Several Roman authors, including Cicero and Pliny the Elder mention this businessman in relation to the first ‘pensiles balineas’ or ‘suspended baths’, although none of these

authors actually speak of him as the ‘inventor’.30 Garrett Fagan even doubted if the term ‘pensiles balineas’ really meant ‘suspended baths’, suggesting that it could be better interpreted as some sort of fish ponds.31 Even if Orata was somehow linked to the spread or the introduction of the new heating system, the archaeological evidence of the first hypocaust systems (see below p. 13) clearly precedes Orata’s suggested business activities between 110 and 91 BC.32 The most important significance of Orata in the story of Roman bathing habits thus lies in the fact that the Romans may have felt the need to attribute the invention of one of their most successful institutions to a Roman citizen.33 The flow of people and ideas through trade, migration and slavery across the Mediterranean must have played an important part in the spread of a public bathing culture.34 The Roman bathing habit could thus be seen as a fusion of different influences, including Hellenistic and local elements.35 The definition that describes the Roman bathing practise as “communal bathing in heated pools deployed in a sequential arrangement of heated rooms”36 covers only part of the specificity. The Romans did indeed expand the water management and heating system, the two core elements of the actual bathing aspect. In contrast to Greek-style baths, these technical aspects were not visible to the bathers.37 But even if there are similarities between Greek and Roman baths on a basic technological level, the most important difference between the two traditions is the role of the building within the society. A Roman bathhouse is a building for the most popular pastime in Roman society, offering facilities for relaxation, bodily hygiene and social contact within a structure centred on the sequential arrangement of heated rooms, set in an interior that reflected the cultural preferences of its users and that reinforced the Roman sense of dominance over other cultures and over nature. The first examples of ‘typically Roman’ baths in the Italian Peninsula do not antedate the 2nd c. BC. As bath-

23  Gasperini (1988); Chellini (2002) 232; Dvorjetski (2007) 95. For sites with hot springs, see Chellini (2002) 17–187. 24  According to most recent knowledge, the evidence is restricted to Lazio and Campania. For pre-Roman water cults and possible bathing practices in the Romagna, Marche and Abruzzo, see the contributions in Pacciarelli (1997). 25  Fabbricotti (1976) 32; Yegül (1992) 50; de Haan (2010) 19–22. 26  Fabbricotti (1976) 33. 27  DeLaine (1999a) 7; DeLaine (1999b); Maréchal (2012) 147–50. 28  Sgobbo (1929); Di Capua (1940) 93–96; De Angelis d’Ossat (1943) 39–40; Crova (1956). 29  Wikander (1996).

30  Cic., Hortensius 68–71; Plin. HN 9.168. See de Haan (2010) 25–28. 31  Fagan (1996a) 59–63. 32  de Haan (2010) 29; Yegül (2013) 87 n. 55. 33  Similarly, modern Arabs consider the hammam as an invention of Islamic culture (Carlier (2000) 1303), even if the earliest Arab literature acknowledges the Roman origin of the baths (Benkheira (2007) 321). It is interesting to notice how the bathing habits are necessarily ‘a fruit of one’s own culture’. 34  Fagan (1999a) 82–83; Yegül (2013) 81. 35  Nielsen (1985); DeLaine (1989) 112–13; Yegül (1992) 49–54; Gros (1996) 390–94; Thébert (2003) 73–83. 36  Fagan (2001) 404. 37  Manderscheid (2000a) 534.

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

ing was one of the most popular activities in Roman society, the frequent restorations of these popular buildings make it very difficult to find archaeological evidence for these oldest phases. The bathhouse of Fregellae (phase II) is one of these extraordinary examples. Dating from the beginning of the 2nd c. BC, this facility of approximately 1,100 m2 was internally divided in a male and female section, each with a dressing room and a room with a collective pool heated by a subterranean channel (fig. 2).38 This type of pool, with its back wall curved for a more comfortable bathing position, is also found in the Hellenistic baths in Sicily. The men’s section had also a small sweat chamber heated by a fully functional hypocaust system, including terracotta tubes for wall heating. By the end of the 2nd c. BC, the Stabian Baths in Pompeii were also equipped with all the basic components of the ‘standard’ Roman-style bathhouse (fig. 3). In both the men’s and women’s sector, the succession of dressing room, cold room, tepid room and hot room (apodyterium-frigidarium-tepidarium-caldarium) ensured the gradation of heat in a fixed route. The caldarium is of the so-called ‘Vitruvian’ type: a rectangular room with a rounded apse (schola labri) on one of the small ends and a rectangular hot pool (alveus) on the opposite end. The extension of a sporting ground (palaestra) and the addition of an open-air swimming pool (natatio) were completed in the middle of the 1st c. BC.39 The round sweat chamber (laconicum) was probably added in 80 BC, but changed function at the end of the same century to become a cold pool. This round room with semicircular apses was clearly connected to the Hellenistic tholoi and has been labelled a characteristic element of early Roman baths. The other public bathhouses in Pompeii and Herculaneum followed more or less this same linear succession of rooms with a flanking exercise yard.40 For the Republican period, there is quite a large number of private baths found in Latium and Campania.41 These facilities have long been treated separately of public baths. However, recent research has shown that relations between private and public baths, especially during the initial development of Roman bathing habits, are more complex than previously thought. It is not yet clear if the popularity of private baths boosted the building of public baths, or vice versa. But the evi38  Tsiolis (2001) 91; Tsiolis (2006); Tsiolis (2013) 94–95. 39  Nielsen (1993a) 33. 40  Pompeii: Suburban Baths (1st c. AD), Republican Baths (end 2nd c. BC.), Central Baths (AD 62–79), Forum Baths (1st c. BC.), baths of Iulia Felix (AD 62–79), Sarno Baths (1st c. AD). Hercu­ laneum: Suburban Baths (1st c. AD), Forum baths (1st c. BC.). 41  Papi (1999a); de Haan (1997); de Haan (2010) 41–42.

13

dence points to a pioneering role of private baths: “In der Entwicklungsphase der römischen Badekultur und der Bädertechnik spielten sie eine sehr wichtige Rolle und waren sogar regelrechte Wegbereiter bei der Verbreitung neuer Technologien.” (“In the development phase of Roman bathing culture and bathing technology these played a very important role and were even downright pioneers in spreading new technologies.”)42 The public baths at Fregellae also display an architectural resemblance to the city’s houses, “perhaps revealing distant roots in the broadly conceived family of kitchen-bath combinations of Italic farmhouses and rustic villas.”43 Imperial Period Bathhouse After the political turmoil that characterized the transition from the Republic to the Principate, the emperor Augustus started a comprehensive propaganda campaign to firmly establish the new form of power within the Roman society.44 A grand building programme in the city of Rome was a major boost for both the economy and his popularity with the people. The availability of large groups of unskilled labour forces and the efficient opus caementicium building technique enabled large structures to be built in a relatively short period of time. It is precisely in this period that Rome became the leading player in bathing architecture. The Baths of Agrippa, named after Augustus’ general and right-hand man, were the first of the so-called ‘imperial baths’ to be built (dedicated in 25 BC). The term ‘imperial’, one should note, has more to do with size, luxury and axial symmetrical layout of the building than with the involvement of an actual emperor.45 Although little is known about Agrippa’s Baths from archaeological evidence, the plan came down to us through a fragment of the Forma Urbis Roma (FUR), a plan of Rome carved out in marble slabs from the 2nd c. AD.46 It is unclear if the plan actually shows the layout of the 1st c. BC, or if it represents a later reorganisation.47 Cassius Dio wrote that the baths were damaged by fire in AD 80.48 Afterwards they were restored and enlarged under Hadrian (first half of 42  de Haan (2010) 37–38; translation by the author. 43  Yegül (2013) 80. 44  Zanker (1990). 45  Brundrett and Simpson (1997) argue that the real innovation of Agrippa’s Baths lay in their size and luxurious decoration, rather than in their innovative plan. 46  Yegül (2010) 105–106. Today, only the Arco della Ciambella, in the similarly named street between the Pantheon and Largo Argentina, are a visual reminder of the monumentality of this Early Imperial bathhouse. The visible remains might well relate to a later restoration phase. 47  Brödner (1983) 51. 48  Cass. Dio 66.24.

14

figure 2

chapter 1

Plan of the baths of Fregellae (Italy) with its early hypocaust (nr. 15) in Tsiolis (2013) 96, fig. 13; with permission

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

figure 3

15

Plan of the Stabian Baths in Pompeii (Italy), with the alleged Greek hipbaths to the north (hatched) after Nielsen (1993b) 98, fig. 75

the 2nd c. AD) and under Constantius and Constans (if EI-54 pertains to these baths).49 In any case, the Baths of Agrippa prove that the imperial family—to which Agrippa certainly belonged—saw the construction of large bath complexes as an act of munificence towards the people and as an ever-lasting reminder of the imperial house. It is also the first time that a large space in the centre of Rome was allocated to public baths.50 The Baths of Nero, built 100 m north of the Baths of Agrippa, are known from the 16th c. drawings of Palladio. Extensive rebuilding under Alexander Severus in AD 226 may have altered the original design. Archaeological investigations in the cellars of the buildings surrounding the modern piazza Sant’Eustachio confirmed the location and the general layout of the baths.51 For the first time, the plan showed the axial symmetry that would 49  Richardson (1992) 386–87. 50  Brundrett and Simpson (1997) 222; Fagan (1999a) 119–20; Gros (1996) 395. 51  The restoration is recorded in SHA. Alex. Sev. 25.3–7. For the excavations, see Ghini (1985); Ghini (1999).

characterize imperial baths throughout the empire (fig. 4). The two-colonnade palaestra seen on Palladio’s drawing probably belonged to the original plan and was incorporated into the building. They flanked a large swimming pool (natatio). The frigidarium had 4 cold pools and was probably cross-vaulted. The caldarium protruded out of the main bloc. The element of space, visible in the lofty halls, became ever more important.52 Under Trajan, the empire reached new heights. The boundaries were pushed further to the east, while tax revenues from the numerous provinces, especially North Africa and Egypt, resulted in unprecedented concentrations of wealth. The emperor and the elite often invested in public infrastructure in their home towns or towns of residence to boost their popularity for upcoming political careers and to immortalize their family name.53 The use of Roman concrete (opus caementicium) and bricks (opus testaceum) enabled the construction of large buildings in a relatively short amount of time. 52  Nielsen (1993a) 46–48; Yegül (2010) 107. 53  Fagan (1999a) 143–44; Saliou (2014) 671–72.

16

figure 4

chapter 1

Plan of the Baths of Nero in Rome (Italy) in Nielsen (1993b) 84, fig. 51 after a drawing by Palladio; with permission

The prosperity of the Roman empire also resulted in a population growth, which in its turn increased the demand for all sorts of public facilities, not in the least for public baths. It was during the 2nd c. AD that a construction boom of both small and large public baths took place.54 The Baths of Trajan in Rome, inaugurated in AD 109, set a new standard for imperial baths and to some extent baths in general, “qui (…) ne connaîtra guère de fantaisie” (“leaving little room for imagination afterwards”).55 Known from the marble FUR plan, 16th and 17th c. drawings and some small-scale excavations in recent years (fig. 5), this Trajanic bath took only 5 years to build.56 The main bath block was now a separate building, detached from the surrounding city by a surrounding wall. This enclosure, with a preference for large exedrae—a feature visible in the forum and market of Trajan—also created spacious gardens around the 54  Fagan (1999a) 79–84. 55  Adam (1984) 296; translation by the author. See also DeLaine (1992). 56  De Fine Licht (1974) 5.

main bath building, housing libraries and nymphaea. The orientation of the baths likewise changed, with the caldarium now orientated to the south to maximise the heating effect of the afternoon sun. The frigidarium became the centre of a strong axial composition, while the proportion of space dedicated to non-bathing activities increased. Other new features included the heated rooms with apsidal endings and the exercise yards (basilica thermarum) flanking the frigidarium instead of the natatio.57 Although there were no imperial baths built in Rome under Hadrian, the bathhouses constructed on the emperor’s private estate at Tibur (Tivoli) exemplify an evolution in bathing architecture that had been evolving since the late 1st c. As space and lighting became ever more important, new oval or polygonal room forms and bold new types of domes expressed a sense of dynamism from both inside and outside the building. “The emperor’s villa near Tivoli was a veritable laboratory for 57  Brödner (1983) 72–74; DeLaine (1992) 259; Gros (1996) 400–401; Weber (1996) 80–81; Yegül (2010) 109.

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

figure 5

17

Plan of the Baths of Trajan in Rome (Italy) with the standing remains highlighted in black after Nielsen (1993b) fig. 53

researching how geometry could be exploited to create new spatial forms and effects of light and shade, of curve and counter-curve”.58 This architectural dynamism is illustrated in the Small Baths of the Villa Hadriana (fig. 6). The basic components, i.e. the succession of palaestra— frigidarium—tepidarium—caldarium are presented in dynamic spatial units, with several additional rooms, including two heated rooms with pools, a calida piscina, and a round solarium, all facing south-west, to benefit from the afternoon sun. There is an obvious preference

for rounded shapes, as seen in the numerous niches and apses.59 The grand imperial baths may have been trendsetters in the architectural development of Roman public baths, but the vast majority of public bathhouses were of the small neighbourhood type such as existed in the 1st c. BC. Pliny the Elder mentions 170 balnea in early Augustan Rome.60 This may not have been the actual figure, but it nevertheless shows that this type of bathing facility was the most common type found in the

58  Jones (2000) 91.

59  Nielsen (1993a) 59; Gros (1996) 402. 60  Plin. HN 36.121.

18

figure 6

chapter 1

Plan of the Small Baths in Hadrian’s Villa at Tibur (Italy) after Nielsen (1993b) 13, fig. 84

capital.61 The layout of these baths is more difficult to categorize, as their planning was above all influenced by the available building space and practical concerns such as water supply and waste water disposal. There are few excavated examples for Rome itself (see below, chapter 3), but the archaeological evidence from the empire is abundant. This type of smaller facility will remain the most important form of the Roman bathhouse during the entire antique period. This has less to do with an actual preference than with a pragmatic choice: the simple succession of rectangular cold, tepid and hot rooms ensured that all the necessary bathing facilities fitted in the allotted building space. “Unlike imperial thermae, whose planning is characterized by a readily recognizable axiality and strong bilateral symmetry, the multitude of these privately owned, small baths (balnea) exhibit diverse planning tendencies and resist an easy classification.”62 Depending on the size of the building space, additional facilities could be added. By contrast, what was considered essential in the bathing process changed over time. In the Early Imperial period, the palaestra and a round sweat chamber often seem to have been an essential part of the bathing experience. By the High Imperial period, the neighbourhood 61  Staccioli (1958); Fagan (1993) 334. 62  Yegül (1979) 110.

baths often lacked a palaestra, but had several heated rooms with or without heated pools. The evolution to more complexity in plan resulting in the different kinds of heated rooms and a heighted importance of rooms without pools is also attested for these standard neighbourhood baths in the 2nd c. AD.63 Continuing on the plan of the Baths of Trajan, the Baths of Caracalla in Rome (fig. 7) added some new features. The main bath block was now completely detached from the perimeter wall, while the cisterns were now attached to the complex and fed by a separate branch of the Aqua Marcia aqueduct, the so-called Aqua Antoniniana.64 The large open air natatio looked onto an impressive marble-veneered wall with several tiers of colonnades, vaulted recesses and niches, housing marble statues, as water gushed out into the pool, giving it a nymphaeum-like appearance.65 The architectural legacy of the Hadrianic period is clearly visible in the handling of space and light and the rounded and oval forms. The round caldarium, however, was an addition of the time of Constantine.66 The neighbourhood baths from the

63  DeLaine (1992) 266–67. 64  DeLaine (1997) 16; Nielsen (1993a) 53; Weber (1996) 83–87. 65  Yegül (2010) 114. 66  Gros (1996) 403. For the caldarium, see DeLaine (1987).

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

figure 7

19

Plan of the Baths of Caracalla in Rome (Italy) after Nielsen (1993b) 87, fig. 56

3rd c. display the same preference for round and curvilinear forms as expressed in the Hadrianic period. The Bathing Itinerary We are informed by several ancient texts, including Lucian, Pliny the Younger and schoolbooks such as the Colloquia Monacensia, about the order of bathing and about some specific habits of the bathers.67 The entire bathing routine was thoroughly influenced by classical medicine, acknowledging the positive effect of sweating and relaxing in warm water before ending with a cold dip. A visit to the baths started with undressing in the apodyterium, stowing away clothes in niches or cupboards to be guarded against thieves by slaves. If there was a palaestra, one could start off with some exercises, including ball games and lifting dumb-bells. After scraping of the combination of sweat, dirt and olive oil with a strigilis, the bather could cut across the frigidarium and go straight to the tepidarium. After acclimatizing to the heat, the bather took a sweat bath in the sudatorium or laconicum, purifying the body by sweating out the bad fluids. However, most bathers probably spent their time 67  Lucian, Hippias; Plin. Ep. 5.11; Corpus Glossariorum Latinorum III, 651–52. See also Fagan (1999a) 12; Dionisotti (1982) 102–103, 116.

chatting or getting a corporal treatment in one of the heated rooms, which were amply illuminated by large windows.68 The importance of these rooms is stressed by Lucian, ‘in which one can sit or stand in comfort, linger without danger and stroll about with profit.’69 The highlight of the experience was a relaxing dip in the hot pools (alvei, solia) of the caldarium, preceded or followed by a quick wash at a water basin on a pedestal (labrum) or in a small pool.70 Before or after entering the alveus, a type of soap made with sodium phosphate or sodium bicarbonate, called nitrum by the Romans, was applied to the skin to induce sweating. The mixture of sweat and nitrum was then removed with a strigilis and with towels.71 One could leave the caldarium by passing through the same tepidarium (in baths with a retrograde layout) or a separate ‘exit-tepidarium’ (in baths with circular layout). One should then finish with a refreshing plunge in the cold pool (piscina) of the frigidarium. This 68  Krencker indicates these rooms with the Roman numerals I, II and III (Krencker et al. (1929) 184–85). 69  Lucian, Hippias 6; as translated by Yegül (2010) 77. 70  For the use of the word solium, see Maiuri (1950). For the habit of washing at the labrum or in a small pool, see Maréchal (2015). 71  Blonski (2014a) 273–76, 315. On soap, see also Blonski (2014b).

20 allowed for the softened and weakened body to regain its strength. At last, the bathers completed their visit with a massage and / or a final perfuming and anointing, to appease the skin from the scraping and rubbing. This could take place in a separate unctorium. Depending on the bathhouse and its facilities, other activities such as a stroll in the gardens or a swim in the large natatio were possible. This general outline could be altered according to personal preferences and habits: “Anywhere along the recommended course, deviations, omissions or repetitions were possible. One bathed as one wished.”72 Regional tastes and habits could also lead to preferences in the architecture, e.g. the importance of sports in the Greek parts of the empire and the bath-gymnasium in Asia Minor. Yet, even with personal or regional habits, the Romans and non-Roman people using Roman baths seem to have had a certain idea of what one could find / do in a bathhouse. Modern Typologies of the Plans The need to identify Latin bathing terminology in the archaeological record has led to the categorization of bathhouses according to the articulation of their separate rooms. Ernst Pfretzschner was the first to categorize Roman bath buildings, but it was the typology of Daniel Krencker that has been followed, adjusted and updated.73 The plan of the baths was analysed by interpreting the disposition of frigidarium—tepidarium—caldarium with the circuit of the visitors in mind. The main types were the row type, the ring type, the double type and the imperial type, each with its own variations.74 Sometimes a fourth type, the ‘block type’, is added. Here, the rooms are arranged in two or more rows next to each other.75 French scholars such as René Rebuffat, Yvon Thébert and Alain Bouet prefer to categorize the baths according to the bathing route.76 The three main possibilities are a retrograde, circular or symmetrical circuit, and their ‘semi’-variants. Although there is great merit in approaching the plans according to the actual use of the rooms, this categorization also forces the variable articulation of the rooms in rigid configurations. Thébert often struggles to assign each room with a unique function, hypothesizing doorways between rooms and their being blocked off to ensure one of the above-mentioned types

72  Yegül (2010) 18; also Weeber (2007) 14–15. 73  Pfretzschner (1909); Krencker et al. (1929). For an overview, see de Haan (2007) 42–45. 74  Krencker et al. (1929) 177–81; Nielsen (1993a) 4. 75  Pfretzschner (1909) 25–29; Weber (1996) 69. 76  Rebuffat (1991); Thébert (1991); Bouet (2003) 163.

chapter 1

can be discerned without two rooms having the same function.77 It is striking that scholars, who are the first to admit that ‘no two bathhouses are alike’, fiercely try to fit every plan into modern typologies. With the diversity of plans in mind, it is much more likely that Roman architects designed new bathhouses by allocating the room types that fitted with the expectations of the bathers into a plan that followed contemporary bathing trends. The classification systems have nevertheless proven their worth, especially by providing scholars a set of standardised models to present a basic description.78 The problems arise when the description of a bathhouse is limited to its typology: “The very nature of the process of classification tends to emphasize the formal architectural similarities of shape and distribution at the expense of geographical or chronological differences.”79 Aspects such as the relation between the heated and non-heated parts, the connection to the water network, the size and location of the building plot or the potential / envisaged number of users are often forgotten in this case.80 Technology of the Baths The core elements for the functioning of a Roman bathhouse were water and fire. Therefore, it is surprising that the technical aspects of Roman baths were neglected for so long in the historical and archaeological research. Hubertus Manderscheid pointed out this important gap in our knowledge: “Im Vordergrund der Thermenforschung haben lange Zeit architektonische Probleme gestanden. Ausgrabungen und daran ansclieβende Untersuchungen konzentrierten sich demgemäβ auf baugeschichtliche Fragestellungen. Die Technik, die ein wesentlicher Bestandteil des römischen Badewesens, ist—wenn überhaupt—häufig nur am Rande behandelt worden.” (“For a long time architectural problems have held a prominent place in research on baths. Excavations and subsequent studies thus focussed on research questions pertaining to building history. Technology, an essential part of Roman baths, has been—if at all—only considered in the margins of research.”)81

77  Thébert (2003). 78  de Haan (2007) 39. 79  DeLaine (1988) 17. 80  de Haan (2007) 42–45. 81  Manderscheid (1988a) 112.

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

figure 8

21

Flow of the hot gasses inside the tubuli according to the experiment by H.-C. Grassmann plan: N. Baeyens, after Grassmann (2011) 32, fig. 72

Hypocaust The link between Greek and Roman forms of floor heating has been interpreted in a wide variety of ways, ranging from a predecessor-link to a mere inspirational neighbour (see above p. 10–12). The fact remains that even if the Greeks came up with the idea of floor heating, the Roman system for heating a bathhouse was far more elaborate. We still do not fully understand the difficult interplay of floor and wall heating, the spread of the heat within the wall or the interplay between tubuli and chimneys. A fundamental difference between Greek and Roman baths was the way in which the heating system was used within the baths. Janet DeLaine remarked that Greek baths never combined the heated communal pool with the heated round sweat room, a feature that only developed in Roman times and changed the bathing habit significantly.82 Garrett Fagan similarly stated that these Greek systems never led to a standardized sequence of rooms, but instead enabled a “variable itinerary.”83 It is indeed essential to remember that the ‘technical aspects’ of a bath were inextricably linked with its use and perception, and even its architecture and decorative schemes. The Roman heating system consisted of three important interacting components: a furnace, a raised 82  DeLaine (1988) 16. 83  Fagan (2001) 406.

floor resting on pillars (suspensura) and a system of wall heating (fig. 8). The term ‘hypocaust’ is a predominantly modern name to indicate the combination of furnace, pillars and suspended floor.84 Vitruvius described how to construct a hypocaust.85 First, one needed to level the building area with mortar and sesquipedales tiles, a square tile of ca. 44.4 cm.86 On top of this levelled floor, pillars (pilae) consisting of layers of bessales tiles (square tiles of ca. 19.7 cm) bonded with mortar were placed some 60 cm apart.87 These pillars were often larger or reinforced near the walls and under pools. They had an average height between 40 and 80 cm, but examples of up to 1 m have also been found.88 Most of the time, the individual tiles of the pillars were square or round, but pillars made of hollow terracotta pipes or small monoliths of a local stone (sandstone, basalt, limestone) also existed.89 Large bipedales tiles (square tiles of ca. 60 cm) were placed on top of the pillars and thus created a surface on top of which several layers of hydraulic mortar (ca. 15–20 cm thick) were poured to make the floor waterproof. In the uppermost 84  Nielsen (1993a) 161. 85  Vitr. De arch. 5.10. 86  Degbomont (1984) 159–64. 87  Nielsen (1993a) 14. 88  Adam (1984) 290. 89  Degbomont (1984) 99–106; Schiebold (2010) 14–17.

22

figure 9

chapter 1

Reconstruction of a hypocaust system with testudo above the furnace, in connection to the alveus after Yegül (1992) 375, fig. 472

layer of very fine mortar, the eventual floor decoration was set. Large marble slabs, smaller fragments of marble in geometric / figurative patterns (opus sectile) or mosaics were very popular.90 This ‘suspended floor’, consisting of bipedales tiles, several layers of mortar and a decorative walking level (mosaics or marble), is called the suspensura. The furnace itself was constructed in one of the sides of the heated room and was connected to the hollow space under the suspended floor (fig. 9). The fire was lit directly on the brick floor, which was sometimes reinforced with blocks of heat resistant stone such as basalt. To reduce the pressure on the overlying wall, the larger furnaces were often covered by a barrel vault.91 Two short walls sometimes ‘guided’ the heat from the furnace under the suspensura. The flow of air that fed the fire was regulated by metal doors or a metal hatch.92 Large cylindrical boilers made of bronze and lead were often placed just above the fire to heat the water for the hot pools. They rested on supporting brick walls and metal bars and were encased in brickwork to reduce heat loss (fig. 9). To clean and refill these boilers, adjacent small stairs enabled one to reach its top.93 Vitruvius described a system with three connected water tanks: one for cold, one for tepid and one 90   Adam (1984) 291; Yegül (1992) 357; Nielsen (1993a) 14; Manderscheid (1994). 91  Yegül (1992) 368–69; Lehar (2012) 37, 45–46. 92  Degbomont (1984) 73, 118–31; Nielsen (1993a) 16. 93  Nielsen (1993a) 16; Lehar (2012) 42–43.

for hot water.94 Alternatively tepid water was obtained by combining water from the cold and hot tanks. The archaeological remains of bath furnaces nevertheless show that this (perhaps ideal) system was overlooked in favour of a system with just one or at most two boilers.95 An additional device to keep the water of the pools warm was popularized in the 1st c. AD. The so-called testudo alvei was a half-cylindrical container in bronze or copper placed directly above the heat duct of the furnace and open to the bottom of the pool on one side (fig. 9). In this manner, the cold water sinking to the bottom of the pool, was heated inside the testudo and again rose upwards.96 For smaller pools, a type of semitestudo, consisting of a metal plate inserted into the bottom of the pool, just above the furnace, might have been used.97 As both boilers and (semi-)testudines were made of metal, these elements were often salvaged once that baths had fallen out of use. Only few examples have come down to us. Their use can sometimes be deduced

94  Vitr. De arch. 5.10.1. Similarly, Faventinus (De Diversis fabricis archietctonae 16.3) and Palladius (opus agriculturae 1.39.3) describe the water boilers of private baths with two and one tank respectively. A rare example of three boilers is found in Fregellae (see fig. 2 nr. 17). 95  Manderscheid (2000a) 495. 96  Literally meaning the ‘tortoise of the hot plunge’ in Latin, due to its form. An early example was found in the Greek-style North Baths at Morgantina. For more examples, see Ginouvès et al. (1998) 107. 97  Maréchal (2017).

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

23

figure 10 Schematic representation of the location of chimney flues in connection to the hypocaust in Lehar (2012) 206, fig. 209; with permission

by examining traces left in the surviving walls above the furnace. Wall Heating The invention of wall heating by leading the draught of hot air from the hypocaust into hollow walls was nothing short of a revolution for heating techniques and subsequently the bathing experience. Larger rooms could be heated and the hot section of the baths could increase in size. “One could well claim that the creation of the great thermae was made possible mainly by the development of sophisticated wall-heating systems.”98 To ensure that the hot air was drawn from the furnace into the hollow space under the suspensura and then into the hollow walls, chimneys leading up to the roof were placed at the corners of the room (fig. 10).99 This system of draught was possibly borrowed from pottery kilns and other similar uses, as illustrated in a combination of a bread oven and a bath furnace in Pompeii.100 Before the spread of box tiles, normal tiles with protruding knobs on their corners were fixed to the walls with T-shaped nails, thus creating a hollow space (fig. 11). These so-called tegulae mammatae or ‘nippletiles’ are found in the Forum Baths and the Stabian Baths in Pompeii.101 This system was probably not that efficient in heating the walls, as the narrow space between tile and wall (around 7–8 cm) did not ensure good cir98  Yegül (2010) 86. See also Rook (1979) 306; Thébert (2003) 75. 99  Degbomont (1984) 146; Rook (1994) 10; Lehar (2012) 83–112. 100  de Haan (2005). For the oven/furnace in Pompeii, see Miselli (2001) 232–35. 101  Degbomont (1984) 136–39; Yegül (2010) 86.

culation, “provoquant au contraire ce que l’on nomme en fumisterie « l’effet de lame », c’est à dire la formation de turbulences freinant la montée de l’air chaud, provoquant meme parfois son retour” (“resulting, on the contrary, in what has been called ‘the blade effect’ in heating engineering, i.e. the creation of turbulences that restrict hot air from rising, sometimes even resulting in its return.”)102 In the 1st c. AD, a different type of wall heating appeared. Several rows of box tiles (tubuli), placed side by side, covered the entire surface of the wall, ending in a horizontal pipe at the start of the vault that led the hot gasses outside through chimneys.103 The tubuli were fixed on the walls with mortar and occasionally with T-shaped clamps (fig. 11). Other sorts of wall-heating are found sporadically in the form of tiles with two opposing raised edges (tegula hammatae) or the combination of tiles and terracotta studs or iron nails with terracotta spacers (fig. 11).104 These alternatives seem to have been local solutions. The system with tiles and spacers is, for example, found mainly but not exclusively in the region of Lycia (Turkey).105 Once the tiles or tubuli were fixed to the walls, the entire surface was covered with successive layers of mortar and plaster. The plaster could then be painted. Often, the lowest part of the walls received a marble 102  Adam (1984) 292. 103  Degbomont (1984) 140–46. 104  Nielsen (1993a) 15; Yegül (1992) 363. 105  Farrington (1995) 102–104.

24

chapter 1

figure 11 Different types of wall heating after Yegül (1992) 364, fig. 455

veneer, probably to protect the walls from water. With the introduction of the tubuli, heavier materials such as marble slabs could adorn the higher parts of the walls as well, as tubuli were more robust than tegulae and could carry more weight.106 The Functioning of the Heating System Although the different components of the heating system are known and their separate function is quite clear, scholars have struggled to understand the precise functioning of the system.107 That said, it is likely that the majority of Romans did not know how their baths functioned any more than we do. The best Roman source we 106  de Haan (2010) 36. 107  For an overview of the scholars that investigated the Roman hypocaust system before the experiments by Kretzschmer in the 1950s, see Degbomont (1984) 201–204.

have is Vitruvius, who gives us a general description of the hypocaust, but does not mention any form of wall heating.108 Some early interpretations of the hypocaust system, such as that of Otto Krell, even considered it a mechanism to dry and consolidate the mortar after the building was erected.109 It was not until the experiments by the engineer Fritz Kretzschmer that the system was scientifically analysed. His pioneering experiments with a recreated hypocaust system in Saalburg tried to explain the relations between the furnace, hollow floor, hollow walls and chimneys. He concluded that the hollow walls did not actively participate in the heating of

108  Vitr. De arch. 5.10. Vitruvius does mention tegulae amatae (De arch. 7.4.2) to describe the insulation of walls in order to keep wall paintings safe from moist conditions. 109  Krell (1901).

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

the room, and only had an “insulating character.”110 The inadequate measuring equipment used in Kretzschmer’s experiment, however, meant that similar experiments were conducted by Heribert Hüser and Dietwulf Baatz.111 Furthermore, Kretzschmer had used charcoal as fuel, which achieves lower operating temperatures during combustion than dry wood.112 Hüser’s measurements concerned not only the temperature of the floor, but also of the flow and the temperature of the hot gasses themselves, their spread pattern across the floor and the amount of fuel that was needed. New experiments in the 1970s in the recreated Roman baths at Xanten (Germany) added new insights into hypocaust construction and heating. Tony Rook emphasised that the hot gasses under the suspensura were stratified, meaning the hot air remained at the top and the cold sank to the bottom. This explains why the pillars of the hypocaust were rather high, so that the heat was transferred to the overlying walking level and not to the ground below.113 The research of Hans-Christian Grassmann in the Xanten baths focused on the interaction between tubuli and floor heating.114 His measurements demonstrated that the hot gasses rise inside the box tiles, cool down near the top, and drop down in neighbouring tubuli thanks to the lateral holes that interconnect the several rows (fig. 8).115 It can clearly be deduced from this experiment that the walls could contribute to the actual heating of the room, and were not necessarily isolating elements. Other important conclusions of his research concern the influence of the dimensions of the box tiles and the thickness of the suspensura. It seems that tubuli, with a larger cross-section and with larger lateral openings, ensure a smoother flow of hot air, as there is less friction resistance on the wall surface. This in turn results in a higher average temperature, which spreads more evenly over the entire heated surface. The height of a box tile has less influence, as long as the cross-section of the separate tubuli is wide enough.116 A very interesting consequence from this heating capacity of the walls is that the inside temperature of the room can be controlled by changing the thickness of the suspensura. As the concrete of the floor has a ‘Speicherverhalten’, a heat storage 110  Kretzschmer (1953) 33–35. However, in analysing the heating system of the Aula Palatina in Trier (Germany), Kretzschmer had to admit that the building could not be heated without the contribution of the tubuli (Kretzschmer (1955)). He saw this case as an exception. 111  Hüser (1979); Baatz (1979). 112  McParland et al. (2009) 181–82. 113  Rook (1978) 279. 114  Grassmann (1994); Grassmann (2011). 115  Grassmann (2011) 32. 116  Grassmann (2011) 32–39.

25

capacity, a thicker floor means that more heat from the hot air below is absorbed and thus less heat is given to the room.117 By synchronizing the measurements of the box tiles, with the thickness of the floor, room temperature could be more or less controlled.118 It is difficult to find out exactly to what degree the Romans themselves made use of this complex interaction between wall and floor heating, as there are no literary testimonies and the archaeological remains are sometimes too damaged or too poorly investigated to make any conclusions. An experiment in Sardis (Sart, Turkey), where a Roman-style bathhouse was constructed for a television documentary, reconstructed several possible scenarios for the transition between the tubuli and the chimneys.119 When the tubuli were open at the top, collecting in one large horizontal flue, the hot gasses would rise through the tubuli closest to the furnace, without heating those at a greater distance. A system in which some of the tubuli were closed at the top and only occasionally connected to the horizontal ‘header’ seems to have been the best option for evenly spread heat. If the tubuli were all closed at the top, acting as a system independent of the flues, the gasses formed convection currents that both insulated and actively heated the interior rooms. In this “passive heating system”, to quote the authors of the Sardis experiment, the hot gasses would rise, while the cold gasses would fall back down into the hypocaust through the neighbouring row of tubuli.120 Hence a circulation pattern was created within the wall in which two rows of ‘hot gas—tubuli’ alternated with two rows of ‘cold gas—tubuli’ (fig. 8). As the Romans did not have any equipment to measure temperatures, there are no indications of what was perceived as ‘the right temperature’ inside the different rooms of a bathhouse. The comparison with Turkish vapour baths made by Erika Brödner can be misleading,121 as the heating techniques and the bathing habits themselves differ from Roman predecessors. It is quite likely that in Roman times, room temperatures differed from region to region and even from bathhouse to bathhouse. It is also difficult to compare the temperature inside the different heated rooms.122 The volume to be heated, the presence of pools and the number and 117  Bansal (1999) 306. 118  Grassmann (2011) 39–40. 119  Yegül and Couch (2003). Similar results can be found in Lehar (2012) 196, 202–203, 207. 120  Yegül and Couch (2003) 207. See also the theory by Grassmann (2011). 121  Brödner (1983) 108. 122  Recent research into the temperatures inside modern hammams confirmed that a certain ‘micro-climate’ exists within the baths, influenced by the location of the rooms and the

26 placement of windows and doors all influenced the temperature.123 Recent studies even suggest that the difference between the rooms may have been less clear-cut than previously assumed.124 A certain ‘thermal blending’ could have occurred between the different heated rooms, especially if the doorways that interlinked them were not closed off. From the experiments mentioned above, scholars learned that there was a significant heat loss around the doors and the windows.125 In contrast to the small and dark Republican baths described by Seneca,126 the baths of the Principate, especially the large imperial ones, had bright lofty halls with large windows. This was only made possible thanks to the invention of wall heating. But windows also meant a higher rate of heat loss. Apparently, these large windows openings were glazed—sometimes double-glazed—and often also equipped with awnings.127 However, windows also provided light and enabled solar heat to enter. In fact, the use of large but securely insulated windows can be seen as a way to reduce the heated surface of the wall, reducing the fuel consumption at the same time.128 With the conclusions of Grassmann in mind (see above), the use of large windows could also have been used to balance the modest heat emitted by a thin suspensura. The door openings posed a similar problem, but often opened onto other rooms of the baths.129 So the doors from the heated rooms to the cold ones were intentionally kept small, while those between heated rooms could be of normal size, to stimulate heat transfer from hot to tepid rooms. The tepidarium, often indirectly heated without a furnace of its own, largely depended on heat transfer through doors and thin walls it shared with directly heated rooms.130 It is difficult to determine which doorways had closing systems. Sometimes, sockets in a threshold can inform the archaeologists about the presence of doors. The use of cloth curtains is more difficult to discern, as such systems have never been preserved. The impact of a simple curtain on heat retention can be significant, as computational models have confirmed.131 outside temperatures (Bouillot (2008); Mahdavi and Orehounig (2008)). 123  Thatcher (1956). 124  Oetelaar et al. (2014) 398. 125  Kretzschmer (1953); Hüser (1979); Grassmann (2011); Oetelaar et al. (2013); Oetelaar et al. (2014). 126  Sen. Ep. 86.8. 127  Broise (1991) 62–64; Weber (1996) 52–53. 128  Ring (1996) 722. 129  Oetelaar et al. (2014) 396. 130  Oetelaar et al. (2013) 63. 131  Oetelaar et al. (2014) 398–400.

chapter 1

A third vulnerable point for heat loss is the ceiling. The computational fluid dynamics (CFD) technique, used to chart temperature distribution within rooms, confirmed that the highest temperatures could be found in the upper parts of the room.132 Vitruvius describes a complex system of suspended vaulting to ensure that the wooden beams of the roof are not damaged by moisture.133 In the archaeological remains, most of the heated rooms were covered with simple concrete vaults. It appears that in most of the cases, especially in doomed rooms, the tubuli did not continue onto the ceiling but stopped at the start of the vaulting. Only in a few examples, such as the special system of a bathhouse in Sussex (England), ‘insulated’ vaulting has been found.134 For barrel vaults, terracotta vaulting tubes (tubi fittili) were sometimes used, mainly in North Africa. Their insulation properties made them especially popular in bath buildings.135 Water Management of Baths Hubertus Manderscheid, author of the article ‘Ohne Wasser kein Bad’ (‘Without water, no bath’),136 has long advocated more research into the water management of Roman baths. As odd as it may seem, the water in Roman baths was neglected until some 30 years ago.137 Nowadays, studies and excavation reports pay more attention to this vital element. The water management of a Roman bath can be subdivided into three important elements: the water supply, the actual use and the water disposal.138 Water Supply It is wrong to assume that all Roman baths were supplied by aqueducts, or that all Roman baths needed vast quantities of water for that matter. Before an aqueduct was constructed for a certain city, bathhouses had to rely on groundwater and rainwater to fill their pools. Even when aqueducts were built, some baths continued to use these supply methods.139 A combination of supply by aqueduct and groundwater and / or rainwater was also frequent, especially in dry regions such

132  Oetelaar et al. (2013) 62, fig. 3. 133  Vitr. De arch. 5.10. 134  Lancaster (2012). 135  Wilson (1992) 108; Storz (1994) 72–89. 136  Manderscheid (2000b). 137  Maréchal (2012) 154–56. 138  Garbrecht and Manderscheid (1994) 11. 139  Nielsen (1993a) 23; Hodge (1991) 267.

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

as Africa.140 The disadvantages of rainwater collection are obvious: the supply depends on the weather, there is a need for large and often expensive storage facilities (cisterns, reservoirs) and the water has to be used rather sparingly.141 The collection of rainwater in tanks on the rooftops allowed for a simple distribution by gravity. With groundwater, a well or a noria system (type of waterwheel) had to be constructed and operated. Sometimes bronze or wood pumps had to be used.142 The quality of the groundwater was variable, depending on the soil type. The groundwater in Pompeii, for example, was of a rather bad quality due to the volcanic soil of the region. Rainwater was often muddy and would have carried along small organic particles (leaves etc.) when being led from rooftops into the cisterns. Hence it sometimes had to be stored, allowing the particles to settle before being used. This raises the question of whether different types of water were used in different parts of the bathhouse, e.g. ‘unfiltered’ water for the open-air pools, or if the quality of the water added to the prestige and the reputation of certain bathhouses.143 The appearance of aqueducts meant that water was now directly taken at source, with less chance of pollution by sands, leaves, insects, etc.144 Furthermore, perennial sources ensured a continuous supply of water all year round. Due to this ‘unlimited’ supply, bathhouses could be equipped with larger pools. The rise of the frigidarium and the natatio could well have been a result of the better water supply provided by the aqueducts.145 According to Vitruvius, one third of the water of an aqueduct that arrived in the city’s distribution tank (castellum aquae) was distributed to the public baths.146 From the castellum, an underground pipe led the water to the different baths, where it was stored in cisterns. A thread wheel—often served by slaves—or a bucket-chain wheel could be used to raise the water to elevated tanks.147 Sometimes a separate branch of the aqueduct was conducted directly towards the reservoirs of the baths, without passing through a castellum. This was the case for the large imperial baths in Rome, such as the Aqua Antoniniana, a branch of the Aqua Marcia that fed Baths of Caracalla.148 Inside the cisterns and res140  Wilson (2001). 141  Manderscheid (2000a) 484. 142  Bouet (1997) 136. 143  Hodge (1991) 267. 144  Different sources, however, were also of a different quality. Large cistern systems not only allowed the water to be stored, but also enabled the ‘impurities’ to settle. 145  Manderscheid (2000a) 488. 146  Vitr. De arch. 8.6.2. 147  Oleson (1984) resp. 336–47 and 361–64; Nielsen (1993a) 23. 148  Nardelli (2012) 120.

27

ervoirs, the sediment and dirt in the water was allowed to sink. Once the water was stored in the reservoirs, it could be piped to boilers in the furnace rooms, where it was heated for the alvei, or it could be led directly to the large cold-water pools (natatio and piscinae). Waste Water Disposal We have no idea how often the water of the pools was changed. Furthermore, cold water pools could be refilled with water coming directly from the reservoir or aqueduct, while the water of the warm pools had to be heated in boilers first. It is likely that the water was changed at least once a day, as the absence of chemicals such as chlorine derivatives must have left the water unclean.149 With all the water splashing over the edges when someone entered or exited, the pools must have emptied rather quickly and hence refills had to be frequent. Since the earliest public baths, the easiest system to empty the plunges was through a drain at the bottom of the pool that gave out straight onto the floor of the room.150 The gentle slope in this floor meant that water could be directed towards the street side of the building, where it would exit through the wall and into the street.151 Once larger pools were built, the drainage system was also enlarged. A network of drainage channels ran underneath the floors and collected the water of the alvei, piscinae and the natatio. These drainage pipes were made of terracotta, lead or sometimes even wood and started at the bottom of the pools. They were closed off with metal stoppers or a system of hinges and moveable lids.152 The drainage network was often connected to the sewers of a city. Before dumping the waste water into the sewers, the drainage channels diverted the water of the large pools, such as the natatio and the piscina, through nearby latrines to flush out the excrement that had collected at the bottom.153 The waste water could also be used to clean out the ashes that had collected under the suspensura.154 Decoration The differences in decoration and luxurious appointment of the buildings must have been significant from 149  Blyth (1995) 3; Manderscheid (2000a) 510; Maréchal (2017) 181–82 . 150  For early examples, see the baths of Megara and Syracuse (both 3rd c. BC.; Broise (1994) 19–22). 151  Manderscheid (2000a) 503. 152  Manderscheid (2000a) 502. 153  Nielsen (1993a) 24; Van Vaerenberg (2011) 78–79. 154  Maréchal (2017).

28 bathhouse to bathhouse, depending on the wealth and the wishes of the benefactors. Ancient texts and archaeological remains make clear that the first bathhouses were rather small and dark (see above p. 13). It was no coincidence that the full development and embellishment of the public bathhouses happened during the High Empire. The prosperity of the pax Romana meant that the elite invested in large building programs, which entailed an unprecedented evolution of technology and architecture, adorned with imported construction materials from long-distance trade. The evolution of decorative schemes in general was obviously intertwined with the technical and architectural development.155 The large thermae in Rome and elsewhere in the empire were always lavishly decorated. Built as a gift to the people, how could they not be? The floors could be paved with marble slabs or mosaics. Especially from the 1st c. AD onwards, mosaics in the Italian Peninsula were often in black and white, representing water-related scenes and figures, such as Venus, Neptune or dolphins.156 In many cases, the topic of the mosaics has been used to name the excavated baths. In the palaestra or auxiliary rooms, floor mosaics depicting athletes were also common, e.g. in the Baths of Caracalla in Rome or the Baths of Constantine in Aquileia (C1). The walls could be adorned with paintings, mosaics, stucco mouldings or marble slabs. The themes could directly or indirectly be related to water, e.g. sea creatures or lush garden scenes with streams and rivers, but also gladiatorial scenes and venationes.157 In general, decorative themes on walls and floors followed the popular decorative trends and styles of that time. Other works of art, such as freestanding sculptures and busts (of benefactors, the emperor, the imperial family, mythological scenes), must have given the thermae a ‘museum-like atmosphere’.158 It suffices to look at the collection of marvellous statues named after the Farnese family and found in the Baths of Caracalla in Rome to understand to what extent care was taken to impress the bathers. The frigidarium and other large halls were the most popular locations to place these statues. Sometimes the statues of benefactors, the imperial family or of gods—especially statues of the Muses—were placed in prominent locations such as the exedrae in the gardens or around the palaestra.159 The ‘educative’ function of the garden and the peripheral structures was thus emphasized. We could perhaps even 155  Colledge (2000). 156  Manderscheid (1994) 64; Dunbabin (1989) 21–32; Weber (1996) 139–40. 157  Dunbabin (1989) 29–30; Manderscheid (1994) 64–65. 158  MacDonald (1986) 211. 159  Weber (1996) 146; Manderscheid (1981) 34.

chapter 1

imagine a sort of ‘garden architecture’. Another decorative category that has already been mentioned, was the luxurious water use found in nymphaea and fountains (see above p. 18). The smaller neighbourhood baths could also have elaborate decorative schemes. The floors could be covered with mosaics or slabs of marble or limestone. The wall paintings were often the work of local artisans. Occasionally, a marble statue or bust of a divinity or local benefactor may have been present. Some baths, however, were hardly decorated, with those at military forts a prime example. These were often strictly utilitarian facilities with only the standard cold-tepid-hot rooms. Floors could be paved with simple opus spicatum or large slabs of the local stone. The walls were covered with a fine but undecorated plaster. Some decorative elements have not survived in the archaeological record, but must have been present in both the large and small bathhouses.160 Wooden objects, such as doors, benches or cupboards, could have been sculpted, thereby surpassing a merely functional role. Other functional objects that could have contributed to the embellishment were suspended bronze oil lamps, water spouts from the pools shaped as an animal’s head, large braziers, textile curtains and cushions.161 Popularity of Bathhouses One of the most popular benefits ascribed to bathing was its therapeutic and preventive effect. Bathing was good for the health.162 Besides the purifying power of water itself,163 the alteration of cold and hot was good for the body. Greek physicians in the Hippocratic text 160  Hoss (2018). 161  Water spouts and other ornamental objects are mentioned in P-71 (see appendix 2). Curtains and benches are mentioned in the Talmud Yerushalmi (Baba Batra 4.6/1.14c). For small finds and furniture in Roman baths, and the related problems with their interpretation, see Hoss (2018); Whitmore (2018). 162  It should not be forgotten that Roman baths were not the clean environments we might associate with places of corporal hygiene according to modern standards (Fagan (2000) 281). There were no disinfectants to add to the water, which could entail serious risks of bacterial infection (Gourevitch (2011) 87–92). Especially in cold water the threat was real. Temperatures above 50° C could kill most of the bacteria (Page (1993) 2–3). Inactive water also increased the risk of infections. As mentioned above, it is not known how frequently the water in the pools was renewed. There are indications that the sick bathed with the healthy (Scobie (1986) 426), so infectious deceases may have spread in the communal pools (Fagan (2000) 281). 163   Guérin-Beauvois (2007) 94–99.

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

corpus (5th c. BC) had already prescribed cures with hot and cold water.164 During the Hellenistic period, more sophisticated treatments involving bathing developed.165 It is probably not a coincidence that we see an increase in both public and private Greek-style bathing facilities in this period.166 Asclepiades of Bithynia, a famous Greek physician working in late 2nd c. BC and said to be the man who introduced Greek medicine to Rome,167 apparently prescribed a therapy with warm and cold water.168 Both Celsus and Pliny the Elder frequently mentioned bathing in their medicinal treatises.169 Whereas Celsus adhered more to Greek traditions, combining a healthy diet, physical exercise and bathing, Pliny prescribed bathing always in combination with medication.170 The influence of these popular medicinal theories on the masses is sometimes seen as a catalyst for the popularity of baths in the 1st c. AD.171 By the time of Galen (2nd c. AD), the popularity of bathing was at its summit. Critical voices doubt the impact of these doctors, as “the less than assured social position of the ancient physicians would have limited their ability to affect popular bathing habits.”172 Yet, it is clear that the prescriptions of these doctors, even if they only followed the trend of an existing bathing culture, must have reinforced the idea of ‘bathing as beneficial for health’. The fact that statues of Asclepius and Hygieia are frequently found at bathhouses also points to a strong link the Romans made between health and bathing.173 The belief that bathing was good for health was one of the driving forces that ensured that the bathing habit persisted for so long, even as other forms of public entertainment such as gladiatorial games had already disappeared.174 It is exactly these ideas about health and hygiene—described by Georges

164  Villard (1994); Flemming (2013) 23–26. 165  Nutton (2004) 128–54. 166  Yegül (2013) 77. In the catalogue of the 70 known Greek public baths, 65 date from the Hellenistic period (Fournet et al. (2013)). We should nevertheless keep in mind that the fragmentary nature of the written medical sources from this period limits the validity of any premature conclusions (Flemming (2013) 27). 167  For traditional Roman medicinal practices, see Scarborough (1969) 19–25, 52–60; Gourevitch (2011) 14–18. 168  Fagan (1999a) 96–103. 169  Jackson (1999) 108–110. 170  Fagan (2006) 195–98. 171  Heinz (1996) 2424–25; Yegül (1992) 345–55; Fagan (1999a) 97–103. 172  Flemming (2013) 30. 173  Dunbabin (1989) 30. 174   Guérin-Beauvois and Martin (2007) 1–11.

29

Vigarello as “l’imaginaire du corps”175—that even today are the prime movers for changes in bathing culture.176 The Romans obviously considered their bathing habit as very Roman and very sophisticated. Attributing the invention of the hypocaust to a Roman citizen (Sergius Orata) can be seen as an exponent of this mentality. Bathing ‘the Roman way’ was an expression of being culturally Roman, while taking care of one’s body was a social obligation, an officium, for Roman citizens, especially those living in the city.177 The fact that Tacitus described how the Britons were ‘Romanised’ by giving them baths is telling.178 This brings us to another point, namely that the bathing habit was used to distinguish Romans from ‘barbarians’. Comments such as this one by Tacitus are often cited by modern scholars in defining the Roman bath as a “key marker for Romanization.”179 In the introduction of her work on Roman and Jewish baths in Palestine, Stefanie Hoss states that the main subject of her research is examining the ‘Romanisation’ of Palestine through the introduction of Roman-style bathhouses.180 Some authors go as far as claiming that the absence of grand imperial thermae must be seen as a failure of Romanisation and imperial propaganda.181 However, one needs to be cautious when interpreting early provincial bathhouses as clear signs of acculturation. First of all, there is a problem of the definition and validity of the term ‘Romanisation’. Too often, it has been used to describe a one-way process of indigenous people taking up Roman customs. Post-colonial thinking has inspired more complex definitions, taking into account the speed of transmission, the nature of the reception and of the recipient and the social and economic levels of this transmission.182 For baths, as with other public buildings, it is not clear if non-Romans made use of these facilities immediately after these were introduced. It is questionable if the indigenous inhabitants would have easily switched to Roman practises. And if 175  Vigarello (1985) 11. 176  The evolution of post-classical bathing habits in Western Europe is described by Vigarello (1985) and de Bonneville (1998). As an example, the recent popularity of showers at the expense of bathtubs can be interpreted as a modern bathing habit that prefers a quick but thorough wash that meets the expectations of our ‘ideal germless society’. The first showers were installed in prisons and army barracks in the second half of the 19th c. to enable several people to have a quick wash using only a small amount of water (Vigarello (1985) 236). 177  Blonski (2014a) 124–49. 178  Tac. Agr. 21. 2. 179  Nielsen (1999) 35. 180  Hoss (2005) 1. 181  Lenoir (1995) 69. 182  Lomas (1995) 109; more recently Versluys (2014).

30 not, would we consider them ‘less involved’ in the new Roman cultural sphere? Greek-style bathing habits persisted in Greece and Egypt long after the introduction of Roman baths. Should the people using these facilities be considered ‘less Roman’ than their neighbours going to the thermae? Besides the importance for health, appearance and ‘Romanitas’, bathing was above all a pleasurable experience, just as it is today. Who does not enjoy a warm bath? ‘Hic corpus reparans mentemque relaxans’ rhymes an epigram about baths.183 What distinguished the Roman bathing habit from previous bathing habits was the social and cultural importance this phenomenon acquired in society. One did not go to the baths only to clean oneself. A visit to the baths meant meeting up with friends, hearing the latest news and gossip, forging social networks, relaxing and enjoying ‘state of the art’ technology while taking care of one’s health, all at the same time. The entrance fees were kept low and free visits were often offered by the wealthy elite as acts of munificence.184 The popularity of public baths also had its benefits for the surrounding social and economic activities. That bathing was such a pleasurable experience was made possible by the services on offer and the pleasant environment in which it took place. As mentioned above, the architecture and decoration of Roman bathhouses ensured that even the most humble servant could have a taste of luxury. Seneca complained that his contemporaries could not bathe without getting a tan by having the sun shining through the large windows.185 When Lucian described the imaginary baths of Hippias, he emphasized the quality of the building materials and the lighting.186 Similarly, Statius sings ‘of the baths that shine with bright marbles.’187 Several inscriptions mention explicitly that the benefactors not only paid for the construction of the baths, but also for their embellishment. The archaeological data confirms these luxurious decorations, whether it is the fine mosaics adorning the piscina of a modest neighbourhood bath or the large granite columns of imperial thermae in Rome. The bathhouse itself also had a strong symbolic connotation. In the eyes of the Romans, the baths were a perfect symbiosis of the opposite elements fire and 183   A L 108. 184  Nielsen (1993a) 131–35. For the province of Egypt, it has been calculated that merchants and public servants could visit the public baths once every two days (Faucher and Redon (2014) 850). Even if these figures pertain to the 2nd and 1st c. BC., there is no reason to assume that this frequency would drop during the High Empire, when the number of baths increased. 185  Sen. Ep. 86 186  Lucian, Hippias 6–8; Yegül (1979) 120–29. 187  Stat. Silv. 1.5.

chapter 1

water. Capturing, canalizing and storing water combined with mastering the fire to heat it was a victory of the Romans over nature.188 An inscription from Phazemon (Pontus, Turkey) commemorating the restoration of a bathhouse, mentions how the nymphs and Hephaistus are awaiting the grand re-opening.189 The personification of the elements through gods (fire = Hephaistus), and mythological creatures, e.g. nymphs with water, is a recurring feature in inscriptions and epigrams. The fact that water ‘gushed out of statues’ and floors and walls radiated heat, without the bather ever seeing the technological achievements that were literally behind the scenes, must have given the Romans a sense of pride, even if in daily life few will have given it much thought. Earlier Research on Late Antique Baths Late Antique Baths in General Works on Roman Baths and Bathing As mentioned in the introduction of this book, the bathhouses of Late Antiquity have only enjoyed limited interest in the archaeological research. The common opinion, as voiced by Julian Marcuse at the start of the 20th c. was that the Roman bathing habit and the building type that went with it had disappeared together with the High Empire. This view persisted in general works on Roman baths. In Daniel Krencker’s seminal work Die Trierer Kaiserthermen of 1929, the author refrained from discussing late antique baths as a specific category. The Imperial Baths at Trier, although late antique in date (Constantinian), were compared to a wide range of bathhouses belonging to different periods and geographical regions.190 As Krencker stood at the commencement of research into Roman baths, we can forgive that he overlooked the chronological evolution of this building, a fact that he himself admitted: “Ich bin mir bewusst, eine rein historische Entwicklung der Thermenanlagen nicht herausgeschält zu haben.” (“I am aware that I did not present a full historical overview of the evolution of Roman bathhouses.”)191 The importance Krencker’s work had for Roman archaeology in the German-speaking world is comparable to Angelis d’Ossat’s work for Italian archaeology. His book, however, focused on the well-preserved baths in the Vesuvian cities and in Rome, without touching upon the ‘post-Roman’ period.192 The evolution of the Roman-style bath was described by Krencker’s student 188  Zajac (1999) 101. 189   S EG IV.729. 190  Krencker et al. (1929) 174–304. 191  Krencker et al. (1929) 305; translation by the author. 192  Angellis d’Ossat (1943) 29–32.

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

Erika Brödner, although she saw the Baths of Trajan in Rome as the apex of the building type, with subsequent constructions only perfecting the basic design (e.g. the Baths of Caracalla) or being inferior derivatives (e.g. the baths built after the reign of Diocletian).193 Werner Heinz considered that in Late Antiquity “keine gründsätzlich neuen Ideen mehr verwirklicht wurden” (“no fundamentally new ideas were realised”)194 and therefore ended his research with the period of Constantine. The publication of Inge Nielsen’s Thermae et Balnea (1990, revised edition in 1993) exemplified the interest in the archaeology of Roman baths that existed at the end of the 1980s and the beginning of the 1990s.195 However, even in Nielsen’s work, the detailed architectural analysis of the evolution of Roman baths stops at the beginning of the 4th c.196 Only the survival of ‘Byzantine baths’ in the eastern empire is briefly touched upon.197 The baths of Late Antiquity, according to Nielsen, are characterized by a reduction in size of the entire building and of the heated parts in particular, as well as a preference for small apsidal pools. This reduction in scale was “because the towns were depopulated, because the water supply was cut, and finally because there was no money for fuel.”198 Fikret Yegül on the other hand, devoted more attention to baths in Late Antiquity and the Byzantine world.199 For the Turkish scholar, the causes of the gradual disappearance lay in the “inability of the secular organization to maintain the vast and complicated water-supply system during this period of economic and political instability.”200 In the archaeological component of his chapter, Yegül only focuses on the eastern empire. The late antique baths of Syria and eastern Turkey constitute the main core of his body of evidence. The reduction of the size of the entire building, the disappearance of the palaestra and the preference for round and apsidal rooms and niches are mentioned as the most important characteristics of late antique baths.201 It is striking that in Hubertus Manderscheid’s bibliography of bath-related research,

193   Brödner (1983) resp. 72–74 (Trajan), 232 (Caracalla), 264 (post-Diocletian). 194  Heinz (1983) 122. 195  See among others Pasquinucci (1987); Manderscheid (1988b); DeLaine (1988); Manderscheid (2004) and the contributions in Anonymous (1991). 196  Nielsen (1993a) 55–57 (for Italy), 92–93 (for North Africa). 197  Nielsen (1993a) 115–16. 198  Nielsen (1993a) 56–57. 199  Yegül (1992) 314–49. 200  Yegül (1992) 315. 201  Yegül (1992) 324–39.

31

covering the period between 1988 and 2001, no entries on late antique baths in general can be found.202 Late Antique Baths in Regional Studies After the wave of all-encompassing studies on Roman baths, the 2000s saw an increase in the number of regional studies. As the geographic context was now delimited, more attention was paid to the chronological evolution. If Andrew Farrington’s research on Roman baths in Lycia still offered too little well-dated material to make any conclusions about bathing habits after the 3rd c. AD,203 Alain Bouet encountered a revival in the construction of bathhouses in Gallia Narbonnensis during the 4th c., after an apparent drop in the 3rd.204 The newly-built baths are smaller and have a reduced heated section in comparison to the cold rooms. During the 5th and 6th c., no new public baths were built in this region, with the only small and strictly utilitarian baths belonging to ecclesiastical property.205 Yvon Thébert remarks for late antique North Africa: “on vit dans une large mesure sur un héritage que l’on se contente d’enrichir d’édififces mineurs” (“they continued to use the inherited baths and just added some smaller buildings.”)206 Besides the obvious reduction in size of the newly-built bathhouses, Thébert also noticed how the basic components (cold, tepid and warm rooms) remained in place, even if the heated section became smaller.207 Maria Pilar Reis pointed to similar evolutions for the 4th c. baths in Lusitania, stressing “la riqueza de formas” (“a richness in forms.”)208 The majority of newly-built baths in this province seems to have been attached to large rural villas. While studying the early hammams of Andalusia (Spain), Caroline Fournier briefly summarizes the evolution of Roman baths, noting the increased importance of the cold section and the resilience of baths related to ecclesiastical complexes.209 For the ‘eastern’ part of the empire, studies into the origins of the hammams often included a ‘prologue’ focusing on the Roman / Byzantine origins of public baths. For Palestine, the work by Martin Dow includes a chapter on ‘The Origins of Early Hammams’.210 The author recognizes some of the fundamental differences between Roman and Islamic baths (pools, windows), 202  Manderscheid (2004). 203  Farrington (1995) 80, 119. 204  Bouet (2003a) 339–41. 205  Bouet (2003a) 342. 206  Thébert (2003) 424; translation by the author. 207  Thébert (2003) 425–26. 208  Reis (2004) 43. 209  Fournier (2016) 26–59. 210  Dow (1996) 32–38.

32 but also mentions the evolution of the layout that had already started in the Byzantine period.211 Stefanie Hoss’ research into baths in Palestina showed that the late antique buildings often had combined frigidariaapodyteria, round or apsidal pools and sometimes a courtyard with a pool or fountain.212 For Syria and the Near East, we are still lacking a general work on Roman baths, although the research published by the Balnéorient project has now provided us with an important framework. In their study about the historical hammams of Damascus, Michel Écochard and Claude Le Cœur linked the origins of Islamic baths to a specific tradition of ‘Byzantine’ baths.213 The Roman-style baths in Egypt were the subject of a Balnéorient conference, the proceedings of which have been published recently.214 In previous articles, Redon had already remarked the high number of late antique baths in Egypt.215 A new summarizing article counts at least 45 late antique (‘Byzantine’) baths.216 The authors identify some regional trends in bath building, such as the small size of the baths, special high-flame furnaces, subterranean service corridors, and the phenomenon of double baths. Moreover, they frame this Egyptian evolution in a wider Mediterranean context, noting the increased importance of multipurpose cold rooms, the disappearance of palaestra and the decrease in size of the hot rooms and pools.217 Specific Studies on Late Antique Baths Publications focusing specifically on late antique baths are rather scarce. Some authors focused on early Christian baths, especially in relation to the sacrament of baptism.218 However, these works mainly assembled literary attestations of baths and bathing in the works of the Early Church Fathers. One of the earliest—and possibly still one of the most important— contributions on the archaeological evidence was an article by René Ginouvès on the late antique phases of three Roman-style bathhouses in Greece.219 The French scholar noted the appearance of small single person baths tubs and tried to explain this new feature in the light of changing bathing habits. In 1982, Albrecht Berger published his doctoral thesis on the Byzantine baths of 211  Dow (1996) 34–35. 212  Hoss (2005) 57–65. 213  Écochard and Le Cœur (1943) 125–28. 214  Redon (2017a). 215  Redon (2012a); Redon (2012b). 216  Fournet and Redon (2017a). 217  Fournet and Redon (2017a) 301–309. 218  Dumaine (1910); Zellinger (1928); more recently Synek (1998); Heinz (2012). 219  Ginouvès (1955).

chapter 1

the eastern empire, in which he relied mainly on literary evidence. In the ‘material’ component of his work, however, he stressed the continuity of the Byzantine baths with their Roman forerunners.220 Marked differences with the ‘classical baths’, were the small size of the entire building and the fact that the frigidarium changed into a multi-purpose hall.221 More recently, the excavations of several late antique baths houses in eastern Turkey and the Middle East have resulted in a number of summarizing articles. For Cilicia and Syria, the disappearance of the palaestra, the transformation of the frigidarium in a multi-purpose lounge-area and the waning importance of pools are mentioned as characteristically late antique.222 In an answer to Fikret Yegül’s chapter on Byzantine baths and hammams in the Near East, Gerard Charpentier subdivided the late antique baths of Syria in two distinct groups, which he called the Apamaean and the Antiochene models.223 The former seems to have been a local model, partially used for bathing, partially for other communal events, and developed in a specific chronological and geographical framework. The latter followed larger regional trends in the Near East and was only used for bathing. More importantly, the author also framed the evolution of public bathhouses during the Byzantine period within the larger evolution of bathing habits that would later lead to the development of the hammam.224 Recently, the work by Thibaud Fournet convincingly demonstrated the resilience of bathing habits and even of imperial type baths in the Near East.225 A colloquium entitled ‘Byzantine and Medieval Islamic baths and their use 600–1200 AD’ organised by the German Institute and the University of Freiburg in Rome in 2014 also focused on the continuity of bathing habits from Late Antiquity into the Middle Ages. For Palestina, a ‘southern type’ of bath is identified during Late Antiquity. It was characterized by an open courtyard with a cold pool, combining the function of frigidarium and apodyterium in the hot summer days. For colder days, there was, however, still a small dressing room inside. The heated rooms were laid out on a single axis parallel to the long side of the courtyard.226 The late antique baths in Egypt, Palestine, Syria, Lebanon 220  Berger (1982) 85–102; also Berger (2011). 221  Berger (1982) 90. 222  Yegül (2003) 59–64; Charpentier and Denoix (2011) 88–90; Fournet (2012) 332–34. However, such lounge areas already existed in the large imperial type baths and in some smaller luxurious baths, as mentioned by Lucian, Hippias 5. 223  Yegül (1992) 314–51 commented by Charpentier (1995). 224  Charpentier (1995) 228, 233; also Charpentier and Denoix (2011). 225  Fournet (2012a); Fournet (2012b). 226  Hoss (2010) 173–74.

General Introduction to Roman Baths and Earlier Research on Late Antique Baths

and Jordan have been studied in the multi-disciplinary, multi-period research project Balnéorient.227 Several important conferences were organised, focusing on public baths and bathing habits in Egypt (Alexandria conference in 2006), Jordan (Amman conference in 2008) and Syria-Lebanon (Damascus conference in 2009). The conference proceedings are an invaluable contribution to our understanding of evolving bathing habits in the Middle East.228 Besides important articles on individual late antique baths, some authors summarized the evolution of baths and bathing habits in Late Antiquity. Denoix stressed the gradual transition from Roman-style baths to hammams in Syria and Egypt.229 Ignacio Arce, while discussing the Umayyad baths of Amman and Hammam as-Sarah, summarizes the evolution of the social role of baths in Syria and Palestine from the Hellenistic to Early Islamic period, mentioning the growing importance of the cold section in social interaction and the reducing size of the heated rooms.230 Berangère Redon and Thibaud Fournet provide us with an interesting summary of late antique baths in Egypt, stressing the longevity of the bathing habit, the peculiar plans (double baths) and unique technological features (high flame furnaces).231 227  See the introduction to the project in Fournet (2012a). 228  Boussac et al. (2009); Boussac et al. (2014); Redon (2017a); contributions on the baths in Jordan were published in the volume of Syria 92. 229  Denoix (2009). 230  Arce (2015) 160–68. 231  Redon and Founet (2017).

33

In Italy, the discovery of the villa baths in Piazza Armerina (Sicily, see Gazetteer) and similar ‘private’ baths had led some scholars to believe that the bathing habit in Late Antiquity took place in the luxurious extra-urban villas of the elite, rather than in the cities.232 When the 12th c. baths of the Crypta Balbi were found in Rome in the 1980s, the excavation’s leading archaeologist Lucia Saguí investigated the link between Roman and medieval baths, with her study relying mainly on written sources.233 Similar studies, framing a specific archaeological site within a wider geographical and chronological context, investigated the late antique bathhouses of a limited region within the Italian Peninsula.234 The common denominator for late antique baths is often their small size and mere ‘hygienic function’. For Italy, research by Francesca Stasolla gathered together much of the (mainly written) evidence, with the focus shifting towards the Early Middle Ages.235 On an architectural level, Stasolla remarks on the continuity of the basic plan of the cold, tepid and hot room.236 On a city-level, only Ostia and the imperial baths of Rome have been the subject of in-depth studies.237

232   Ward-Perkins (1984) 148; Baldini Lippolis (2001) 64. 233  Saguí (1990a); Saguí (1990b). 234  Haug (2003) 204–15; Volpe et al. (2007); Cera (2008). 235  Stasolla (2002a); Stasolla (2002b); Stasolla (2008). 236  Stasolla (2002a) 44–57; Stasolla (2008) 878. 237  Ostia: Poccardi (2006); DeLaine (2006); Rome: DeForest (2013).

chapter 2

Written Evidence of Baths Late Antique Literature Methodology and Approach The sheer abundance of literary sources from Late Antiquity makes it a daunting task to find references to baths and bathing habits. The different backgrounds of the authors, including divergent regional developments within the empire—not only on a political, but also on a cultural and religious level—demand a certain precaution when assembling these various testimonies. Furthermore, several authors turned to writing about events in a long-lost past—often entitled ‘Chronicles’ or ‘Histories’—meaning the veracity and plausibility of certain events, and their citations must be approached with a sound sense of perspective. The fact that some lesser-known authors, especially in Greek, Hebrew and Syriac, have not yet been translated makes the task even more strenuous. Coming to an exhaustive overview of all references is therefore an impossible endeavour. In the following chapter, an attempt is made to present informative references from the most authoritative writers of the period between AD 300 and AD 700 in the whole of the Mediterranean world.1 Contrary to the historical and archaeological evidence, there will be no limitation to a predefined geographical context. This is due to the nature of the evidence, which is often less context-bound than archaeological and epigraphic finds. An author such as Procopius, for example, wrote about several regions of the empire, making it impossible to confine his work to one specific area. For obvious reasons of comparison, the authors will be chronologically grouped according to their own lifespan, even though their narrative might refer to different periods. The logic behind this is that they wrote for contemporary audiences, thus reflecting and anticipating contemporary ideas. The Ancient Hebrew and Syriac sources will only be explored on the margins of the Latin and Greek texts.2 The earliest 1  Latin sources from north-western Europe, such as the 6th c. writings of Caradoc of Llanczarvan will not be included in this research. For the references to baths in the writings produced in these regions, see Dumaine (1910) 93–97. The late antique pas­ siones will not be included either, as these are often restricted to the toponym referring to a bath. The appropriate references can be found in the Lexicon Topographicum Urbis Romae (ed. Steinby, 1999). 2  For an overview of the most important Hebrew texts about baths and bathing, see Hoss (2005) 67–80.

© koninklijke brill nv, leiden, ��20 | doi:��.��63/9789004419421_004

attestations of hammams in Arabic literature are discussed separately at the end of this subchapter. 4th c. Latin Authors Ammianus Marcellinus There is no better author to start with than the man who largely shaped our historical knowledge of the 4th c. Ammianus Marcellinus (AD 330–395) has been called “the last great Latin historian of the Roman empire.”3 A native from Antioch-ad-Orontes (Roman Syria, modern Antakya in Turkey), he was originally educated in Greek, but wrote his Res Gestae in Latin. As a military officer, he served in Northern Italy, Gaul, Germania, Illyricum and Mesopotamia, where he survived the fall of Amida and the failed Persian campaign of Julian. After his military career, he visited Egypt, southern Greece and the Black Sea coast, before settling in Rome (AD 380), where he eventually completed his magnum opus. The Res Gestae consists of 31 books, covering the reign of Nerva up until the death of Valens, thus picking up where Tacitus’ Historiae had stopped. The first 13 books have been lost, meaning the narrative starts in AD 353 and ends with the battle of Adrianople (AD 378). His personal involvement in some of the events compromises the objectivity of his work, although large sequences are mainly based on written sources. Adhering to the traditional religion, he does not openly attack Christianity, but rather fulminates against the debaucheries of the elite in Rome, especially concerning sexuality and sensual pleasures. It is in this context that we find the baths as a setting for displaying such decadence. Tales ubi comitantibus singulos quinquaginta mi­ nistris tholos introierint balnearum, ubi sunt nostri? minaciter clamant.4 When such men, each attended by 50 servants, have entered the vaulted rooms of a bath, they shout in threatening tones: “Where on earth are our attendants?”5 The same impious men parade at thermal baths, where they have difficulty in choosing the best garment out of 3  Matthews (1996a) 73. This entry in the OCD acts as key source for the following synopsis of Ammianus’ life. 4  Amm. Marc. 28.4.9. 5  Translation by Rolfe (1952) 140.

35

Written Evidence of Baths

the huge quantity they brought with them, enough to clothe 11 men.6 However, the baths are not the instigator of such conduct, but merely a setting that enabled it. On other occasions, the baths are cited as an essential element of any great city. When describing Constantius’ visit to Rome in AD 356, the baths are named as one of the wonders that inspired the princeps: Deinde intra septem montium culmina per adcli­ vitates planitiemque posita urbis membra conlus­ trans et suburbana, quicquid viderat primum, id eminere inter alia cuncta sperabat: Iovis Tarpei delubra, quantum terrenis divina praecellunt: lava­ cra in modum provinciarum exstructa: amphithea­ tri molem solidatam lapidis Tiburtini compage, ad cuius summitatem aegre visio humana conscendit: Pantheum velut regionem teretem speciosa celsitu­ dine fornicatam: elatosque vertices scansili suggestu consulum et priorum principum imitamenta por­ tantes, et Vrbis templum forumque Pacis et Pompei theatrum et Odeum et Stadium aliaque inter haec decora urbis aeternae.7 Then, as he surveyed the sections of the city and its suburbs, lying within the summits of the 7 hills, along their slopes, or on level ground, he thought that whatever first met his gaze towered above all the rest: the sanctuaries of Tarpeian Jove so far surpassing as things divine excel those of earth; the baths built up to the measure of provinces; the huge bulk of the amphitheatre, strengthened by its framework of Tiburtine stone, to whose top human eyesight barely ascends; the Pantheon like a rounded city-district, vaulted over in lofty beauty; and the exalted heights which rise with platforms to which one may mount, and bear the likenesses of former emperors; the Temple of the City, the Forum of Peace, the Theatre of Pompey, the Odeum, the Stadium, and amongst these the other adornments of the Eternal City.8 There can be no doubt that the baths still had an important function within a city such as Rome and still defined the urban topography. In two passages, the imperial baths (of Anastasia in Constantinople and of Constantine in Rome) are used to locate other structures within the city.9 Furthermore, when Valens punished Chalcedon (Kadiköy, Turkey) for allegedly plotting against him, he 6  Amm. Marc. 28.4.19. 7  Amm. Marc. 16.10.14. 8  Translation by Rolfe (1950) 249. 9  Amm. Marc. 26.6.14; 27.3.8.

ordered the walls of the city to be torn down in order to build a bathhouse in Constantinople.10 An important defining building in the city of Chalcedon was destroyed in order to erect a different one in the capital. According to the legend, a stone was discovered amidst the rubble that foretold these events. Such links with magical events are not exceptional. The death of Valentinian I (AD 375) was announced by an owl on top of the roof of the imperial baths in Savaria (Szombathely, Hungary).11 A good example of the link between baths and ‘the occult’ takes place in Asia Minor. Visus adulescens in balneis admovere marmori manus utriusque digitos alternatim et pectori, sep­ temque vocales litteras numerasse, ad stomachi re­ medium prodesse id arbitratus: in iudicium tractus, percussus est gladio post tormenta.12 In the bath a young man was seen to touch alternately with the fingers of either hand first the marble and then his breast, and to count the 7 vowels, thinking it a helpful remedy for a stomach trouble. He was hauled into court, tortured and beheaded.13 The fact that the poor boy tried to cure his stomach ache in the baths is of course very normal. The baths were indeed seen as places to cure diseases (see above, chapter 1), but also to prevent them from spreading. And so, we read that the aristocracy in Rome demanded that their slaves be bathed when they returned from inquiring about the health of some ill friend.14 A last remark about the attestations of baths within the work of Ammianus concerns the recurrent stress on the strong link between thermae and the emperors. Besides the passages about the portentous owl and the baths built with Chalcedonian spolia, the emperor-bath relation is best exemplified by the story of an angry mob in Antioch, targeting the unpopular Valens. (..) et Antiochiae per rixas tumultusque vulgares id in consuetudinem venerat, ut quisquis vim se pati existimaret “vivus ardeat Valens” licentius clami­ taret, vocesque praeconum audiebantur adsidue mandantium congeri ligna ad Valentini lavacri suc­ censionem, studio ipsius principis conditi.15

10  Amm. Marc. 31.1.4. 11  Amm. Marc. 30.5.16. 12  Amm. Marc. 29.2.28. 13  Translation by Rolfe (1952) 233. 14  Amm. Marc. 14.6.23. 15  Amm. Marc. 31.1.2

36

chapter 2

(..) at Antioch, in quarrels and riots of the common people, it became usual that whoever thought that he was suffering wrongs shouted without restraint: “Let Valens be burned alive!” and the words of public criers were continually heard, directing the people to gather firewood, to set fire to the baths of Valens, in the building of which the emperor himself had taken such interest.16 On a linguistic level, it is interesting to note how Ammianus frequently uses the word lavacrum next to the traditional terms balneum and thermae to denote baths.17 Already in the 2nd c., Aulus Gellius had used the term to denote a private bathhouse or more probably the pools of a private bathhouse.18 His contemporary fellow writer Apuleius had used the term as a synonym for the act of washing. In the latter meaning, Tertullian (AD 160–240) also used it as a term for the Christian holy washing, i.e. baptism. It is in these 4 meanings—a pool, a bathhouse, the act of washing and baptism—that the word became current in the Latin vocabulary.19 Scriptores Historia Augusta There has been much debate about the authorship, date and veracity of this collection of imperial lives. Written in the style of Suetonius’ De vita Caesarum, these texts recount the lives and deeds of the emperors and usurpers from Hadrian (AD 117–138) to Numerian (AD 282–284). Formerly ascribed to 6 authors in the first half of the 4th c.—the so-called Scriptores Historiae Augustae— the authorship has also been attributed to a single author writing at the end of the same century.20 The aims of the books are not entirely clear, and neither is the religious preference of the author(s). There is, however, a clear objection against hereditary monarchy and decadence.21 The attestations of baths and bathing in the Historia Augusta have been thoroughly studied by Elke Merten, even if her study focuses on bathing practices in general.22 When assessing the work as a document of the 4th c., interesting observations in comparison to other contemporary documents can be made. 16  Translation by Rolfe (1952) 379. 17  Unlike balneum and thermae, the term lavacrum did not have a Greek etymology. It stemmed from the Latin verb lavare (to wash), and was hence closely linked to the terms labrum (wash basin on a pedestal), lavatio (privy) or lavatrina (privy / bathroom). 18  Aul. Gell. Noct. Att. 1.2.2. See Maréchal (2015) 143–44. 19  Apul. Met. 2.19; Tert. Ad ux. 1.6.2. For a more detailed analysis of the evolution of lavacrum, see Maréchal (2015). 20  For the debate, see Johne (2006). 21  Matthews (1996b) 713. 22  Merten (1983).

As in the work of Ammianus, the stress on the link between emperor and thermae is omnipresent. The ‘gift’ of large baths to a city is always praised, even if it concerns otherwise wicked leaders, such as Caracalla (reigned AD 198–217). vixit denique in odio populi diu Antoninus, nomen­ que illud sanctum diu minus amatum est, quamvis et vestimenta populo dederit, unde Caracallus est dictus, et thermas magnificentissimas fecerit.23 For a long time, finally, the people hated Antoninus, and that venerable name was long less beloved, even though he gave the people clothing (whence he got his name Caracallus) and built the most splendid baths.24 Other emperors that erected or restored public baths include Hadrian, Antoninus Pius, Septimius Severus, Elagabalus, Alexander Severus, Gordian III, Aurelian, and Tacitus.25 Alexander Severus (reigned AD 193–211) raised taxes for certain tradesmen such as linen weavers, glass-makers or goldsmiths in order to maintain the baths, both the newly built and the old.26 Tacitus (reigned AD 275–276) even had one of his houses torn down to build baths on that land.27 Although these acts of euergetism allude to events outside our period, it is nevertheless meaningful that the author praises the building and maintenance of public baths as a gift towards the people and as a sign of good government.28 Another imperial action that was considered praiseworthy was opening up a private bathhouse to the general public by Antoninus Pius (reigned AD 138–161).29 It is telling that the exact same action by the ‘wicked’ Elagabalus (reigned AD 218–222) is explained as a ruse, ‘that by this means he might get a supply of men with unusually large organs.’30 Using bathing habits or bath-related actions to distinguish the ‘good’ (pious) emperors from the ‘bad’ is a recurring feature in the 23   S HA Sev. 21.11. 24  Translation by Magie (1921) 423. 25  Hadrian: SHA Hadr. 19.10; Antoninus Pius: SHA Ant. Pius 8.3; Septimius Severus: SHA Sev. 19.5; Elagabalus: SHA Heliogab. 17.8–9 and 30.7; Alexander Severus: SHA Alex. Sev. 25.3–6 and 39.3; Gordian III: SHA Gord. 32.5; Aurelian: SHA Tyr. Trig. 45.2; Tacitus: SHA Tac. 10.4. 26   S HA Alex. Sev. 24.5–6. 27   S HA Tac. 10.4. 28  There is some doubt whether some of the construction recorded in the SHA actually took place or if this was more a literary topos, as a sign of good government (Merten (1983) 24). 29   S HA Ant. Pius 7.6. 30   S HA Heliogab. 8.6; translation by Magie (1924) 123.

Written Evidence of Baths

Historia Augusta. The most cited example concerns the ‘good’ Hadrian, who took his bath with the common people and even presented a slave to an old, poor veteran to assist him in bathing.31 Even if the episode probably never happened, it is important that in the eyes of the author and the readers, it unmistakably pointed to Hadrian’s humble and righteous nature.32 Similarly, measures against mixed bathing for men and women (balnea mixta) are attributed to the virtuous emperors such as Marcus Aurelius and Alexander Severus,33 while ‘bad’ emperors such as Elagabalus reinstated the practice. Whether these measures were actually introduced by these emperors or not is unclear, but the passage shows that by the time the Historia was written, mixed bathing was frowned upon to the point that it could be forbidden by law. Other bath behaviour that was considered improper and thus representative for impious emperors was bathing too long and too often, especially in combination with abundant eating and drinking. Of Commodus (reigned AD 177–192), it is said that he spent his time bathing with 300 concubines, that he took up to 8 baths a day and had the habit of eating while bathing.34 These bathing excesses are also mentioned for Gordianus Junior (reigned in AD 238) and the tyrant Celsus.35 Hadrian, however, denied furlough to one of his staff members because he had heard of the man’s addiction to bathing.36 The shameful bathing habits hit new lows with Elagabalus.

37 using the razor with his own hand—with which he would then shave his beard.38 The ‘decadence of bathing’ as a sign of corruptness of body and mind was nothing new. Suetonius had already derided the extravagant bathing habits of ‘wicked’ emperors such as Caligula.39 The idea that hot baths in particular could be harmful if not used with moderation already existed in earlier times.40 Alexander Severus is said to have preferred swimming in a cold piscina to lingering in the caldarium.41 In a speech, he also called his troops to order. Milites Romani, vestri socii, mei contubernales et commilitones, amant, potant, lavant, et Graecorum more quidem se instituunt. Hoc ego diutius feram? Et non eos capitali dedam supplicio?42 Soldiers of Rome, your companions, my comrades and fellow-soldiers, are whoring and drinking and bathing and, indeed, conducting themselves in the manner of the Greeks. Shall I tolerate this longer? Shall I not deliver them over to capital punishment?43 Tacitus was rarely seen in a bathhouse, but was all the stronger in his old age.44 The private baths of Carinus (reigned AD 283–285) were as cold as the basement rooms.

in balneis semper cum mulieribus fuit, ita ut eas ipse psilothro curaret, ipse quoque barbam psilothro ac­ curans, quodque pudendum dictu sit, eodem quo mulieres accurabantur et eadem hora. rasit et virilia subactoribus suis ad novaculam manu sua, qua pos­ tea barbam fecit.37

balneis ita frigidis usus est, ut solent esse cellae sup­ positoriae, frigidariis semper nivalibus. cum hiemis tempore ad quendam locum venisset, in quo fontana esset pertepida, ut adsolet per hiemem naturaliter, eaque in piscina usus esset, dixisse balneatoribus fertur, ‘Aquam mihi muliebrem praeparastis.’45

In the public baths he always bathed with the women, and he even treated them himself with a depilatory ointment, which he applied also to his own beard, and shameful though it be to say it, in the same place where the women were treated and at the same hour. He shaved his minions’ groins,

The baths which he used were as cold as the air of rooms that are under the ground, and his plungebaths were always cooled by means of snow. Once, when he came in the winter to a certain place in which the spring-water was very tepid—its wonted natural temperature during the winter—and

31   S HA Hadr. 17.5. Bathing with the people was already considered a sign of good character by Suetonius (Tit. 8.2). 32  Merten (1983) 131. 33  Marcus Aurelius: SHA Marc. 23.8; Alexander Severus: SHA Alex. Sev. 24.2. 34  Respectively SHA Comm. 5.4 and SHA Comm. 11.5. 35  Gordianus Junior: SHA Gord. 19.5; Celsus: SHA Tyr. Trig. 29.1. 36   S HA Hadr. 11.6–7. 37   S HA Heliogab. 31.7.

38  Translation by Magie (1924) 169. 39   S HA Calig. 37.1–2. 40  See the remarks by Seneca about Scipio Africanus and hot baths (Sen., Ep. 86). S HA Alex. Sev. 30.4–5. 41   42   S HA Alex. Sev. 53.7. 43  Translation by Magie (1924) 287. 44   S HA Tac. 11.3. 45   S HA Car. 17.4–5.

38

chapter 2

he had bathed in it in the pool, he shouted to the bath-attendants, it is said, “This is water for a woman that you have given me”.46 Clearly, the opinion that hot baths were for women, the weak, and the Greek, while ‘true’ Roman men bathed in cold water, still existed in the late 4th c.47 The negative connotation of baths was not restricted to the softening effect of the heat. There are also several passages in which bathhouses are the setting for murders and executions. Commodus, aged 12, supposedly ordered the balneator (responsible for the baths) to be thrown into the furnace because the water was too cold.48 Both Caracalla and Elagabalus ordered the execution of political adversaries in the bathhouses, while Pertinax (reigned AD 193) was murdered in the bathhouse by men loyal to his predecessor Commodus.49 The baths as a location for execution seemed logical, as people were at their most vulnerable here.50 Such despicable crimes had already been mentioned in the ancient Greek literature and are found throughout the Roman period, e.g. Suetonius (AD 69–122) or Plutarch (AD 46–119).51 Besides the passages in which baths were a catalyst to describe the character of the emperor, the Historia Augusta also teaches us some interesting facts about bathhouses and bathing in general. We learn about the prostitutes that solicited in circuses, the theatres and baths, that eunuchs were employed in the women’s baths of the imperial palace, that slaves served as bath-

46  Translation by Magie (1932) 443. 47  The idea that cold water was more beneficial for the body due to its reinforcing nature was commonly accepted. It is found in many medical treatises, especially in the work of Galen (going back to ancient Greek medicine). Writers such as Seneca had already pointed out the weakening effect of hot baths (Blonski (2014a) 217). See also the work of Oribasius (below p. 44). 48   S HA Comm. 1.9. 49  Caracalla: SHA M. Ant. 4.4; Elagabalus: SHA Heliogab. 13.8; Pertinax: SHA Pert. 12.8. 50  Alexandre Dumas, travelling in the Sinai in the 19th c., experienced the same feeling when visiting a hammam: ‘je voulus me débattre, mais ma résistance avait été prévue; je n’étais d’ailleurs ni en costume, ni en situation favorable pour soutenir la lutte, aussi m’avouai-je vaincu.’ (‘I wanted to struggle, but my resistance had been anticipated; besides, I had nor the clothes, nor a favourable situation to strengthen me in my battle, and thus I declared defeated.’) (Dumas (1855) 14; translation by the author). 51  Suet. Galb. 10.5; Suet. Dom. 17.1; Plut. Vit. Sull. 31.5; Plut. Vit. Crass. 9.4. For passages in the ancient Greek literature, see Blonski (2014a) 182.

attendants and that bathers used a bathing costume.52 Furthermore, the balnea for the plebs were poorly equipped and some thermae remained open even after dark, causing a nuisance during the night.53 In contrast to the information about the building and restoration of several baths in Rome (see above), only two passages mention the decline of baths. The Thermae of Nero were apparently out of use at the time of writing, while the Baths of Septimius Severus (reigned AD 193–211) only functioned for a short time, as the aqueduct that supplied it fell into ruins shortly after its dedication.54 The opposing ideas about public baths that are expressed in the Historia Augusta already existed in the preceding centuries. On the one hand, public baths were a necessary facility for every city as they fulfilled the basic hygienic needs and played an important role in medical treatment. On the other hand, the baths formed the perfect setting to indulge in excesses. It is this corruption of body and mind that is central to the writings of the authoritative Christian author St. Jerome. Jerome (Eusebius Hieronymus) Born to a Christian family in Dalmatia (Croatia), Jerome (lived AD 347–420) studied rhetoric in Rome and stayed in Augusta Treverorum (Trier, Germany), before heading east to Antioch. After a short spell living as an anchorite monk in the desert, he returned to Rome to become a trustee of the archbishop Damasus (around AD 382). Here, he dedicated himself to translating the Bible into colloquial Latin (the so-called ‘Vulgate’), but was forced to leave the Eternal City after his relations with the widowed Paula came under suspicion. He eventually settled with Paula in Palestine, where they founded a monastery and a convent.55 Jerome is one of the authors cited on numerous occasions to demonstrate the Christian ‘resistance’ to the Roman bathing habit, especially his letter to an anchorite monk. Scabra sine balneis adtrahitur cutis? sed qui in Christo semel lotus est, non illi necesse est iterum lavare.56

52  Prostitutes: SHA Heliogab. 26.3; eunuchs: SHA Alex. Sev. 23.5; slaves: Alex. Sev. 42.2; bathing costumes: SHA Alex. Sev. 42.1. 53  Plebs balnea: SHA Gord. 32.5; night nuisance: SHA Tac. 10.2. 54  Thermae of Nero: SHA Alex. Sev. 25.3; Baths of Septimius Severus: SHA Sev. 19.5. 55  Scourfield (1996). 56  Jer. Ep. 14.10.

Written Evidence of Baths

Is your skin rough and scurfy without baths? He who has once washed in Christ needs not to wash again.57 This statement about bathing comes from the New Testament (Joh. 13, 10). Yet, the fact that this letter is addressed to a monk is often overlooked. Rather than an ‘official Christian stance’, these words should be interpreted as a reminder to a very select group on how to overcome ‘the burden of the flesh’. Similarly, Jerome’s disapproval of the bathing habits of widows and virgins fitted within his view of these special categories of Christians. He explains his vision on the bathing habits of virgins in a letter to Laeta, daughter of Paula (dated around AD 403). Scio praecepisse quosdam, ne virgo Christi cum eu­ nuchis lavet, ne cum maritis feminis, quia alii non deponant animos virorum, aliae tumentibus uteris praeferant foeditatem. Mihi omnino in adulta vir­ gine lavacra displicent, quae se ipsam debet erubes­ cere et nudam videre non posse. Si enim vigiliis et ieiuniis macerat corpus suum et in servitutem redi­ git, si flammam libidinis et incentiva ferventis aeta­ tis extinguere cupit continentiae frigore, si adpetitis sordibus turpare festinat naturalem pulchritudi­ nem, cur e contrario balnearum fomentis sopitos ignes suscitat?58 I know that some people have laid down the rule that a Christian virgin should not bathe along with eunuchs or with married women, inasmuch as eunuchs are still men at heart, and women big with child are a revolting sight. For myself I disapprove altogether of baths in the case of a full-grown virgin. She ought to blush at herself and be unable to look at her own nakedness. If she mortifies and enslaves her body by vigils and fasting, if she desires to quench the flame of lust and to check the hot desires of youth by a cold chastity, if she hastens to spoil her natural beauty by a deliberate squalor, why should she rouse a sleeping fire by the incentive of baths?59 The fact that some Christians believed a virgin could bathe—be it with eunuchs or married women—implies that the other Christian women had no specific bathing restrictions. Jerome’s own ultra-puritan views on virgins clearly contrast with common Christian beliefs. 57  Translation by Wright (1933) 51. 58  Jer. Ep. 107.11. 59  Translation by Wright (1933) 363–65.

39 In the same letter, Jerome agrees that little girls should take baths before the start of their marriage to God, thus acknowledging the ‘hygienic’ role of bathing.60 The dangers of bathing clearly lie in the sensual pleasures. There is a strong link between the heat of the baths and the internal heat that arouses passion. The threat of sexual temptation must be avoided at all cost. Balnearum fomenta non quaeras, qui calorem cor­ poris ieiuniorum cupis frigore extinguere.61 Avoid hot baths: your aim is to quench the heat of the body by the help of chilling fasts.62 The difficulty of controlling desire was not an exclusively Christian problem. In the classical cults, sex was thought to exhaust the male body, draining away vital life spirit through every ejaculation.63 Self-imposed abstinence in order not to waste important energy was a recurring phenomenon among the aristocracy, especially the politically active classes.64 However, the anti-sexual attitude of the Christian ascetics and in particular the choice of girls to remain a virgin was something new. This is not to say that the traditional cults were unfamiliar with the idea—the Vestal Virgins being a good example—but the ‘popularity’ (among young Christian women) and ardour (e.g. the self-mortification of the ascetics) was different. Even if this ‘problem of sexuality’—the impossibility to control basic, animal-like sexual desire—was central in Christian doctrine, this did not mean that Christianity in general strove to abolish all man-woman relationships. The family unit remained the core of ancient society and if Christianity wanted to gain any acceptance among the population, it could not afford too severe a rupture with the traditional way of life. The innumerable Christian movements all had their own responses to the ‘problem of sex’ and within most movements, moderation was the rule and abstinence was reserved for a ‘selected few’.65 It is in such a context that the writings of ‘fundamentalists’ like Jerome should be read. He himself is fully aware that he is writing for these ‘chosen few’ and—probably even with a sense of superiority—despises the looseness of the average ‘faithful’.

60  Jer. Ep. 107.8 61  Jer. Ep. 125.7. 62  Translation by Wright (1933) 407. 63  Rousselle (1988) 14–15. 64  Brown (1988) 18. 65  Brown (1988) 54–55 (importance of the family unit), 60–67 (selected few).

40

chapter 2

In ipso quoque convivio, ut vescaris carnibus, quasi invita cogeris, ut vinum bibas, Dei laudabitur crea­ tura, ut laves balneis, sordibus detrahetur; et omnes te, cum aliquid eorum, quae suadent, retractans fe­ ceris, puram, simplicem, dominam et vere ingenuam conclamabunt.66 At the dinner itself they will pretend you are unwilling and will force you to partake of the meat; to get you to drink wine, they will praise it as the gift of the Creator. To induce you to visit the baths, they will speak of dirt with disgust. And when you reluctantly do something of what they would have you do, they will cry out in chorus: “What a frank, innocent girl she is! What a genuine lady!”67 Even if Jerome considers the baths to be among the recurring temptations (including wine, meat, and luxury items)68 that try to prevent widows and virgins of keeping on the right path to God, he still feels obliged to mention the dedication of the large imperial thermae in Rome and selected cities in his Chronicle.69 Apparently, his personal disgust with public baths could not justify omitting these important buildings from a chronicle for a larger public. Ausonius The importance of baths in both the public and private spheres, is stressed in the work of Ausonius (AD 310– 395). This statesman and teacher of Burdigala (modern Bordeaux, France) reached the upper echelons of Roman society as a tutor of Gratian at Trier and later consul in AD 379. Even if his Christian allegiance filters through some of his work, he displayed a lively interest in the traditional pantheon and succeeded in combining Christian values with the classical heritage.70 In his Ordo urbium nobilium, a catalogue of famous cities, Ausonius describes the cityscape of contemporary Milan. Et Mediolani mira omnia, copia rerum, innumerae cultaeque domus, facunda virorum ingenia et mores laeti; turn duplice muro amplificata loci species po­ pulique. voluptas circus et inclusi moles cuneata theatri; templa Palatinaeque arces opulensque mo­ neta et regio Herculei Celebris sub honore lavacri.71 66  Jer. Ep. 117.6. 67  Translation by Wright (1933) 385. 68  See also Jer. Ep. 128.3. 69  Jer. Ab Abr. 2079, 2109, 2204, 2218, 2235, 2245, 2317. 70  Green (1996) 223. 71  Auson. Ordo nob. urb.7, Mediolanum.

At Milan also are all things wonderful, abundant wealth, countless stately houses, men able, eloquent, and cheerfully disposed; besides, there is the grandeur of the site enlarged by a double wall, the Circus, her people’s joy, the massy enclosed Theatre with wedge-like blocks of seats, the temples, the imperial citadels, the wealthy Mint, and the quarter renowned under the title of the Baths of Herculeus.72 The Herculean baths clearly had an important role in the organization of the city, as it gave its name to a whole quarter. The same regard to baths is found in a famous poem about a villa near the river Mosel (Germany).73 Although here the description is of a private bath, it nevertheless emphasizes how indispensable a bathhouse was to a pleasant lifestyle: ‘Such comfort and splendour is attractive, yet its delight breeds no excess’.74 The poem shows strong resemblances to a poem of Statius of the 1st c. AD, meaning the text could be a style exercise in classic Latin hexameters.75 However, the description of the heating system (l. 339: ‘per cava flammas’),76 the elaboration on the different pools (lavacri, lacus, pisci­ narum) and the praise of the bath’s refinement all point to an obvious familiarity with baths and bathing habits. The references to the thermal baths of Baiae near Naples (l. 346) meant the reputation of this health resort was also still very much alive. Theodorus Priscianus and the Pseudo-Theodorus The medical author Theodorus Priscianus lived in the late 4th / early 5th c. and was a pupil of Helvius Vindicianus, a 4th c. physician in North Africa.77 His medical compendium, the Euporista (written around AD 400), comprised three books (Phaenomena, Logicus and Gynaecia) and was originally written in Greek, according to Theodorus himself. The therapies described by Theodorus are inspired by Greek medicine—especially by Hippocratic texts—the Roman physician Galen and the so-called ‘methodist’ physicians.78

72  Translation by Evelyn-White (1919) 273. 73  Auson. Mos. 335–348. 74  Lines 347–348, translation by Evelyn-White (1919) 253. Note again the concern over possible excesses (luxus). 75  Stat. Silv. 1.3.43–57. See Busch (1999) 60–61. 76   ‘flames (..) through the channelled walls’ (translation by Evelyn-White (1919) 251). 77  The medical works written by Vindicianus are now lost, except for some small fragments and quotations. 78  Nutton (2004) 299. The ‘methodists’ believed in a constant process of observation and adapted treatments, without considering the patient’s natural condition (Nutton (2004) 193).

Written Evidence of Baths

Several treatments for small disorders and serious diseases included the use of baths. Special ointments in the bathhouse are recommended for infection to the scalp and against head lice.79 Frequent washing was also necessary when having scabies.80 When dark or white spots appear on the skin, one should rub ointments, natron and alum onto the skin during a visit to the bathhouse.81 The heat of the hot pools was beneficial in cases of hydrophobia and epilepsy, while swimming in the pools of the bathhouse could help overcome atrophy.82 The combination of bathing and drinking medical beverages was one of the key elements in the treatment of jaundice.83 Drinking certain mixtures just after bathing could also help with abortions.84 Another medical treatise, written in the style of Theodorus and possibly dating from the same period, has come down to us under the name of Pseudo-Theodorus. The baths play a pre-eminent role in several treatments, often in a combination of hot water, specific oils and special beverages. In the so-called Additamenta (‘the Accessions’), the treatments for gout, scabies, spasms, jaundice and kidney failure include visits to the baths. For abortion, the combination of ‘softening’ hot baths and drinking special mixtures is recommended.85 Itinerarium Burdigalense The so-called Itinerarium Burdigalense (or Itinerarium Hierosolymitanum, around AD 333), the travel account of a pilgrim from Burdigala to Jerusalem, mentioned the baths of a certain centurion Cornelius, ‘who gave many alms’, in the city of Caesarea Maritima (Sdot Yam, Israel).86 A miraculous power was ascribed to a spring near Mount Syna ‘where there is a fountain, in which, if a woman bathes, she becomes pregnant.’87 Throughout the Itinerarium, the mysterious power of water is omnipresent. In the valley at the foot of Mount Sion, a spring ran into two pools (one had 4 porticoes) 6 days a week, but on the 7th day (Sabbath), it did not run at all. At the ‘fountain of Elisha’ near Jericho (Israel), there used to be a spring that made women infertile. The prophet 79  Scalp: Theod. Prisc. Phaen. 2; head lice: Theod. Prisc. Phaen. 5. 80  Theod. Prisc. Phaen. 35. 81  Theod. Prisc. Phaen. 36. 82  Hydrophobia: Theod. Prisc. Log. 8; epilepsy: Theod. Prisc. Log. 15; atrophy: Theod. Prisc. Log. 23. 83  Theod. Prisc. Log. 27 84  Theod. Prisc. Gyn. 6. 85  Gout: Pseudo-Theodorus, Ad. 1.34; scabies: Pseudo-Theodorus, Ad. 1.35; spasms: Pseudo-Theodorus, Ad. 2.10; jaundice: Pseudo-Theodorus, Ad. 2.27; kidney failure: Pseudo-Theodorus, Ad. 2.35; abortion: Pseudo-Theodorus, Ad. 3.6. 86  I tin. Burd. (PL 8, col. 790). 87  I tin. Burd. (PL 8, col. 790).

41 Elisha himself is said to have ‘purified’ the source. The healing aspect of water is represented in natural hot baths. A healing facility is found in the city of Jerusalem itself: two twin pools (piscinae gemelares) with 5 porticoes cure people who have been sick for many years.88 It is clear that the author refers to baths fed by natural springs: ‘Aquam autem habent hae piscinae in modum coccini turbatam.’ (‘the pools contain water which turns scarlet when it is disturbed.’). The mysterious powers of this location were reinforced by the remark that there was a crypt there where Solomon used to torture demons. The palace of Solomon is also said to have had large subterranean reservoirs and ‘pools constructed with great care’.89 There were also two large pools on each side of the temple, both made by Solomon. To conclude the visit of Jerusalem, it is interesting to notice how the author described the baptistery adjacent to the basilica of Constantine as a ‘balneum a tergo, ubi infantes lauantur.’ (‘a bath behind it, where children are washed’).90 Other 4th c. Latin Authors There are several other authors of the 4th c. who refer to baths or bathing in their texts, but none really elaborate beyond short mentions.91 In the works of Hilary of Poitiers (lived ca. AD 315–367) and Sulpicius Severus (lived ca. AD 363–420/25), no references to baths or bathing have been found. In his Breviarium ab urbe condita, the historian Eutropius (writing around AD 360) mentioned the dedication of the Baths of Nero, which were called the Thermae Alexandrinae in his day, and of the Baths of Caracalla, called the Thermae Antoninianae. He also talks about the lavacrum built by Decius.92 Similarly, Aurelius Victor, an African native who became governor of Pannonia Secunda (parts of modern Croatia, Serbia and Bosnia-Herzegovina) in AD 361 and praefectus urbi in AD 389, wrote a biographical work (De Caesaribus) in which he ridiculed Commodus’ baths in Rome (‘hardly worthy of Roman power’), but mentioned the wonderful baths of Caracalla and of Constantine.93 In the anonymous Epitome de Caesaribus—sometimes attributed to Aurelius Victor—the dedication of thermae by Nero and 88  I tin. Burd. (PL 8, col. 790). 89  I tin. Burd. (PL 8, col. 791). 90  I tin. Burd. (PL 8, col. 791). 91  We should remember that it was not always opportune to talk about baths. 92  Baths of Nero: Eutr. 7.15; baths of Caracalla: Eutr. 8.20; baths of Decius: Eutr. 9.4. 93  Baths of Commodus: Aur. Vict. Caes. 8.3; baths of Caracalla: Aur. Vict. Caes. 21.4; baths of Constantine: Aur. Vict. Caes. 40.27.

42

chapter 2

Trajan94 are mentioned. Just as in the Historia Augusta, the idea of baths as a setting for murder and execution was present. Commodus is said to have been poisoned when he got out of his bath, Julian was beheaded ‘in abditas palatii balneas’ and Constantine killed his wife Fausta by throwing her in the hot baths.95 Vegetius, probably a member of the imperial administration, wrote an account of Roman military practices (Epitoma rei militaris). When talking about recruitment, he considered peasants to be the best possible soldiers, as they were used to the scorching heat of the sun and were not acquainted with the use of baths, nor to the other luxuries of life.96 Just as Alexander Severus’ speech to his soldiers in the SHA, a good soldier should not weaken himself by bathing. In the works of Ambrose, bishop of Milan (AD 340– 397), we find few negative references to public baths. For one of the most influential figures of a religion that is said to be against bathing, this is rather surprising. Ambrose himself did clearly not oppose bathing, as he found it acceptable for Christians to bathe during daytime, but warned for bathing at night.97 In a famous letter to Valentinian II (reigned AD 375–392) against the restoration of the Altar of Victory to the Roman Senate house, he lamented that Christians were constantly confronted with ‘pagan’ images in baths, colonnades and streets.98 Statues, mosaics and wall paintings with depictions of the traditional gods in the baths bothered him, or better, could bother pious Christians. Commodianus (probably 4th c. AD), a poet of strong Christian conviction who attacked the traditional cults as well as Judaism, allegedly wrote the Carmen Apologeticum or Carmen de duobus populis. In the verses 670–674, the Jews are accused of infatuating the common people with baths: Adhuc infatuant profanos balneis ire, Quos faciunt mundos ipsi Deo summo placere. Aqua lauat sordes, non intima cordis iniqua, Nec sacra sacrilega po­ terunt lauare curata.

94  Nero: Epit. de Caes. 5.3; Trajan: Epit. de Caes. 8.6. 95  Commodus: Epit. de Caes. 17.5; Julian: Epit. de Caes. 19.3; Constantine: Epit. de Caes. 41.12. See also Sidonius Apollinaris (Ep. 5.8.2). Philostorgius, writing in the first half of the 5th c., gives a different version. He claims it was his son Crispus who was suffocated in the hot baths after he had an affair with his stepmother Fausta (Hist. Eccl. 2.4). The text is preserved as an epitome by Photius (9th c.). 96  Veg. Mil. 1.3. 97  Amb. In Luc. 7.158 (PL 15.1741). 98  Amb. Ep. 18.31.

As long as they [the Jews] infatuate the common people with baths, they themselves think these worlds please the Almighty Lord. Water washes away the filth, but not the innermost vices of the heart, and the observance of the rites cannot cleanse sacrileges.99 Commodianus clearly dismissed the idea that a simple wash in a public bathhouse could cleanse the spiritual filth. Even if the Jews did seem to have bathed in Roman-style bathhouses,100 Commodianus’ reproaches seem unjustified, as the Jews performed their ritual wash in special bath tubs (miqveh) with pure and unpolluted water.101 An epigram found in the Anthologia Latina is sometimes attributed to Commodianus.102 It incites a certain Florens to bathe in the balnea Pontica, which was fed by the well-odorous waters. It is interesting to note that bathers could bathe in a pool or could simply ‘rigat in pluviam’ (‘douse at the shower’).103 4th c. Greek Authors Eusebius of Caesarea When discussing public baths in the works of Greekspeaking Christian authors in Late Antiquity, the oeuvre of Eusebius (lived ca. AD 260–339) seems a good place to start. As bishop of Caesarea (Palestina), he played an important role in the contemporary ecclesiastical politics of the east. His works can be described as apologetic, defending Christian doctrines against philosophy, with an important influence from Origen (lived AD 185–254). His Chronicle, on which Jerome based his version (see above p. 40), has not survived, but in his Ecclesiastical History there are several references to baths and bathing. A passage about the martyr James claims that he ate no meat, drank no wine, never shaved, nor anointed himself and that he did not use the bath.104 The refusal to bathe is considered unusual, but at the same time a remarkable achievement, only reserved for truly pious men. That Eusebius believed other Christians should not shun the baths is illustrated by the following fragment. Quoting a work of Polycarp, himself repeating Irenaeus (2nd c. AD), the apostle John is said to have entered a bathhouse, but learning that the heretic Cerinthus was inside, he fled 99  Translation by the author. 100  Hoss (2005) 67–80. 101  Only water that had been collected without human interference (e.g. not water carried in buckets) and had not been used for other purposes could be used in the miqveh (Hoss (2005) 109). A L 744R. 102   103  Busch (1999) 344. 104  Euseb. Hist. eccl. 1.23.5.

Written Evidence of Baths

from the building, urging his fellow bathers to do the same: ‘Let us flee, lest the bath fall; for Cerinthus, the enemy of the truth, is within’.105 In a passage about the persecutions under Lucius Verus (reigned jointly with Marcus Aurelius in AD 161–169), Christians were said to have been banned from houses, markets and baths.106 During persecutions under Maximinus (reigned AD 310– 313), soldiers were stationed before the bath entrance ‘in order to defile with the abominable sacrifices those who went to wash in them.’107 The measure presumes that Christians were among those who frequently entered the baths and they may have been troubled when requested to offer to the imperial cult, as statues of emperors were present in most baths.108 Furthermore, in the Life of Constantine, the ‘first Christian emperor’ is credited for building the Church of the Apostles in Constantinople, which was adjoined by a portico, many apartments, baths and promenades.109 The pious emperor also visited hot baths when he was feeling ill.110 When the news of Constantine’s death reached Rome, the baths and markets were closed for a period of mourning.111 Libanius and Julian The rhetorician Libanius (lived ca. AD 314–393) spent most of his life in Antioch (Syria), where he committed himself to education and literature. His admiration for Julian was closely connected to his passion for classical literature and philosophy. In his oeuvre—consisting of speeches, declamations, rhetorical exercises and letters—the references to baths mainly concerned the thermae in Antioch. According to Libanius, the magnificence of the city was the result of its oratory schools, its multitude of temples and its fabulous baths.112 Again, the possible danger lay in excess. The severity of a certain Alexander at the head of the Antioch government is praised, as he exhorted the idle citizens to work rather than spend their days bathing and sleeping.113 The vulnerability while bathing (see above, SHA) is illustrated 105  Euseb. Hist. eccl. 1.28.6. The passage is repeated at 4.14.6 where we learn that the events took place at Ephesus (Turkey). The same story is told by Epiphanius but with a certain Ebion instead of Cerinthus (see below). 106  Euseb. Hist. eccl. 5.1.5. 107  Euseb. Hist. eccl. Mart. Pal. 9.2. 108  For imperial statues at baths, see Brödner (1983) 132–35. 109  Euseb. Vit. Const. 4.59. 110  Euseb. Vit. Const. 4.61. The link of baths and healing treatments is also apparent in Eusebius’ Onomasticon, in which several thermal baths in Palestine are mentioned, including well-known Gadara. 111  Euseb. Vit. Const. 4.69. 112  Lib. Or. 1.11. 113  Lib. Ep. 622.

43 by a passage about soldiers being slaughtered while bathing in a river.114 Libanius also complained about the poor state of the cities’ councils. The lack of personnel was made clear with the example of the person who was responsible for supervising taxes was also responsible for starting up the baths and even had to attend to the bathers’ individual wishes.115 If a person asked for hot water and someone else for cold, he could not help both at the same time and therefore had to bare the insults of at least one of them. It is interesting to notice that in at least some bathhouses bathers could chose the temperature of the water according to their preferences. Julian (reigned AD 361–363), called the Apostate by later Christian writers because of his adherence to the traditional cults, was a correspondent of Libanius. His writings include several letters, some theological works and satires. In a letter dated to AD 358–359, Julian spoke of ‘his first bath’ after a severe sickness,116 while in a letter of AD 361, he recalled his childhood in Bithynia (Turkey) where he enjoyed ‘πηγὰς οὐ φαύλας καὶ λουτρὸν οὐκ ἀναφρόδιτον καὶ κῆπον καὶ δένδρα’, ‘excellent springs and a charming bath and garden and trees’.117 Epiphanius Epiphanius (ca. AD 315–403) was born in Palestina and became an ascetic at a young age and in around AD 366, he was elected bishop of Constantia on Cyprus. During his episcopate, he focused his intellectual energy on combating Greek philosophies and Christian heresies, especially against Origenist factions. This ideological battle is reflected in his works, the Ancoratus and the Panarion (Adversus haereses). Epiphanius was clearly not an advocate of bathing. In his Panarion, he criticized the Gnostics for their ‘culture of the body’ involving anointing, bathing, feasting, whoring and drinking. At the thermal baths of Gadara (Umm-Queis, Jordan), men and women even bathed together!118 Epiphanius saw this as the work of the devil: ‘For where wonders have been given by God the adversary has already spread his deadly nets—the bathing there is mixed!’119 A naive Christian girl was dismayed at the manners of the men, making the cross 114  Lib. Or 1.17. 115  Lib. Or. 2.34. 116  Julian, Ep. 44. 117  Julian, Ep. 46; translation by Wright (1923) 79. Sozomen, writing in the 5th c. (see below p. 55), recounts that Julian was not executed like his father, but confined to a magnificent residence with baths, gardens and fountains when he was just a child (Soz. 5.2). 118  Epiph. Adv. haeres. 26.5.8. For more literary reference on the thermal baths of Gadara, see Dvorjetski (2007) 355–59. 119  Epiph. Adv. haeres. 30.7.5.; translation by Williams (2009) 136.

44 gesture to ward off unsavoury characters. She should not have bathed with men.120 The critique on these bathing habits fits within Epiphanius’ attacks on the Jewish bathing rituals. The ‘hemerobaptists’ did not recognize the power of baptism, as they bathed daily to get rid of their sins. Similarly, the Ebionites washed after each possible contamination.121 The heretic Ebion featured in a passage about St. John, who was called to the public baths by the Holy Spirit to find Ebion there.122 The story is the same as the one told by Eusebius and seems to have been a popular way of discrediting clerics. The alousia—as found with such holy men as Joseph, James or monks—was considered a wonderful virtue.123 Furthermore, Joseph found the shell of an unfinished temple at Tiberias (Israel) and, instead of making it into a bath, as the inhabitants had wished, made it into a church.124 Incidentally, Epiphanius mentioned that the Manichaeans prohibited bathing, but does not seem to find it praiseworthy.125 Oribasius Oribasius (lived ca. AD 320–400) was a physician of Pergamum (Turkey) who studied medicine in Alexandria. He was the personal physician of Julian and followed him in his campaigns to Gaul and Mesopotamia. An adherent of the classic cult, Oribasius was exiled after Julian’s death, only to be reinstated by Valens. His writings included a collection of excerpts from Galen— now lost—and the encyclopaedic Ἰατρικαί συναγωγαί (Collectiones Medicae), a collection of the writings of earlier medical authors dedicated to Julian (about one third survived). Contrary to the North African physicians writing at the end of the 4th and beginning of the 5th c., Oribasius limited his work to assembling these texts, without adding his own opinion. The work nevertheless reveals the medical practices and theories of the mid-4th c. Baths and bathing are included in many therapies mentioned in the Collectiones, from curing simple cramps to combating elephantiasis.126 Often bathing is prescribed in combination with a specific diet and phys120  Epiph. Adv. haeres. 30.7.6. 121  Hemerobaptists: Epiph. Adv. haeres 17.2.2; Ebionites: Epiph. Adv. haeres. 30.2.5. 122  Epiph. Adv. haeres. 30.24.2–5. 123  Resp. Epiph. Adv. haeres. 78.13.2; Adv. haeres. 78.14. 2 and De fide 23.6. 124  Epiph. Adv. haeres. 30.12.2. 125  Epiph. Adv. haeres. 66.28.5; also McDonough (2009). On the interdiction of bathing in the Manichean cult, see also the works of Augustine (below p. 49). 126  Cramps: Or. Coll. Med. 8.6.32–33; elephantiasis: Or. Coll. Med. 45.29.42–44.

chapter 2

ical exercise such as walking or swimming.127 This is also underlined in the book ‘About a salubrious lifestyle’.128 Our most important source about the medical use of baths, however, comes from the book entirely devoted to baths. It is subdivided into several chapters regarding bathing in ‘normal water’ (book 10.1 after Galen), bathing in ‘artificial water’ (book 10.2 after Antyllus), bathing in ‘mineral water’ (book 10.3 after Antyllus and 10.5 after Herodotus), steam baths (book 10.4 after Herodotus), cold baths (book 10.6 after Galen) and the difference between hot and cold baths (book 10.7 after Agathinus).129 At the beginning of these chapters, it is explained what a complete visit to the bathhouse should be like.130 First one should stay in the heated rooms, which harmonizes the body and enables bad fluids to exit the body. Secondly, one should submerge oneself in the heated water of the pools, which introduces beneficial warmth into the whole body. Thirdly, the bather should dive into the cold-water pool in order to contract the skin and close all the pores. Fourthly and lastly, he should rub himself dry with linen and anoint with oil, removing the bad fluids from the skin and making the body soft and tender.131 This succession was based on the primary properties of hot and cold water. Hot water has the power to humidify and soften, while cold water has a contracting and reinforcing power.132 Hence it is of great importance that a person knows which type of bath he should take and for how long. The task of the physician is to prescribe the appropriate bath therapy according to the affliction, the nature of the patient, his habits, his age, the season, country of residence and the air of the region.133 Such considerations were already made by Greek physicians, in particular in the Hippocratic tradition.134 Besides the temperature, the type of water is also considered important. We find 7 categories, all with their specific properties: natural mineral water, alkaline water (against heartburn, tumours, dropsy), aluminous water (against haemorrhages, vomiting, irregular menstrual periods, haemorrhoids), sulphurous water (against all sorts of pain), bituminous water (heating effect), vitriolic water (for rinsing mouth and eyes) and ferruginous water (against stomach aches).135 It is crucial not to disturb these types of water, as sprays and vapours may 127  Or. Coll. Med. 7.2.19; 7.6.5–6; 8.1.39; 8.6.5. 128  Or. Coll. Med. 54.22.19–23. 129  All these physicians were of Greek origin, living between the 1st and 3rd c. AD. 130  Or. Coll. Med. 10.1.4. 131  See also Bourdy (1992) 31. 132  Or. Coll. Med. 10.1.14–15. 133  Or. Coll. Med. 10.1.16. 134  Villard (1994) 50–51; Flemming (2013) 24. 135  Or. Coll. Med. 10.3.3–8.

Written Evidence of Baths

cause nausea. It is therefore advised to simply submerge oneself in these waters without making movements.136 Taking the properties of the temperature and the type of water into account, the physicians should prescribe different therapies for different age groups and health categories (pregnant women, person with a weak health, etc.). Small children are too often bathed in warm water, as wet nurses find it easy to calm toddlers in this way. Such a habit makes the children soft and vulnerable. Young men should prefer the cold baths, starting with physical exercises and limiting the stay in the heated rooms to a minimum. Old and feeble men, on the contrary, should resort to hot baths and should pay special attention to massages and rubbing the body with linen. Apparently, these men had a separate bathhouse, quite possibly one without an exercise yard. Pregnant women should also refrain from baths in the first weeks after impregnation.137 The beneficial effect of bathing could be enhanced by adding certain decoctions or additives to the water. Bathing in a mixture of water and olive oil was often prescribed. Other ‘ingredients’ include sunflowers, fenugreek, chard, Egyptian acorn and sodium carbonate. The so-called aphronitrum, a sort of nitre, was used as soap. Apparently, the dirt of the palaestra was recycled as a medicine that induced transpiration.138 We can imagine that the combination of dust, dirt, oil, medical powders and ointments must have made the water inside the pools quite dirty, reminding us of the quote of Marcus Aurelius: Ὁποῖόν σοι φαίνεται τὸ λούεσθαι· ἔλαιον, ἱδρώς, ῥύπος, ὕδωρ γλοιῶδες, πάντα σικχαντά What is bathing when you think about it: oil, sweat, filth, muddy water, everything repugnant.139 In fact, it is quite clear that the medicines should be applied in the public baths: bloodletting should preferably be carried out ‘ἐν βαλανείῳ’ (‘inside the bathhouse’),140 as should the application of several decoctions.141 A 136  Or. Coll. Med. 10.3.9. 137  Children: Or. Coll. Med. Libr. Incert. 54.20.1.; young men: Or. Coll. Med. Libr. Incert. 54.18.4–7; old and feeble men: Or. Coll. Med. Libr. Incert. 54.22.12; pregnant women: Or. Coll. Med. Libr. Incert. 54.6.10. 138  Oil: Or. Coll. Med. 10.37.35; 10.37.16–17; 10.38; sunflowers: Or. Coll. Med. Libr. Incert. 54.24.17; fenugreek and other: Or. Coll. Med. 54.29.59; nitre: Or. Coll. Med. 14.27.6; palaestra dirt: Or. Coll. Med. 14.62.1. See also Blonski (2014b). 139  M. Aur. Med. 8.24. 140  Or. Coll. Med. 7.9.10–11. 141  Or. Coll. Med. 45.29.59,71–72.

45 mustard plaster should also be administered in the sun, or in a heated room, preferably in a bathhouse.142 Severe cases of itchy skin should be treated by a mixture of honey and liquid alum, first applied in the sun and then in the baths or the vestibule of the bathhouse.143 The resulting sludge on the floors and in the pools must have necessitated a frequent changing of bathing water or maybe a limited volume of water per bather. Basil of Caesarea Basil of Caesarea (ca. AD 330–379) studied in Athens, but spent his episcopal career in Caesarea (Cappadocia). His ascetic lifestyle and writings had a mayor influence on the development of monasticism in the Greek Church. As a churchman, he pleaded unity, but nevertheless attacked Arianism.144 His writings include letters, homilies and his famous sermons called the Hexaemeron (‘The six days’). In his ‘Homily against the rich’, Basil deplores the greed of some of his fellow Christians, who only cared about accumulating wealth and acquiring every possible commodity, from baths in the city to baths in the country.145 In his ‘Homily On the Forty Soldier Martyrs of Sebaste’, the officers of Licinius (ca. AD 320) refused to offer to the traditional pantheon and were punished by being plunged into a frozen pond while looking out onto the hot baths. When one of the condemned got out of the cold water and ran into the baths to dive into the hot water, he died on the spot.146 In homily III (‘On the firmament’) of the Hexaemeron, Basil compares the external outlook of the vaults of a bathhouse with the celestial spheres: the domed form of the interior does not prevent the roof from having a flat surface on the outside.147 It is clear that the architecture of bathhouses provided Basil with a recognizable comparison to describe the structure of the heavens to his reader. Yet, one cannot rule out a possible symbolic connotation of the bathhouse, where the life-giving water was domed by a steady construction, just as the earth and its waters were domed by the firmament. It is not a coincidence that Basil starts this passage with ‘And God said, let there be a firmament in the midst of the waters, and let it divide the waters from the waters. And God made the firmament, and divided the waters which were under

142  Or. Coll. Med. 10.13.16. 143  Or. Libr. Incert. 116; as quoted in Bussemaker and Daremberg (1854) 649. 144  Rousseau (1996a) 234. 145  Basil, Hom. 7 in PG 31, col. 285. 146  Basil, Hom. 19 in PG 31, col. 520. 147  Basil, Hex. 3.4 in PG 29, col. 60.

46

chapter 2

the firmament from the waters which were above the firmament.’148

tion grievous; nor a frequenter of Zeuxippos, that new Jerusalem?153

Gregory of Nazianzus Gregory of Nazianzus (ca. AD 329–389) studied in Athens alongside Basil of Caesarea, but never shared the latter’s enthusiasm for ascetism. His ecclesiastical career evolved in the shadows, yet he had an important role in fighting Arianism after the Councils of Nicaea (AD 325) and Constantinople (AD 381) by further promoting the doctrine of the Trinity. His work consists of several letters, orations and poetry. In a gripping funeral oration for his friend Basil, Gregory recalled how he managed to relieve his friend from the usual initiation rites for students in Athens.149 Normally, a procession took the freshman to the bathhouse, where initially the elders teasingly denied access, shouting loudly and knocking on the doors. Eventually, the ‘victim’ could enter and bathe, after which he was accepted as a student. The passage reminds us that water and in extension bathing also had a ‘cleansing’ connotation in a less spiritual way, such as in a ‘rite of passage’ for (Christian) students. We also find attestations of baths in their curative capabilities. The springs that bring forth hot water are a gift from God, ‘supplying our need for hot baths in many parts of the earth, and in conjunction with the cold, give us a healing which is without cost and spontaneous.’150 In his letters to Olympius,151 we learn that Gregory himself visited the baths of Xanxaris (unknown location in Cappadocia, Turkey) on the advice of his ‘medical men’. Gregory obviously lived in a society in which baths were an active part of daily life, especially in large cities like Constantinople. In his oration Against the Arians, Gregory defends his ‘villageattitude’ against the virulent attacks of the citizens of the capital who boast of their racecourses, stadiums, high walls and marvellous baths.152 Gregory reproached the Arian cosmopolites their lust for luxury and gossip:

Gregory of Nyssa Gregory of Nyssa (lived ca. AD 330–395) was the brother of Basil of Caesarea. Like his sibling, he was a devout opponent of Arianism, though his writings also contain an important influence from Neo-Platonist philosophy and of the work of Origen. He became bishop of Nyssa (in Cappadocia, Turkey), but was deposed by the provincial governor during the Arian offensive under Valens. In AD 381, he took part in the Second Ecumenical Council in Constantinople, becoming one of the leading theologians of the ‘orthodox’ faith. His works are an important step in the ‘Christianisation’ of almost every literary genre, combining both ‘classic’ and Christian traditions.154 The testimonies on baths and bathing in the works of Gregory do not point to a rejection of the practice, with only excess seen as injurious to a person’s health. In a comparison to explain why one should follow the right doctrinal path in life, Gregory speaks of the baths as part of an imprudent lifestyle.

And where will you put the fact that I am not full of small talk, nor a jester popular with company, nor great hunter of the marketplace, nor given to chatter and gossip with any chance people upon all sorts of subjects, so as to make even conversa-

148  Translation by Jackson (1895) 67. 149  Greg. Naz., Or. 43. 16. 150  Greg. Naz. Or. 28. 26; translation by Browne (1893) 298. 151  Greg. Naz. Ep. 125 and 126. 152  Greg. Naz. Or. 28. 7–8.

Let us suppose two persons suffering from an affection of the eyes; and that the one surrenders himself most diligently to the process of being cured, and undergoes all that medicine can apply to him, however painful it may be; and that the other indulges without restraint in baths and winedrinking, and listens to no advice whatever of his doctor as to the healing of his eyes. Well, when we look to the end of each of these, we say that each duly receives in requital the fruits of his choice, the one in deprivation of the light, the other in its enjoyment.155 In the same text, the baths are counted among the most favourite pastimes of men, next to hunting, going to the theatre and to banquets.156 However, it seems that bathing was still part of a daily routine, even among Christians. Gregory describes how a certain priest called Helladius received him in a rude manner. This Helladius had a bath prepared in the evening and organized a banquet, 153  Translation by Browne (1893) 331. Zeuxippos refers to the Zeuxippos-baths in Constantinople, often praised as the city’s most glamorous and sumptuous thermae. See also below, John Malalas. 154  Rousseau (1996b) 657. 155  Greg. Nyss. infant. in PG 46, col. 177; translation by Moore and Wilson (1892) 376. 156  Greg. Nyss. infant. in PG 46, col. 180.

Written Evidence of Baths

all on a Sabbath day and on a martyr’s commemoration.157 Gregory also acknowledges the spiritual connotations of water. In his ‘Answer to Eunomius’ Second Book’ the name of the prophet Moses is explained as being the Egyptian term for ‘water’. The princess who found the baby in the river while taking a bath, called the child Moses, as he was a gift from the water ‘a name by which God Himself did not disdain to address His servant, nor did He deem it beneath Him to allow the name given by the foreign woman to remain the Prophet’s proper appellation.’158 In the same text, it becomes clear that Gregory was well informed about the different activities that took place within a bathhouse, as he makes a comparison with bath-attendants, listing some of their specialties like depilation, giving foot baths and distributing towels.159 John Chrysostom Educated at Antioch by Libanius, John Chrysostom (lived ca. AD 354–407)—meaning literally ‘John with the golden mouth’—briefly lived as a monk and later as a priest before becoming the bishop of Constantinople in AD 398. Doctrinal disputes with the empress Eudoxia and the archbishop of Constantinople ultimately led to his exile to Armenia, where he died in AD 407. The references to baths and bathing in the extant work of John Chrysostom are numerous and would merit an independent study. Besides John’s personal opinions about bathing, we can also find a mass of information about Christian opinions and general details concerning daily bathing habits at the end of the 4th c. Chrysostom does not hide the fact that he himself occasionally enjoys the benefits of a bath. He does not share the opinion of some extremist Christian about the required lifestyle for priests. If he cannot raise the dead, he is of no worth, they say: such a one is pious, but this man is not. If he eats a moderate meal, for this he is accused, he ought to be strangled, they say. If he is seen at the bath, he is much censured. In short, he ought not to look upon the sun! If he does the same things that I do, if he bathes, eats and drinks, and wears the same clothing, and has the care of a house and servants, on what account is he set over me?160

157  Greg. Nyss. Ep. 18 (to Flavian) in Moore and Wilson (1892) 542. 158  Translation by Moore and Wilson (1892) 279. 159  Greg. Nyss. Answer to Eunomius’ Second Book in Moore and Wilson (1892) 291. 160  Joh. Chrys. Hom. in Tit. 1.4; translation by Schaff (s.d.) 523.

47 Baths form a part of daily life, just as the evening meal.161 According to John, it is perfectly permitted for Christians and priests to enjoy baths. Ἀλλὰ λούεσθαι, φησὶν, οὐ χρή. Διὰ τί, εἰπέ μοι; ποῦ τοῦτο κεκώλυται; οὐδὲ γὰρ ὁ ῥύπος καλόν.162 He ought not to frequent the bath, you say. And where is this forbidden? There is nothing honourable in being unclean.163 While travelling to exile in Cappadocia, John complains about the miserable conditions, especially the fact that he could not bathe.164 A revolt against Theodosius in Antioch in AD 387 was followed by a closure of the public baths, much to the despair of the local citizens.165 The public baths nevertheless only provide a superficial cleaning, removing sweat and dust. Washing in the baths can do nothing about the sins one has committed. But there are some who after committing numberless sins during the day, wash themselves in the evening and enter the churches, holding up their hands with much confidence, as if by the washing of the bath they had put off all their guilt. And if this were the case, it would be a vast advantage to use the bath daily! I would not myself cease to frequent the baths, if it made us pure, and cleansed us from our sins!166 In an interesting comparison, Roman bathing habits are weighed against Jewish ritual washing and Christian baptism.167 Only the holy washing (baptism) of the Church can cleanse a man’s sins. Βαλανεῖον γάρ ἐστιν ἡ Ἐκκλησία πνευματικὸν, οὐ ῥύπον σώματος, ἀλλὰ ψυχῆς ἀποσμῆχον κηλῖδα τοῖς πολλοῖς τῆς μετανοίας τρόποις.168

161  Joh. Chrys. Hom. ad pop. Ant. 14–18; Hom. in Joh. 18.4; Hom. in Ep. ad Rom. 31.5. 162  Joh. Chrys. Hom. in Tit. 1.4 163  Translation by Schaff (s.d.) 523. 164  Joh. Chrys. Ep. ad Olymp. 17.4. 165  Joh. Chrys. Hom. ad pop. Ant. 14.6; 17.2; 18.4. 166  Joh. Chrys. Hom. in Ep. II ad Tim. 6.4; translation by Schaff (s.d.) 499. See also Hom. in Ep. I ad Cor. 18.2 and De mutat. Nom. Hom. 4.1. For Christian literature on bathing after sins, see Dumaine (1910) 85–86. 167  Joh. Chrys. Ad Ill. Cat 1.2. See already Commodianus (above p. 42) for attestations of Jewish attitudes towards bathing. 168  Joh. Chrys. Hom. in Ep. II ad Cor. 15.5.

48

chapter 2

For the Church is a spiritual bath, which wipeth away not filth of body, but stains of soul, by its many methods of repentance.169 The baths are necessary to maintain personal hygiene, without exaggerating.170 Enjoy thy baths, take care of thy body, and throw thyself freely into the world, and keep a household, have thy servants to wait on thee, and make free use of thy meats and drinks! But everywhere drive out excess, for that it is which causes sin, and the same thing, whatever it be, if it becomes excessive, becomes a sin.171 Just as in the Historia Augusta or the works of Ausonius, the danger of excess is stressed. Bathing with women was out of the question.172 Indeed, a person was at his weakest in the baths.173 Some persons like peasants, soldiers or wrestlers should avoid baths altogether.174 The prolonged bathing of women also explains why they are so much weaker than men.175 Besides spending too much time in baths, it is also malicious to display wealth (jewellery, fine linen) and status (a group of slaves) while bathing.176 Spending money to build a bathhouse can also be seen as an act of vanity. And Christ indeed hath not where to lodge, but goes about a stranger, and naked, and hungry, and you set up houses out of town, and baths, and terraces, and chambers without number, in thoughtless vanity; and to Christ you give not even a share of a little hut, while for daws and vultures you deck out upper chambers. What can be worse than such insanity as this?177 There also numerous other references which inform us about many bathing habits and the perceptions of baths. We learn that sometimes highly pregnant women went into labour in the baths and that some poor homeless 169  Translation by Chambers (1889) 355. 170  Joh. Chrys. Hom in Joh. 18.4; 61.3; ad Ill. Cat. 1.2. 171  Joh. Chrys. Hom. in Eph. 13.3; translation by Schaff (s.d.) 115. 172  Joh. Chrys. De inani gloria et de educandis libris 60. 173  Joh. Chrys. Hom. in Act. 31.3. 174  Resp. Joh. Chrys. Hom. in Act. 18.5; Hom. in Ep. I ad Tim. 13.4 and Hom. in Hebr. 5.4. 175  Joh. Chrys. Hom in Hebr. 29.5. 176  Joh. Chrys. Hom. in Joh. 80.2; Hom. in Hebr. 28.13; Hom. in Ep. I ad Cor. 23.6; Hom. in Eph. 23.6; Hom. in Ep. ad Col. 10.4; Hom. in Ep. I ad Tim. 2.2. 177  Joh. Chrys. Hom. in Ep. ad Rom. 14.11; translation by Morris and Simcox (s.d.) 452.

people huddled in the service rooms of public baths, even covering themselves with the ashes of the furnace to stay warm. Other passages give a glimpse of the bustling activity inside a bathhouse: women chattering aloud (as they did in church!), doctors helping their patients, the unbearable temperature in over-heated baths (even compared to the flames of hell!) and rich people with their throngs of slaves. Lepers were forbidden to enter public baths.178 The link between baths and superstition is also mentioned, as many Christians made a cross-gesture when entering a bathhouse179 and mothers or wet-nurses entering the baths smeared a mark with mud on the foreheads of children to avert ‘the evil eye, witchcraft and envy’.180 In a text sometimes attributed to Chrysostom, Christians are encouraged to pray anywhere they can, even in the baths.181 There are some allusions to (partial) nudity inside the baths,182 notably when a slave-girl had been beaten and the bruises on her nude back could be seen by all.183 In a rather surprising comparison, Chrysostom mentions how martyrs are afraid to be the first to face the stake, just as each individual in a group of bathers is reluctant to be the first to dip in the hot water.184 Eunapius The historian and sophist Eunapius (lived ca. AD 340– 420) was a fervent opponent of Christianity. His opinions are clearly reflected in his surviving works, of which Lives of the Sophists is the only one surviving in full. In this idealized picture of prominent 4th c. Neoplatonist philosophers, the baths are a recurrent setting to illustrate their miraculous powers. Of Porphyry, it is said that he expelled a demon from a bathhouse. Iamblichus on the other hand conjured up two boys as the personification of two hot springs in the thermal baths of Gadara (Palestine) to convince his pupils of his powers. The philosopher Chrysantius always seemed fresh from the 178  Pregnant women: Joh. Chrys. Hom. in Joh. 34.3; poor people: Joh. Chrys. diab. 3.5; Consolatio ad Stagyram 3.13; chattering: Joh. Chrys. Hom. in Ep. I ad Cor. 36.6; Hom. in Ep. I ad Tim. 9.1; doctors: Joh. Chrys. De Diab. Tent. 1.5; temperature: Joh. Chrys. Thdr. 1.10; lepers: Joh. Chrys. Stag. 3.13. 179  Joh. Chrys. Hom. in Act. 10.5. See also Tertullian, De Corona, 3 (PL II, col. 99) stating the newly baptized Christian made a cross-sign before their every act between Easter and Pentecost: ‘ad lavacra…frontem crucis signaculo terimus.’ There is also evidence that Christians believed demons housed in classical buildings and statues (Saradi-Mendelovici (1990) 55–56). 180  Joh. Chrys. Hom. in Ep. I ad Cor. 12.7. 181  Joh. Chrys. De Chananaea 11. 182  Joh. Chrys. Hom. in Matth. 81.4; Hom. in Ep. I ad Tim. 2.4. 183  Joh. Chrys. Hom. in Eph. 15.3. 184  Joh. Chrys. De Sanctis Martyribus Sermo 2.

49

Written Evidence of Baths

bath, although he seldom took one. To conclude, in a story about Prohaeresius, family members arriving in Athens are welcomed by taking them to the baths and new disciples of Prohaeresius’ teachings are initiated in public baths.185 Other 4th c. Greek Authors In the treatise Against the Donatists, Optatus (writing ca. AD 360) bishop of Milevis (Numidia, Africa), stressed that the only true washing of a person was an ‘orthodox’ baptism.186 In a passage about Donatists who stripped orthodox virgins of their veils—the sign that befitted their status—Optatus recounts with indignation that their heads were then sprinkled with unclean ashes and they were ordered to wash with salt water.187 There is no indication that this washing took place inside a bathhouse—the sea seems the logical option—but this nevertheless shows that salt water could have purifying properties ascribed to it.188 Asterius (lived ca. AD 350–410), bishop of Amasia (Bithynia, Turkey) and a contemporary of John Chrysostom, lashed out at the greed and thirst for luxury. In his sermon ‘Against Covetousness’, he sketches a gripping picture of the difference between the rich and the poor by using the baths as a medium. Some lie down under gilded roofs and live in houses that are like small cities, adorned with sumptuous baths and chambers, and most extensive porches, and every kind of extravagance, while others have not the shelter of two boards. When they cannot live in open air, they either take refuge beside the furnaces of the baths, or, finding the attendants of the baths inhospitable, they dig into the dung like swine, and so contrive to get for themselves the needful warmth.189

185   Porphyry: Eunap. 358–59; Iamblichus: Eunap. 368–69; Chrysantius: Eunap 550–51; Prohaeresius: Eunap. 482–83. 186  Opt. 5 (several passages). 187  Opt. 6.4. 188  There are examples of public baths making use of salt water, for example the inscription advertising for a bathhouse in Pompeii that had both aqua marina and aqua dulci (‘seawater and fresh water’; CIL 10.1063). See also SHA, Heliogab. 23, 7 for swimming pools with seawater. For a discussion about seawater in baths, see Fagan (1999a) 62, esp. n. 75. For an attestation of salt being used at the baths, see also the papyrological evidence (P-44). 189  Asterius Amas. Hom. (Against Covetousness Serm. 3; translation by Anderson and Goodspeed (1904) 101).

The possession of a private bath was clearly seen as sign of immense wealth.190 The antithesis of the rich who enjoyed their private baths and the poor who could not even afford to enter public baths is already encountered throughout the work of Chrysostom, and stands in glaring contrast to the ode on private baths by the upper-class Ausonius. 5th c. Latin Authors Augustine Arguably the most important Latin author of Late Antiquity, Augustine (lived ca. AD 354–430) was born to a ‘non-Christian’ father and Christian mother in Thagaste (Souk Ahras, Algeria). After an education in rhetoric, he taught at Carthage, Rome and Milan. His quest for spiritual salvation led him from Manichaeism to a Christianized Neoplatonism (fostered by Ambrose at Milan). He was ordained priest and later bishop of Hippo Regius (Annaba, Tunisia), where he developed his personal Christian beliefs in his ‘battle’ against the Donatist movement. His impressive oeuvre comprises the autobiographical Confessiones, a teleological explanation of the sack of Rome (AD 410) in De Civitate Dei and several doctrinal expositions in the form of letters and sermons. Throughout Augustine’s work, bathing is presented as normal for human beings, necessary for a good health, but often only appreciated as a pleasurable pastime. When mourning over the death of his mother, Augustine let himself be enticed to have a bath only to take his mind off his grief. visum etiam mihi est, ut irem lavatum, quod audie­ ram inde balneis nomen inditum, quia Graeci ba­ laneion dixerint, quod anxietatem pellat ex animo. ecce et hoc confiteor misericordiae tuae, pater or­ phanorum, quoniam lavi et talis eram, qualis prius­ quam lavissem.191 It appeared also good to me to go and bathe, having heard that the bath had its name (balneum) from the Greek Balaneion for that it drives troubles from the mind. And this also I confess unto Thy mercy, Father of the fatherless, that I bathed, and was the same as before I bathed.192 Even if the bath did not have the effect Augustine had hoped for, it shows us that his new found Christian beliefs did not prevent him from visiting public baths. In 190  See also de Haan (2010) 130–34. 191  August. Conf. 9.12. 192  Translation after Pilkington (1943) 213.

50 an earlier passage, he remembered how he once bathed with his father when he was just 16.193 It had delighted his father to see that his boy was finally becoming a man, a thought that had horrified his pious mother. Besides forging a link between baths and (awakening) sexuality, this episode also alludes to nude bathing. In discussing the Fall and the subsequent shame of nakedness, Augustine recounts about some ‘barbarians’ even bathed with all their clothes on.194 That nudity in baths was not a sin is repeated on several occasions.195 In a letter to Publicola (dated AD 398), Augustine addressed the question if Christians could bathe in a bathhouse that was the setting for ‘pagan’ sacrifices or festivities.196 In his opinion, as long as Christians did not actively contribute to the sacrifices, there was no harm in going to these baths. This same concern about the presence of ‘idols’ in baths has already been encountered with Ambrose, while the act of sacrificing at the baths reminds us of the passage by Eusebius. There are numerous other references to ‘Christian’ bathing habits. We learn that Christians in some cities had to refrain from baths during Lent, that it was exhausting to bathe while fasting, and that a person who would receive baptism should have bathed the day before.197 This last point is especially interesting as it positions Christian baptism in a long tradition of cleaning the body before starting specific religious actions (see below p. 75). Even if bathing is thus justifiable for Augustine, he shares the common opinion about the dangers of excess.198 In a letter to nuns, he advises them not to make bathing part of a daily routine; once a month should suffice. If a nun falls ill however, additional baths may be necessary if the illness justifies it. Baths to boost morale, for pure enjoyment, should be out of the question. The additional remark that at least three monks / nuns should bathe at the same time is probably proposed as a kind of ‘social surveillance’ mechanism.199 The underlying ideas are obvious: bathing should be restricted to the act of

193  August. Conf. 2.3. It was usual for children to go with their mothers to the baths until they reached puberty (En. in Ps., psalm 34.19), as is still the case in the present day hammams (Grotzfeld (1970) 99). According to Cicero, it was prohibited for fathers to bathe with their adolescent sons (Cic., Off. 1.129). Plutarch says fathers-in-law avoided bathing with their sonsin-law (Plut., Vit. Cat. Mai. 20.5). 194  August. De Civ. D. 14.17. 195  August. De Doct. Christ. 3.12; De Natura Boni Contra Manichaeos 23. 196  August. Ep. 45 with the answer in Ep. 47. 197  August. Ep. 54. 198  August. In Evang. Iohan. 15.4.16; En. in Ps., psalm 37.11. 199  August. Ep. 211.

chapter 2

washing, i.e. to maintain a good health.200 The sensual pleasures of bathing were only superficial, just as the joys of money, the spendthrift in taverns or the impurity of intercourse.201 The passages about Christians at the baths stand in contrast to the references about Manichaeans. As a former follower, Augustine wrote several treatises against this religion. The so-called ‘Manichaean Elect’, who practised a strict ascetism, refrained from baths, although ‘many were caught with wine and animal food, many at the baths’.202 Apparently, these Elect took offence at the bathing habits of the ‘Hearers’, the common believers, who held to simple moral rules.203 A trial against such an Elect was—coincidence or not—held in the baths of Sossius in Hippo.204 Trials were not the only occasion when baths served as assembly locations. The ecclesiastical congress of AD 411 in Carthage was held in the ‘baths of Gargilius’.205 The bishops assembled ‘in tam spatioso et lucido et refrigeranti loco’ (‘in such a spacious, bright-lit and cold room’). The room in question was not necessarily the frigidarium, but could have been the basilica thermarum or the oblong halls preceding the frigidaria which have often been recognized in North African baths (see below, chapter 4). Sidonius Apollinaris The political career of Sidonius Apollinaris (lived ca. AD 430- after 480) exemplifies the tension between the traditional power of the Roman Senate and the expanding role of the Germanic courts during the 5th c. AD. Born at Lugdunum (Lyons, France), Sidonius worked his way up to the imperial courts of Avitus (reigned AD 455–456) and Majorian (reigned AD 456–461). After the latter’s fall, he returned to his native Gaul, maintaining a difficult relationship with both the Senate in Rome and the Gothic kingdom in Gaul. He eventually became bishop of Clermont in around AD 470. In his literary oeuvre, Sidonius praises the delights of contemporary private villas, just as Ausonius had done a century earlier (see above). In his letters, which were actually conceived as a book, much like Pliny the Younger’s Epistularum,206 Sidonius invites a friend to 200  August. Soliloquiorum 1.17. Augustine also mentions the beneficial effects of bathing in seawater (de moribus manichaeo­ rum 8). See also above, n. 188 and Oribasius. 201  August. De catechizandis rudibus 16. 202  August. De moribus manichaeorum 19. See also Mc Donough (2009) 264; Baker-Brian (2011) 122–28. 203  August. c. Faust. 20.23. 204  August. acta seu disputatio contra fortunatum manichaeum, introduction. 205  August. ad Don. post Coll. 35.58. 206  Gibson and Morello (2012) 234–64.

51

Written Evidence of Baths

his villa in Avitacum (Aydat, France) and tries to convince him by describing his private baths—labelled balneum or balineum—in full detail.207 He walks us through the several rooms, beginning with the hot room and its semicircular pool (solium), the adjoining unguentarium (anointing room), the frigidarium with its pyramid roof, and a piscina or—as the Greeks would call it—baptisterium. Great care is taken to describe the inner ambience, stressing the sound of gushing water, the abundant sunlight entering through the windows and the plain, yet stylish decoration. No disgraceful paintings adorn the walls and no imported marbles colour the floors. He is very clear which figures are not to be found: no nude figures, no actors with masks and coloured garments and no anointed wrestler with intertwined limbs. The link of baths to carnal pleasures and indecency, often mirrored in the decorative themes, and maybe even the depiction of traditional gods (the nude figures?) are clearly disapproved off. It is the water management, especially that of the piscina, that merits praise. in hanc ergo piscinam fluvium de supercilio montis elicitum canalibusque circumactis per exteriora na­ tatoriae latera curvatum sex fistulae prominentes leonum simulatis capitibus effundunt, quae temere ingressis veras dentium crates, meros oculorum fu­ rores, certas cervicum iubas imaginabuntur.208 Into this piscina, then, a stream lured from the brow of the hill is conducted in channels curving round the outside of the swimming basin; it issues through 6 pipes terminating in lions’ heads which, to one entering rapidly, seem to present real fangs, authentic fury of eyes, indubitable manes.209 The overwhelming sound of the water coursing through lead pipes and gushing out of various spouts and taps allowed one to have a private conversation in the middle of a crowd. ita sonitu pressus alieno ridiculum affectat publicus sermo secretum.210 The interference of this alien sound forces conversations which are quite public to assume an amusing air of secrecy.211 207  Sid. Apoll. Ep. 2.2.4–9. 208  Sid. Apoll. Ep. 2.2.8. 209  Translation by Dalton (1915) 38–39. 210  Sid. Apoll. Ep. 2.2.9. 211  Translation by Dalton (1915) 39.

The entire description of the baths searches for a compromise between the minimalistic baths as applauded by Seneca and the baths as a marvel of technique found in Ausonius.212 The same baths are the subject of a poem.213 Here, their delights are compared with those at the thermal baths of Baiae, reminding us again of Ausonius’ poem. A separate poem also describes the piscina of the private baths,214 obviously a source of great pride to the owner since it had also been described separately in letter II.2. In the poem, the delights of a cool dip after the scorching heat of the hot rooms are brought to the fore. Intrate algentes post balnea torrida fluctus ut solidet calidam frigore lympha cutem et licet hoc solo mergatis membra liquore, per stagnum nostrum lumina vestra natant. Enter the cool waves after the hot baths, that the water by its coldness may strengthen your heated skin; and even if you plunge only your limbs in this water, your eyes swim through our pond. The fact that Sidonius estimates the capacity of this piscina at around 20,000 modii,215 equalling roughly 175 m3, means this pool must have been quite large. The Latin term stagnum (lake, pond) that is used in the poem further points to impressive dimensions.216 On other occasions, Sidonius also speaks of private baths as part of a delightful villa. In a letter to his friend Consentius (dated AD 478), he describes the villa of Octaviana with its fantastic location near the town, the sea and the river, its perfect architectural symmetry, its colonnade and its baths.217 Although the facilities described above are not public, the formal description of the architecture and of the perception on baths is directly relevant for the study of public baths. Sidonius even expresses the similarities himself, claiming his frigidarium can stand the comparison with public 212  Sen. Ep. 86; Auson. Mos. 335–348. 213  Sid. Apoll. carm. 18. For a comparison between the letter and the poem, see Busch (1999) 69–74. 214  Sid. Apoll. carm. 19. 215  Sid. Apoll. Ep. 2.2.8. 216  S. Busch (1999) calculates the surface of the pool at around 100 m2, considering the volume of 125 m3 and a depth of around 1.75m. Even if Sidonius makes a literary exaggeration, such large piscinae are archaeologically attested, for example in the villa of Poppea in Oplontis (near Naples, destroyed in the eruption of AD 79). 217  Sid. Apoll. Ep. 8.4.1.

52 facilities: ‘piscinas publicis operibus extructas non impudenter aemularetur’ (‘it might without impertinence challenge comparison with baths built as public undertakings.’).218 Deliberate bragging or not, these private baths were apparently conceived to host several people at a time. The description of the large piscina, where a master and his guests could have a private conversation even in a crowd, makes clear that bathing was a social occasion in the villas too.219 Besides singing the praises of private baths, Sidonius also refers to baths and bathing in other contexts. In reply to a friend, Sidonius wards off criticisms of his native province by slamming his friends’ home town Caesena (Cesena, Italy) in return. In that swampy marsh of his, everything is upside down: the towers float and the ships stand still, business men turn soldiers while soldiers turn business men and the baths are cold while the houses are on fire.220 The passage reminds us that baths were still seen as basic facilities of a town and that balnea still implied heated forms of bathing.221 In the poem to Consentius, Sidonius reminds his friend of the good old days in Rome when they visited small private baths together, instead of the large public baths of Agrippa, Diocletian or Nero.222 In a different letter, Sidonius recounts how he observed a ball game while waiting for Mass to be celebrated.223 Challenged by a somewhat older player to write a poem about a towel, he composed some verses in which the sweat of a ferven­ tia balnea (hot bath) is mentioned. The passage again stresses the link between hot baths and the use of towels within or after the baths. This fondness of steaming hot baths is also encountered in Ep. 2.9.8–9. Sidonius, finding the private baths of two of his hosts not ready for use, ordered his servants to dig a pit, cover it with branches and goat skins and put glowing hot stones inside it to create an improvised sweat hut. By pouring water over the stones, a vapour-bath was created. The sweating session was followed by a dip in hot water and finished off with a shower of cold water (from the nearby fountain, well or river). It is difficult to assess whether the passage reflects existing contemporary habits or a liter-

218  Sid. Apoll. Ep. 2.2.5. Translation by Anderson (1936) 421. 219  See also de Haan (2010) 129–30. 220  Sid. Apoll. Ep. 1.8.2. 221  The strict correlation between ‘balneum’ and a heated environment had been present since the ‘Roman’-style of bathing took shape in the 2nd–1st c. BC. (Blonski (2014a) 200–201). For thermae as characteristic elements of a town, see also Carmina 22, 23 and 24. 222  Sid. Apoll. Carm. 23, 495–500. 223  Sid. Apoll. Ep. 5.17.

chapter 2

ary allusion to known ‘barbarian’ bathing habits such as described already by Herodotus and Strabo.224 Caelius Aurelianus A physician native of Sicca Veneria (Le Kef, Tunisia) writing at the beginning of the 5th c. AD, Caelius Aurelianus modelled his medical treatises on the work of earlier writers, especially Soranus of Ephesus (lived around AD 100). Among his most important works are the Gynaecia and De morbis chronicis et acutis. His writings reveal a methodist approach of medicine.225 As Caelius was the heir of a long medical tradition, we find Greek-Hellenistic theories about the effects of bathing reflected in his works.226 Bathing in cold water, called psychrolusia by Aurelius, had an astringent effect, while bathing in hot water, called simply lava­ crum (bathing), had a relaxing and softening effect.227 Showers, called cataclysmus or illisio, are prescribed for the massaging effect the tumbling water could have on specific body parts. They are included in the therapies for epilepsy, insanity, inner ear problems, asthma, and back problems.228 Therapies combining bathing with a certain diet—for curing haemorrhages, arthritis and gout—or bathing and physical exercises such as swimming (against dropsy) and walking were also popular.229 Often ointments and decoctions were prescribed to use inside the baths.230 Bathing in olive oil or a mixture of olive oil and water was thought to be beneficial, for instance against constipation, bladder problems, paralysis, coeliac disease, and sexual urges.231 The quality of the water also played an important role, as sometimes seawater (against dropsy and against chronic headaches) or mineral waters (against arthritis and gout) were preferred.232 That seawater was probably used in public baths is implied by Aurelius’ critique of some physicians 224  Hdt. 4.73–75; Strabo Geogr. 3.3.6–7. 225  Nutton (2006). 226  Garzya (1994) 114. 227  Psychrolusia: Cael. Aur. Morb. chr. 1.47; lavacrum: Cael. Aur. Morb. ac. 2.50; see already Oribasius (above p. 44). 228  Epilepsy: Cael. Aur. Morb. chr. 1.99; insanity: Cael. Aur. Morb. chr. 1.168; inner ear problems: Cael. Aur. Morb. chr. 2.70; asthma: Cael. Aur. Morb. chr. 3.10; back problems: Cael. Aur. Morb. chr. 4.104. 229  Diet: Morb. chr. 2.182; Morb. chr. 5.37; swimming: Cael. Aur. Morb. chr. 3.117; walking: Cael. Aur. Morb. chr. 3.94. 230  Cael. Aur. Morb. chr. 4.18. 231  Constipation: Cael. Aur. Morb. ac. 3.151; bladder problems: Cael. Aur. Morb. chr. 5.75; paralysis: Cael. Aur. Morb. chr. 2.23; coeliac disease: Cael. Aur. Morb. chr. 4.45; sexual urges: Cael. Aur. Morb. ac. 3.184. 232  Seawater: Morb. chr. 3.117 and Morb. chr. 1.42.; mineral water: Cael. Aur. Morb. chr. 5.37.

53

Written Evidence of Baths

who prescribed seawater vapour baths (‘aquae mari­ nae vaporationem’) and swimming in pools of seawater (‘implendum solium aqua marina ferventi’).233 There are few direct references to public baths, as Caelius speaks of bathing with the general term lavacrum.234 However, it is very plausible that therapies necessitating full immersion in hot or cold water took place within (public) bathhouses, as these were the only facilities to have both such facilities. Furthermore, earlier medical writers such as Oribasius leave little doubt that their therapies must take place in public baths (see above p. 45). Caelius also insists that in some cases, the body of the patient has to be kept warm during the therapy, preferably in a room with a portable stove or heated walls (‘ex carbonis vel parietibus ignitis’).235 This reminds us of the heated rooms (resp. the laconicum and the caldarium or sudatorium) in bathhouses. Cassius Felix Cassius Felix lived in the 5th c. AD and was a medical writer who composed a handbook called De Medicina. A native of North Africa, he continued a long tradition of medical authors in the region, following in the footsteps of Vindicianus (end of the 4th c.), Theodorus Priscianus and Caelius Aurelianus (both beginning of the 5th c.). Building on the principles of Greek medicine such as Hippocrates, Cassius Felix obviously attributed an important part of his work to water-related therapy.236 The public bathhouse was the logical place to undergo these therapies, as water and bath tubs were not readily available in every dwelling. So we learn that one should shave off all his hair ‘in balneo’ (‘in the bathhouse’), massage his head with a special ointment and rinse it afterwards ‘in sole’ (‘in a hot tub’) to counter severe headaches.237 Other treatments that include the use of baths, especially the pools, include therapies against dandruff, pimples, black spots on the skin and sunburn.238 However, the therapy against scabies does not involve bathing.239 The effects of hot (dehydrating) or cold (contracting) baths, was often compensated by drinking specific beverages or anointing with special oils just after the dips. Pain in the kidneys could be alleviated 233  Cael. Aur. Morb. chr. 3.117. 234  The use of the word lavacrum is rather complex, as it could mean bathing, bathhouse and possibly also pool. See Maréchal (2015). 235  Cael. Aur. Morb. chr. 3.40. 236  Fraisse (2002) xxix. 237  Cas. Fel. De Med. 1.13–15. 238  Dandruff: Cas. Fel. De Med. 6.2; pimples: Cas. Fel. De Med. 7.2; black spots and sunburn: Cas. Fel. De Med. 9.3. 239  Cas. Fel. De Med. 15.

with a beverage after a visit to the baths.240 To counteract jaundice, one should submerge in a warm bath, wash with dill-oil and have a drink ‘in lavacro’ (‘in the bath tub / bathhouse’).241 A warm bath with oil fights dehydration after gastroenteritis.242 In several cases of severe or recurring fever, bathing is prescribed, especially in combination with anointing with oils and breathing healthy air.243 It is clear that bathing still played an important role in the medicine of the age of Cassius Felix, with the hydrating or dehydrating effects were put to particular good use. If the acts that accompanied these therapies, such as shaving or bathing in special oils, all took place inside the public baths, one can imagine that there must have been adapted facilities that enabled such personal treatments. The thermal baths that have been excavated, such as Hammat Gader in Israel or Aquae Flavianae in Algeria, display a number of individual tubs to ensure the personal treatment of the visitors.244 Other 5th c. Latin Authors In the letters of Paulinus of Nola (lived AD 353–431), the son of a wealthy aristocratic family of Burdigala who chose an ascetic lifestyle after his political career, no attestations of baths have been found.245 In his Laterculus (dated AD 448), Polemius Silvius included a list of famous sights in Rome. He probably based this on the Notitia Regionum Urbis Romae and the Curiosum. For the thermae, he mentioned the baths of Diocletian, Caracalla (Antoninian baths), Alexandrina, Commodus, Agrippa and Sura. The baths of Caracalla are listed among the 7 marvels of the city. The poet Claudian (lived AD 370–404) presents a roadside bathhouse in the form of an inscription in which the pleasure and reinvigorating character of baths, especially on a long journey, is something to be thankful for. Fontibus in liquidis paulum requiesce, viator atque tuum rursus carpe refectus iter lympharum dominum nimium miraberis, hospes, inter dura viae balnea qui posuit.246 240  Cas. Fel. De Med. 45.3. 241  Cas. Fel. De Med. 49.2–6. 242  Cas. Fel. De Med. 51.13. 243  Cas. Fel. De Med. 55, 57, 60 and 61. 244  Hammat Gader: Hirschfeld and Amitai-Preiss (1997); Aquae Flavianae: Gsell and Graillot (1893). 245  The works of Paulinus are rather under-represented in the translations of the early Christian writers. For the present research, only the English translation of Paulinus’ letters (Ancient Christian Writers vols. 35 & 36, transl. by P. G. Walsh) were consulted. 246  Claudian, Carm. Min. 12.

54

chapter 2

Stay awhile and bathe in these waters, traveller then set forth again upon thy journey refreshed. And thou become its guest, warm will be thy gratitude towards him that built this bath and set it by the side of this long dusty road.247 Similar praise for baths can also be found in the Anthologia Latina. Some of the bath-related epigrams have been identified as building inscriptions which were probably visible inside the bathhouse (see below, EA-52–55, 64, 65).248 Other epigrams seem to have been part of a cycle by one or more authors in the style of Commodianus.249 They have been tentatively dated to the 5th c. There are references to Baiae, to the classical mythology and to the positive effects of water on the body.250 Cold water is most beneficial: ‘Una salus homini est gelidum captare lavacrum, / ne tepidus red­ dat morbida membra vapor.’ (‘The only salvage for man is taking a cold bath, as not to let warm vapour make the members sick’)251 The light inside the baths is also worthy of praise: ‘Denique succedit nostris lux tanta lavacris, / ut cernas nudos, erubuisse sibi.’ (‘Finally, so much light enters our baths, that one can see the naked bathers blushing about their own nakedness’)252 5th c. Greek Authors Socrates Scholasticus Little is known about the life of this lawyer living in Constantinople at the end of the 4th and beginning of the 5th c. (lived ca. AD 380–450). His magnum opus was the continuation of Eusebius’ Ecclesiastical History, covering the period AD 305–439. Untroubled by doctrinal disputes, Socrates aimed to couple ecclesiastical to secular, historical events. Considering the aim of the work, it is not surprising to find most information on public baths is about benefactions by important historical figures concerning specific buildings. So we learn that the famous Zeuxippos Baths in Constantinople had a passageway that connected it straight to the imperial palace.253 On command of Constantius II, the praetorian prefect of Constantinople ordered the bishop to meet him at the 247  Translation by Platnauer (1922) 185–87. 248  However, AL 108 (EA-64) and AL 109 (EA-65) may also have been part of this epigram cycle (Busch (1999) 332). 249  Busch (1999) 331–44. 250  Baiae: AL 110; classical mythology: AL 111 and AL 112; effects of water: AL 110, 113 and 164. A L 164. Translation by the author. 251   252   A L 111. Translation after Busch (1999) 335. 253  Socrates Hist. eccl. 2.16.

aforementioned baths under the pretext of a public gathering, but secretly intended to have him arrested. When realizing an angry crowd had formed outside the baths, the prefect escaped via the passage to the palace. The baths feature here as a setting for discussing public affairs (see also the public hearing of a Manichaean mentioned by Augustine). Such a hearing was not the only public event that took place in a bathhouse: when John Chrysostom was banned from the city, we learn that his followers were celebrating Easter inside the Thermae Constantianae.254 Offerings to the traditional cults such as encountered in the texts of Eusebius and Augustine were apparently not the only religious dealings that could be performed inside bathhouses.255 The Zeuxippos Baths and the Thermae Constantianae are not the only baths mentioned in the text: the Baths of Achilles were destroyed in a fire that hit large parts of Constantinople; the Baths of Anastasia were said to have been built on the order of Valens, and Eusebius’ anecdote about the baths being built from the stones of Chalcedon’s walls is repeated.256 The negative connotations such as excess and promiscuity also filter through in Socrates’ account. A certain Justina, daughter of the governor of Picenum (Italy), was able to join the entourage of Severa, wife of Valentinian II. Justina’s subsequent intimacy with the empress reached such an extent that they bathed together. Seeing Justina nude at the baths, the empress was struck by her beauty and talked about it with her husband, something she would later regret, as the emperor ended up marrying Justina.257 An anecdote about Sissinius, a bishop of the Novatian Church,258 exemplified how strange an ascetic this man was: his table was full of food, even though he was on a diet, and when asked why he bathed twice a day at the public baths, he said ‘because it is hard to 254  Socrates Hist. eccl. 6.18 255  See also the conference in the baths of Carthage in AD 411 mentioned by Augustine, p. 50. These religious gatherings in bathhouses were not restricted to the Christian community. In 3rd-c. Palestina, rabbi Haninah bar Hama had summoned other rabbis and notables to discuss the halakhah (Jewish law) in the thermal baths of Hammat Gader (Talmud Yerushalmi, Eruvim 6, 23c). 256  Baths of Achilles: Socrates Hist. eccl. 7.39; baths of Anastasia: Socrates Hist. eccl. 4.9; reuse of wall stones: Socrates Hist. eccl. 4.8. 257  Socrates Hist. eccl. 4.31. 258  Novatianism was a rigorist schism that developed after the persecutions of Decius (AD 249–250), when the Roman presbyter Novatian was unpleased with the election of archbishop Cornelius (AD 251). Novatian became a rival bishop of Rome, refusing to collaborate with those who had made compromises during the persecutions (see Cross and Livingstone (2005) 1172).

55

Written Evidence of Baths

bathe three times’.259 Although the bystander in the fragment thought it rather excessive to bathe twice a day, Sissinius, and with him perhaps Socrates, did not find it harmful at all. Sozomen The Ecclesiastical History by Sozomen (lived ca. AD 380–450), a lawyer of Gaza (Palestine) writing in Constantinople, borrowed heavily from Socrates. It covered the same period, still continuing where Eusebius had stopped, but is considered more stylish and more detailed than Socrates’ version. It should not come as a surprise that several passages pertaining to baths match those in Socrates. So we hear the same anecdote on Sissinius and are reminded of the arrest of bishop Paul in the Zeuxippos Baths and the Easter celebrations in the baths of Constantius in Constantinople by Chrysostom’s followers.260 In addition, we get some new or more detailed accounts of certain passages. When Chrysostom was arrested for example, we learn that his followers shunned public life, refraining to go to the markets or public baths.261 The Baths of Anastasia, also mentioned by Socrates, are referred to as the Baths of Anastasia and Carosa.262 Instead of the great fire in Constantinople that destroyed the Achillean Baths, it was an earthquake that sparked a fire in the furnace room of the baths in Nicomedia.263 In an additional chapter on monasticism, the well-known image of the perfect ascetic is presented as Antony, who ‘rejected the practice of anointing with oil, and the use of baths and of similar luxuries likely to relax the tension of the body by moisture; and it is said that he never at any time saw himself naked.’264 The self-imposed ban on baths was not restricted to ascetics, as apparently abstaining from baths and several kinds of meats was sometimes demanded by the bishop as an act of penitence.265 All these passages clearly demonstrate that bathing was a pleasurable activity, only to be missed in extraordinary circumstances (ascetism, grief, penitence). The curative aspect of bathing also resurfaces in Sozomen. The anecdote about the ill emperor Constantine going to the baths of Helenopolis is clearly copied from Eusebius.266 A passage about Christ washing his feet in the fountain of Emmaus explains why this 259  Socrates Hist. eccl. 6.22. 260  Sissinius: Sozom. Hist. eccl. 8.1; bishop Paul: Sozom. Hist. eccl. 3.9; Easter celebrations: Sozom. Hist. eccl. 8.21. 261  Sozom. Hist. eccl. 8.23. 262  Sozom. Hist. eccl. 6.9. 263  Sozom. Hist. eccl. 4.16. 264  Sozom. Hist. eccl. 1.13; translation by Hartranft (1886) 249. The passage refers to Athanasius V. Ant. 47. 265  Sozom. Hist. eccl. 7.16. 266  Sozom. Hist. eccl. 2.34 contra Euseb. Hist. eccl. 4.61.

water has been healing every disorder ever since.267 This shows a readiness of Christians to reinterpret and reinstate the ‘magic’ powers of thermal springs.268 Theodoret Theodoret (ca. AD 393–466) was the third important chronicler who wrote a continuation of Eusebius. His Ecclesiastical History was later edited with those of Socrates and Sozomen to form the Latin Historia Tripartita. As a monk and later bishop of Cyrrhus (Syria), Theodoret himself was more involved in the Church’s doctrinal disputes than his predecessors. He actively supported his friend Nestorius against Cyril, bishop of Alexandria. We find references to baths in the context of these religious quarrels. Thus Theodoret relates a story about an Arian bishop of Constantinople who had wished to bathe and therefore instructed his servants to close the baths to the public.269 When a crowd had formed before the closed doors, he quickly changed his mind and invited all to enter. When he saw the people were hesitating to share the hot pool with him, he rose and left, thinking it was out of respect. Theodoret informs us that the people did not want to enter the pool because it had been polluted by the bishop’s heresy and asked the servants afterwards to fill the tub with a fresh supply of water. The passage is very informative on several aspects of bathing habits in Theodoret’s community. Apparently, it was not uncommon for a bishop to bathe—even to share a pool—in public baths with his fellow men. The fact that this particular Arian bishop closed the baths for his own use is quoted as an example of his selfish ways. That a person of a certain status could close the baths for his private use was clearly not appreciated by fellow citizens. The positive alternative for a wealthy person was opening the baths for free for everyone to use, a strategy adopted at a later stage by the Arian bishop to regain some popularity. A second interesting fact is that people believed that water could be ritually polluted by a person. We should perhaps not see this as an entirely ‘spiritual’ contamination, as it could very well be that people believed the bad humours (fluids) which were expelled in the hot baths mingled with the water of the pools. It also shows that bathers could request the water be refreshed if they found it too dirty. A further example of the relationship between bishops and baths is found in the letters from Theodoret to someone at court and to a consul as a reaction against certain accusations which entailed his eventual transfer 267  Sozom. Hist. eccl. 5.21. 268  See also DeLaine (2007); Dvorjetski (2007) 393–424. 269  Theod. Hist. eccl. 4.13.

56

chapter 2

to Cyrrhus. To the nobleman at court, he stressed all his achievements as leader of his congregation: It is on the contrary well known to your excellency that I have spent a considerable portion of my ecclesiastical revenues in erecting porticoes and baths, building bridges, and making further provision for public objects.270 By the time of Theodoret, it is clear that the Church had acquired an important role in managing the city’s public life and finances. In a different letter, Theodoret mentions also the building of a water supply system in addition to the baths and other amenities.271 The fact that baths remained an important landmark in the city is again stressed by a passage about the deportation of ‘orthodox’ Christians by ‘heretic’ Arians in 4th c. Alexandria. Christians had stood trial ‘in the neighbourhood of the public baths’.272 In his Historia Religiosa, a compilation of biographies of 30 famous monks, a certain Limnaios was cured from diarrhoea through prayer, as bathing apparently did not help.273 We have again confirmation that ascetics visited the baths when they fell ill. A last interesting passage in Theodoret’s oeuvre brings us back to the association of baths with danger. The story takes place in the palace baths of Valens (reigned AD 364–378) in Antioch.274 The servant who was in charge of the baths had mistreated the monk Aphraates. When the bully tried to heat up the furnace, he suddenly felt dizzy and fell into the hot plunge. When the emperor grew tired of waiting for his bath, he sent some servants to investigate, only to find the brutal servant dead in the boiling water. Although Theodoret presents the anecdote as an example of divine justice, it also informs us about the preparation of the baths, as the boiling water in the passage obviously had to be mixed with cold before being conducted to the pools. Palladius of Helenopolis The monk Palladius (lived ca. AD 364–431) was born in Galatia, but having taken up an ascetic lifestyle, he settled in Egypt and Palestine—first in Alexandria, later in Nitria and Kelia—before returning north to Bithynia. Around AD 400 he was appointed bishop in Helenopolis 270  Theod. Ep. 79, dated around AD 449. Translation by Jackson (1892) 275. 271  Theod. Ep. 81. 272  Theod. Hist. eccl. 4.19. 273  Theod. Hist. Rel. 22.4. 274  Theod. Hist. eccl. 4.23.

(Turkey) and rallied with John Chrysostom. He was later sent to the see of Aspona (unknown location in modern Turkey), where he wrote his Lausiac History (around AD 420). The work reflects much of Palladius’ own ideas about ascetism, drawing from his own experiences. The references to baths in the Lausiac History are limited to the ‘alousia’, the abstention of washing, of the venerable monks. From Isidorus, it is said that he never wore linen, never ate meat and never took a bath.275 Similarly, the hermit Evagrius Ponticus, with whom Palladius lived for 9 years in Egypt, is quoted as saying to his disciples: From the time that I took to the desert, I have not touched lettuce nor any other green vegetable, nor any fruit, nor grapes, nor meat, nor a bath. (…) For three years I have not been troubled by fleshly desire (…) after so long a life and toil and labour and ceaseless prayer.276 In the Dialogue on the life of John Chrysostom, the opinions of Palladius are presented as a Plato-like dialogue between Chrysostom and a priest. The ideas about ‘alousia’ for widows, virgins and ascetics are the same as those encountered in Jerome, Eusebius, and Augustine. Chrysostom supposedly reformed ‘the order of widows’ and recommended fasting, abstinence from baths and plain clothing to fight the carnal desires.277 The pious Olympias only bathed when she was ill, and even then kept her robe on.278 Chrysostom himself did enjoy the baths, as he was cruelly denied access to the road station baths by the soldiers as he was travelling to his exile destination.279 His adversaries even planned to arrest him at the public baths.280 His associated priests and followers had no trouble in celebrating Easter in the public baths.281 Even if bathing was permitted, Palladius, speaking through the mouth of Chrysostom, nevertheless condemned the bishops ‘who waste the money of the poor on hanging corridors, and water-cisterns raised into the air three storeys high, and disreputable baths, hidden from sight, for effeminate men.’282 The act of building public facilities for the city is praiseworthy 275  Pall. Hist. Laus. 1.2. 276  Pall. Hist. Laus. 38.12–13; translation by Lowther Clarke (1918) 137. 277  Pall. Dial. 5.140–144. 278  Pall. Dial. 12. 279  Pall. Dial. 11. See also Chrysostom’s own writings, above p. 47. 280  Pall. Dial. 10. 281  Pall. Dial. 9; see also above p. 54–55, Socrates Scholasticus and Sozomen. 282  Pall. Dial. 13; translation by Lowther Clarke (1918) 115.

57

Written Evidence of Baths

however (‘I do not include in my condemnation those who build reasonably, and of necessity, or beautify Church property’).283 The ‘disreputable baths, hidden from sight, for effeminate men’ are probably what most critics of baths, Christian or not, had in mind. A good example of such a despicable man in search for vain glory is further given in the same passage, with a certain bishop of Ephesus (Turkey) carrying away the marble from the baptistery to adorn his own private bath.284 Other 5th c. Greek Authors The monk John Cassian (lived ca. AD 360–425) was probably of Roman birth, but was born and raised in the Middle East. He became a monk in Egypt and later a deacon under John Chrysostom in Constantinople. In around AD 415, he founded the abbey of St. Victor at Massilia (Marseille). It is probably during his time as a monk that he wrote several important works on ascetic life. These texts, reflecting the teachings of the Egyptian Desert Fathers, proved to be very influential in the development of western monasticism.285 His only attestation of baths refers to the idle vanity of some men who only care about their looks and their good name, in comparison to the monks who live by the grace of God amidst filth and dirt.286 Bishop Nilus of Ancyra (died in AD 430), a friend and ally of John Chrysostom, warned his Christian fellow men of bathing together with women.287 Those who did should be expelled from the community. The historian Olympiodorus wrote about the period between AD 407 and 425. His Histories are cited by the 9th c. bishop Photius, where we find a reference to the baths of Rome. According to Olympiodorus, the Baths of Caracalla (called Antoninianae) were made out of marble and could accommodate 1,600 persons, while the Baths of Diocletian could welcome twice as many.288 In the Apophthegmata Patrum, a collection of sayings about the Desert Fathers in Egypt, a passage recounts how the admirable Abba Macarius found himself staying the night in an ancient tomb. When he used one of the mummies as a pillow, some demons tried to scare him and called out in a woman’s voice ‘Come with us to the baths’.289 The brave man, however, ignored the demons.

283  Translation by Lowther Clarke (1918) 115. 284  Pall. Dial. 13. 285  Frank (2006). 286  Cassian, Conference 4 (Abbot Daniel), 11. 287  Nil. epp. 2.211 in PG 79, col. 12. 288  Phot., Bibl. 80. 289  A  pophth. Patr. 7.15.

5th c. Syriac Literature Zacharias of Mytilene The Ecclesiastical History by Zacharias of Mytilene (lived ca. AD 465-after 536) was originally composed in Greek, but has only survived in a Syriac revision. As a rhetorician of Gaza (Palestine), he attended the famous school of rhetoric in Berytus (Beirut), before moving to Constantinople. He was appointed bishop of Mytilene on Lesbos in AD 551. He addressed public baths in a context of war. The city of Dara (Oguz, Turkey), built as a frontier town by Anastasius, raised its walls when learning that the Persian king Kawad approached, providing also wonderful cisterns, spacious storehouses and large baths to withstand a siege.290 When the Persians had conquered the city of Amida (Diyarbakir, Turkey), their king passed his time ‘in the baths of Paul Bar Zainab (steward of the governor)’.291 The Roman baths did not only charm the Persians, as the usurper Vitalian, a Goth by birth, even received envoys while in the baths of Constantinople.292 The involvement of the Church in public baths is also apparent in a passage about a presbyter in Amida who was overseeing the construction of a bathhouse.293 Of a monastery connected to the heretic Macedonius, it is said that the monks enjoyed luxuries such as baths. The emperor reacted against these heretics by shutting off their water supply.294 Philoxenos of Hierapolis Philoxenos was a Syrian bishop and theologian who preached monophysite teachings. Expelled from Edessa, perhaps along with the ‘School of the Persians’ in 489, he became bishop of Hierapolis (near modern Aleppo, Syria) with the support of the monophysite patriarch of Antioch, Peter the Fuller. He wrote a Syriac version of the New Testament. In Constantinople, he was supported by Anastasius, before being exiled by the latter’s successor, the orthodox Justin. In a discourse on gluttony, Philoxenos represents bathing as a vain attempt to clean the body when the spirit has been befouled by sin. All these things happen unto gluttons, even though others may bring nigh helps for their bodies continually, purgatives, and cleansing draughts, and other means for relieving the body, and bloodletting, and medicines which open the bowels, and 290  Zach. Myt. Hist. eccl. 7.6. 291  Zach. Myt. Hist. eccl. 7.4. 292  Zach. Myt. Hist. eccl. 8.2. 293  Zach. Myt. Hist. eccl. 7.6. 294  Zach. Myt. Hist. eccl. 7.7–8.

58

chapter 2

violent washing with water at all hours; but overeating overcometh all these beneficial means, and createth for them in their bodies severe pains, and sore sicknesses, which it is difficult even for wise physicians to heal.295 Bathing had clearly no effect on cleaning the soul. It could even be the impulse for impious thoughts and deeds. It was precisely at the moment that king David saw Bathsheba bathe that he succumbed to desire. He went up to the roof of his royal house, and saw a woman washing, and he lusted after her, and he sent and brought her, and committed adultery with her. If he had looked simply, he would not have lusted; and if he had not lusted, he would not have committed adultery.296 These attestations to the dangers of bathing are not the only link between Philoxenos and baths. In an old liturgical text (14th c.) discovered in the monastery of Haboanas in Turkey’s Tur’ Abdin region, we learn that Philoxenos was exiled to the town of Gangra (Çankiri, Turkey), where he was imprisoned in a small room above the furnace room of a bathhouse. With all the windows and doors closed off, he was eventually asphyxiated.297 Joshua Stylites The monk Joshua the Stylite presumably composed a chronicle in Syriac about the period AD 495–506, focusing on a conflict between Constantinople and Persia. We learn that the brother of the deceased Persian king was hated by his own people, as he had no money to support the army. The priests also despised him because he tried to abolish sacred laws and wanted to build a bathhouse.298 The Zoroastrian religion firmly disapproved of communal bathhouses. Yet, there is a second reference to a Persian king enjoying the pleasures of a good bath. It is said that Kawad, when he conquered Amida (in AD 504), discovered the benefits of bathing in a public bath (see also above, Zacharias), consequently ordering baths built in all the towns throughout the Persian empire.299 The baths obviously carry strong symbolism in these passages, as ‘markers’ of Roman culture. It is not a coincidence that the Persian kings who ‘succumbed’ to the baths were considered traitors by the Persian people. 295  Philox. Discourse 10; translation by Wallis Budge (1894) 361. 296  Philox. Discourse 13, translation by Wallis Budge (1894) 575. 297  Mingana (1920) 155. 298  Josh. Styl. 19. 299  Josh. Styl. 70.

In other passages, the erection and dedication of bathhouses is considered a praiseworthy undertaking. When applauding the activities of the new governor of Edessa, Joshua mentions the construction of public baths next to the granary, which had been planned for years.300 He also refers to summer and winter baths in Edessa.301 When a powerful earthquake hit the city in AD 500, two colonnades and the tepidarium of these summer baths fell down, while two people died at the door of the frigidarium, presumably by falling blocks. The governor Eulogius later restored the baths to their former glory.302 From the winter baths, we hear that the colonnade was blocked up in an effort to create lodgings for the sick and the homeless after the earthquake. We also hear of a bathhouse beside the Great Gate and the Church of the Apostles that served as an improvised hospital to accommodate the sick.303 All these references clearly indicate that the baths still were important public facilities in a city, even to the point that their reconstruction after a disaster was considered a priority. 6th c. Latin Authors Cassiodorus Magnus Aurelius Cassiodorus (lived ca. AD 490–585) served as a politician in Italy and then Constantinople, before retiring back to his family estates in southern Italy. He served in the royal administrations under the reigns of Theoderic the Amal, Athalaric and Amalasuntha, being consul in AD 514 and prefect of Italy in AD 533. He retained this position under Theodahad and Witigis, but moved east during the Gothic Wars. From early on, he was closely involved in religious matters, planning a school of Christian higher education with archbishop Agapitus. His estate in southern Italy hosted the monastery of Vivarium, for which he wrote a rule.304 Important literary works include a Chronicle of Rome and the World, a now lost History of the Goths—extensively used by Jordanes for his Getica (see below)—panegyrics on the Gothic kings, exegetical works and the Variae, a collection of state papers. As the Variae mainly consist of imperial letters sent to state officials, the baths are often mentioned in a legal context. In a letter to the praetorian prefect Fustus, the citizens of Spoletium (Spoleto, Italy) were to be offered free admission to the baths as ‘the Kingdom and the

300  Josh. Styl. 29. 301  Resp. Josh. Styl. 30 and 43. See glossary for winter and summer baths. 302  Josh. Styl. 87. 303  Josh. Styl. 43. 304  Barnish (1996a) 298–99.

59

Written Evidence of Baths

revenues prosper’.305 It is interesting to notice the reason for this generosity. cupimus enim libenter impendere quae ad salubrita­ tem novimus civium pertinere, quia laudes sunt nos­ trorum temporum celebrata gaudia populorum.306 We wish to pay freely for anything that tends to the health of our citizens, because the praise of our times is the celebration of the joys of the people.307 The same preoccupation for maintaining public health facilities is found in a letter from Theoderic to an architect, in which the emperor sings the praises of the thermal baths at Aponum (Abano Terme, Italy) and urges the architect to restore these important structures with the funds provided.308 In what seems to be an attempt to justify these expenses, the extraordinary ‘pious’ character of these baths is stressed. sed ut ipsum quoque lavacrum mundius redderetur, stupenda quadam continentiae disciplina in undam, qua viri recreantur, si mulier descendat, incenditur, propterea quia et ipsis altera exhibitio decora collata est: scilicet ne ardentium aquarum fecundissimum locum non crederent habuisse, unde plurima lar­ giretur, si uterque sexus uno munere communiter uteretur.309 Yet more wonderful is the moral purity of this fountain. Should a woman descend into the bath when men are using it, it suddenly grows hotter, as if with indignation that out of its abundant supply of waters separate bathing-places should not be constructed for the two sexes, if they wish to enjoy its bounty.310 The continuous popularity of these thermal baths is encountered in two different letters in which the king grants a furlough to visit such baths.311 However, the healing capacity of these waters is only made possible by the benevolence of God Himself.312 305  It could be that these thermae in Spoleto were the Turasi ther­ mas (the Baths of Turasius) mentioned in a different letter (Cassiod. Var. 4.24.2). 306  Cassiod. Var. 2.37. 307  Translation by Hodgkin (1886) 191. 308  Cassiod. Var. 2.39. 309  Cassiod. Var. 2.39.6. 310  Translation by Hodgkin (1886) 192. 311  Baiae in the case of Var. 9.6 and the baths of Bormio in Var. 10.29. 312  Cassiod. Var. 10.29.4.

The ‘salubriousness’ of a city depended on the baths and the aqueducts that provided the water. The farmers near Ravenna were urged to maintain the aqueduct by cutting down shrubs and vegetation contaminating water for drinking and the public baths.313 For Rome, the water flowing from the Aqua Claudia to the different baths and private houses was presented as a more wondrous sight than the flooding of the Nile.314 How different would the city look if it were not for the large public thermae and their glimmering pools of crystal clear water?315 In an interesting remark, the size and splendour of the empire and its cities in long lost days of glory are deduced by the impressive size of its thermae. Apparet, quantus in Romana civitate fuerit populus, ut eum etiam de longinquis regionibus copia provisa satiaret, quatenus circumiectae provinciae peregri­ norum victui sufficerent, cum illi se ubertas advecta servaret. nam quam brevi numero esse poterat, qui mundi regimina possidebat! Testantur enim turbas civium amplissima spatia murorum, spectaculorum distensus amplexus, mirabilis magnitudo therma­ rum et illa numerositas molarum, quam specialiter contributam constat ad victum.316 It is evident how great was the population of the city of Rome, seeing that it was fed by supplies furnished even from far off regions, and that this imported abundance was reserved for it, while the surrounding provinces sufficed to feed only the resident strangers. Never could a people that ruled the world be small in number. For the vast extent of the walls bears witness to the throngs of citizens, as do the swollen capacity of the buildings of entertainment, the wonderful size of the baths, and that great number of water-mills which was clearly provided especially for the food supply.317 It is clear that the size of the public baths in Cassiodorus’ time could not match the large imperial thermae. Yet, going to the baths with one’s acquaintances remained one of the pleasures of city life, just as strolling through the forum or attending a banquet.318 Even the monastery on Cassiodorus’ estate had a bathhouse to ensure

313  Cassiod. Var. 5.38.2. 314  Cassiod. Var. 7.6.4. 315  Cassiod. Var. 7.6.3. 316  Cassiod. Var. 11.39.1–2. 317  Translation by Barnish (1992) 161. 318  Cassiod. Var. 8.31.8.

60 a proper hygiene and some worldly pleasure among the monks.319 Gregory the Great Gregory (lived ca. AD 540–604) earned the epithet Magnus for his valiant administration of the city of Rome as archbishop between AD 590 and 604. Already prefect of Rome in AD 573 and later adviser to archbishop Pelagius II, his experience in state matters enabled him to navigate Rome through turbulent years of war, flood, famine and pestilence. He ensured Rome’s food supply through papal estates and negotiated with Lombard leaders to maintain peace. In ecclesiastical matters, he tried to convert the Lombards (who had adopted a form of Arianism) and sent missionaries to Britain. His writings include moralizing and mystic homilies and Dialogues, while his prose reveals a rhetorical training.320 In the works of Gregory, bathing appears as a manner of maintaining personal hygiene. Just as in other Christian authors, taking a bath was, however, not enough to clean away sin. A passage tells the story of a man who molested his god-daughter and went to the baths the following morning ‘si acqua balnei lavaret macuiam peccati,’ (‘as though the water of the baths could have washed away the stains of his sin’).321 The ‘superficial cleaning’ by bathing is also reflected in a letter to a bishop named Augustine: ‘the usage of the Romans from ancient times has always been for a man after intercourse with his own wife both to seek the purification of the bath and to refrain reverently for a while from entering the church.’322 It is obvious that the bath did not ensure to be ‘spiritual clean’ as one still had to wait before entering a sanctuary. The baths, as Gregory put it, were for maintaining the body, not for the mind or for the senses.323 In a different work, a bishop of Padua is said to have visited the public baths on advice of his physician.324 Besides the confirmation that bishops did indeed visit the baths for health-related purposes, the rest of the story is even more interesting. When the bishop arrived at the bathhouse, he found the ghost of Paschachius, an opponent of archbishop Symmachus, standing in one of the pools. This former deacon had been condemned to 319  Cassiod. Inst. 1.29.1. Cassiodorus was fully aware of the relaxing, even weakening effect of bathing. He uses the metaphor of ‘softening pools’ to describe a gentle approach in state matters (Var. 12.3.1). 320  Barnish (1996b) 657. 321  Greg. Dial. 4.32. Translation after Gardner (1911) 216. 322  Greg. Ep. 11.64; translation by Barmby (1898) 79. 323  Greg. Ep. 1.13. 324  Greg. Dial. 4.40.

chapter 2

haunt the baths as punishment for his sins. The idea of a wandering ghost inside a bathhouse is a literary topos also found in a passage about a priest that washed frequently at a thermal bath and one day visited the public baths.325 He was welcomed by a man who subsequently provided the services of a capsarius. The next time the priest visited the baths, he wanted to reward the good man with some loaves of bread, but was surprised to find out that the capsarius was in fact the ghost of the deceased owner of the bathhouse who had been condemned to wander around his establishment.326 A last bath-related fait divers worth mentioning concerns the use of baths by different ethnic groups. In a letter to Domitian, bishop of Mytilene during the reign of Maurice, Gregory hints at the fact that black people were admitted to certain public baths. The passage goes as follows: Moreover, though I grieve that the Emperor of the Persians has not been converted, yet I altogether rejoice for that you have preached to him the Christian faith; since, though he has not been counted worthy to come to the light, yet your Holiness will have the reward of your preaching. For the Ethiopian, too, goes black into the bath, and comes out black; but still the keeper of the bath receives his pay.327 We cannot be sure if the Ethiopian was a slave or not, nor if he could bathe at the same time or the same bathhouse as Roman citizens.328 Gregory probably used the statement just as a figure of speech to point out ‘an obvious fact’, i.e. black people remained black, even when they washed themselves. Nevertheless, the image of a black person at the baths, who had paid to enter and thus was not part of the personnel, was a mental picture that could be summoned by Gregory’s audience. Gregory of Tours With Gregory of Tours (lived ca. AD 538–594?), we find ourselves in the Frankish kingdom in northern Gaul. Born into a prominent religious and political family, 325  Greg. Dial. 4.55. 326  The idea that public baths were haunted by ghosts or demons is a topos that was quite widespread in Late Antiquity and after, see for example the Acta Iohannis 24, 40 and 44 (Yegül (2010) 202–203). A 17th c. drawing of Luca Ciamberlano still shows St. Philip Exorcising the Thermae of Diocletian in Rome (now at the British Museum). For demons in the Medieval hammams, see Benkheira (2003), 423–28. 327  Greg. Ep. 3.67; translation by Barmby (1895) 142. 328  For the difficulties of identifying slaves at the baths, see Fagan (1999b).

Written Evidence of Baths

young Gregory soon became deacon of Lyons and later bishop of Tours by the appointment of king Sigebert in AD 573. He was able to maintain his position as bishop despite the rapid usurpations and successions of kings. His most important work is the Libri Historiarum (Histories). Concerning baths, Gregory repeats the anecdote that Constantine ordered the execution of his wife Fausta by throwing her in a hot bath.329 Similarly, king Theodahad is said to have condemned his cousin and co-ruler Amalasuntha to death in a hot bath: Quae nec mora inter arduos vapores ingressa, in pavimento conruens, mortua atque consumpta est.330 And when she entered the hot vapours she fell at once on the pavement, and died, and was consumed.331 Furthermore, the baths are presented as markers for a decadent lifestyle, especially for monks, nuns and the clergy in general. The sinful bishops Salunius and Sagittarius bathed and dined until dawn, while the nuns in the monastery of lady Radegunda were accused of bathing with men in the monastery’s bathhouse.332 Radegunda defended her case by claiming that the baths had been opened to the servants until the local limestone—with which the building had been erected—no longer emanated a bad smell. Other 6th c. Latin Authors Jordanes, an historian writing around AD 550, wrote a history of the Goths, the Getica, which, so the author claims, borrowed heavily from the History of Cassiodorus. When discussing the Gothic victories in Thrace at the beginning of the 4th c., a passage recounts how the Gothic troops halted for several days at the thermal baths close to Anchiali (Pomorie, Bulgaria).333 The anecdote seems to stress how the Goths, even before settling on Roman territory, already enjoyed ‘civilized’ pastimes such as bathing. The popularity of Roman-style bathhouses is later reflected in an episode in which king Theodahad orders Amalasuntha to be strangled in his private baths.334

329  Greg. Tur. Hist. 1.36. See already Sidonius Apollinaris, the Epitome De Caesaribus and Philostorgius. 330  Greg. Tur. Hist. 3.31. 331  Translation by Brehaut (1965) 69. 332  Salunius and Sagittarius: Greg. Tur. Hist. 5.20; nuns: Greg. Tur. Hist. 10.16. 333  Jord. Get. 20.109. 334  Jord. Get. 59.306. See also Greg. Tur. Hist. 3.31.

61 In the 6th c. texts known as the Anonymous Valesianus, king Theoderic is attributed the erection of public thermae in Verona and Ticinum (Pavia, Italy).335 The continued popularity of large thermae in Constantinople is made clear by the chronicler Marcellinus Comes, who mentioned the dedication of the Thermae of Arcadius in AD 394, of Theodosius II in AD 427 and of the so-called Thermae Achillae in AD 443.336 The Christian poet Venantius Fortunatus (lived ca. AD 540–600), an acquaintance of Gregory of Tours and of the aforementioned Radegunda, praised the baths of two villas near Bordeaux.337 In the anonymous Antidotarium Bruxellense, a medical treatise, bathhouses are mentioned as still the preferred locations for drinking special mixtures to cure certain disorders.338 The prescription is based on Hippocratic ideas about the relaxing virtues of hot water, repeated by Theodorus Priscianus and Cassius Felix. 6th c. Greek Authors John Malalas Little is known about the life and work of John Malalas, a Christian chronicler who probably lived and worked in Antioch and later Constantinople (lived ca. AD 490– 570).339 His Chronicle covered the history of civilization from Adam to Justinian.340 The emphasis, however, lies on the emperors Zeno, Anastasius, Justin and especially Justinian, all of whom John devoted an entire book. His sympathies lay with Justinian and the ‘orthodox’ faith. As Malalas wrote his chronicle for an Antiochene audience, it is not surprising that most of the references to baths in his work concern the numerous bathhouses in Antioch itself. Malalas consistently mentioned the great baths that had been built through the ages on the command of the various emperors. Starting with Julius Caesar, the imperial baths were presented as examples of imperial largesse towards citizens. Hence, according to Malalas, Augustus, Tiberius, Domitian, Trajan, Hadrian, Commodus, Septimius Severus, Diocletian, Valens and Justinian all ordered public bathhouses to be built in Antioch. In some cases, these emperors also 335  Anon. Vales. 12.71. 336  Marcell. com. Chronicon, in PL 51, resp. col. 920; col. 925 and col. 927. 337  Venantius Fortunatus Carm. 1.18.15–16 and Carm. 1.20.21–22. For Radegunda and the critique on her monastery, see above Gregory of Tours. 338  Rose (1894) 366. 339  Jeffreys et al. (1986) xxi–xxiii. 340  The references to the texts of Malalas (e.g. 9.14) follow the subdivision of the Chronicle by Jeffreys et al. (1986), while the numbers in italic (e.g. 222) refer to the pages of Dindorf’s edition in the Bonn Corpus of 1831.

62

chapter 2

urged wealthy senators to finance the construction of public baths.341 Imperial munificence is also mentioned for other cities of the empire, such as Laodicea (near modern Denizli, Turkey), Caesarea Maritima (Sdot Yam, Israel), Ephesus (Efes, Turkey) Nicomedia (Izmit, Turkey), Dara (a frontier town in south-west Syria), and especially Constantinople.342 In the capital, the famous Zeuxippos Baths are often referred to. Built by Septimius Severus on the location of an ancient Greek temple dedicated to Apollo Zeuxippos (‘the horse-yoking Apollo’), it was called the Zeuxippos Baths in reference to the bronze statue of the god, rather than with its official name of Severianum.343 These baths were finished by Constantine, who added columns, different marbles and bronze statues.344 Furthermore, Constantine decreed that admission should be free on the ‘birthday’ of the city. A part of these baths was later damaged by fire during the infamous Nika Riots in AD 532.345 Some interesting information can be deduced from Malalas’ attestations about baths. Apparently, the name of bathhouses was often dedicated to the ‘patron’ who financed its construction. In many cases, this was the emperor, so that we find names as the Commodium, the Severianum or the Diocletianum.346 The names could also refer to private citizens, as was the case for the baths of Thermos in Alexandria or the Livianum in Antioch.347 A second important element for naming a bath seems to have been the presence of a specific feature, such as a well-known statue.348 The case of the Zeuxippos Baths—as a popular name, rather than the official one—has already been mentioned.349 Malalas also referred to a ‘private bath’ of a certain Paris, dancer and lover of Domitian, which was called ‘Paradise’. The same emperor also built a bath in Antioch that was nicknamed ‘Medeiea’, after a statue of Jason’s unfortunate lover found inside this building. Similarly, the baths of Icarus in Tripolis (modern Tripoli, Lebanon) were

named after a statue of Icarus and his father Daedalus on display in the baths. A last interesting remark about names concerns the fact that baths could change names over time. Malalas mentioned the baths of Ampelion in Antioch, which initially were named the Agrippaion, after Augustus’ general.350 For newly built baths, Malalas speaks about prefects that were sent from the capital to oversee the works and had to make sure that the money was well spent.351 In some cases, senators were appointed to carry out this supervision.352 The construction site was sometimes chosen following the discovery of a new spring, or often near existing public buildings, such as a hippodrome or basilica.353 Not surprisingly, the construction of baths was often linked to the construction of an aqueduct.354 The fundamental interaction of the two structures is stressed in a passage about a general water shortage in AD 563, prompting the closing of the public baths.355 Besides the erection of new bathhouses, Malalas also stresses the important restoration works carried out on the orders of emperors and wealthy patrons. Damage caused by earthquakes, consistently called the ‘wrath of God’, were indicated as the prime origin of such needed repair.356 New constructions and restoration works on baths were sometimes part of a more general urban program. Under Septimius Severus, officials in Antioch bought the property of rich lady called Livia so they could build a large bathhouse on this newly acquired land.357 We learn that Marcus Aurelius demanded the restoration of the baths of Centenarium in Antioch, damaged by an earthquake during the reign of Trajan, while ordering at the same time the erection of a nymphaeum and a mouseion.358 Commodus built his Commodianum while starting work on a temple of Athena just across the street.359 After an earthquake, Theodosius II seized the opportunity to start an urban renewal program in Antioch by building and restoring baths, colonnades,

341  Malalas Chron. 10.19; 244 and 11.14–15; 279. 342  Laodicea: Malalas Chron. 12.21; 294; Caesarea Maritima, Ephesus and Nicomedia: Malalas Chron. 14.20; 363; Dara: Malalas Chron. 16.10; 399. 343  Malalas Chron. 12.20; 291–292. 344  Malalas Chron. 13.8; 321. 345  Malalas Chron. 18.71; 474. 346  C  ommodium: Malalas Chron. 12.2; 283.; Severianum: Malalas Chron. 12.21; 294; Diocletianum: Malalas Chron. 12.38; 307. 347  Baths of Thermos: Malalas Chron. 12.21; 293; Livianum: Malalas Chron. 12.21; 295. 348  See also Dunbabin (1989) 16. 349  The coexistence of an official and a popular name is still a well-attested fact for modern sport stadia. A good example is the mythical Estádio Jornalista Mário Filho in Rio de Janeiro (Brazil), better known as the Maracaña.

350  Paradise: Malalas Chron. 10.49; 263; Medeiea: Malalas Chron. 10.49–50; 263; baths of Icarus: Malalas Chron. 14.29; 367; baths of Ampelion: Malalas Chron. 9.14; 222. 351  Malalas Chron. 10.18; 243. 352  Malalas Chron. 10.19; 244. 353  Spring: Malalas Chron. 9.14; 222 (Antioch); 10.10; 243; 12.21; 293– 295.; hippodrome: Malalas Chron. 12.38; 307–308 (Antioch); basilica, such the baths of Olbia in Antioch: 16.6; 397. 354  Malalas Chron. 9.5; 216; 10.18; 243; 11.9; 276; 11.14–15; 278–279; 16.20; 409; 18.17; 435 355  Malalas Chron. 18.147; 496. 356  In Antioch 10.18; 243; 11.30; 282; 17.19; 423; on Crete 14.12; 360; in Tripolis (Palestina) 14.29; 367; in Constantinople 18.118; 487. 357  Malalas Chron. 12.21; 295. 358  Malalas Chron. 11.30; 282. 359  Malalas Chron. 12.2; 283.

Written Evidence of Baths

churches, public spaces and the harbour.360 Anastasius likewise provided the city of Dara with baths, cisterns, warehouses, etc.361 In some rare passages, Malalas even informs us about the secondary use of baths. Constantine is said to have built a large church on the site where a bathhouse used to stand. The bathhouse was, however, not destroyed without reason: ‘the bath was old and ruined by time and unfit for bathing,’362 although it is difficult to assess whether the baths were already closed to the public or not. We also learn that the baths of Commodus, the Commodianum, in Antioch were later transformed into the praetorium of the consular governor in Syria.363 One could imagine that the lofty walls of the frigidarium would make great audience halls, while the luxurious decoration also must have fitted its new purpose. The positive attitude with which Malalas approaches the ‘pagan’ emperors, some of whom had persecuted Christians, may come as a surprise, yet the respect shown towards rulers of a long-lost past is not uncommon. For Byzantine chroniclers, the ‘Roman’ roots of their empire were something worth remembering and repeating. Besides, praising the munificence of these emperors did not prevent Malalas from criticizing their anti-Christian measures. When Trajan ordered the execution of some Christian women, he also commanded that their ashes be mingled with the brass that was to be used in the boilers of his new baths.364 But the vapours that arose in the heated parts intoxicated the bathers, prompting the emperor to remove the boilers from the baths, melt them down and create 5 brazen pillars, which he had set up in the baths. Trajan sneered that it was he, not the Christian God, that had ensured the eternal life for these women. The spiritual aspect of water in a Christian context is also found in a different passage. A mimeactor had to play the part of a Christian in a parody of baptism.365 He was submerged ‘in a large bathhouse tub full of warm water’ to mimic the baptismal font, but when he re-emerged, the mime-actor claimed he really had been converted to Christianity and refused to play out his part. He was eventually stoned to death by the furious spectators, hence becoming a true Christian martyr.

360  Malalas Chron. 14.20; 363; see also Justinian: 17.19; 423. 361  Malalas Chron. 16.20; 409. 362  Malalas Chron. 13.3; 318. Translation by Jeffreys et al. (1986) 172. 363  Malalas Chron. 13.30; 338. 364  Malalas Chron. 11.10; 277. 365  Malalas Chron. 12.50; 314–315.

63 Alexander of Tralles Alexander (writing around AD 565) was a travelling physician who eventually settled in Rome. His Therapeutics has a clear Galenic approach of medicine, without falling back into a dogmatic stance.366 It included popular remedies, spells, chants, amulets and different religioussuperstitious elements. Just as in other medical writings, Alexander stresses the heating and humidifying properties of baths as well as their cooling and contracting properties.367 The temperature of the water has an impact on the different fluids in the body and these fluids appear in different ratios in different people.368 Hot water makes these more liquid, augmenting the volume. In these cases, hot baths can be dangerous, especially to patients with a ‘moist nature’ that are suffering from high fever.369 Tepid baths are presented as the best option in most cases,370 yet we do not know what temperature was considered tepid. A cold bath—recommended for a severe 4-day fever—is called a ‘soldiers’ bath,’ reminding us of the passages in the SHA and Vegetius.371 The emphasis in Alexander’s therapies often rely on a healthy diet, physical exercise and an adapted bathing routine.372 However, bathing should never take place just after eating.373 After the baths, a light snack and some wine is considered beneficial.374 Even if Alexander was not a supporter of surgery, he does prescribe decoctions and ointments that should be administered before, during, or after bathing. Different types of oil are recommended, sometimes in combination with natrium or with the foam that develops on top of the water boiler.375 The oils are often applied in the tepid rooms, but are washed away before entering the pools.376 The bathing routine often ends with anointing the whole body with oil and wrapping it in linen. Honey-based mixtures and soap are also mentioned.377 It is interesting to remark that Alexander recommends that some therapies take place in portable tubs in private 366  The text was edited and translated by T. Puschmann (1878 and 1879). The relevant passages mentioned in this book refer to the Greek text found in this edition. 367  Puschmann (1878) 363. 368  Puschmann (1878) 375–77. 369  Puschmann (1878) 293–95. 370  Puschmann (1878) 363; 603. 371  Soldiers’ baths: Puschmann (1878) 439; SHA Alex. Sev. 30.4–5 and 53.7; Veg. Mil. 1.3. 372  Puschmann (1878) 309; 411; 597. 373  Puschmann (1878) 326; 541. 374  Puschmann (1878) 553; 603; 615. 375  Oil: Puschmann (1878) 375; natrium/foam: Puschmann (1878) 405. 376  Puschmann (1878) 363; 375. 377  Puschmann (1878) 308; 427.

64 houses, rather than at the public baths.378 In the case of narcolepsy, the patient should first bathe in a tub at home, because the temperature in public baths is too intense.379 The tub should preferably be made of wood and should have a seat. Some of his therapies should be carried out in public bathhouses, particularly those requiring steam baths.380 Evagrius Scholasticus Evagrius (lived ca. AD 535–600) was a scholar and aid to Gregory, ‘Chalcedonian’ patriarch of Antioch (AD 571–593). He wrote a Church History covering the period between AD 431 and 594, focusing mainly on the east. Evagrius obviously borrowed from John Malalas.381 The public baths are presented as standard facilities of a town throughout the work. We learn that Theodosius II and his wife Eudocia spent 200 pounds of gold for the restoration of the Baths of Valens in Constantinople, which had been damaged by fire.382 The same concern for repairing the baths of Trajan, Severus and Hadrian is found in a passage about an earthquake at Antioch in AD 474.383 It is interesting to notice that a bathhouse that had been out of use for a while was restored now that all the others baths of the city lay in ruins. In AD 589, Antioch was again hit by an earthquake, and again, the public baths—the ones that had separate divisions according to the seasons—suffered hard.384 Numerous fires also started in structures with hearths and ovens, such as kitchens and baths.385 In newly founded cities such as Anastasius’ Dara (Syria), bathhouses were still part of the essential infrastructure, which also included town walls and churches.386 Important members of the local elite still adorned the public baths with statues.387 The continued popularity of public baths can also be deduced from a punitive measure in Alexandria after riots following the Council of Chalcedon. The distribution of provisions was limited, all spectacles were suspended and all public baths were closed.388 Even if bathhouses were thus regarded as normal parts of the

378  Puschmann (1878) 327. 379  Puschmann (1878) 533. 380  Puschmann (1878) 296. 381  Markschies (2006). 382  Evagr. 1.20. 383  Evagr. 2.12. 384  Evagrius makes a reference to the city’s summer and winter baths (see glossary, thermae aestivae and thermae hiemales). 385  Evagr. 6.8. 386  Evagr. 3.37. 387  Evagr. 3.28. 388  Evagr. 2.5.

chapter 2

townscape, the restrictions for monks and clergy are still present.389 Procopius The historian Procopius (ca. AD 490/507- after 555) studied law and rhetoric before obtaining a post on the staff of Justinian’s general Belisarius. He loyally accompanied his superior during the campaigns in Persia (AD 527–531), Africa (AD 533–536) and Italy (AD 536–540), returning to Constantinople in AD 542. His most important work is entitled History of the Wars of Justinian (De Bellis, published in AD 551),390 an account of the military campaigns described above. The Secret History (Anecdota), a harsh critique on Justinian’s policy, must have circulated clandestinely, as “a kind of sub-text to the History of the Wars.”391 It provides a counterbalance to the panegyric tone of On the buildings of Justinian (De Aedificiis, ca. AD 553–555), which praises the emperor’s building activity and was composed at Justinian’s request. There are numerous attestations to baths and bathing in the different works of Procopius. It should come as no surprise that the majority are found in On Buildings. In total, Procopius speaks of 12 newly built baths and 4 restorations of existing baths.392 It is very characteristic that the baths are mentioned in combination with markets, stoas and theatres ‘καὶ εἴ τι ἄλλο πόλεως μεγάλης ἐς κόσμον διήκει.’ (‘and whatever else contributes to the embellishment of a great city’)393 To build and restore these baths, special ‘master-builders’ were put into service, as is mentioned for the baths in the military camp of Zenobia in Mesopotamia.394 That baths still were an essential part of a town’s fabric is illustrated by a passage recounting how the Persian king Khusro I ordered a city to be built for the captured Antiochenes. It was equipped with baths and a hippodrome for their enjoyment.395 The inability to bathe was nothing short of a catastrophe. In the Secret History, the scandalous behaviour of the elite is exemplified by stressing how they refused to repair aqueducts and hence deprived the outraged citizens of their baths.396 A well-tried siege tactic that cut off a city’s water supply by destroying or diverting the main aqueduct not only led to a shortage 389  Evagr. 1.21. 390  Although written in Greek, the texts are referred to by their Latin title (following the OCD). 391  Browning (1996) 1251. 392  For newly built baths, see Aed. 1.10.3; 1.11.20; 2.8.24–25; 3.4.18; 4.1.23; 4.10.21; 5.2.4; 5.3.19; 5.4.17; 6.1.13; 6.2.6; 6.4.11; 6.5.10. For restorations, see Aed. 2.6.10–11; 5.3.3; 5.3.7; 5.9.34. 393  Procop. Aed. 3.4.18. Translation by Dewing (1940) 201. 394  Procop. Aed. 2.8.24–25. 395  Procop. Bell. 2.14. 396  Procop. HA 26.23.

65

Written Evidence of Baths

of drinking water, but also deprived the despairing citizens of their baths. This was the case when the Goths laid siege to the army of Belisarius in Rome.397 The public baths built in military camps and frontier cities seem to have embodied ‘the Roman city’. On two occasions, Procopius recounts how the Persians sent a message to the Roman command, asking them ‘to prepare the baths’.398 This declaration of war meant as much as ‘we are going to take your city and then enjoy your typical Roman luxuries’. The Persians were not the only non-Romans that acquired a taste for Roman-style bathing, as Procopius proclaims that the Vandals in Africa ‘used to indulge in baths, all of them, every day’.399 The wife of a Gothic general even incited her husband against his rival in command, as the latter’s wife had humiliated her in the bathhouse by showing off her jewellery and her throng of attendants.400 Besides the importance of baths for the appearance of a city and the popularity of bathing among both Romans and non-Romans, Procopius also outlined the hygienic and curative aspect of bathing. When discussing the Justinianic plague of AD 542, Procopius remembered how some people were bathing to cure the disease and recovered, while others did just the same but died anyway.401 Furthermore, the thermal baths of Anchiali are used as a geographical reference point in the description of Thrace, hinting at the fame of this curative centre.402 Justinian also ordered a bathhouse to be built at the natural spring of Pythia in Bithynia (possibly modern Termal, Turkey) so that the hot waters were put to good use. An aqueduct provided sweet water to this facility, so that bathers could also enjoy the pleasures of normal baths.403 Yet, the disapproval of excess is also reiterated, for example when commenting on 397  Procop. Bell. 5.19–20. 398  Procop. Bell. 1.13 and 1.14. Some of these Roman commanders were accompanied by their personal bathing assistant while on campaign (Bell. 1.13). Such a declaration of war could also be formulated as ‘make your palace ready for me’, cf. John Malalas quoting Atilla’s ambassador to Valentinian (Chron. 14.10). 399  Procop. Bell. 4.6.6. The popularity of baths in Vandal circles is also attested in the epigrams of the Anthologia Latina (Busch (1999) 240–65). See also Isidorus of Seville, mentioning a Vandal king that died in the baths when he mocked the Holy Trinity (Chron. Mai. 392). 400  Procop. Bell. 7.34–40. 401  Procop. Bell. 4.6. However, Procopius does not specify where they bathed. For curative bathing in public bathhouses, see above the passages in Theodorus Priscianus, Oribasius, Cassius Felix and Alexander of Tralles. 402  See also the attestations of Anchiali in Jordanes (above) and John of Ephesus (below p. 66). 403  Procop. Aed. 5.3.16–20.

the empress Theodora, who ‘used to enter the bath very early and quit it very late’.404 The Epigrams in the Anthologia Palatina The Anthologia Palatina is a compilation of some 3,700 Greek epigrams covering no less than 16 centuries. The collection of 15 books was arranged according to subject. In book IX, the so-called epideictic epigrams treat a wide variety of subjects, including baths.405 Of these bath-related epigrams, some can be attributed to specific poets, mainly of the Justinianic era. Book II is entirely devoted to the statues found at the Zeuxippos Baths in Constantinople, described by Christodorus of Coptus. Leontius Scholasticus wrote three epigrams involving baths. Two epigrams praised small yet luxurious baths, one next to the Zeuxippos Baths and another next to the gates of the public baths, built ‘for excellence, not competition’.406 The third epigram concerns ‘royal hot baths’, where ‘the spring is born hot of its own accord’.407 Agathias Scholasticus (lived AD 532–580), an historian that continued the work of Procopius, wrote an epigram about a bathhouse in Constantinople involving the judgement of Paris. Just as Leontius, he also praised a thermal bath near Smyrna (near Izmir, Turkey).408 Paulus Silentiarius, a court official and poet in Constantinople who wrote the Ekphrasis—the detailed literary description—of the restored Hagia Sophia (AD 562), composed several erotically tinted epigrams. One recounts how men yearned for the women in the women’s baths, lying next to the men’s bath, only separated by a little door.409 Marianus Scholasticus wrote similar epigrams involving the classical personification of ‘Love’ and the nymphs.410 Such allusions to classical divinities are also apparent in the epigrams of Damacharis Grammaticus. A bathhouse is brought up in connection to the judgement of Paris, as the three goddesses prepared themselves to convince the young judge.411 Macedonius the Consul, a court official and occasional poet, praised the service of the doorman of a bathhouse in the Lycian quarter (Constantinople?), noting the time when each bather entered and preventing voyeurs from peeping. Besides alluding to the Naiads 404  Procop. HA 15.6–7; translation by Dewing (1935) 177. 405  Ant. Pal. 9.606–640. 406  Resp. Ant. Pal. 9.614 and Ant. Pal. 9.624. 407  Ant. Pal. 9.630. 408  Constantinople: Ant. Pal. 9.619; Smyrna: Ant. Pal. 9.631. 409  Ant. Pal. 9.620. 410  Ant. Pal. 9.626–627. 411  Ant. Pal. 9.633.

66 and the Graces, he ends the epigram with a reference to Homer.412 The classical heritage was indeed still very much alive in the higher circles of society. Only one epigram, by Johannes Grammaticus, praised the restoration of a bathhouse. In this case, the Baths of the Horse in Alexandria enjoyed a facelift thanks to the benevolence of the emperor.413 An entire book of epigrams (II) was dedicated to the statues that adorned the Zeuxippos Baths, called ‘τὸ δημόσιον γυμνάσιον τοῦ ἐπικαλουμένου Ζευξίππου’ (‘the public gymnasium called Zeuxippos’). The composer, Christodorus of Coptus, did not actually describe the statues themselves, but rather the accomplishments of the represented characters. These include mainly wellknown Greek philosophers (Plato, Aristotle, etc.), writers (with special attention to Homer), statesmen (Pericles, Demosthenes, Sarpedon), and gods (Apollo, Aphrodite, Hermes, Artemis). Only two Roman statesmen, Caesar and Pompey, were mentioned, while only Virgil and Apuleius represented the Roman literary scene.414 Other 6th c. Greek Authors The historian Zosimus, an adherent of the classical cults, wrote a New History of the Roman empire, starting with the first emperor and ending with the sack of Rome by Alaric in AD 410. He repeated the anecdote of Constantine commanding the assassination of his wife Fausta in the baths.415 The act of murdering someone who is bathing has already been encountered with numerous other authors and is repeated in a passage about one of Valens’ generals, who killed the indigenous people of Thrace while they bathed in the river.416 The Alexandrian merchant Cosmas Indicopleustes wrote a self-illustrated Christian Topography around AD 540, in which he combined geography, astronomy and Nestorian theology. He referred to baths on two occasions. In an exposé ‘against those who, while wishing to profess Christianity, think and imagine like the pagans

412  A  nt. Pal. 9.625.  nt. Pal. 9.628. 413  A 414  For two statue bases found in excavations in the vicinity of the Zeuxippos Baths in modern Istanbul, see Basset (1996). The inscriptions on the bases read Αἰσχήνης and Ἑκάβη (dated AD 330). Both of these statues were mentioned by Christodorus. 415  Zos. II. Such an attack on Constantine infuriated Evagrius, who called Zosimus a liar who still held on to the accursed and foul religion of the Greeks (Evagr. 3.40). Zosimus was not the first to mention this fact, see above Sidonius Apollinaris, Epitome de Caesaribus, Philostorgius and even Gregory of Tours. 416  Zos. IV.

chapter 2

that the heaven is spherical’,417 the baths are mentioned as an example of how vapour (water) could not be ‘drawn up’ from the air by heat, as baths are heated from below. The water is thus rather likely to be ‘pushed up’.418 In a description of Heaven as a vaulted chamber, the comparison with the spacious roof over a bath is made.419 The use of baths in comparisons points to the fact that Cosmas and his readers were still familiar with the architecture and the functioning of these structures. John Moschos (AD 550–619) recounts how a pious virgin was haunted by a demon, which incited her to go to the baths and seduce men.420 In a different passage, the son of a rabbi is punished for his conversion to Christianity by throwing him in the furnace room of a bathhouse. Instead of burning alive, the new Christian aura of the child put out the flames and hence the bath water stayed cold.421 6th c. Syriac Authors John of Ephesus John of Ephesus (ca. AD 507–588), also called John of Asia, was a Syrian monk. With the support of empress Theodora and hence also her husband Justinian, John was sent to the region of Ephesus as a missionary. He was later ordained bishop of the same city. Under Justinian’s orthodox successor Justin II, John was imprisoned and later banished. His writings include an Ecclesiastical History and Lives of Saints. John mentions baths in a context of healing and medicine. A passage recounts how an opponent of the bishop of Constantinople was incarcerated, but was nevertheless permitted to visit the hot baths when he fell ill.422 On several other occasions, the emperor Justin himself is said to have visited thermal baths.423 The destruction of the city of Anchiali, also mentioned by Jordanes and Procopius is accompanied by a description of its thermal baths.424 The fondness for baths at the imperial court is also reflected by a passage in which Tiberius II (reigned AD 578–582) ordered a new wing to be added to the palace in Constantinople, including spacious stables and a top-end bathhouse.425 417  As translated by J. Crindle (1897) 7, from the Greek title of chapter one. 418  Cosm. Ind. Book 1 as found in PG 88, col. 63. 419  Cosm. Ind Book 4 as found in PG 88, col. 185. The comparison of the celestial sphere with the vaults of a bathhouse are already found in the writings of Basil of Caesarea (see above p. 45). 420  Joh. Moschus Prat. 244. 421  Joh. Moschus Prat. 227. 422  Joh. Eph. Hist. eccl. 1.15. 423  Joh. Eph. Hist. eccl. 1.26–27 and 1.40. 424  Joh. Eph. Hist. eccl. 6.49. 425  Joh. Eph. Hist. eccl. 3.23.

Written Evidence of Baths

67

The erection of such baths clearly demanded a certain expertise and John mentioned that the Avar khagan sent an envoy to Justin II to ask for skilled artificers and masons to build him baths.426

ies of water, he also mentions the hot springs that were often used in bathhouses, referring to thermal baths.432 In discussing the use of ceramic dishes, the hydria is mentioned as a vessel used in the baths.433

Other 6th c. Syriac Authors In the anonymous Chronicle of Edessa, dated after AD 540, the fact that the so-called ‘warm bath of the Iberians’ failed to function for three days was important enough to include it in the chronicle of the city.427

John of Nikiu John, a Coptic monk and then bishop of Nikiu in Egypt, lived during the Islamic conquests of the second half of the 7th c. He wrote a Chronicle starting with Adam and Eve and ending with the Islamic conquest of Egypt. The work has been badly preserved, with the only two surviving 17th and 18th c. copies being ancient Ethiopic (Ge’ez) translations of an Arabic translation of the original Greek text. According to John, King Solomon (son of David) was the first to build baths and places for study and education.434 It is interesting to notice how baths are linked to both the greatest king of Israel and to places of knowledge and culture. This link is also found in a passage about Antoninus Pius, who built many baths and academies in the east.435 The baths also serve as the setting for a miraculous event: the same story of the ashes of 5 executed Christian women being mixed with the bronze of the water boilers in the baths and causing nausea among the bathers, which has already been mentioned in the Chronicle of Malalas.436 This is not the only link between Christians and bathhouses in John’s work. During the reign of Justinian, the sinful bishop of Alexandria was deposed from office, as he was accused of sodomy with a deacon in the baths.437 In a more pious context, the father of the Persian king Khusro is said to have been converted to Christianity and was apparently baptised in a pool of the public baths.438 In other passages, the importance of the bathhouses in a city’s fabric is also stressed. The Tainadonhus Baths and the Baths of the Syrian Nation are mentioned to demonstrate the extent of the damage done to Antioch by an earthquake during the reign of Justin.439 When some corrupt governors in Egypt burnt down the city of Busir, John specified that they even destroyed the public baths.440 Even for

7th c. Authors Isidorus of Seville Isidorus of Seville, also called Isidorus Hispalensis (lived ca. AD 560–636), was bishop of Seville during the Visigoth reign in the Iberian Peninsula. He wrote several historical and theological works, but his most influential was the incomplete Etymologiae, an encyclopaedic text on language, arts, law, religion and other topics.428 In the chapter about public buildings, Isidorus used three different words for baths. Thermas appellatas quod caleant; Graeci enim THERMON calorem vocant. Balneis vero nomen indi­ tum a levatione maeroris; nam Graeci BALANEION dixerunt, quod anxietatem animi tollat. Haec et gymnasia dicuntur, quia ibi athletae uncto corpore et perfricato manibus exercitantur; nam GUMNASION Graece, Latine exercitium dicitur.429 Hot baths (thermae) are so called because they are warm, for the Greeks call heat θέρμον. Baths (bal­ neum) are assigned their name from the idea of the lifting of sorrow, because the Greeks called it βαλανεῖον since it takes away one’s anxiety of spirit. Gymnasiums (gymnasium) are so called because there athletes are trained, with their bodies anointed and massaged, for γυμνάσιον in Greek means ‘training’ in Latin.430 He also mentioned the etymology of the word apo­ dyterium, after the Greek verb ἀποδύειν (to undress), and the word popina (a food stall) in connection with bathhouses.431 In a passage about the different variet426  Joh. Eph. Hist. eccl. 6.24. 427  Cowper (1865) 36. 428  Of Isidorus’s extant oeuvre, only the Etymologies has been considered for this study. 429  Isid. Etym. 15.2.39–40. The use of gymnasium to denote a bathhouse is also explained in Etym. 15.2.30. 430  Translation by Barney et al. (2006) 307–308. 431  Isid. Etym. 15.2.41–42.

432  Isid. Etym. 13.13.11. 433  Isid. Etym. 20.4.3. 434  Jo. Nic. 38.1. 435  Jo. Nic. 74.8. 436  Jo. Nic. 72.6–9. For baths as a background for miraculous events, see already above p. 66, John Moschos. 437  Jo. Nic. 92.7. 438  Jo. Nic. 95.23–25. Similarly, in a passage found in the treatise On Christmas by the Armenian author Ananias of Shirak (beginning of the 7th c.), Bishop Cyril of Alexandria is said to have ordered the public baths to be purified in order that a large group of ‘pagans’ could be baptised. 439  Jo. Nic. 90.24. 440  Jo. Nic. 97.5.

68 an Egyptian monk writing at the end of the 7th c., public baths were still an integral part of society and hence made a recognizable setting for different anecdotes. Sophronius of Jerusalem The monk Sophronius, who later became patriarch of Jerusalem (AD 634), wrote about the miracles of St. Cyrus and St. Johannes. These saints appeared to the faithful in a bathhouse near their sanctuary at Alexandria and their cult replaced that of Isis at the same thermal spring.441 One of the miracles recounted how a certain Theodora had slipped in the baths, possibly because of a mush of ointments, and suffered severe trauma to her neck and back. The saints intervened and cured the poor woman, thus triumphing over the evil demon that had been responsible for the accident.442 The passages not only remind us of the dangers lurking inside the baths—whether real (slippery floors) or symbolic (demons)—but also illustrates how Christian authors still tried to justify the use of thermal baths. It is interesting to note that all the sick had first consulted ‘conventional physicians’ before turning to the holy sources of Menouthis.443 Summarizing the Literary Attestations of Baths and Bathing Habits To summarize the late antique literary attestations of baths is a difficult task. The time span covers 4 turbulent centuries in which changing social and religious contexts gave rise to divergent text traditions. Regional circumstances, also evolving at a fast pace, further complicate the picture. However, some ideas seem to have been shared in different regions and over large periods of time. A first obvious observation is that these buildings were still important enough to write about. More than that, it has become clear that bathhouses still retained their place at the centre of towns and social life. From Ammianus Marcellinus, through Socrates Scholasticus and Joshua Stylites to John Malalas and Procopius, baths are mentioned as key locations within a town, giving their names to whole neighbourhoods and acting as rendezvous points for various official and trivial occasions. Even religious and juridical matters could be settled in the halls of bathhouses (Augustine, Socrates Scholasticus). In the work of chroniclers such as Ammianus Marcellinus, the Scriptores Historiae Augustae or John Malalas, the large imperial complexes—still denoted with the term thermae in Latin—are frequently linked 441  Montserrat (2005) 230. 442  Sophron. Narr. de mir. Cyr. et Joh. 9.8–11. 443  Montserrat (2005) 233.

chapter 2

to imperial patronage, even if the emperors in question were Christian. This is not only the case for the capitals Rome and Constantinople (Ammianus Marcellinus, SHA), but also for other important centres like Antioch (esp. Malalas) and Carthage (Procopius). Providing a city with bathing facilities was still presented as an important act of imperial benevolence, even if these baths may only have been mere shadows of the large thermae of the days of yore. It is also interesting to note that the ‘social aspect’ of the baths also had a more gloomy side in ancient literature. Chroniclers like Plutarch or Suetonius had already mentioned bathhouses as the setting for conspiracy and murder. The topos of baths as crime scene continued to be used in Late Antiquity, especially concerning the assassinations of emperors or high dignitaries (see amongst others SHA, Basil of Caesarea, Theodoret, Gregory of Tours or Jordanes). Besides historical accuracy, the popularity of this topos may have been inspired by the obvious vulnerability (nakedness) inside bathhouses, the large choice of ‘lethal possibilities’ (water, fire, slippery floors, suffocating steam) the buildings offered, or even a literary desire to stress the paradoxical duality of baths, bringing both joy and death, hence also being a reflection of the patrons who built them. Besides the social and monumental aspects of baths that benefitted the cities (and the good name of the patrons), the idea that bathing was good for the health also had its roots in earlier times. Medical writers such as Celsus (1st c. AD) and Galen (2nd c. AD), who themselves integrated Hippocratic principles, were quoted and commented upon. North Africa and Egypt in particular brought forth several medical writers, with Alexandria, traditionally the ‘capital of medicine’, likely still playing a major role. The integration of bathing into several therapies is based on the idea that hot water had a relaxing and humidifying effect, while cold water had a contracting and reinforcing effect. The succession of hot and cold pools served to restore an imbalanced harmony of bodily fluids, which was a prerequisite for a healthy body. This basic idea is found from the work of Theodorus Priscianus down to the writings of Alexander of Tralles. Cold baths in particular seem to have been prescribed for young men. Often, bathing was part of a therapy comprising a special diet and physical exercises (esp. Oribasius and Caelius Aurelianus). These therapies, especially the water-related parts, seem to have taken place inside public baths (esp. Cassius Felix). The popularity of using oils and decoctions, as well as the use of special types of water (seawater, alkaline water, sulphurous water, etc.) may even have encouraged a more individual form of bathing (see below, chapter 4). The continued popularity of natural thermal sources such as

Written Evidence of Baths

Baiae (Ausonius and Cassiodorus), Anchiali (Jordanes, Procopius or John of Ephesus) or near Alexandria (Sophronius of Jerusalem) is in line with this belief in the healing properties of water.444 Following this ‘medical appreciation’ of baths, we can remark that the idea of ‘decadence’ also persisted into Late Antiquity. First of all, one should not overdo bathing, as hot water had a weakening effect on body and mind. The examples of the ‘wicked emperors’ in the SHA spending all day inside the baths are a case in point. It should be noted that this critique did not come exclusively from a Christian perspective and had already been vented in earlier times (for example Libanius and earlier Seneca). The Christian writers may have attacked a lifestyle of decadence and luxury more virulently, because Christian doctrine paid so much attention to the poor and a sober way of life. As seen in the texts, bathing is often condemned when mentioned together with feasting and drinking. Displaying wealth by showing off jewellery and throngs of slaves also fitted this image of decadence (amongst others Ammianus Marcellinus, Basil of Caesarea, John Chrysostom, Palladius and Procopius). However, some writers, especially wealthy members of the senatorial elite (Ausonius, Julian, Sidonius Apollinaris), did not neglect the opportunity to praise the delights of bathing in sumptuous private baths, often as a part of a comfortable estate. The epigrams of the Anthologia Palatina, mainly composed in the east under Justinian, submerge us in a world of watery pleasure by describing baths in eloquent verses interspersed with references to the classical pantheon. This contrast between the warnings (of excess) and (at least poetical) praise is a long-lasting, yet hardly surprising debate for a phenomenon as important as bathing. The Christian attitude towards bathing has been the subject of much debate. Initially, modern scholars were eager to expand the harsh puritan thoughts of writers like Jerome to the whole Christian community. The evidence of pious Christians taking their baths alongside other Roman citizens, as found in the works of Eusebius, Gregory of Nazianzus, Gregory of Nyssa, Augustine, or Theodoret, seems to have been overlooked. There are also clear references to clergy being directly involved in the building of bathhouses (Theodoret, Palladius, Zacharias). Criticism of baths was nevertheless voiced by some Christian authors such as Jerome, Philoxenos, Cassiodorus and John of Nikiu. However, it mainly concerned the practice of men bathing together with women (balnea mixta) and the sexual temptations that lurked in the steamy corners of the baths. This last point of critique was especially aimed at virgins and ascet444  For the east, see Dvorjetski (2007) 393–424.

69 ics. The strong condemnation of baths in the writings of authors like Jerome, Eusebius, Sozomen, Palladius, Zacharias or Gregory of Tours must be understood as warnings towards these special categories of Christians, rather than as interdictions for all Christians. It is not surprising that such critique was voiced most virulently in the east, where the ascetic movement had found most support. Decadent bathing—for hours on end, assisted by slaves and followed by banquets—must have represented the sort of lifestyle that the ascetics renounced. An opinion that was widely promoted by the Church and its intellectual protagonists stressed that baths only provided a ‘superficial’ cleaning of the body, but could not wash away sins. The passages in Commodianus, John Chrysostom, Optatus, Philoxenos or Gregory the Great all stress this important limitation of bathing. This did not mean that Christianity did not attribute a purifying power to water: their own purification act, baptism, was in itself nothing more than a ritual bath. There is even some evidence that baptism was sometimes performed in the pools of a public bathhouse (John of Nikiu, Socrates Scholasticus, Sozomen, Palladius, and John Malalas). Furthermore, the terminology used to denote baptismal fonts is derived from bath-related vocabulary (see below p. 79). There are no indications that the Church formally prohibited the use of (public) baths. In fact, the aforementioned attestations of bishops and other members of the clergy inside baths, the involvement of clergy in the erection of bathhouses, and the fact that some monasteries had bathhouses (Zacharias) confirm that bathing was an accepted part of daily life in most Christian circles. To conclude, there are some aspects of bathing that surfaced more frequently in the literature of Late Antiquity. Thus, several authors mention some kind of magic or divine phenomenon taking place in or around bathhouses. In a non-Christian context, this mainly involves omens or divine interventions (Ammianus Marcellinus, Eunapius), while in Christian narratives, these pertain to miraculous events, or, on the contrary, demons and ghosts lingering in the pools (Itinerarium Burdigalense, Gregory of Nyssa, John Chrysostom, Theodoret, Sozomen, Gregory the Great, John Malalas, John of Nikiu). Natural hot springs in particular were attributed to some divine force, even by Christians (Cassiodorus). The curative powers of thermal sources, which had been ascribed to nymphs and local deities in the imperial period, were often reinterpreted in a Christian sense. Martyrs and saints were now responsible for the healing and protection of the bathers (Itinerarium Burdigalense, Cassiodorus). For man-made baths, the presence of the elemental fire and water justified the links with nymphs and demons. This mystical

70 aspect of baths was also reflected in the role that baths could play in cults (Eunapius), rites of passages (Gregory of Nazianzus) and Christian celebrations (Augustine, Socrates Scholasticus, Sozomen, Palladius, Gregory the Great). The changes in religion were not the only challenges Roman society had to cope with during Late Antiquity. Changes in the ethnic composition of society also meant new ways of defining what it meant to be ‘Roman’. Just as it had for Tacitus,445 bathing was seen as marker of Romanitas (Joshua Stylites, Procopius). From the Goths (Jordanes), through the Vandals (Procopius) to the Persians (Zacharias, Augustine, Joshua Stylites, Procopius) and Avars (John of Ephesus), non-Roman groups enjoyed the bathhouses and even tried to build them themselves (requiring the help of specialized Roman architects, see John of Ephesus). We can end the summary of the late antique literature by mentioning the few attestations to the secondary use and destructions of bathhouses. The work of John Malalas in particular is very informative, as it teaches us that in some cases baths were transformed into churches or administrative buildings when they had ceased to function. Malalas, Evagrius Scholasticus and John of Nikiu point to the devastating effects of earthquakes on bathhouses. While such natural disasters were not confined to Late Antiquity, this period and especially the Middle East seems to have been one of on-going seismic activity. Hammams in Early Medieval Arabic Literature As there are no contemporary accounts of the 7th and 8th c. cities in Arab literature, the written evidence concerning hammams dates from the Medieval period. Even if the study of these sources would lead too far within the framework of this book, it is nevertheless interesting to have a quick glance at some of the most important research that has been carried out in the last two decades. The earliest text that mention hammams is from the 9th c. The word seems to derive from the Aramaic root h-m-m, meaning ‘heat’,446 just as the word thermae had done in the Roman period (see above, chapter 1). Other words such as ballān or ballānāt are borrowed from the Greek terminology, namely balaneion.447 Scholars often emphasized the religious character of the hammam, enabling Muslims to perform quick ablutions before praying (wudû) or a more thorough wash (gusl) 445  Tac. Agr. 21. 446  Pagani and D’Amora (2011) 7. 447  Pagani and D’Amora (2011) 7, n. 2.

chapter 2

after certain events (sex, menstruation, birth) or before special occasions (Friday prayer, holidays, pilgrimage, marriage).448 Furthermore, the epilation of the body and pubic areas—a habit that was linked to the natural state of the body as intended by Allah (fitra)—probably contributed to the acceptance and continued importance of bathing.449 These bath-visits, however, did not purify the body (just as in the Christian context, see above p. 69), but merely got rid of the dirt and made the body clean. The first texts mentioning hammams were of a religious nature, underlining the importance of a strict separation of the sexes in the baths.450 Some passages claim to go back to the Umayyad caliphs of the 8th c., imposing bans on mixed bathing and obliging the bathers to wear loin cloths. Some juridical texts even went as far as banning women from the baths. Issues of sex and desire and the fear of unacceptable behaviour, including adultery and same-sex relations, would continue to dominate juridical and religious texts in the Medieval period.451 Even if the baths—as places of nudity and bodily pleasure—had to be neatly framed in a religiousjuridical code, there has never been a negative or extremist text in the canonical hadith against bathing.452 But the warnings of exaggeration and of possible dangers lurking in the baths were never far away. Just as had been the case for Roman baths (see above), evil spirits (djinns) were thought to have inhabited the hammams.453 It is clear that the link between hammams and religion was strong. But the hammams were also praised for their hygienic, relaxing and therapeutic effects. Several medical treatises from the 9th and 10th c. prescribe bathing to maintain good health.454 The same basic principles on the functioning of baths in relation to Hippocratic medicine were still present: the opposite effects of relaxing / strengthening and dehydrating / humidifying for hot and cold environments of the baths respectively (see above, chapter 1).455 The Arab medical authors did not simply content themselves with copying the classical texts, however. They expanded the medical framework to other domains of medicine, including new treatments in the field of ophthalmology. Later 448   Grotzfeld (1970) 8–9; Denoix (2009) 17–20; De Miranda (2010) 9. 449  Benkheira (2007) 369–70. 450  The hammam-related content of these early texts has been studied by Benkheira (2007). 451  Benkheira (2007); Cuffel (2009). 452  Grotzfeld (1970) 21–22; Benkheira (2007) 355. 453  Benkheira (2003) 423–28. 454  Bessard (2008) 111–12. 455  For a comparison of both medical traditions, see Koetschet (2014).

71

Written Evidence of Baths

medical authors such as al-Magusi (10th c.) or Avicenna (AD 980–1037) emphasized the effects of sweating, with little attention to submersion in water. These passages remind us of the evolution in the architecture of the hammams, with pools gradually disappearing from the 8th c. onwards (see below, chapter 3). While these hygienic and medical prescriptions affected the body in first instance, they also had an important effect on the soul: “En domptant la chair, le bain apaise l’âme et libère l’esprit. La douce griserie du corps qui « lévite » conduit le croyant à l’abandon de soi au Seigneur: « starfillah » (pardon, mon Dieu), ne cesse de murmurer le baigneur à la sortie de l’étuve.” (“By subjugating the flesh, the bath soothes the soul and frees the mind. The soft euphoria of the body which ‘levitates’, leads the faithful to deliver himself to the Lord: ‘starfillah’ (forgive me, my Lord), the bather repeatedly murmurs while leaving the steam room.”)456 The nudity and the particular state of mind of a bather made the hammam an unsuited place to pray or even pronounce the name of God.457 Recent studies have argued that the religious gusl was not the only reason to visit the baths; a more profane visit to the baths may also have been common.458 However, the textual evidence for such ‘bathing for pleasure’ is lacking for the 8th and 9th c. The archaeological evidence for baths in Umayyad ‘hunting lodges’ (see below chapter 3), in combination with the textual evidence from the Mamluk or Ayyubid era,459 point to the existence of a more ‘recreational’ bathing habit. By the 12th c., Damascus had no less than 40 hammams intra muros and another 17 extra muros, according to the historian Ibn Asakir (AD 1105–1175).460 This same author also claimed that some companions of the Prophet who settled in Damascus in the late 7th c. had a residence with a private bath.461 These were probably the luxurious mansions with private baths, perhaps with a semi-public function, that had been taken from the Roman upper class.462 The alleged rupture with the Roman period was obviously less dramatic than previously assumed.

456  Carlier (2000) 1305; translation by the author. 457  Benkheira (2003) 417–18. A similar ban forbade Jews to recite passages of the Torah in Roman-style bathhouses (Mishnah Avodah Zarah 3, 4; see Sartre (2006) 417–25). 458  Benkheira (2003) 405; Koetschet (2014) 1022–24. 459   Piraud-Fournet (2014) 699–700; Koetschet (2014) 1022–24. 460  Quoted in Grotzfeld (1970) 14. 461  As quoted in Piraud-Fournet (2014) 700. 462   Piraud-Fournet (2014) 701.

Legal Documents To consider the legal texts of Late Antiquity, we must turn to the two collections of laws that were commissioned by Theodosius II and Justinian.463 Both corpora brought together pre-existing laws in an effort to harmonize the legal system of a disunited empire. There is still debate over the actual impact of these laws and to what extent these codices were put into practise. The fact that several laws were repeated in subsequent decades is sometimes seen as evidence for their ineffectiveness. Yet, one could also argue that these reiterations show how the issues involved in these laws were very much alive, notwithstanding an official stance of the imperial administration. The laws concerning Roman baths should also be approached in this respect. They reveal how the government in Constantinople reacted to existing and even recurring problems. Codex Theodosianus The Codex Theodosianus contained some 2,500 imperial laws, collected and edited between AD 429 and 438. Theodosius II ordered this new collection of laws to update and supplement the existing Codex Gregorianus and Codex Hermogenianus that had been published a century earlier. The idea was to create a harmonious law collection by which forgeries could be eliminated and conflicting laws could be resolved.464 The most frequent attestations of bathhouses concern issues of management and operation. In a law issued under Honorius, Arcadius and Theodosius II (AD 406), the curia and the municipalities were not obliged to heat private baths if tribunes and minor counts demand it. Only the noble comes and the magister militum have this privilege.465 In the following law, we learn that the dukes of the Euphrates frontier regions exacted a specific sum of money per day for wood and baths.466 They were therefore sentenced to repay double the sum they had collected. Such laws teach us that some officials took advantage of their status by demanding that baths were at their disposal. A law dated AD 400 mentions that the taxes to be paid for one year in all civitates, municipia, vici and castella should come from the income of storehouses, baths, workshops, taverns, 463  No attestations of baths or bathing were found in the socalled Sirmondian Constitutions and the Novels of Theodosius, Valentinian, Majorian, Marcian, Severus and Anthemius, also translated and discussed by Pharr (2001). 464  Honoré (1996a). 465  Cod. Theod. 7.11.1 466  Cod. Theod. 7.11.2.

72 etc.467 Running a bathhouse could indeed be a lucrative business.468 An interesting addition follows: ‘salinis (..) quae populi romani lavacris inserviunt (..) nullius excepta persona unius anni’ (‘No person shall be exempted, but the saltworks shall be excepted (..) which serve the baths of the Roman people’).469 This link between baths and salt extraction is repeated in a law of Valentinian and Valens (dated AD 368 or 370). The supervisors who were responsible for the thermae in Rome ‘should share in the operation of the salt works and the profits therefrom’.470 There were some restrictions for taxing the working class: ‘sed sollicita inspectione prospiciatur, ne a quoquam amplius postuletur quam necessitas exegit lavacrorum vel instituta iam dudum forma praescripsit.’ (‘But provision must be made with solicitous inspection that no more be demanded from any one person than is required by the needs of the baths or is prescribed by the general rule formerly instituted.’).471 In some cases, the operation of the baths could be financed by assigning a third of the income of farms belonging to the municipality to the heating of baths and repair of public buildings.472 In the case of the Zeuxippos Baths in Constantinople, the shops and adjoining houses that had filled its portico were to be taxed on their income to provide lightning (oil lamps), repairs and new roofing for these baths.473 The cost of operating baths did not always affect the cities. In a response to the praetorian prefect Olybrius, the emperors Theodosius, Arcadius and Honorius reassured him that they are pleased to continue paying for heating the thermae in Antioch.474 Other laws did not concern the management of baths, but responded to more specific problems. A law of AD 415 addressed to the city of Carthage ordered the removal of all objects that consecrated by sacrifice (e.g. statues) from the baths.475 On a totally different matter, prisoners should be able to wash and could be accompanied by guards to the bathhouse.476 A law of Theodosius II and Valentinian III (dated AD 432) prohibits arms inside public buildings such as baths, gardens and churches.477 If we remember the numerous attestations of violent murders at the baths, this law was not a mere precaution. An interesting law concerns the relation between 467  C  od. Theod. 11.20.3. 468  Nielsen (1993a) 124–25. 469  Translation by Pharr (2001) 311.  od. Theod. 14.5.1; translation by Pharr (2001) 412. 470  C 471  C  od. Theod. 13.5.13: Translation by Pharr (2001) 395. 472  C  od. Theod. 15.1.32 (dated AD 395).  od. Theod. 15.1.52 (dated AD 424). 473  C 474  C  od. Theod. 12.1.131. 475  C  od. Theod. 16.10.20.3.  od. Theod. 9.3.7. 476  C 477  C  od. Theod. 9.45.4.

chapter 2

private baths and the urban water supply. The whole text is given below. Imppp. gratianus, valentinianus et theodosius aaa. clearcho praefecto urbi. summas quidem domus, si lavacris lautioribus praesententur, binas non amplius aquae uncias aut, si hoc amplius exegerit ratio dignitatis, supra ternas neutiquam possidere, mediocres vero et inferioris meriti domus singulis et semis contentas esse decernimus, si tamen huiusce­ modi balneas easdem habere claruerit. ceteros vero, qui mansionem spatio angustiore sustentant, ad mediae unciae usum tantum gaudere praecipimus neque obreptionem cuiquam patere, ita ut quod tibi paret officium sex librarum auri multa feriatur, nisi prodiderit usurpantes et is qui fefellit careat impe­ trato. dat. x kal. iul. constantinopoli antonio et sya­ grio conss.478 Emperors Gratian, Valentinian, and Theodosius Augusti to Clearchus, Prefect of the City. If the greatest houses are furnished with very elegant baths, We decree that they can have no more than two inches of water, or if by reason of high rank, more than this amount is required, by no means shall they possess more than three inches each. We also decree that houses of mediocre or inferior merit shall be content with an inch and a half, provided that it appears that they have such baths. We order that all other persons who maintain houses of smaller dimensions shall enjoy the use of only one half of an inch. No opportunity for any surreptitious undertaking shall be open to anyone. Unless the office staff who obey your orders should betray the unlawful users, they shall be stricken by a penalty of 6 pounds of gold, and the obtainer of a grant by deception shall forfeit the use of what he was granted. Given on the tenth day before the Kalends of July at Constantinople in the year of the consulship of Antonius and Syagrius—June 22, 382.479 The fact that the volume of water for private baths was limited probably reflects excessive use and the illegal tapping of aqueducts by private individuals. These problems were already known during the empire, as is testified by Livy.480 Private individuals normally had to

478  Cod. Theod. 15.2.3. 479  Translation by Pharr (2001) 430. 480  Livy 39.44.4. The same concern is encountered in the work of Frontinus, see Vladu (2017).

73

Written Evidence of Baths

pay for a license (tributum or vectigal) to use a certain amount of public water.481 Corpus Iuris Civilis The Corpus comprises three collections of laws (Codex, Digesta or Pandectae and Institutiones), collected on the authority of Justinian, and a book of later laws (Novellae). Just as Theodosius II had done before, Justinian aimed to assemble and harmonize pre-existing laws. He ordered a commission of 10 holders of public office to prepare an updated and concise edition of Roman law, omitting repetitions and avoiding conflicts. A first edition, now lost, was presented in AD 529, comprising laws from the codices Gregorianus, Hermogenianus and Theodosianus and other laws promulgated since AD 300. Besides the codices, a compilation was made of some 1,500 old law books in an effort to compose a revised syllabus (Digesta) for the most important law schools, for example in Berytus or Constantinople. Henceforth the opinions of earlier respected lawyers could be cited in court. For the same purpose, Justinian also ordered an updated edition of the lectures by the famous law teacher Gaius (2nd c. AD), called the Institutiones. After the publication of both Digesta and Institutiones in AD 533, a second edition of the Codex appeared in AD 534. These three collections were all written in Latin and were consequently more easily introduced in Italy and North Africa.482 The legislation enforced by Justinian after the codification was completed, were edited later, mainly in Greek (Novellae). As the Codex Iustinianus incorporated some of the laws already promulgated in the Codex Theodosianus, several laws concerning baths were repeated: the laws about prisoners being allowed to visit baths, the law prohibiting minor officials to demand baths of a private citizen and the law about repairing the Zeuxippos Baths in Constantinople with taxes levied on its portico stalls.483 The law that no minor officials could demand the use of private bathhouses from their hosts is also extended to the African territories.484 The other laws mainly concern building regulations. One can only build on top of a bathhouse if the baths as well as the superstructures are built upon arches.485 Furthermore, it is forbidden to build baths on commonly owned ground without the approval of the co-owner. It shall 481  de Haan (2010) 95–113, esp. 108–109. 482  Honoré (1996b) 803–804. 483  Prisoners: Cod. Iust. 1.4.9pr = Cod. Theod. 9.3.7.; minor officials: Cod. Iust. 1.47.1 = Cod. Theod. 7.11.1.; repair of the Zeuxippos Baths: Cod. Iust. 8.11.19 = Cod. Theod. 15.1.52. 484  C  od. Iust. 12.40.6.  od. Iust. 8.10.1. 485  C

be considered an attempt to illegally acquire the entire property.486 It is also illegal to plant trees near the water conduits of public baths, as the roots could damage the pipes.487 Specially appointed officials have the permission to search out illegal use of public waters or the illegal planting of trees.488 A specific law on the water conduits of Hadrian in Constantinople specifies that the water should only serve the baths adjacent to the palace and the nymphaea.489 The sanction against the illegal use of public water is confiscation of property (public gardens, baths, houses, etc.). However, a law going back to the reign of Constantine prohibits curators from confiscating bathhouses in order to settle the debts of individuals.490 Baths are also mentioned in a law of Zeno against monopolies, preventing bath proprietors from making agreements with possible competition.491 In a different context, the fact that a man’s wife has bathed with other men is a legitimate motive for divorce.492 The attestations of baths in the Digesta are numerous. Most of the juridical cases involving bathhouses concern inheritance and servitude. A case-study about the will of a rich citizen of Tibur (Tivoli, Italy) is very informative.493 We learn that the deceased put it in his will that his fellow citizens would be able to bathe for free during 10 months of the year in the bathhouse built against his house. In the legal issue that followed this case, we learn that the heirs contested the fact that they had to pay for the yearly cleaning and maintenance on top of the fuel costs and entrance fees. That the costs of operating a bathhouse were a heavy burden is confirmed by a case in which the deceased left a stretch of forest that would supply firewood, for example to use as fuel for the baths.494 There are more examples of wealthy members of the elite hiring a bathhouse so citizens could bathe for free.495 If the bath operator, for any reason (e.g. damage by fire), was unable to open the bathhouse, he had to make a financial contribution for each day that the baths were closed. An additional cost for maintenance of the furnace and the conduits could also be demanded from the person that wanted to hire the baths.496 A slave of the renter could also be taken in

486  487  488  489  490  491  492  493  494  495  496 

Cod. Iust. 8.10.5. Cod. Iust. 11.43.6.2. Cod. Iust. 11.43.6.3. Cod. Iust. 11.43.6. Cod. Iust.5.37.22.1. Cod. Iust.4.59.2.1. Cod. Iust.5.17.11.2. Dig. 32.35.3. Dig. 32.55.3; see also Blyth (1999). Dig. 19.2.30.1. Dig. 19.2.58.2.

74

chapter 2

pledge.497 Sometimes, a bathhouse was bequeathed to the heirs. These baths could be considered as a part of the house under certain circumstances: Balneas legatae domus esse portionem constabat: quod si eas publice praebuit, ita domus esse por­ tionem balneas, si per domum quoque intrinsecus adirentur et in usu patris familiae vel uxoris non­ numquam fuerunt et mercedes eius inter ceteras meritoriorum domus rationibus accepto ferebantur et uno pretio comparatae vel instructae communi coniunctu fuissent.498 It was determined that a bathhouse was part of a bequeathed house: if the testator never opened the baths to the public, it is only part of the house if there was an internal passage through the house, if it was frequently used by the testator or his wife, if the earnings from the baths were taken up as accepted among the other earnings of the house and if the house and the baths were bought for the same price or if these were inserted into one communal expense.499 It is interesting to note that a private bath could be (partly) open to the public, even if it was built as a part of the house and a passage connected the house with these baths. Apparently, the proprietor of the house could choose if he opened the baths to outsiders. Legally speaking, these were then no longer considered to be part of the house. If a house with a private bath was inherited, it was prohibited to make these baths publicly accessible. Furthermore, if a person inherited a house, he could not divide it into several apartments or make it into a public bathhouse.500 It was also stated that the balneator, the person in charge of the daily functioning of the baths, was part the bequeathed bathhouse, as baths could not function without balneatores.501 Other bath-related items could also be bequeathed, such as firewood or bathing apparel. This bathing apparel must have consisted of flasks, sandals, towels, robes and other objects for personal grooming (see below, the papyrological 497  D  ig. 20.4.9pr. 498  D  ig. 32.91.4. 499  Translation by the author after Spruit (1997).  ig. 7.1.13.8. 500  D  ig 33.7.13.1 and 33.7.17.2. In a different law case, it is said 501  D that the balneator could not be used for other purposes, as this would be considered an improper use of property (Dig. 7.1.15.1). For more attestations of the term balneator, see Wissemann (1984) 81–83.

evidence). Silver bathing apparel, possibly situlae, were often the subject of much discussion among the heirs. Such legal disputes concerning luxurious bathing apparel also reminds us of the social segregation between the rich and the poor that existed inside the baths.502 Another important type of legal cases in which baths played a prominent role is connected to building activities. A person who built tubuli or water conduits against a communal partition wall could be summonsed, as the walls could suffer from the heat.503 The construction of bathing rooms against a common wall was not illegal as such, but the works could be put to a stop if moisture started to cause nuisance to the neighbour. The smoke or vapour from a bath could also cause problems, yet one could not be forbidden to start a fire on one’s own property, even if this was for heating a bath.504 If a bathhouse was turned into a house by an heir, the usufruct of this bathhouse naturally came to an end.505 When a person destroyed a house on someone else’s land in order to build a bathhouse, the building belonged to the owner of this land and the builder was suable for damages done.506 As already mentioned above, the operating costs of a bathhouse could weigh heavily on private proprietors. The same goes for the cost of city-owned baths, to the extent that money intended to buy corn was spent on the maintenance of baths. The costs of heating the baths fell under municipal obligations. Sometimes, a curator oversaw these heating costs, provided that he received the money from the municipal treasury.507 Some miscellaneous attestations of baths inform us in what manner baths and bathing could make local headlines. The capsarius, the cloakroom attendant, was sometimes accused of stealing or negligence in guarding personal belongings.508 Such cases against 502  Dig. 34.2.32.7 and 34.2.40.1. For archaeological evidence of such late antique situlae as part of bathing apparel, see Arce (2005). For social segregation at the baths, see Zajac (1999) 104; Yegül (2010) 37. 503  Dig. 8. 2.13.pr. Another legal case mentions the existence of a special interdict for repairing the conduits that allow steam to circulate ‘in balnearis vaporibus’ (‘in steam baths’) (Dig. 43. 21.3.6). 504  Moisture: Dig. 8. 2.19.pr; smoke: Dig. 8.5.8.6–7. 505  Dig. 7.4.12.pr. 506  Dig. 9.2.50. 507  Maintenance: Dig. 50.8.2.4; heating costs: Dig. 50.4.1.2; curator: Dig. 50.4.18.5. 508  Dig. 16.3.1.8. The same legal case mentions the fact that these attendants were paid for their services. In some cases, the cloakroom attendants were women who doubled as prostitutes (Dig. 3.2.4.2).

Written Evidence of Baths

75

the capsarii fell under the jurisdiction of the praefec­ tus vigilum. Thieves that were active inside bathhouses were tried in exceptional court procedures, the same as for ‘nocturnal thieves’. It seems that the penalties for stealing when someone was vulnerable (asleep or in the baths) were more severe. Soldiers that were caught stealing inside the baths were dishonourably discharged.509 A less ‘object-related’ but maybe even more scandalous crime was to provide baths as a location for adultery.510 If a person were to build on public space, thus preventing others from going to public baths or attending games, he could also be summonsed.511 The ‘right to bathe’ was obviously important enough to find its way into legal cases. Bathing, buying, trading, and attending games and festivals was what made a person a member of the city.512 There are few attestations of baths in the Novellae. We learn that a part of the provincial taxes should serve for the maintenance of public buildings such as baths.513 In the case of a city named Aphrodisium, the public treasury was controlled by the members of the elite, on the condition that they each paid regular sums to the city. When they failed to do so, the maintenance of the city and of its baths suffered greatly.514 A legal case also mentions the will of a wealthy person who left his grandson a suburban estate that included baths.515 The Latin epitome of the Novellae mentions the existence of praetorian warrants that could demand the restoration of bridges, baths and public washing facilities.516 The Greek text, however, only mentions the restoration of aqueducts. The only attestation of baths in the Institutiones, is found in a passage about the manumission of slaves, who could be freed almost everywhere, even while accompanying their masters to the baths.517

of church officials, were more concerned with daily matters and administration. However, few formal statements concerning baths were formulated. In the 4th c. Synod of Laodicea, canon 30 prohibited priests, clerics, ascetics, Christians and laymen to bathe together with women.518 The canon, the actual impact of which is unknown, is clearly aimed at men and women bathing together, not at bathing itself. This canon is repeated as canon 76 of the Council in Trullo (Constantinople, AD 692), adding that the cleric who was caught should be deposed. The fact that this canon was repeated, may indicate that it was often ignored. Canon 11 of the same council advises Christians not to eat the unleavened bread of the Jews, nor have intercourse with them, nor bathe together with them. The penalty for doing so is again deposition for clerics and the casting out of layman. The 4th c. Constitutiones Apostolorum, probably written as guidance for the clergy in Antioch, orders men and women to use their respective baths. We should nevertheless remember that mixed bathing was also problematic in the non-Christian society.519 Some canons or rulings were never discussed during a synod or a council, yet were approved later. In a somewhat odd phrasing, the question whether a Christian could partake of the communion after accidentally swallowing a drop of water while bathing or washing is answered in the affirmative.520 The question probably related to a certain fear of ‘pollution’ or contamination by impure water.521 The concept of ritual purity before attending cult acts and the fear of pollution is an ancient concept that was already present in ancient Greek societies (the so-called μίασμα) and in the Jewish religion, which obviously had an important influence on early Christian ideas.

Ecumenical Councils and Synods Although Ecumenical councils and synods may not have had the same legal status as the codices, they did have an important role in delimiting the boundaries of accepted social behaviour within the Christian community. As the Ecumenical councils concentrated on important doctrinal issues, no guidelines concerning baths and bathing were issued. The synods, local or provincial assemblies

Summarizing the Legal Documents The attestations of bathhouses in legal documents of Late Antiquity can be divided into three large categories:

509  P raefectus vigilum: Dig. 1.15.3.5; thieves: Dig. 47.17.1; soldiers: Dig. 47.17.3. 510  D  ig. 48.5.10.1.  ig. 43.8.2.9. and 47.10.13.7. 511  D 512  D  ig. 50.1.27.1. 513  N  ov. 149.2.  ov. 160. 514  N 515  N  ov. 159. 516  N  ov. 25.4.2. 517  I nst. Iust. 1, Tit. 5.2.

518  The early Church Fathers such as Clemens of Alexandria (end of the 2nd c. AD; Clem. Al., Paid. 3.5) or Cyprian of Carthage (middle of the 3rd c. AD; Cypr., De habitu virginum, 19–21) had already complained about Christian men and women bathing together in shame. 519  Const. Apost. 1.6.13. See already Pliny the Elder (Plin. HN 29.26) or Quintilian (Quint. Inst 5.9.14); Ward (1992) 134; Schöllgen (1995) 188–91. 520  ‘The Canonical Answers of Timothy the Most Holy Bishop of Alexandria, Who was One of the Fathers Gathered Together at Constantinople, to the Questions Proposed to Him concerning Bishops and Clerics, Question 16’ in Schaff and Wace (s.d.) 613. 521  See also August., Ep. 46 in CCSL 34.2, 123–29 in which Publicola wrote to Augustine about his fear that he might drink water from a fountain used for ‘pagan sacrifices’.

76 financial aspects, building activities and social aspects. Not surprisingly, most attestations belong to the first category. Operating a bathhouse could be both a costly and lucrative operation. It required a heavy investment, especially in fuel, but could be profitable if run correctly. In the two most important codices of Late Antiquity, we find several allusions to how baths were financed. Money for balnea publica could come from taxes on the income of farms (Cod. Theod., Nov.) or from the stalls that were erected on space belonging to the baths (Cod. Theod.). In more important cities, emperors sometimes contributed from their own pocket (Cod. Theod.). The fact that municipal money for buying corn was known to be diverted to the baths could reveal that even cities had trouble paying bath-bills. For the balnea meritoria, the cost of the fuel must have been important, with some citizens bequeathing firewood to their heirs (Dig.). The laws that prohibited officials from demanding free access to privately owned baths were probably a result of complaints over the costs made by bath-owning citizens (Cod. Theod. and Cod. Iust.). The costs of operating bathhouses are also stressed by legal cases involving the donation of free baths through testament. The sum left by the deceased should cover fuel costs, staff, entrance fees, repairs, maintenance of furnaces and conduits and yearly cleaning (Dig.). On the other hand, running a bath could also be a lucrative business. The income of the baths could even be taxed to contribute to the city’s treasury (Cod. Theod.). Private investors seem to have established illegal agreements in order to acquire a monopoly of baths in certain neighbourhoods (Cod. Iust.). In an effort to gain more money, heirs converted houses into baths or opened up an inherited private bath to the public (Dig.). The second important category of bath-related attestations in legal documents revolves around regulations of building activities. The stability of a bathhouse was a primary concern, especially when houses were built on top (Cod. Iust.). Furthermore, the erection of a bathhouse should not be a nuisance (smoke, heat, moisture) to the surrounding residents (Dig.). However, building a bathhouse against a communal wall was not illegal as such. Another important issue, was the water supply. There was a restriction on the volume of water for private baths (Cod. Theod.) and the water conduits for public baths had to be protected (against vegetation and illegal tapping) at all times (Cod. Iust.). The third category embraced the attestations of baths in their social context. Bathing was important enough to grant even prisoners the right to bathe (Cod. Theod., repeated in the Cod. Iust.). Obstructing a fellow citizen to visit public buildings, such as baths, was a suable action (Dig.). Visiting the baths was an essential part of the

chapter 2

city life (Dig.), even if their use could entail some small risks, such as being robbed (Dig.). The fact that thieves at the baths and the responsibility of the capsarius were mentioned in separate case-studies of the Digesta suggests that this was a recurrent problem. The importance of the balneator for the proper functioning of the baths is stressed on several occasions (Dig.). With so many polemical literary passages about mixed bathing, it is also somewhat surprising to find no formal laws banning the phenomenon. The law that enabled men to divorce their wife if she had bathed with other men (Cod. Iust.) is the only measure found in both codices. The liability for offering baths as a location for adultery is a more general measure that also applied for houses or taverns (Dig.). In the synods (Laodicea, repeated in Trullo), mixed bathing was indeed forbidden for Christians. However, no general interdictions of bathing were ever formulated by the Church. Epigraphic Evidence Methodological Problems and Approaches The research into the epigraphic evidence of Roman baths and bathing encounters many problems. Firstly, there are some methodological issues. The first step of the research—finding specific inscriptions in such a vast amount of data—is already a daunting task. Furthermore, the period under consideration here, between the 4th and 7th c. AD, is problematic within the longue durée of the epigraphic habit. Inscriptions, as Ramsey MacMullen stressed, are a cultural phenomenon which has its own evolution.522 The choice whether or not to commemorate an event on an inscription is in itself a process that was prone to external circumstances. In Late Antiquity, these external circumstances (politics, religion, social values) changed drastically. There are also text-internal problems—a lack of datable elements, confusing terminology—that further complicate interpretation.523 Finally, the post-depositional processes cloud our understanding of the primary context of the inscription and may have partly destroyed the text. Keeping these problems in mind, we must accept the fact that the overview of the bath-related inscriptions for the geographical and chronological areas under consideration, cannot be exhaustive. As mentioned in the introduction of this book, this study will not attempt to collect all possible epigraphic data concerning baths in Late Antiquity. Rather, it combines and reassesses

522  MacMullen (1982). 523  Rebuffat (1991); Fagan (1999a) 130.

Written Evidence of Baths

the data collected previously by other scholars, such as Berger, Nielsen, Fagan, or Thébert.524 Inscriptions of the Italian Peninsula The first remark that can be made while browsing through the corpus of late antique inscriptions from the Italian Peninsula (appendix 1, EI-1 to EI-54) is the progressive decrease in numbers from the 4th to the 7th c. (graph 1). This evolution fits within the larger framework of the epigraphic habit, which saw a general decrease from the 4th c. onwards.525 Most Christian inscriptions are either funerary or upper class. The ‘middle class’ gradually disappeared from the record, while members of the elite—be it secular or ecclesiastical—continued to immortalize their deeds in stone.526 4th c. More than half of the total number of inscriptions date to the 4th c. (EI-1 to EI-40) and the majority of those again date to the second half of the century, mentioning the subsequent reigns of Valentinian / Valens (AD 364–378) and of Gratian / Valentinian II / Theodosius (AD 378– 395). In the first half of the century, the inscriptions referring to the reigns of Diocletian, Constantine and Constantius II are mainly found in Rome. It is difficult to examine the geographical spread of the inscriptions within the Italian Peninsula (map 1). Some regions have enjoyed more archaeological research than others. So it is normal that the Roman Regio I (Latium et Campania) has yielded more epigraphic finds than for example Regio V (Picenum). Latium and Campania form the traditional core of Roman Italy, with the area between the capital and Neapolis (Naples) being highly urbanized. The Urbs aeterna also had a strong symbolic power as the cradle and cultural centre of the empire. When looking at the dataset of Appendix 1, we can observe that Regio I is best represented, both in the first and the second half of the 4th c. The weight of epigraphic evidence from cities such as Rome or Ostia plays an important role in this. In other cities, the limited epigraphic and archaeological information is due to the continuous urban activity from Roman to the modern times, making it difficult for the archaeologists to reach the Roman layers (e.g. in Verona, Brescia, Milan). In the case of Ravenna, subsidence—the lowering of the terrain by tectonic activity and the weight of sediments—saw the pre-medieval layers slip under the water table. These circumstances hamper any assump524  Berger (1982); Nielsen (1993a) 203–208; Fagan (1999a) 233–347; Thébert (2003) 486–521. 525  Cardin (2008) 388. 526  Cardin (2008) 27.

77 tions to be made about the number of inscriptions in relation to geographical areas. Instead of focusing on the absence of inscriptions in certain regions, we shall focus on the inscriptions that we do have. There is a high number of inscriptions commemorating the restoration of bathhouses in contrast to the commemoration of newly built baths.527 Often, the restoration took place after a period of decay, or after damages caused by a fire or an earthquake.528 A long exposure to heat and moisture would have been damaging to building materials, especially in the hypocausts. One can easily imagine how the hypocaust pillars would crumble during an earthquake or how fires could start in the furnace rooms. The extent of damage and the subsequent repairs mentioned in inscriptions is, however, difficult to assess, as the clients ordering the inscriptions must have been keen to emphasize the importance of their interventions. A good example is the formula ‘restituit a fundamentis’ (‘restored from the ground up’).529 There was certainly a level of exaggeration involved. Similarly, the word conlapsus may have meant ‘dilapidated’ rather than ‘utterly collapsed’.530 However, we cannot assume that all these restorations were insignificant, as the bathers would certainly evaluate these repairs in relation to the damage. The client could not afford to ridicule himself in public by praising the replacement of a broken window as a thorough restoration of the entire baths.531 Besides restoring baths, adding to the decoration also seems to have been worth commemorating, whether it be repairing marble or mosaic floors or donating or relocating statues.532 If we look at the clients who ordered the inscriptions, we can observe that in the majority of the cases, the local city council or one of its members was involved. Most of the benefactors are presented with the title v(ir) c(larissimus), meaning ‘a man of most noble rank’ and de facto a member of the senatorial class. Sometimes, the title was vir perfectissimus, which was often awarded to prefects and men of the municipal administration.533 527  Restorations: EI-1–4; 6; 9; 11–12; 18; 20–26; 28–30; 35; 37; New baths: EI-7; 14. 528  Decay: EI-3; 6; 9; 12; 17?; 19–20; 23–24; 26–28; 37; fire: EI-4; 15; 21?; earthquake: EI-11; 19; 28. 529  See EI-11; 24. 530  Fagan (1996b) 84–85. For the term conlapsus, see EI-6; 12; 19; 20?; 24; 37. See also deformatas, EI-3. 531  Fagan (1996b) 89–90 contra Thomas and Witschel (1992). 532  Decoration: EI-10; 16; 37; 38? For restoration of paintings and statues, see EI-37. See also EI-16, where statues are brought ‘ex abditis / locis ad celebritatem / thermarum Severianarum’ (‘from concealed places to the famous Baths of Severus’). 533  Vir clarissimus: EI-8; 10–14; 16; 21; 23; vir perfectissimus: EI-16; 28; 37.

78 For some of these men, their careers have been reconstructed on the basis of other inscriptions and sometimes by references in ancient authors.534 These citizens still felt the need to invest in their home town or city of residence and were eager to record and announce their munificence in stone. Often, these men acted on behalf of the local council, i.e. with money of the communal treasury.535 The specific title of the benefactor could be further clarified as a consul, a curator (responsible for local taxation), or rector (provincial governor).536 In Ostia, the praefectus annonae—the man responsible for the food supply of Rome—seems to have contributed to the maintenance of public baths.537 In other cases, these wealthy benefactors acted on their own behalf, stressing this fact with the line ‘de sua pecunia fecit’ (‘made it with their own money’).538 There are few examples that mention the emperors as benefactors,539 although several inscriptions start with the imperial titulature to announce that the activities mentioned in the inscription were made possible by the prosperous reign of the princeps.540 This does not mean that the emperor did not finance the restoration or building of bathhouses. The literary attestations and the inscriptions from other provinces (see below) make clear that the emperors were still very much interested in baths as a vehicle for imperial propaganda. Bath-related inscriptions come from both large and smaller towns. It is not surprising that cities such as Rome, Ostia, Spoletium, Aquileia and Regium Iulium saw continued building activity in the 4th c.541 The smaller towns that have produced inscriptions were frequently located near important man-made or natural strategic points (roads, rivers, bays, mountain passes). From the moment of foundation, these locations had ensured the success of the town or city and continued to do so during Late Antiquity. Tarracina on the Via Appia, Cures Sabini on the Via Salaria, or Allifae on the Via Latina all lay on important roads.542 Rusellae, Ocriculum and Tellesia benefitted from river traffic, while Antium, Tarentum

534  See the notes accompanying the inscriptions in Appendix 1. 535  On the thin line between acting as a representative of the local council and its members acting as individuals, see Fagan (1999a) 143–44. 536  C  onsul: EI-21; 23; curator: EI-26; 27; rector: EI-11–13; 17. 537   E  I-9; 18; 38. E  I-7; 12; 28–29; 32–34. 538   539   E  I-3; 9; 15; 18–19. 540   E  I-1–2; 5; 8–9; 20; 22–24; 32; 37. Often, a vir clarissimus saw to the work, see EI-3; 9; 18–19. 541  Rome: EI-1; 2; 4; 14; 35–37; Ostia: EI-9; 18; 25; Spoletium: EI-15; Aquileia: EI-6; 8; Regium Iulium: EI-19; 20. 542  Tarracina: EI-21; Cures Sabini: EI-29; Allifae: EI-11.

chapter 2

and Fanum were coastal towns.543 Some of these towns, such as Cures Sabini or Rusellae, even became the seat of a bishop. When looking at the terminology used in the inscriptions, it is remarkable that the word thermae is more frequently found than bal(i)neum. Of the inscriptions mentioning a balneum, one refers to the dedication of a newly built bath, one to what seems to be the name of a bathhouse and one to a profession related to baths.544 The fact that there are, up to now, no inscriptions mentioning the restoration of a balneum may reveal that only restorations of the more luxurious thermae were worth commemorating. Assuming that the only baths being restored were thermae would be dangerous. The literary and juridical evidence from the 4th c. in any case illustrated that the word balneum did not disappear. On the contrary, it seems that the word was the most prevalent term to denote bathhouses. A difference between a literary and an epigraphic ‘vocabulary’ may explain this disparity. However, the restorations of balnea may not have been ‘worth’ commemorating. Minor or rudimentary interventions, without using luxurious building materials, may have simply passed unnoticed. This fits in with the thesis that balnea were ‘basic’ neighbourhood baths in contrast to the more luxurious thermae (see above, chapter 1). One could then argue that the members of the elite preferred to connect their names to the superior baths, deeming only these restorations worthy of an inscription. A closer look at the attestations of thermae teaches us that their names were often connected to an emperor, meaning a certain standard could be expected.545 A more likely explanation is that the term balneum designated a privately owned bath, for which it was not usual to commemorate restorations, as this could not be considered an act of munificence towards the city’s built public heritage. Other interesting uses of terminology concern the terms thermae aestivae (summer baths) and thermae iemales (winter baths). It reminds us of the difference existing between different types of baths according to the seasons. The difference between these two types of bath has been difficult to discern in the archaeological record.546 As the winter baths are often a smaller set of rooms attached to the larger summer baths, it is easy

543  Rusellae: EI-17; Ocriculum: EI-31–34; Tellesia: EI-13; Antium: EI-23; Tarentum: EI-28; Fanum: EI-30. 544  Resp. EI-7, 36; 14. 545   E I-2; 8; 16; 22; 26; 35. See also the literary attestations of thermae. 546  Thermae aestivae: EI-24; thermae iemales: EI-31–34. See Nielsen (1993a) 138–40.

Written Evidence of Baths

to misinterpret them as ‘the women’s section’.547 Some uncommon terms for separate parts of baths can also be identified. The word baptisterium comes from the Greek verb βαπτίζειν (to dip, to drown) and was used in Greek to denote a pool (βαπτιστήριον).548 The word is also used by Pliny the Younger and Sidonius Apollinaris.549 Similarly, the term colymbus is a Latinisation of the Greek κολυμβήθηρα for pool. The term was also used in this sense by Galen.550 Finally, the word lavacrum also seems to have had the meaning of pool, as the Thermae Pentascinenses were equipped with a better water supply, because the baths “had ceased to be visited in their lavacra due to the interruption of the water supply by an earthquake.”551

79

5th c. During the 5th c., the number of inscriptions further decreased. The political turmoil in the Italian Peninsula had a serious impact on central authority. The sack of Rome in AD 410 (and again in AD 455) shattered the capital’s untouchable status. The loss of North Africa to the Vandals was a hard blow to the economy and the incursions of the Huns plundered cities such as Mediolanum, Aquileia and Ticinum (Pavia). After the reign of Honorius, the imperial presence in the west withered. In the second half of the century, a succession of puppet rulers sent by the east or installed by the western magister was finally halted by Odoacer. Building activity suffered in these periods of crisis. From the small sample in Appendix 1 (EI-41 to EI-50), more or less the same pattern as in the 4th c. arises. It mainly concerns inscriptions of the central regions (see graph 1) commemorating the restoration of thermae carried out on the order of a vir clarissimus.552 In Rome, the praefectus urbi was under the direct command of the emperor and had been given more power. They saw to the restoration of the Baths of Trajan and the Baths of Constantine.553 However, there are no more inscriptions commissioned by the emperor, nor are the inscriptions by local councils or officials dedicated to the emperors. The offices involved are consuls, corrector, curator and praefectus urbi. In one case, the specification ‘comite divinor[um]’ (‘companion of the divine rulers’) is found.554

As a result, most of the inscriptions are difficult to date. Only a broad period of ‘the end of the 4th and beginning of the 5th c.’ can be proposed on the basis of palaeography. Some private initiative was still taken, as a vir laudabilis in the town of Interamna Lirenas restored and decorated the t[h]ermas tivas (summer baths) ‘ex proprio’ (‘out of his own pocket’).555 An inscription from Beneventum explicitly refers to war damages, as a wealthy benefactor restored the thermae Commodianae after the ‘enemy had burnt the entire city’.556 For such a small sample, it is difficult to discuss the geographical spread. Rome is evidently still an important city where construction must have continued, even if this was on a reduced scale. The other locations still lay on traffic arteries, be it roads or rivers. Liternum was connected to the Via Domitiana, Volsinii to the Via Clodia, Interamna Lirenas to the Via Latina and Beneventum lay on the crossroad of the Via Traiana and the Via Appia.557 Good accessibility was crucial to the survival of a town in such turbulent times. The decay of a road could spell the decline of a town. This is what scholars have suggested for Ocriculum on the Via Flaminia during the Gotho-Roman wars, when the city lay on key military routes.558 Regarding the terminology, we can remark that the word thermae is still the most frequent. In one inscription, the restoration of a balneum is mentioned.559 The work was commissioned by a vir clarissimus, the consul of Campania, and dedicated by the curator of Capua, Liternum and Cumae. These ‘Baths of Venus’, as they are called, must have been popular baths worth repairing and connecting one’s name to. Alternatively, this privately owned balneum may have come into the hands of the city council at some point in time.560 We also find a reference to summer baths, spelled ter­ mas extivas.561 The same building was presumably the subject of a 3rd c. inscription commemorating the restoration and adornment of summer baths by a certain M. Sentius Crispinus, probably a member of the same family as the 5th c. benefactor. The family may have had a special connection to this bathhouse.562 With this connection, it is worth remembering the laws and legal case-studies about families bequeathing bathhouses

547  A case in point are the Large Baths in Sbeitla, see Gazetteer. 548   E  I-29. Nielsen (1993a) 155. 549  Plin. Ep. 2.17.11 and 5.6.25; Sid. Apoll. Epist. 2.2.8; see above p. 51. E  I-37; Gal. De meth. med. 7.6 and 11.10. 550   551   E  I-28; Fagan (1999a) 312; Maréchal (2015) 153. 552  T  hermae: EI-39; 41; 46–47; 49; 50; vir clarissimus: EI-39–40; 45–46; 49. For decoration of baths, see EI-40; 41; 44. 553  Resp. EI-45 and 46. 554  C  onsul: EI-39; 49; corrector: EI-40; curator: EI-50; praefectus urbi: EI-43; 45–46; comite divinorum: EI-49.

555   E I-41. 556   E I-47. 557  Liternum: EI-49; Volsinii: EI-50; Interamna Lirenas: EI-41; Beneventum: EI-47. 558  Pietrangeli (1943) 33; although Millet (2013) 11 has reservations. 559   E I-49. 560  See Foulché (2013) for such transfers in Rome. 561   E I-41. 562   C IL 10.5348. See Fagan (1999a) 272.

80

chapter 2

Bath-related Inscriptions 90 Number of Inscriptions

80 70 60 50 40

Africa

30

Italy

20 10 0

4th c.

5th c.

6th c.

7th c.

Africa

40

11

6

0

Italy

39

10

2

1

graph 1

Inscriptions mentioning baths or bathing from Italy and North Africa between the 4th and 7th c. AD. The numbers are based on the inscriptions included in Appendix 1. The number of inscriptions from Italy is 52 and from Africa 57, giving a total of 109. The undatable inscriptions (2 in Italia; 8 in Africa) were not included in this graph. For insecure dates (e.g. 4th or 5th c.), the oldest date was preferred, except for EA-52 to EA-56, as the time-span of the reign between AD 496 and AD 523 favoured a date in the 6th c.

to heirs (see above, Digesta). An interesting line on the epitaph of the ‘[r]estauratori thermarum Tusciani’ (‘the renovator of the Baths of Tuscianus’) in Volsinii reveals the Christian background of the deceased: ‘pax tibi cum sanctis’ (‘Peace to you with the Saints’).563 6th and 7th c. The number of bath-related inscriptions after the 6th c. further decreased. The reign of Theoderic the Amal saw a limited renewal of building activities in specific cities (especially in Ravenna and to a lesser degree in Rome and Ostia). This did not translate into a strong revival of the epigraphic habit. War in the Italian Peninsula made the political situation volatile and left the cities in vulnerable positions. The Lombard expansion in the north further shifted power at the end of the 6th and during the 7th c. Under these changed socio-political and economic circumstances, the epigraphic habit slowly disappeared. Only some rare inscriptions in an ecclesiastical and funerary context persevered. An inscription of Tarquinii, dated AD 504, mentions a curator restoring a balneum with his own money after a long period of decay.564 Two other inscriptions were found in Rome: one commemorated the restoration of a balineum by a member of the clergy, while the other praised the virtues of a bath for the body, yet warned for 563   E  I-50. 564   E  I-51.

the possible danger of awakening sexual desire.565 This inscription, not securely dated, recalls the passages of Jerome and Augustine in which they discourage monks and virgins to bathe. Just as these authors’ advice did not condemn bathing as such, the inscription ends with the remark that it is a slippery lifestyle that brings problems, not the baths themselves. Inscriptions of North Africa 4th c. The 4th c. produced the lion’s share of bath-related inscriptions (graph 1), especially in the second half of the century, under the reign of Valentinian and Valens (appendix 1, EA1–65). The socio-economic situation changed little in North Africa during the 4th c. Mauretania Tingitana had become more isolated from the other provinces and Roman influence here had further decreased. The traditional magistracies and priesthoods, however, continued to exist in the other African provinces. The Donatist schism and periodic raids from discontented tribes, especially in Tripolitania, caused some upheaval, but did not drastically change the situation from the preceding century. When studying the geographic spread of the inscriptions in North Africa (map 2), we encounter the same problems that have occurred for the Italian Peninsula. Archaeological research has traditionally paid most 565  Resp. EI-52 and EI-53.

81

Written Evidence of Baths

map 2

4th c. inscriptions mentioning baths or bathing found in North Africa. The numbers refer to the EA-numbers in Appendix 1. The Roman numerals refer to Roman provinces. plan by the author

attention to the large coastal sites. Furthermore, the early clearing works of these important sites destroyed much of the stratigraphic information. In the absence of chronological clues within the text, this could have provided crucial information in dating the inscription. The different historic trajectories of the North African provinces, especially the turmoil in Mauretania Tingitana and Tripolitania, could also have influenced the epigraphic habit of each individual region. The epigraphic evidence as presented in Appendix 1 (EA-1 till EA-65) shows a clear concentration in Proconsularis / Zeugitana, Numidia and Mauretania Sitifensis / Caesariensis (map 2). The absence of inscriptions in Mauretania Tingitana is not surprising. Even during the first three centuries of our calendar, little bath-related epigraphy has been found there.566 The scarce evidence from Tripolitania seems more puzzling. Certainly, the raids by desert tribes such as the Austoriani affected the stability of the province; hence building efforts were probably focused on defensive measures. However, even if the large coastal cities Sabratha, Oea and Leptis Magna were thriving settlements, the hinterland of Tripolitania was not urbanized. In other words, the number of potential sites is more limited than in highly urbanized regions such as Proconsularis. Indeed, the only cities in which archaeological remains of public baths have been found are Gigthis (Boughrara, Tunisia), Sabratha, Khoms, Leptis Magna and the military camp 566  See Thébert (2003) 521.

of Gholaia (Bu Njem).567 If we look at the total corpus of 4th c. inscriptions, we can see that the epigraphic habit in Tripolitania did not disappear.568 Maybe it was just bad luck that we have not identified bath-inscriptions from this century. But other important cities in North Africa such as Thamugadi or Bulla Regia, still flourishing during the 4th c., have not yielded datable bath-related inscriptions either.569 Just as in the Italian Peninsula, the inscriptions record primarily the restoration of bathhouses.570 The main reasons for these interventions seem to be the decay of the building after a long period of use, an occasional fire or problems with the water supply.571 The restoration of aqueducts, pipes and pools (see below) is a recurrent feature. The problems of water shortage in North Africa must have made such repairs an especially ‘commemorative’ act, as was the case with other water related interventions.572 There are no recorded restorations after earthquakes, even though several are known to have hit 567  For the Roman baths in modern Tripolitania, see Maréchal (2013) 206–13. 568  Around 160 inscriptions dating from the 4th c. are included in the IRT. Most of these are honorary inscriptions for emperors or local officials. 569  For Thamugadi, we should note that few evergetic inscriptions have come to light (Corbier (2009) 196–97). 570   E A-6; 7?; 9?; 15; 17–28; 30; 32–34; 36; 38; 62? 571  Long use: EA-6; 20; 25; 30; fire: EA-18; water supply: EA-23; 32; 36. 572  Shaw (1984) 167; Wilson (2001) 93.

82 the region, e.g. in AD 306 or 310, 362–363 and 365.573 The fragmentary nature of many inscriptions may account for this apparent absence. This also makes it difficult to understand some inscriptions, as the line describing the main action has disappeared.574 It is interesting to note that on several occasions, the precise nature of the restoration is mentioned. It concerns mainly the restorations or addition of pools or how a special room, such as the cella soliaris (room with heated pool), was added.575 The combination of restoration and decoration was also popular: wonderful ornaments, marbles and mosaics, statues, and paintings and floors.576 A special type of inscription concerns the texts that were laid out in mosaics. Such text often speaks directly to the bather, praising the baths (or its waters) he is about to enter. The comparison to Baiae, the famous thermal baths in the bay of Naples, highlights the beneficial powers of the water.577 In contrast to the Italian Peninsula, there are no inscriptions in North Africa attributing benefactions to the emperor.578 However, a standard formula to begin inscriptions is to praise the reign of the emperors (e.g. Beatissimo saeculo or Saeculo felicissimo, ‘During the blessed / prosperous reign of’).579 Most of the works were commissioned by local officials belonging to the senatorial rank (vires clarissimi). They were often a curator rei publicae or a flamen perpetuus (priest for life).580 Besides the titles of the emperor, the proconsuls, who often had the legal and administrative powers of a legatus, are sometimes mentioned.581 In only a few cases, it is emphasized that the works were carried out with pecunia publica (public money); while some wealthy benefactors stressed the fact that they alone covered the charges.582 573  Suleiman et al. (2004) 547, table 1. 574   E  A-1, 5, 7–9, 16, 31, 33, 39. 575  Pools: EA-15, 17?, 23, 24, 25?, 32, 36, 38; special rooms: EA-6, 15, 24, 36 .In EA-36 the bronze and lead elements were also recast. For other rooms, see EA-10 (absida), 26 (atrium). 576  Ornaments: EA-20, 28; marbles and mosaics: EA-24; statues: EA-25; paintings and floors: EA-37. 577  Praise: EA-2, 12, 13, 35, 37; Baiae: EA-2, 35; Allen (2009) 169–171; for the popularity of Baiae and other thermal baths in late antique literature, see above p. 69. For the popularity of thermal baths in Africa, even in Late Antiquity, see Jouffroy (1992); Pettenò (1996); in general also DeLaine (2007). 578  During the High Empire, imperial benefections are for example recorded for Volubilis (ILAfr 614 = IAM 404) and Cyrene (AE (1960) 198 and AE (1928) 2). 579   E  A-1, 6–9, 11, 12, 15, 17, 19, 20–25, 27, 28, 30, 31, 33, 34, 38, 39, 61. 580  C  urator rei publicae: EA-10, 15, 17, 20, 23–26, 36, 40; flamen per­ petuus: EA-17, 20, 23, 26, 32, 36, 40. 581   E  A-15–17, 23–26, 28, 31, 34, 40. 582  Pecunia publica: EA-20, 32, 40; only private funds: EA-5, 10.

chapter 2

When looking at the cities where the inscriptions were found, we can remark that little evidence comes from sites in Tripolitania (v on map 2). The same goes for cities in Mauretania Tingitana (I) and large parts of Mauretania Caesariensis (II) and Sitifensis (III). The absence of evidence from large towns such as Thamugadi, Cuicul or Lambaesis is striking; especially since these sites were equipped with several important bathhouses (see Gazetteer). However, this image is put into perspective when we look at the epigraphy of the preceding centuries in these cities. In Cuicul only one of the three large public baths yielded any inscriptions at all. These 4 inscriptions are all datable between AD 183 and AD 200.583 Only two inscriptions,584 mere copies of each other, commemorate a dedication, while the other two are inscriptions on statue bases, found in baths but not referring to them. In Thamugadi, a city with over 10 large public baths, only three inscriptions refer to the dedication of baths.585 The other bath-related inscriptions were found on statue bases, mainly dating between AD 250 and AD 300 and coming from the Large South baths.586 In Lambaesis, three inscriptions commemorate the restorations of unknown balnea.587 From the Camp Baths (‘Thermes du Camp’) one inscription dates from the period between AD 161 and AD 200, while from the Large Baths (‘Palais du Légat’) two undatable inscriptions bear little information about the action that was commemorated.588 The situation described for these three cities make clear how difficult it is to assess the presence or absence of bath-related inscriptions and how low the survival rate of such inscriptions is, even in well-preserved sites. The cities that did produce bath inscriptions were on the one hand large regional centre with important economic roles, such as Carthage, Thuburbo Maius, Thysdrus, Sufetula, Cirta or Iol Caesarea,589 and on the other hand smaller towns that nevertheless lay on important roads, such as Auzia, Ammaedara, Calama, Musti or Capsa.590 The fact that in smaller settlements like Limisa or Hippo Diarrhytus the bathhouses were 583  Thébert (2003) 505–506, nr. 102–106. 584   A E (1920), 16 and AE (1935), 45. 585  Thébert (2003) 516, nr. 169 dated AD 167 found in the Large East Baths; AE (1894) 44 dated AD 198 found in the Large South Baths; CIL 8.2369 dated AD 213 of unknown provenance. 586  Thébert (2003) 517–19, nr. 172–188. C IL 8.2692; CIL 8.2706; AE (1971) 508. 587   588  Camp baths: Thébert (2003) 508, nr. 121; Large baths: AE (1912) 19; AE (1913) 12. 589  Carthage: EA-5, 16, 31; Thuburbo Maius: EA-17, 34, 36; Thysdrus: EA-7; Sufetula: EA-1; Cirta: EA-14, 22; IOL Caesarea: EA-39. 590  Auzia: EA-3; Ammaedara: EA-10; Calama: EA-23; Musti: EA27; Capsa: EA-33.

Written Evidence of Baths

still restored once more underlines how widespread and important bathhouses were.591 We also know of Cirta being sacked in AD 310 and Iol Caesarea in AD 370, with both cities sporting inscriptions to commemorate restorations that took place after these events.592 On terminology, we can remark the same trend as in the Italian Peninsula: bathhouses are mainly denoted as thermae, while balneum is only rarely mentioned.593 In one case, balnea is used to speak about baths in general. In the other case, the restoration of a balnea (sic) is commemorated.594 The word oceanum seems to be a rather exceptional and poetic alternative to designate a pool.595 Abbir Maius, the site where the inscription was found, lay some 60 km from the sea, making it unlikely that these baths made use of seawater. Alternatively, the term may refer to a room with a mosaic of Neptune / Oceanus, most probably the frigidarium.596 In the use of thermae, there is sometimes the specification of thermae hiemales and thermae aestivae. In a mild climate with hot summers, it would be normal to find more ‘summer variants’ of baths, which possibly had more and / or larger pools.597 Remaining in the ‘pool section’, it has already been mentioned that several inscriptions specify the restorations of pools (see above). The restoration of the piscina, solium or both was worth mentioning in an inscription. In one case, the new piscina was given the name Cochlea (Snail) and the cisterns (exceptoria) were restored. The use of cochlea can perhaps be linked to the word concha (shell), in some exceptional cases used to denote pools.598 The term lavacrum is used in water related contexts, probably denoting the pools rather than the whole bathhouse.599 The mention of an atrium of the baths is also attested in the east. It was probably a type of lounge—dressing room, more 591  Limisa: EA-30; Hippo Diarrhytus: EA-35. 592  Resp. Aur. Vict. Caes. 40.28 and Amm. Marc. 29.5.18 and EA-14, 22 and EA-39. 593  Two more attestations of balneum may date from the 4th c. (EA-59, 61). 594  Resp. EA-3 and EA-27. Interestingly, inscriptions dated to the late 2nd and early 3rd c. found in the same city mention the gymnasium in connection to the sportulas populo (‘public feast for the people’) (CIL 8.1574; CIL 8.1577; CIL 8.1587). In the East, the term gymnasium was often used to denote baths (Yegül (1992) 307–13). E  A-25. 595   596  See AE (1975) 873.  hermae hiemales: EA-10, 34; thermae aestivae: EA- 17, 20, 21, 597  T 28, 38. Nielsen (1993a) 139. There is some debate about these terms. Thébert links the distinction between summer and winter baths to the presence of a palaestra (Thébert (2003) 461–63), and not to the size and number of pools. 598   E  A-36. See Leclerq (1910) 387. E  A-13, 17, 32. 599  

83 commonly referred to as vestibulum.600 The inscription from Madauros also speaks of a porticus and a pronaum (porch).601 The patina (large shallow dish) mentioned in the same text was probably a part of the boilers.602 5th c. The 5th c. in North Africa was marked by the invasion of the Vandals, crossing the Straits of Gibraltar in AD 429 and seizing Carthage in AD 439. It is obvious that from the Vandal reign onwards, we find no more inscriptions dedicated to Roman emperors. In fact, there are very few written sources about the Vandal rule.603 Epigraphic evidence of building or restoration activity is scarce, but could be explained by the Vandals being unfamiliar with commemorative inscriptions.604 The number of inscriptions is too low to make conclusions about the geographical spread of the evidence (map 3). We can only remark that an inscription was found in modern Aïn Temouchent (Algeria), a rare example of a bath-related inscription in western Mauretania Caesariensis.605 The inscriptions that commemorated building activities concern repairs after a long period of use, decoration and maybe even the dedication of new buildings.606 The inscriptions found in mosaics, impossible to date precisely if there are no references to historical events or persons, are more poetic in nature and often carry a message for the bathers.607 In one case, a reference to Baiae is made.608 The mosaic found in Sidi Ghrib is especially interesting.609 The clever phrase ‘si placet commune est, si displicet nostrum est’ (‘if it pleases, it is communal. If it displeases, it is ours.’) could refer to a general wisdom, or it could hint more specifically at the baths built by the dedicator: if it is a good bath, everybody wants access, but if it is no good, it can remain private. Incidentally, we can doubt the excavator’s interpretation of these baths as an exclusively private bath.610 The commissioners of the restorations and decorations are still the vires clarissimi and curatores acting as

600  A  trium: EA-20; 26. For an example in the East: Ephesus, AE (1898) 1121. For vestibulum: Nielsen (1993a) 163. 601   E A-20. 602  Fagan (1999a) 278, nr. 129. 603  Liebeschuetz (2003a) 58. 604  Courtois (1955) 365–90; Leone (2013) 91–99. E A-51. 605   606  Long use: EA-42, 47; decoration: EA-45; new building: EA-46; 48; 50. E A-48–50. 607   608   E A-49; but possibly also EA-60. 609   E A-50. 610  Enabli (1986) 50; see also Gazetteer.

84

map 3

chapter 2

5th and 6th c. inscriptions mentioning baths or bathing found in North Africa. The numbers refer to the EA-numbers in Appendix 1, the Roman numerals refer to Roman provinces plan by the author

local officials of the emperors and proconsuls / legates.611 The mosaic inscriptions do not refer to the benefactors directly, although there are allusions to construction activity.612 These smaller baths were often built on private initiative. The building inscriptions come from larger towns such as Carthage or Madauros, while the mosaic inscriptions come from small settlements of which the Roman names are not known.613 The structures in which the mosaics have been found, are often interpreted as private baths, the only argument being their small size. However, there are no villas found in the vicinity. The terminology fits with the pattern of the 4th c. inscriptions. The building inscriptions mainly mention thermae and only rarely balnea.614 In the inscription of Madaurus, the repairs are specified as restorations of the ‘cameram cum suspensuris’ (rooms with suspended floors, or the heated rooms) and the ‘lavacra’ (here probably pools).615 In the inscription of Membressa, the statues of the piscina were arranged.616

611  V  ir clarissimus: EA-42, 45; curatores: EA-42, 45, 47. References to emperors: EA-42, 43, 45, 46. 612   E  A-50, 51. 613  Carthage: EA-48; Madaurus: EA-42–44; smaller settlements: EA-49–51. 614  T  hermae: EA-43, 46, 47; balnea: EA-42. 615   E  A-42. 616   E  A-45.

6th and 7th c. The 6th c. was a period of turmoil in North Africa. The internal disputes among Vandal rulers ended when Justinian sent Belisarius and his army to conquer the region (AD 533). Despite the numerous attestations of Justinianic building activity in the works of Procopius (see above, p. 64–65), there are few inscriptions concerning baths dating from this period. After the Arab conquest in the 7th c., there are no more bath-related inscriptions found in North Africa. The number of inscriptions dating from the 6th c. is somewhat misleading, as 5 inscriptions are actually epigrams by a poet named Felix, who praised the ‘Thermae Allianae’ built by King Thrasamund (AD 496–523).617 The location of these baths—if these actually existed, rather than being literary inventions—is unknown, although they must have been situated in Carthage.618 An inscription dated to the reign of Gelimer describes the baths of Gebamund, also of unknown location in Carthage.619 These 6 texts commemorate how the Vandal king—or a member of the royal family620—ordered the erection of a new and luxurious bathhouse for the benefit of the people. One inscription specifies that the baths were built in just one year.621 It is also remarkable how the 617   E A-52–56. 618  Busch (1999) 244. 619   E A-57. 620   E A-57. 621   E A-53.

85

Written Evidence of Baths

inscriptions often make references to classical mythology, even though the Vandals were Christian. The gods Vulcan and Neptune are used as metonyms to describe the natural elements fire and water.622 Such references stress once more how important classical heritage, especially in literature and poetry, still was. The references could point to the intention of the Vandal kings to present themselves as legitimate heirs of Rome.623 Building Roman bathhouses for themselves and their subjects can also be seen in this respect. The Thrasamund texts are characterized by the focus on the salubrious aspect of the baths and the perfect temperature of the water. The comparison with Baiae is again made.624 The baths are seen as a perfect symbiosis between the contrasting natural elements of water and fire. Furthermore, the luxury—often stressing the marble decorations and the effects of light625—is elaborately praised. The only inscription dating to the Byzantine phase of the 6th c. comes, not surprisingly, from a fortress.626 A person named Solomon can be identified as the magister militum of Africa. The t(h)ermas seems to have been enclosed by the walls of the fortress. The absence of restoration inscriptions seems remarkable, but the low number of 6th c. inscriptions does not permit further conclusions. The terminology used in the texts does not vary from preceding centuries. The place-name Baiae is used as a synonym for bathhouse.627 Otherwise, the term thermae is used. The word lavacrum seems to have been used both for pools and bathhouse. The sentence ‘Hic fes­ sos artus viva lavacra fovent’ (‘here the lavacra keep the exhausted limbs warm’ with the adverb hic referring to the bathhouse) reminds one of the mosaic inscription found in Castellum Tingitanum. In the other case, the ‘praecelsa lavacra’ (‘towering lavacra’) clearly refer to an entire building.628 Inscriptions of Cyrenaica In Cyrenaica, the only inscriptions that can be related to bathhouses do not mention the actual building. Two statue bases, one of Heracles and the other unknown, were found in the frigidarium of the City Baths in Ptolemais.629 In the Byzantine Baths of Taucheira,

622   E  A-52, 54, 57. 623  Busch (1999) 243. 624   E  A-53, 57. 625   E  A-52, 55. 626   E  A-58. 627   E  A-57. 628  Resp. EA-53, EA-13 and EA-56. 629  Kraeling (1962) 214, n. 36 and 37.

an Arabic inscription praising Allah was found in the threshold of the entrance.630 Inscriptions of Palestina The inscriptions of ancient Judaea and Palestina are in the process of being assembled in the volumes of the Corpus Inscriptionum Iudaeae / Palaestinae (CIIP). Its first three volumes, focusing on Jerusalem, the region around Caesarea and the southern coast were published in 2010, 2012 and 2014 respectively. The majority of inscriptions found in the region, especially those recovered during the numerous excavations of the last 40 years, have been published in separate articles, making them difficult to trace. The CIIP and efforts to digitize some inscriptions (Inscriptions of Israel / Palestine project of Brown University) will undoubtedly improve the accessibility of the inscriptions in the years to come. The bath-related inscriptions that have been traced concern texts laid out in mosaics—or in an exceptional case painted—mentioning good health or referring to the figures depicted in the mosaics.631 The style of figurative mosaic often permits a broad dating of the inscription, but if only text is laid out in tesserae, dating the inscription is an impossible task. An inscription found in Scythopolis commemorates the restoration of the ‘Baths of the Lepers’, giving us a rare example of a bathhouse specifically conceived for the sick.632 In the thermal baths of Hammat Gader, several Arabic inscriptions of the Medieval period point to the prolonged use of the complex.633 Summarizing the Epigraphic Evidence After this brief overview of the epigraphic evidence of late antique baths, we can attempt to draw some basic conclusions. Obviously, the inscriptions followed larger regional trends in urban evolution and euergetism. The continued economic and building activity that has been identified by archaeological research for 4th c. Italy and North Africa is mirrored by a continued production of inscriptions (graph 1). The reigns of Valentinian and Valens seem to have been a period of renewed activity. The abandonment of several cities in the Italian Peninsula during the 5th c. (see below, chapter 4) is also recognizable in a drop in the number of inscriptions. In North Africa, the ill-documented Vandal rule also left few inscriptions, although archaeological evidence suggests that trade and building continued on some scale. The inscriptions mentioning Roman emperors 630  Jones (1984) 111. 631  Painted text: EP-6; good health: EP-3, 5; figures: EP-4. 632   E P-1. 633   Amitai-Preiss (1997); Dvorjetski (2007) 145–58.

86 disappeared altogether. The Byzantine reconquest of first North Africa and then the Italian Peninsula did little to reverse the trend. Few bath-related (re)construction activities were commemorated in inscriptions. The form of inscriptions did not change. A commemorative text basically mentioned the name of the benefactor, often including his honorary titles, and what exactly had been done. Other information included how the project was paid for and when it was carried out (name of the emperor or consul). The people who paid for the building activities were still the same as in the preceding periods, i.e. to a limited extent the emperor, the city councils and their officials, and rich benefactors acting on their own behalf. As was the case for the High Empire, the interventions of the emperor were rather scarce and mainly concerned restoration works.634 For our period, these interventions only concerned the Italian Peninsula. In some cases, we can imagine that these imperial benefactions probably came about when a city sent envoys seeking financial support. In other cases, the emperor himself might have decided to help with the reconstruction of a bath, out of political considerations (propaganda) or even out of genuine altruism. The majority of benefactions, however, were carried out by local city councils or by private citizens. There are few attestations of ecclesiastical authorities acting as benefactors for the construction of a bathhouse (EI-52), even if several ‘church baths’ have been discovered in the regions under investigation (see below, chapter 4) and literary attestations point to bishops being involved in bath building (see above). The inscriptions mainly concern restoration works, embellishment or enlargement. New construction is rather exceptional. Most of the repairs were carried out after a long period of use or after a (natural) catastrophe, such as a fire or an earthquake. Pools, heated rooms and water supply systems (pipes, aqueducts) were most often restored. Embellishment often involved the use of marble or the addition of statues, sometimes brought over from other public buildings in decline. Enlargement could concern almost every part of the building, but new pools or lounges areas seem to have been named in particular. Besides construction activities, other content found its way into the epigraphic record. In North Africa and Palestina, mosaic inscriptions, often evoking a world of leisure and watery pleasure, were popular. In contrast the Vandal kings and the clergy in Rome evoked the pleasures and possible dangers of the bathing habit respectively in eloquent epigrams. On a terminological level, we can remark the continued popularity of the word thermae (and its vari634  Fagan (1999a) 118.

chapter 2

ants) and surprisingly little attestations of balneum. However, it must be stressed that balneum is mainly found in inscriptions pre-dating the imperial thermae in Rome—more particularly pre-dating the Baths of Nero635—when this seems to be the only word for a bathhouse. With the introduction of the new term ther­ mae, the word balneum seems to be restricted to small neighbourhood baths or to privately owned baths. The thermae, whether one interprets these as the more luxurious baths or the city-owned baths, seem to have been the only ‘category’ of baths worth making an inscription for. This would mean that a drop in the number of inscriptions—naming almost exclusively thermae— would correspond to a drop in this specific ‘category’ of baths, rather than a drop of the number of baths in general. The word lavacrum also appears to denote bathhouses, just as it did in Latin literature. However, the term seems to have been used for both ‘bathhouse’ and a specific part of the bathhouse, in some cases a pool. Other terms are rather uncommon (colymbus, cochlea, patina) and remind us how diverse and local the bath-related vocabulary could be. Papyrological Evidence Introduction The papyri come from a limited number of sites. The vast majority was found in Oxyrhynchus, although Hermopolis Magna, Karanis, Kysis and Arsinoe have yielded important finds. We should remember that the papyri were not always written in the same city as they were found. A letter found in Hermopolis Magna may well talk about a bathhouse in a different city (e.g. P-5; P-13). In spite of the limited geographic spread of the papyrological evidence—especially the absence of papyri from the Nile Delta—the information found in these texts gives us a good insight into some otherwise lesser-known aspects of late antique bathing habits. The papyri were collected by consulting online papyrological databases such as the Heidelberger Gesamtverzeichnis der Griechischen Papyrusurkunden Ägyptens (http://www.rzuser.uni-heidelberg.de/~gv0/) and Papyri.info (http://papyri.info/), as well as the database of the CEDOPAL at the University of Liège. The research of scholars such as Béatrice Meyer and Berangère Redon, as well as the general list by Inge Nielsen has also proven to be very useful.636 Just as with the inscriptions, the post-depositional problems 635  Fagan (1999a) 17. 636  Meyer (1994); Redon (2009); Redon (2012a); Redon (2012b); Nielsen (1993a) 207–208.

87

Written Evidence of Baths

Bath-related Papyri and Ostraca

Number of Texts

50 45 40 35 30 25 20 15 10 5 0

Ostraca Papyri

4th c.

5th c.

6th c.

7th c.

Ostraca

7

1

1

0

Papyri

36

10

15

10

graph 2

Late antique papyri and ostraca mentioning baths and bathing found in Egypt. The number of papyri is 71 (3 papyri are undatable), the number of ostraca 9, giving a total number of 80. Undatable data was not included in this graph. For insecure dates (e.g. 4th or 5th c.), the oldest date was preferred.

also cloud our understanding of the contents of certain papyri (e.g. P-34; 35), making it sometimes difficult to date the papyrus. As a rule, it is impossible to link the bathhouses mentioned in the papyrus to surviving remains. Papyri 4th c. The majority of the papyri and ostraca mentioning baths or bath-related subjects date to the 4th c. (see Appendix 2, graph 2). The papyri were mainly found in Oxyrhynchus and Hermopolis Magna, while the ostraca come from Karanis and Kysis. The papyri are mainly official notes, receipts and private letters. The ostraca are mainly administrative notes. The business letters and notes, often addressed to the λογιστής,637 mention construction or repair works carried out on public baths, with subsequent expenses.638 The works include the restoration of wall paintings, wooden elements, lead pipes for the pools and (window?) glass.639 Another responsibility of the local council was fuel supply.640 In one papyrus, this was charcoal and flax; in an ostracon it was chaff, reminding us of the scarcity of firewood in

637   P-8; 10–12; 19; 21. 638   P-1; 8; 24; O-7. 639  Wall painting: P-10; wooden elements: P-11, 21, 58; lead pipes: P-12; glass: P-19. 640   P-1; 16; 20.

Egypt.641 The ostraca of Kysis mention rent and a license for the baths paid for in wheat.642 The running of a public bathhouse, built with money out of the public treasury, was probably outsourced to a private individual who paid an operating tax to the city. The exact nature of the ‘license’ is unclear. The papyri and ostraca also inform us about the staff working at the baths: the furnace-operator (ὑποκαυστής), cloakroom attendants (καψάριοι), a general bath-attendant (παραχυτής), a bathoperator (βαλανεύς) and guards (φυλακαί).643 In addition to this administrative information, the personal letters mainly inform us about bath-related items, sometimes specified as wooden clogs (σόλγεια), a bucket or bath towels (βαλανάρια).644 In the archive of Theophanes, we get a rare glimpse of the monthly expenses that a person would spend visiting the baths, although it is difficult to interpret the entrance fee of the baths compared to, for example, the average daily wage.645 When a bathhouse is mentioned, it is often used as a reference point within the city.646 In some 641  Charcoal and flax: P-16; chaff: O-2. Also Meyer (1989) for earlier periods. 642  Rent: O-4; License: O-6. Furnace-operator: P-6; cloakroom attendants: P-7, 25; general 643   bath-attendants: P-9; bath-operator: P-18; guards: P-33, O-1. 644   Bath-items: P-17, 30; clogs: P-4; bucket: P-31; towels: P-32. 645   P-5, 13–15. For entrance fees, see Faucher and Redon (2014) 848–49. 646   P-2, 3, 12, 23, 35?

88 cases, the baths are named after an emperor or the benefactor, sometimes they are just referred to as ‘the hot baths’.647 The magic bulla that had to be thrown into the furnace of a bathhouse are a good reminder of the popularity of magic and superstition that still existed in Late Antiquity. The bath furnace, however, only seems to act as an appropriate location as fire was the essential element to make the spell work.648 On a terminological level, the standard word for a public bathhouse seems to have been το βαλανεῖον. The Latin term thermae is found under θέρμαι. The use of γυμνάσιον for bathhouse, a common practise in papyri, is only encountered once on an ostracon.649 However, in one papyrus, the baths and the gymnasium are mentioned as two different buildings, casting doubt on the interpretation of bathhouse on the ostracon.650 The title of gymnasiarch mentioned in a papyrus was most probably related to the gymnasium in its ‘sporting ground’ meaning, as baths seems to have fallen under the jurisdiction of a λογιστής.651 In a papyrus mentioning restoration works, several rooms of the bathhouse are mentioned, including θόλος (a round room, probably heated) and ξυστός (a walking courtyard, perhaps even a palaestra).652 5th c. The number of bath-related papyri dating from the 5th c. is small (graph 2), fitting within a larger pattern of scarce 5th c. papyri.653 Just as in the preceding century, the evidence comes from Oxyrhynchus and Hermopolis Magna. The topics of the bath-related text differ little from the 4th c. examples. A receipt mentions nails and other unspecified building material for construction works inside a public bath; a document mentions the fuel supply (chaff) and there is an example of a lease contract for a bathhouse covering a period of 10 years and 3 months.654 It apparently had a special room for women (ματρώνικον). Concerning the staff, only a bathattendant (περιχύτης) is mentioned.655 Personal letters 647  Emperor/benefactor: P-2, 3, 10, 18; hot baths: P-2, 12, 23. 648   P-28, 29. Meyer (2001) 940. 649  T  hermae: P-10; gymnasium: O-2. For the use of gymnasium in papyri, see Meyer (1997) 691. 650   P-19. Nevertheless, we should remember the fact that gymna­ sia often had a bath suite (see above, chapter 1). The fuel supply for the gymnasium mentioned in P-19 may well have been intended for this part of the building. 651   P-26. 652   P-10. For xystus, see Nielsen (1993a) 164. 653  Hickey (2008) 122–23. 654  Building material: P-37; fuel: P-40; lease contract: P-39. 655   P-38.

chapter 2

talk about bath towels and clogs.656 The term λουτρόν is found next to βαλανεῖον to denote a public bathhouse, whereas θέρμαι is not attested.657 The low number of papyri, however, excludes any type of conclusion about terminological preferences.658 The matronikon is an otherwise unknown room in bathhouses and may have denoted a room restricted to women, in analogy to the matronika in ecclesiastical architecture.659 6th c. The number of bath-related papyri dating from the 6th c. rose in comparison to the 5th c. (graph 2). Oxyrhynchus and Hermopolis Magna are best represented in the sample, even if some other sites such as Antinopolis and Aphroditopolis yielded a text. The texts pertaining to construction work, whether restorations or not, are mainly receipts for building material, including lead and tin, glass and nails.660 A tax still existed to ensure the fuel supply of the baths, specified on one ostracon as brushwood.661 The staff mentioned in these texts include the bath-attendants and a water carrier (ὑδροπάροχος).662 In a papyrus found in the area around Hermopolis Magna, there is an attestation of an architect of baths.663 There are other miscellaneous facts about baths. A bathhouse in Oxyrhynchus was located in the suburbs.664 The entrance fee is deemed too high in a bathhouse in Antinopolis.665 In a papyrus of unknown provenance, the salt sellers talk about salt for the baths, perhaps a type of soap or ‘salt for the baths’ indirectly refers to the taxing of salt sellers to partially cover the expenses of the baths.666 On a totally different matter, the monks of a monastery in Oxyrhynchus bought a rope for the baths, possibly in connection to a well supplying the water.667 Besides βαλανεῖον, there are several attestations of λουτρόν. Just as with the apparition of the term lavacrum 656  Towels: P-36, 41; clogs: P-42. 657  Loutron: P-37. 658  Contra Denizeau and Redon (2009) 401. 659   P-39. See Meyer (1992) 56. 660  Building material: P-57, 60; lead and tin: P-49, 50; glass: P-51; nails: P-62. 661  Fuel: P-47, 55; Brushwood: O-9. 662   Bath-attendant: P-53; βαλανεὺς: P-43, 52 and possibly P-54; water carrier: P-45. 663   P-56. 664   P-49. 665   P-48. 666   P-44. For salt used as soap, see above Oribasius, p. 45; also Blonski (2014a) 267–84; Blonski (2014b). For salt to cover expenses, see above, Cod. Theod. 11.20.3 and 14.5.1. 667   P-46.

89

Written Evidence of Baths

to denote bathhouses in late antique Latin terminology, λουτρόν seems to have been introduced in late antique Greek terminology.668 The stables of the baths (στάβλου τοῦ / δημοσίου λουτροῦ) mentioned in a papyrus from Arsinoe may refer to the stables of a private house, attached to private baths.669 Alternatively, one could think of a stable for the pack animals that brought in the fuel (mentioned in P. Lond 3, 1166R of the 1st c. AD) or turned the saqiyah. 7th c. Bath-related papyri from the 7th c. are few in number again (graph 2). Besides papyri from Oxyrhynchus, the evidence from Arsinoe is well represented in this century. Glass panes for the construction works are intended for a private bathhouse.670 Public baths were, however, still operational, as there are accounts of the expenses of visiting the baths.671 Some texts mention the bath personnel, including a bath-operator (βαλανεὺς), a bathattendant (περιχύτης) and a σύμμαχος (clerk?).672 On an administrative level, there is evidence for the payment of rent for baths and for the expenses of the water supply.673 The word λουτρόν was often used to denote a bathhouse, even though βαλανεῖον is also found.674 In an early 7th c. letter, a person apparently has the proper name Βαλανέος (Balaneos), possibly referring to some affiliation with the baths.675 In a list of furniture addressed to a σύμμαχος, several items of the bathhouse were listed, including bronze waterspouts in the form of a lion’s head.676 Summarizing the Papyrological Evidence Even if the papyrological evidence of baths and bathing is rather restricted in geographical space, it covers a wide spread in time. Unlike the epigraphic evidence (see above), there are still papyri informing us about the 6th and especially the 7th c. Nevertheless, it is risky to come to any conclusions about the popularity of baths and bathing based on the number of attestations in the

668   P-52, 54, 59; O-9. Also Denizeau and Redon (2009) 401; Maréchal (2015). 669   P-62. For more attestations of στάβλον, see Husson (1983) 254–56. 670   P-65. 671   P-69, 74. 672   Bath-operator: P-70; bath-attendant: P-67; clerk?: P-66, 71. 673  Rent: P-64; Water expenses: P-72. 674  Loutron: P-64–66, 68, 71; balaneion: P-67, 69, 72–74. 675   P-63. 676   P-71. See also above, the description of such water spouts by Sidonius Apollinaris (Ep. 2.2.8).

papyri, as changed depositional and post-depositional processes heavily influence the survival chances of papyrus. It is sometimes difficult to distinguish new construction from restoration, as the notes and receipts often only pertain to the building materials used during the works. They do, however, inform us about the several types of specialized labour that were employed, such as glass workers, painters, carpenters and lead workers. The fact that there are only a small number of sites that yielded papyri and that no late antique baths have been found in these sites, makes it impossible to link the baths mentioned in these papyri to archaeological remains. We can nevertheless state that the baths are often mentioned as topographical reference points within the city fabric. There is some interesting information about the actual operation of a bathhouse. It seems that in Oxyrhynchus and Kysis, private individuals could rent a bathhouse, presumably from the city, sometimes for a period of up to 10 years. This rent could be paid in kind, that is in volumes of wheat.677 The operation of the public baths, including fuel supply and maintenance was the responsibility of a λογιστής (the Greek equivalent of the Latin curator), at least in the 4th c. His main task was collecting taxes to cover the expenses of fuel supply and repair works. The actual procurement of the fuel and the building materials was delegated to specialized guilds and craftsmen, who later formulated the letters to the λογισταί asking for remuneration of expenses. There are few attestations on the expenses of the water supply, which can most probably be explained by the fact that the water supply system was a one-time cost during the construction of the baths (e.g. as a cistern, a saqiyah, a well or a connection to an aqueduct). The maintenance and operation (in the case of a saqiyah) of the water system fell on the operator of the bath. The type of fuel, obviously an important cost, is often identified as chaff, flax or brushwood instead of firewood. The personnel that worked in the baths are often indicated in rather general terms such as bath-operator (βαλανεὺς) or attendant (περιχύτης), who fulfilled a variety of tasks (massages, epilation, washing, scrubbing). More specialized occupations include cloakroom attendants, guards, furnace operators and water-suppliers. At least in some cases, the personnel were paid by the local council out of the public treasury. The personal letters mainly inform us about bath utensils such as towels and clogs. They confirm a logical assumption that at least some people owned these 677  See also Dig. 50.8.2.4.

90 items, rather than renting or borrowing them at the baths. The lists with daily expenses found in some personal archives cannot give us a definite answer on the entrance fee of publicly accessible baths, but they do show us that a certain amount of monthly expenses was

chapter 2

spent on bathing. These may not have been restricted to entrance fees, but could also include oil, soap and other utensils. The entrance fee does not seem to have been too high, because exorbitant prices met with the fierce resistance of the townspeople.

chapter 3

Archaeological Evidence Introduction Archaeological evidence of Roman-style bathhouses is ubiquitous across the entire Mediterranean. Not only the larger Roman cities, but also the smaller settlements were equipped with public baths. As bathhouses often had sturdy walls and foundations—especially in the heated parts where insulation was of utmost importance—their chance of survival is quite high. Furthermore, hypocausts lay underneath Roman walking level, which increased the survival chance of this very characteristic element. Hence the bathhouses were often identified among the mass of buildings, in contrast to others with less ‘recognizable’ features. The most important problem about the remains of baths is the lack of reliable excavation data and secure dating evidence. Roman sites were often discovered long before the standardization of archaeological field methods and typo-chronological studies of material culture. Late antique levels were also considered of lesser importance, hardly worth mentioning in concise excavation reports. The latter only offer very basic descriptions of the standing remains and the extraordinary finds. In the absence of inscriptions, the date of buildings was determined on the basis of design, construction technique or decorative schemes (mosaics, wall paintings). An additional problem for Roman sites is the continued occupation of many Roman cities until this day, meaning that Roman remains lie buried underneath several metres of occupied ground. This is especially problematic in the Italian Peninsula, where important late antique cities such as Mediolanum (Milan), Ravenna and Rome, remain concealed under the modern cities. In contrast, the archaeological remains of Roman cities in North Africa are among the best-preserved examples in the Mediterranean. The so-called ‘ghost towns’ offer unique opportunities to study urban transformation from foundation to eventual decline. This rich archaeological heritage caught the interest of the colonial powers, who set out to unearth treasures for the various national museums in their home countries. It was these early ‘excavations’ at the end of the 19th and beginning of the 20th c. that revealed most of the large monuments that are central in importance to these archaeological sites today. In Egypt and Cyrenaica, the late antique and Early Islamic layers were often brushed aside in search of Greek or pharaonic phases of a site. Several small bathhouses have been identified and subsequently hastily © koninklijke brill nv, leiden, ��20 | doi:��.��63/9789004419421_005

attributed to the ‘Roman or Byzantine’ period, solely on the basis of construction technique. More recently, surveys and small excavation campaigns in Egypt have provided much-needed reliable archaeological data, mainly based on ceramic studies. In Palestina, the urban (re)development of the region began only after the creation of the state of Israel, meaning important building activity mainly took place in the second half of the 20th c. and still continues today. The recent decades saw an increase in salvage archaeology, generating a wealth of archaeological data, especially for the late antique period. Each separate bathhouse is discussed at length in the Gazetteer. Besides the baths that were built in Late Antiquity, a concise overview is given of the existing baths that remained in use. When possible, late antique changes to the original layout are highlighted. For a select number of sites, the corpus of newly built baths and baths with a continued use is studied within its larger urban context. The focus is hence shifted from the architectural, technological and decorative programme to the functioning of the baths within a complex fabric of streets, public and private space, water supply and disposal, etc. In the following chapter, the baths in Rome, Ostia, Cuicul, Thamugadi, Carthage, Sufetula, Ptolemais and Scythopolis will be presented. At the end of this chapter, there is an overview of the archaeological evidence of early hammams. Rome Concise History of the Urban Fabric The history of Rome cannot be summarized in just a few pages. The 3,000 years of history that started with Bronze Age settlements perched on the mythic hills and continue to this day have been told and retold in an incredible amount of scholarly literature. Simply enumerating the most important works would be a daunting task.1 For the present research, a concise overview of the most important urban developments up until Late Antiquity will suffice. As capital of the Roman empire, Rome had always benefitted from the munificence of its rulers. From the 1  For a select bibliography on the classical history of Rome, see the entry ‘Rome (history)’ in Oxford Classical Dictionnary (4th ed., 2012). See also Storia di Roma (Giardina and Schiavone (1991)).

92 time of the first kings, the city was continuously changing, with new buildings being erected while others were restored or demolished (map 4). The urban fabric of Rome looked—and admittedly still looks—like a big building wharf, permanently under construction. The forum Romanum, more so than any other area in Ancient Rome, has seen continued building activity and remodelling. It embraces the Augustan regiones VIII and part of IV. The valley between the Campidoglio and Palatine hill was first ‘urbanized’ in the 6th c. BC, when the swampy area of the Velabrum river was drained by the Cloaca Maxima. From this oldest phase, the Temple of Jupiter Optimus Maximus on the Capitolium was the most important monument. The forum was hence divided into two distinctive parts: a religious-political area at the foot of the Campidoglio and a commercial zone further east.2 All of the buildings on the forum were frequently restored and remodelled during their long period of use. At the end of the Republic, the forum Romanum had become too small to cope with the political, religious and commercial needs of the city. Julius Caesar was the first to plan a new forum, just north of the forum Romanum.3 Several other emperors would expand the economic and political heart of Rome by adding their own forum: the Forum of Augustus, the forum transitorium or the Forum of Nerva and the large Forum of Trajan with an adjacent commercial complex (‘Trajan’s market’). To the south of the forum Romanum, the Palatine hill seems to have been the preferred location for the residences of the elite.4 Especially in the 2nd and 1st c. BC, the large domus of consuls, such as Cicero, were located here. Octavian was even born here and decided to stay and live there even when he became emperor.5 Under Tiberius, the Flavians (especially Domitian) and the Severans, large domus were built and expanded until most of south-eastern part of the hill belonged to the imperial palace.6 Elagabalus ordered the construction of a large temple dedicated to the sun god Elagabalus. On the southern and eastern slopes, towards the forum boarium, several horrea were built. The valley between the Caelian, Palatine and Esquiline hills was used as the site of Nero’s ‘golden house’, but was returned to the people of Rome by the Flavian dynasty. The artificial lake of the imperial complex was filled up and a gigantic amphitheatre, the amphitheatrum 2  Purcell (1995a); (1995b); Coarelli (2008) 47. 3  Morselli (1995) 299–306. 4  Tagliamonte (1999) 14–22. 5  Papi (1999b) 22–28. 6  Papi (1999c) 28–33.

chapter 3

Flavium (or Colosseum) was built.7 To the north, the Baths of Titus (see below, p. 97) may have replaced the private bath complex of Nero.8 Under Domitian, three gladiatorial schools (ludi) were constructed east of the Colosseum. The Baths of Trajan were a milestone in the architecture of the imperial type bathhouse (see above, chapter 1), built on the filled-in remains of Nero’s Domus Aurea. Some 80 m to the east, the large water reservoir now called ‘Sette Sale’ received water from the Aqua Iulia to feed the baths. To the north, the porticus Livia is only known through the marble Forma Urbis.9 South of the Palatine, the Circus Maximus was restored and expanded several times during the 1st c. AD. North-west of the Circus, the forum Boarium (cattle market) was surrounded by several temples. A small arch, dedicated to the argentarii (bankers), dates from the early 3rd c.10 Regio I evolved around two important traffic arteries: the Via Appia and the Via Latina. Along both roads, the funerary monuments of important Romans and columbaria were erected. Regio II was dominated by the temple of Divus Claudius on the Caelian hill, dedicated to the deified Claudius after his death in AD 54 and finished under Vespasian. In the vicinity must have been situated the macellum magnum, built under Nero, although its exact location remains unknown.11 The area between the Caelian and the Lateran was dotted with residential buildings and crossed by the Aqua Claudia.12 Towards the porta Asinaria stood a residential neighbourhood developed in the 2nd and 3rd c. with two bathhouses: the Lateran Baths and the Baths under the Lateran baptistery (see below, p. 101). At the end of the 2nd c., the imperial cavalry was housed in the Castra nova equitum singularium.13 Between the Lateran and the porta Praenestina, the Castra priora equitum singularium stood in a largely residential neighbourhood. Around the porta Maggiore, the Castrum amphitheatre (Severan era) may be identified with the Vivarium, a park-like area for games.14 The Baths of Helena (see below, p. 97) date from the Severan period. By going north towards the Esquiline, one crossed residential neighbourhoods with several parks (horti), which were incorporated in the imperial estates from 7  Rea (1993). 8  Coarelli (2008) 204. 9  Slab VIII-3, 10o-r in Rodríguez Almeida (1981) plate 18. 10  Coarelli (1995a) 295–97. 11  Coarelli (2008) 278. 12  Besides the Aqua Claudia, three other aqueducts crossed the Celio, be it underground: the Aqua Appia, the Aqua Marcia and the Aqua Iulia. 13  Buzzetti (1993) 246–48. 14  Richardson (1992) 431–32.

Archaeological Evidence

map 4

Location of the most important buildings and public areas in Rome Imperial Period and Late Antiquity; plan by the author 18: Tiber Island 1: Castra Praetoria 19: Theatre of Marcellus 2: Porticus Liviae 20: Forum of Trajan 3: Basilica S. Croce in Ger. 21: Crypta Balbi 4: Castra prioria eq. sing. 22: Saepta Iulia 5: Basilica Salvatoris 23: Theatre of Pompey 6: Colosseum 24: Stadium of Domitian 7: Temple of Divus Claudius 25: Villa of Agrippina 8: Church SS. Nereus et Achilleus 26: Circus of Caligula / S. Peter 9: Circus Maximus 27: Mausoleum of Hadrian 10: Palatine palaces 28: Naumachia vaticana? 11: Basilica nova 29: Mausoleum of Augustus 12: Forum Romanum 30: Horti Luculliani 13: Imperial fora 31: Horti Sallustiani 14: Forum Boarium 32: Temple of Serapis 15: Church S. Sabina 33: Church S. Agatha dei Goti 16: Navalia and horrea 17: Naumachia of Augustus

93

94 the Julio-Claudian period onwards.15 Near the porta Esquilina, the macellum Liviae was probably constructed (1st c. AD).16 More to the south, the Aqua Iulia ended in an impressive nymphaeum from the time of Alexander Severus. On the Cispius, along the clivus Suburanus, large domus from the imperial age were discovered,17 while on the summit of the clivus, there was a large fountain of Orpheus. On the north-eastern edge of regio VI, the castra pretoria was built under Tiberius to house the praetorian guard.18 In the area of the modern Termini station, underneath the Piazza dei Cinquecento, a neighbourhood with houses, shops and public baths dating from the 2nd c. was discovered (see below, p.98–99). To the north of the Viminal, the villa estate of the historian Sallust (horti Sallustiani) originally belonged to Caesar and was later acquired and extended by Tiberius.19 On the eastern slope of the Viminal, the living quarters continued, while the Baths of Novatianus were also found (see below, p. 99–101). Regio VII centred around the Via Lata (modern via del Corso), where large apartment blocks were constructed during the reign of Hadrian. Shops and warehouses have also been found.20 In this commercial-residential neighbourhood, a bathhouse has been found during recent construction works for a new Rinascente mall (see below, p. 101).21 On the location of the modern Galleria Sciarra, the Porticus Vipsania may have been located (Augustan period), and has been proposed as the office of the cursus publicus, the imperial postal system.22 Regio IX is the area of the Campus Martius, originally the plain where soldiers and young men exercised. It was also the area that was used for electoral purposes during the Republic. The temples of Apollo and Bellona and the sacred area of Largo Argentina— albeit with later restorations—are the only remains from the mid Republic. At the end of the Republic, the theatre of Pompey was built. Under Augustus, an important building program was carried out, resulting in the dedication of the theatre of Marcellus, the theatre of Balbus and its square, the porticus Octavia with two temples and libraries, the Baths of Agrippa (see below, p. 95–97), the Pantheon, the saepta Iulia and the Ara Pacis.23 After Augustus’ death, a round mausoleum was 15  Coarelli (2008) 255–57. 16  Pisani Sartorio (1996) 203–204. 17  Coarelli (2008) 244. 18  Lissi Caronna (1993) 251–54. 19  Richardson (1992) 202–203. 20  Coarelli (2008) 333. 21  It is unclear if the baths belonged to regio VI or VII. 22  Coarelli (2008) 336. 23  Coarelli (2008) 346–49.

chapter 3

constructed on a parcel of land between the Tiber and the Via Lata.24 Nero added large thermae (see below, p.97) near the Pantheon, while Domitian added a stadium (modern Piazza Navona), an odeum, the porticus Minucia Frumentaria and the porticus Divorum sanctuary.25 The temple of the Divus Hadrianus was added in the late 2nd c. To the south of the theatre of Marcellus, the forum holitorium (oil market) was equipped with three small temples in the Republican age.26 To the east, the temples of Fortuna and Mater Matuta were 3rd c. BC reconstructions of older republican temples. Along the Tiber, towards the forum boarium (cattle market), large horrea date from the reign of Trajan.27 The name of regio XII, Piscina Publica, probably referred to a lake that was used for bathing and public use in pre-Augustan times.28 It was only under Caracalla that a large public bathhouse was built there (see below, p. 97). Besides this enormous structure, the area seems to have been made up of living quarters and some larger domus.29 Nearby regio XIII, the Aventine hill, was a mercantile area in Republican times.30 It was only included in the pomerium of the city under Claudius. Since the 3rd c. BC, a temple dedicated to Minerva stood on the hill top, while in the 1st c. AD, a temple of Diana was constructed nearby. From the Forma Urbis, a bathhouse by the name of balneum Surae is known to have stood to the south-east of the temple of Minerva.31 In AD 252, the thermae of Decius were inaugurated (see below, p. 98). Two small baths, possibly private, were discovered on the western flank of the Aventine.32 Little is known about their chronology. To the south of the Aventine, the city port (emporium) developed from the 2nd c. BC onwards. Along the Tiber, a 487 m long structure should possibly be identified as the navalia, used to harbour the military fleet in the Republican period and later transformed into warehouses. Several horrea have also been identified in this area.33 The mercantile character of this regio is also represented by the Testaccio hill, made up entirely out of amphora sherds.34 The important building on Tiber Island was a temple dedicated to Asclepius (3rd c. BC).35 On the other side 24  Richardson (1992) 247–49. 25  Coarelli (2008) 348. 26  Coarelli (1995b) 299. 27  Coarelli (2008) 405. 28  Coarelli (1999) 93–94. 29  Coarelli (2008) 406. 30  Andreussi (1993) 147–50. 31  Slab VII-14, 21a-d in Rodríguez Almeida (1981) plate 15; see also below p. 102. 32  Coarelli (2008) 461. 33  Coarelli (2008) 464–65. 34  Maischberger (1999a) 28–30. 35  Degrassi (1996) 99–101.

95

Archaeological Evidence

of the Tiber, the modern Trastevere neighbourhood had become the most densely populated part of the capital at the end of the High Empire.36 Our knowledge of the archaeological remains, however, is quite limited. The largest public structure was probably Augustus’ Naumachia, an artificial lake used to re-enact epic naval battles.37 No remains of this large structure have been found. Several smaller sanctuaries, some of which were dedicated to ‘eastern’ deities, were dispersed throughout the residential quarters.38 Outside of the porta Septimiana, the large villa estate of Agrippina (1st c. BC) was discovered.39 The circus of Nero (or Gaianum) was the most important public building in the ager vaticanus. To its north-west, a burial ground has yielded several 2nd and 3rd c. sepulchres. Following the Tiber to the east, one could not miss the mausoleum of Hadrian.40 North of it, some 2nd c. remains have been interpreted as the Naumachia Vaticana, built by Trajan to replace one of Augustus.41 Newly Built Late Antique Baths At the beginning of the 4th c., the emperors Diocletian and Constantine still deemed it important to construct large public baths (C21, C22) to gain popularity with the people of the Eternal City, even though they themselves resided elsewhere. The Baths of Diocletian were even the largest public baths ever built in the Roman empire. Other late antique baths were rather small, not exceeding 500 m2. The continued use of the imperial thermae, as well as dozens of others smaller neighbourhood baths (see below, p. 102–105), may have fulfilled all bathing needs, especially after the 4th c., when population numbers were dropping. With the relocation of court to Constantinople, imperial investment in Rome dwindled and no new imperial baths were erected. The small Baths of Elagabalus (C24) and on the Palatine were possibly private investments, perhaps even linked to specific associations. The growing grip of associations on the fabric of the city during Late Antiquity has also been seen in the construction of fountain houses, which may have been used as assembly halls.42 To construct the new imperial thermae, existing buildings and insulae were levelled. For the smaller baths, pre-existing buildings were often partially destroyed or partially transformed (C23, C24, C25, C26). Putting existing buildings to new use had the advantage 36  Maischberger (1999b) 77–83. 37  Richardson (1992) 265. 38  Coarelli (2008) 452. 39  Coarelli (2008) 479–80. 40  Richardson (1992) 249–51. 41  Coarelli (2008) 489. 42   Schmölder-Veit (2010).

of recycling building materials or even entire walls, but constrained the available building space to a predefined lot. However, this did not prevent the architects from adhering to classical building schemes and contemporaneous taste for round and semicircular forms, especially for pools, apses and exedrae. Furthermore, the limited building space did not result in a reduction of heated rooms, as even the smaller baths had several hot rooms (C23, C24). The small size of pools in the neighbourhood baths are in line with the overall small size of the bathhouses themselves. We can note that a somewhat larger bath (C25) could still have large communal pools, with several small single-person pools in addition. The plans of the imperial baths fit within the general evolution of this type in Rome. The importance of the garden area, with a large semicircular exedra, seems to stress the growing importance of the non-bath related activities that took place within the enclosure wall. The cavea-like exedrae in the imperial baths may well have hosted certain types of performance in an area of the city otherwise without of theatres or odea. The decoration in the imperial thermae was luxurious, just as it had been in the older imperially funded baths. Pools, floors and walls were embellished with marble slabs and decorative elements, while vaults were supported by granite and marble columns. Some of the smaller baths also boasted marble-lined pools (C24, C26, Porta San Lorenzo Baths), while floors could be covered in mosaics depicting geometric motifs or marine scenes (C23, Porta San Lorenzo Baths). The heating system still relied on the traditional hypocaust with pillars made of square or round tiles. The preferred type of wall heating was by means of tubuli. In at least one bathhouse, the use of a testudo can be deduced from the ‘Beckenausbruch’ (C23), but it can be imagined that the large alvei in the imperial thermae also had such devices. The water for the imperial baths was presumably supplied by aqueducts. For the small neighbourhood baths, the catchment of rainwater, supply by an aqueduct (C24?) or wells (C26?) or a combination of the above probably sufficed. Waste water disposal was mainly ensured through (pre-existing) sewers in the streets surrounding the baths (C21?, C22?, C23, C24). Continued Use and Abandonment of Existing Baths The Baths of Agrippa The Baths of Agrippa (map 5, nr. 1) were the first large public baths of the imperial type built in Rome. The plan, mainly known through Renaissance drawings,43 underwent several structural changes. Due to a lack of 43  Reproduced in Hülsen (1910) 10–14.

96

map 5

chapter 3

Location of the public baths in Rome C-numbers refer to the gazetteer, plan by the author 1: Baths of Agrippa 2: Baths of Nero 3: Baths of Titus 4: Baths of Trajan 5: Baths of Helena 6: Baths of Caracalla 7: Baths of Decius 8: Baths of P. dei Cinquecento 9: Baths of Novatianus 10: Baths of the Lateran Bapt. 11: Lateran Baths 12: Baths of the New Rin. Mall 13: Baths near the navalia 14: Balneum Surae 15: Minerva Medica 16: Baths near Porta S. Lor. 17: Baths near S. Croce in G. 18: Baths on Palatine

Archaeological Evidence

archaeological remains, the chronology of the building cannot be fully ascertained. According to the Notitia regionum urbis Romae (see below, p. 103–104) and Sidonius Apollinaris (see above, chapter 2), the baths were still in use in the 4th and the late 5th c. The Baths of Nero The Baths of Nero, sometimes called the thermae Alexandrinae, set the standard for imperial thermae in Rome (fig. 4; map 5, nr. 2). The inauguration of the baths is still included in the Chronicle of Jerome (see above, chapter 2) and Cassiodorus’ Chronica (ad a. 227). Apparently, the baths were still in use during the lifetime of Sidonius Apollinaris (Carm. 23, 495; see also the Notitia regionum, regio IX). The archaeological investigations in the area of the baths could not supply any conclusive evidence about their last phase.44 Several architectural elements, such as columns, were reused in various churches in the vicinity of the ruins.45 The Baths of Titus The Baths of Titus (map 5, nr. 3), located on the grounds of Nero’s Domus Aurea, were mentioned by the 4th c. Notitia Regionum (Regio III). However, the excavations in the 1980s revealed that the baths were out of use in the 5th or 6th c., when occupation layers were found inside the building.46 The site was probably reused to construct the palace of the praefectus urbi.47 The Baths of Helena The remains of a large public bathhouse near the porta Maggiore and Santa Croce in Gerusalemme could still be seen in the 16th c. (map 5, nr. 5). The plan was drawn by Palladio, but the structural remains were destroyed when the neighbourhood was reorganized in the early 20th c.48 The baths were named after Helena, the mother of Constantine, who restored the baths in the early 4th c. (EI-4). The construction probably dated to the 3rd c., when the entire quarter was reorganized.49 The only remains still existent are the large reservoirs that provided the baths with water and were themselves fed by the Aqua Alexandrina. Studies of construction techniques and the type of mortar point to an early 4th c. date for this reservoir.50

44  Ghini (1985) 399. 45  Ghini (1999) 62. 46  Caruso et al. (1990) 63; Caruso (1993). 47  Caruso et al. (1990) 67. 48  Palladino (1996) 857. 49  Palladino (1996) 869–70. 50  Palladino (1996) 867.

97 The Baths of Trajan The Baths of Trajan had an important impact on the layout of imperial baths (fig. 5; map 5, nr. 4). An inscription mentioning the curia A(th)letarum, dated to AD 367–375, may have refered to the athletic facilities in the Baths of Trajan.51 In AD 465, the praefectus urbi still invested in the baths (EI-45) and they are also mentioned in the Notitia Regionum (Reg. III). Archaeological investigations of their water supply, especially of the large reservoirs known as the Sette Sale, confirmed that the baths were restored in the 4th c. and fell out of use only in the 5th c.,52 possibly in the third quarter if we combine stratigraphic evidence from the water supply and the inscription of AD 465–467.53 However, as there are only a few remains of the baths preserved, it is not possible to reconstruct any late antique additions or changes made to the original plan. The Baths of Caracalla The famous Baths of Caracalla (fig. 7; map 5, nr. 6) were in use until at least the reign of Theoderic. Tiles with stamps of the Gothic king were used to carry out restorations.54 Attestations by Polemius Silvius and Olympiodorus (see above, chapter 2), both writing in the 5th c., confirm this late phase of use. Diocletian (see EI-2) and Constantine55 had ordered restorations at the beginning of the 4th c. However, it is difficult to identify these interventions in the archaeological record, especially since the excavations of the 19th and early 20th c. were poorly documented. The supporting walls in the service corridors and behind the reservoirs of piscinae possibly date to the Constantinian era.56 Some statue bases in the frigidarium mention the reign of Valentinian and Valens.57 The polychrome geometric mosaics with large tesserae may also date from the late 3rd or 4th c.58 The cold water pools in the southern side-rooms of the palaestra and possibly the semicircular pools in heated rooms on the south-eastern and south-western corner of the bath block were also added at a later date, perhaps when the aqueduct was temporarily out of use.59 51   C IL 6.10154; see Volpe (2007a) 431–34. 52  Carboni (2003) 76. 53  F. Carboni, oral communication. 54   C IL 15.1665; Lanciani (1879) 15; Lanciani (1881) 90. A headless statue of the goddess Artemis was reused as a paving stone in the underground galleries at this time (Piranomonte (2012) 26). 55  Piranomonte (2012) 18–20. 56  DeLaine (1997) 38. C IL 6.794; 117–1173. 57   58  DeLaine (1997) 28, esp. n. 59. 59   DeLaine (1997) 39–40. Janet DeLaine even suggests that the new pools may have replaced the larger cold pools of the frigidarium in times of disrupted water supply, or, alternatively,

98

chapter 3

figure 12 Plan of the baths under the Piazza dei Cinquecento in Rome (Italy), (left), and the excavation during the construction of modern Termini train station (right) in Santagnelli Valenziani and Meneghini (1996) 176, figs. 5–6; with permission

The baths are thought to have been abandoned in the 6th c., when the Gothic king Vitiges cut the aqueducts during his siege of the city in AD 537 (see above, chapter 2, Procopius). Excavations in the gardens revealed burials dating from the 6th to 8th c.60 These may have been connected to a nearby xenodochium.61 Some of the peripheral halls in the gardens may also have been used as granaries after the 6th c.62 The Baths of Decius According to the written sources,63 these imperial baths were dedicated in AD 252 on the Aventine hill (map 5, nr. 7). The remains were sketched by Palladio and show a compact plan (ca. 70 × 35 m) with a cruciform frigidarium and several round rooms, confirmed by a survey in

became the new cold sections of the baths when parts of the large heated section could not be heated any more due to fuel shortage (DeLaine (1997) 40). 60  Cecchini (1985) 588–94. 61  Now the church of SS Nereus and Achileus, see Mulryan (2014) 35–38. 62  Cecchini (1985) 593. 63  Cassiod., Chron. ad a. 252.

the 1980s.64 Inscriptions found on site recall restoration works carried out by the praefectus urbi in AD 353–354.65 They may have been restored under Honorius in AD 414.66 The Baths under the Piazza dei Cinquecento The bathhouse under the Piazza dei Cinquecento (fig. 12; map 5, nr. 8) was discovered during the construction of the first Termini train station in the 1860s. The remains were subsequently reburied until the new Termini train station was built in 1947–1948, which offered another opportunity to investigate the building. However, the observations by the archaeologist Pietrogrande were not published until the 1990s.67 The bathhouse was inserted into an insula, next to a rich domus with peristylium. The complex was dated to the 2nd c. on the basis of wall paintings and mosaic floors.68 At the turn of the 3rd and 4th c., the baths underwent important changes.69 As the study of the 64  La Folette (1985); La Folette (1994). 65   C IL 6.1159; 1160. 66  Nielsen (1993a) 55, n. 128. 67  Barbera and Paris (1996) 117–71. 68  Barbera and Paris (1996) 125. 69  Meneghini (1999) 172.

99

Archaeological Evidence

figure 13 Drawing of the caldarium of the baths under the Piazza dei Cinquecento in Rome in Barbera and Paris (1996) 132, fig. 3; with permission

remains was done on the basis of photographs and excavation diaries, it was difficult to detect late antique restorations. It is possible that the small semicircular pool, lodged in one of the niches of the frigidarium, belonged to this phase. It is an element that seems out of place in the otherwise strict symmetrical layout. The fact that only one of the niches was equipped with a pool, and that the pool itself was larger than the niche (F on fig. 12), suggests that it may have been a later addition. Similarly, the small heated pool in the caldarium could also be a later addition (fig. 13). The walls of this round single-person pool were made in a different wall technique, which were obviously of a later date, according to Pietrogrande.70 The niche in which the pool was lodged, however, was certainly contemporaneous with the construction phase of the building. Perhaps it used to be a doorway to yet another heated room, which was abandoned in a later stage and turned into a praefurnium. Alternatively, the niche was originally a schola labri, which was then transformed into a small pool. Unfortunately, the limited data does not permit a verification of this hypothesis. The basic restorations of the 70  As quoted by Barbera and Paris (1996) 148, n. 4.

mosaics should also be attributed to this later phase of the baths. Simple marble slabs were inserted where the tesserae had disappeared.71 In the second half of the 5th c. or in the early 6th c., the baths were abandoned, as was the domus north of it.72 Several statues were found on the frigidarium floor. These were possibly stored here after the baths fell out of use and did not necessarily come from the bath itself. The presence of poorly-built walls, typical for the 6th and 7th c., reveals that the baths were reused for housing or industrial purposes at this time.73 The Baths of Novatianus under Santa Pudenziana During excavations in the 1930s, the remains of a bathhouse were identified underneath the Church of Santa Pudenziana (map 5, nr. 9). On the basis of brick stamps, the remains have been dated to the reign of Hadrian.74 71  Barbera and Paris (1996) 132, fig. 3 and 140, fig. 1. 72  Santagnelli Valenzani and Meneghini (1996) 173. 73  Santagnelli Valenzani and Meneghini (1996) 173; Meneghini (1999) 172. 74  Krautheimer and Corbett (1967) 277–79 with anterior bibliography; for doubts about the interpretation as a bathhouse, see Mulryan (2014) 29–30.

100

chapter 3

figure 14 Plan of the Baths under the Lateran Baptistery in Rome (top) and plan of the Lateran Baptistery (bottom), with a small pool to its north after Pelliccioni (1973) 60, fig. 93 and 94, fig. 136

Archaeological Evidence

A large basilical hall (9 × 27.5 m) was equipped with oval cold pools, and should consequently be interpreted as the frigidarium. The baths were probably remodelled into a church at the turn of the 4th and 5th c., but it cannot be ruled out that the baths ceased functioning in an earlier period and were reused as a Christian meeting place without any significant modifications.75 According to the Liber Pontificalis, a church was already built in the Novatian baths under Pius I (AD 142–155).76 Even if the archaeological evidence cannot confirm this claim, the Baths of Novatianus are still the only known example of a secular public building being reused as an early Christian church in Rome. The Baths under the Lateran Baptistery Excavations underneath the Lateran baptistery (1960s or 1970s?) revealed the remains of a substantial bathhouse (ca. 1500 m2, fig. 14 top), dated to the reign of Trajan on the basis of brick stamps (map 5, nr. 10).77 Later renovations took place under Hadrian, Antoninus Pius and at the end of the 3rd c. Under Constantine, the baths were razed to the ground and a baptistery was constructed on top of the remains (fig. 14 bottom). The sewers and the water supply of the baths made an interesting argument in the choice of the site. The southern piscina was also reused to create the actual font. During the 4th c., the chapel of San Giovanni Evangelista was created using tubi fittili as a building material. A small room heated by a hypocaust and tubuli was added north of the octagonal baptistery. The small pool, decorated with marble, was interpreted by the excavators as a ‘winter baptismal font’ or a reservoir of tepid water.78 It would be strange, however, if a small pool, lodged in an angle of a room and hence not fit for the rite of baptism, was preferred above the sumptuous font. The lush marble decoration and the internal steps make it highly unlikely that this was a simple reservoir. They seem to indicate that this small pool served as a heated bathing tub in which the catechumen could wash before receiving the sacrament of baptism.79 The combination of a baptistery with bath rooms or even a bathhouse was not uncommon (see below C27, Cuicul, Thamugadi, Tipasa). The Lateran Baths The bathhouse was discovered during construction works on the Lateran hill, near the modern episcopal complex (map 5, nr. 11). The structure partially reused 75  Krautheimer and Corbett (1967) 299–300. 76  L ib. Pont. I, 132. 77   The single most important source on the excavations is Pelliccioni (1973). 78  Pellicioni (1973) 73–75 (later restorations), 86–87 (piscina as font), 104–105 (tubi fittili), 106–17 (winter font). 79  See Augustine about bathing prior to baptism, Ep. 54.

101 some older walls that were dated by brick stamps to the 2nd c. and the Severan period. The bathhouse itself probably dates to the 3rd c.80 An inscription was dated to the 4th c., which could point to a continued use of the baths.81 The plan does not reveal any important alterations that can be linked to this later phase of use. The Baths of the New Rinascente Mall During the construction of a new shopping mall in the centre of Rome on the junction between the via del Tritone and the via Due Macelli in 2011, the remains of a bathhouse were discovered (map 5, nr. 12). What appears to be a frigidarium had two semicircular piscinae and a heated trilobate room, possibly the caldarium. The trilobate design could point to a late antique phase of the building.82 Private Baths Although this research does not focus on private bathhouses and the difference between private and public baths is often difficult to discern, it can be interesting to give a brief overview of the private bath installations that have been attributed to Late Antiquity. The ‘privatization’ of the bathing habit has sometimes been mentioned as a characteristic development of Late Antiquity.83 Therefore, it is surprising that only a limited number of private baths has come to light during excavations in a city such as Rome. The Baths of the House of the Sette Sale are a clear example of a private bath.84 The domus, built on top of the large reservoirs of the Baths of Trajan, had a small suite of bathing rooms that could only be accessed from the interior of the house. The remains are dated to the late 4th c.85 On the Caelian hill, a room with a hypocaust, tubuli and two semicircular alvei was interpreted as part of a small private bath of a domus. The private character was, however, only deduced from its size and its location within a residential neighbourhood. The bathing rooms were added to an existing domus during the 4th c. and remained in use until at least the 5th c.86 The Baths under Palazzo Valentini, just north of Trajan’s forum, were attached to a 3rd c. domus.87 At the 80  Colini (1944) 335 (construction date), 339 (brick stamps). 81   C IL 6.31394. Colini (1944) 338. 82  Unfortunately, the site was still under investigation at the time of writing and no publication was available. For a comparison, see the via della Foce baths in Ostia (C14). 83   Ward-Perkins (1984) 148; Sagui (1990b) 100. 84  Volpe (2000a) 161–62. 85  Cozza (1975) 92. 86  Carignani (1993) 502–505. These baths should not be confused with the heated rooms mentioned by Pavolini ((1993) 447–48), which should probably be interpreted as the heated rooms of a private villa. 87  Gatti and De Spagnolis (1981).

102 table 1

chapter 3 Buildings in each of the regiones of Rome as presented by the Curiosum and the Notitia Regionum I

II

III

IV

V

VI

VII

thermae

2

0

2

0

0

2

0

balnea

86

85

80

65/75

75

75

75

lacus

81/87

65

65

71/78

74

73

76

insulae

3250

3600

2757

2757

3750

3403

3805

domus

120

127

60/160

88

180

146

120

streets

10

7

12

8

15

17

15

shrines

10

7

12

8

15

17

15

horrea

16

27

17

18

22

18

25

bakeries/mills Circumference

20 12211/12219

15

16

15

15

16

15

12200

12350

13000

15600

15700

13300

turn of the 3rd and 4th c., the baths were enlarged.88 As the total extent of the baths could not be determined, it cannot be ruled out that these baths had street access and were hence semi-public in use. Besides these confirmed examples, the baths on the via Ariosto, the forum Romanum, near Santa Croce in Gerusalemme and near Porta San Lorenzo may or may not have been part of a private residence. The archaeological evidence is supplemented by written sources. Here, the Liber Pontificalis is especially informative.89 The archbishops of Rome sometimes donated a house with baths to a certain church. The Titulus Equitii, located near Santa Maria Maggiore, was given a ‘domum in urbe cum balneum’ (‘a house in the city, with baths’) by Silvester (archbishop between AD 314–335). Innocent I (AD 401–417) gave a similar house to the Church of San Vitale and Sixtus III (AD 432– 440) gave a ‘domus cum balneum’ to the basilica of Santa Maria Maggiore.90 Even if these baths were clearly a part of the house, integrated into the general layout or as a separate building that was connected to it (compare the evidence from the Digesta in chapter 2), it is not certain that these baths were for the private use of the owners of the house. On the contrary, it seems likely that these baths were open to the public.91 The baths could attract worshippers and pilgrims, while the generated income was probably a welcome financial support for the titulus.92

88  Baldassari (2008) 368. 89  Stasolla (2002a) 23–31; Hubert (2007) 128–32. 90  Silvester: Lib. Pont. 1.170–171; Innocent I: Lib. Pont. 1.221; Sixtus III: Lib. Pont. 1.233. 91  Stasolla (2002a) 28. 92  Sagui (1990b) 100.

Baths on the Forma Urbis Roma Several bathhouses can be identified on the fragments of the Severan marble plan. Four are marked with the word balneum or balineum, while other buildings can be recognized on the basis of their plan.93 The example best known is the balneum Surae in regio XIII (map 4, nr. 14). Besides the plan on the Forma Urbis, it is also known through various literary sources.94 However, no archaeological remains have been found. The balneum Caesaris (FUR frag. 43ab), the balneum Ampelidis (FUR frag. 47) and the balneum Cotidis (FUR frag. 48) are known from Renaissance drawings of the now lost fragments of the FUR. The first has been identified as the Domitianic bath of the imperial palace on the Palatine, while the other two have not been located.95 Some fragments without an accompanying text have been identified as baths by the presence of apsidal rooms (FUR frag. 377ab) or by a ‘Pompeian-type plan’ consisting of a palaestra flanked by a suite of rooms (FUR frag. 330, 647). Only one of these baths could be located, in an insula south of the horrea Lolliana near the Testaccio hill.96 It has been proposed that these baths belonged to the headquarters of a collegium or should be identified with the balneum Tigillini mentioned by Martial.97

93  Staccioli (1961) 97–98. The numbers of the fragments used throughout this book refer to the numbers of fragments as found in Rodríguez Almeida (1981) and the revised version of Stanford’s Digital FUR project (http://formaurbis.stanford .edu/index.html). 94  Staccioli (1961) 94, n. 2. 95  The balneum Ampelidis was located in regio XIV according to the Notitia regionum (see below). 96  Staccioli (1961) 98. 97  Mart. 3.20; see Rodríguez-Almeida (1995–1996). For the interpretation as collegium, see Staccioli (1968).

103

Archaeological Evidence

VIII

IX

X

XI

XII

XIII

XIV

0

2

0

0

1

2

0

86/85

63

60

15

63

44/60

86

89/90

120

120

3480

2777

2742/2642

20

80/81

89

180

2600

2487

2487

4405

130

140

89

88/89

113

130

150

34

35

20

21

17

18

78

34

35

20

21

17

18/17

78

18

25

48

16

27

35

22

20

20

20

16

25

20

24

14067

32500

11510

11500

12000

18000

33388

In some cases, continued use can be ascertained by consulting the 4th c. Notitia Regionum.98 The balneum Ampelidis and the balneum Surae are mentioned for regiones XIII and XIV respectively. If the identification of the balneum Caesaris with the palace baths of the Palatine is correct, we can propose that the baths in their ‘FUR-shape’ ceased to exist when they were transformed under Maxentius. For the fragments without accompanying text and without known location, it is impossible to find out if they had a late antique phase of use. The Evidence for Baths in the Written Sources The Notitia Regionum, also called the Notitia Romae, is a late antique catalogue listing the most important buildings in the 14 regiones of the capital (table 1). There is much debate about the meaning and date of this work. A plausible theory connects the Notitia to the reforms of Diocletian, with later additions under Constantine and his sons.99 The main argument is that no Christian monuments are listed and that there are still references to the barracks of the praetorian guard, which were destroyed by Constantine. However, the inclusion of Constantinian monuments means that a revised version must certainly have been made.100 The list also existed in a later version, called the Curiosum urbis Romae regionum XIV, which differed on some points from the Notitia. It included the obelisk added to the Circus Maximus by Constantius II (AD 357). The Breviarium is a short list simply enumerating the different building types in the capital (13 thermae and 856 balnea). The numbers do not correspond to the sum of the buildings mentioned in the Notitia or the Curiosum. This is probably because

some of the numbers in both of these works have been altered during medieval copying.101 Later authors such as Polemius Silvius (see above, chapter 2, p.53) or the anonymous Syriac chronicler of AD 596 seem to have used the Notitia and the Breviarium. The Syriac chronicler does not mention the thermae, but records 924 balnea in Rome. The division of the city in 14 regiones was the work of Augustus in AD 7.102 In the Notitia, we find a catalogue of the most important buildings for each regio, followed by the number of streets, shrines, vicomagistri (streetwardens), curators, insulae, domus, horrea, balnea, reservoirs and bakeries / mills (table 1). For each regio, the circumference in Roman feet is also given. Although the Notitia informs us in surprisingly great detail about the urban fabric of late antique Rome, we should proceed with caution when using it as a tool to reconstruct the urban fabric. Firstly, the exact delineation of each regio is unknown. The division was based on the Augustan regiones, but through time each regio had grown and its form was often irregular.103 A comparison between regiones is also difficult, as the numbers given for the different types of buildings does not give any indication about their size. For example, the 65 reservoirs of regio III could have contained the same volume as the 76 of regio VII, depending on the size of the individual reservoirs. As mentioned above, the Breviarium also reveals that some of the numbers in the Curiosum and Notitia may have been altered. For baths, both the Notitia and the Curiosum mention several buildings by name. In the plethora of balnea, we can assume that those mentioned by name were

98  Foulché (2011); see below. 99  Nordh (1949) 60. 100  Nordh (1949) 60, 64.

101  Foulché (2011) 602. 102  Robinson (1992) 9–13. 103  Platner and Ashby (1929) 444–47.

104 somewhat exceptional.104 Perhaps they were the most popular, the most luxurious or the most famous of a regio. Alternatively, they could have had a specific topographical importance (centrally located, on crossroads, on boundaries between regiones) or may have been privately owned baths that came into imperial hands, perhaps as a result of confiscation.105 Unfortunately, the balnea mentioned in the Notitia are not known archaeologically or from the FUR.106 As the size of the different baths and the total surface of each regio cannot be known, it is difficult to compare the regiones based on the number of baths alone. The number of baths in regio VIII (the imperial fora and the Campidoglio) seems rather high (85) for a largely commercial and ceremonial area.107 However, we do not know the size of the baths, perhaps all very modest, nor do we know the 4th c. extension of this regio. A similar problem arises for the baths in regiones IV to VII, all numbered at 75 balnea. Besides the remarkable coincidence of this repeated number, one could conclude that regio IV, which was smaller in size according to the Augustan division, had proportionally more baths than the other regiones (ratio baths / size of regio). However, regio VII had two large imperial thermae that could accommodate several thousand bathers at a time, while the baths in regio V could, for example, be larger than those in a densely populated regio VII. If we consider the Notitia as a very schematized representation of the city, rather than as a reflection of an actual count of 4th c. baths, and as an internally coherent document, we can attempt a very careful comparison between the regiones.108 By calculating the percentage of baths for each regio and by plotting these on a map of the city with the regional divisions, we can get a schematized view of the spread of balnea as envisaged by the composer (map 6). A comparison of the percentage of baths with the percentages of other buildings and then comparing the different regiones with each other can lead to interesting results. If we compare the share of baths with the share of insulae for each regio, we can notice how a low number 104  Foulché (2011) 600–601. 105  Foulché (2013). 106  The only exception is the balneum Ampelidis, known from a Renaissance drawing of the FUR (see above, p. 102). The fragment is too small to make any conclusions about size or topographical importance. 107  Foulché (2011) 604. 108  Already Fagan (1999a) 358, map A3. Some numerical values such as the 75 balnea in the regiones IV to VII or the rounded numbers of insulae or domus could underpin such a schematic intent (Fagan (1999a) 357). Alternatively, they could point to adjustments made by the copyists of the text, for example to complete lost or illegible parts of the text (Foulché (2011) 604).

chapter 3

of insulae did not necessarily mean that the regio had a low number of baths. For example, regiones XII and XIII only had 5.6% of the total number of insulae, but had an average number of balnea (resp. 6.5% and 6.2%). Regiones with slightly more insulae, such as X (6.3%) or XI (5.9%), had a lower of share of balnea (resp. 4.6% and 1.6%). If we compare the number of baths for each regio with the number of domus, we can notice how the regiones with the lowest number of baths (X and XI) also had the lowest number of domus. Perhaps a considerable number of these private houses had private bath suites, making public baths less popular. Nevertheless, such a hypothesis cannot be applied for other regiones, for which a high number of baths (e.g. IV) existed for a low number of domus. The fact that the Notitia does not differentiate between publicly and privately accessible balnea blocks any further exploration of this hypothesis.109 The number of reservoirs for each regio is proportional to the number of balnea. The only important differences are seen in regiones VIII to X and XII to XIV, which can perhaps be explained by a higher demand for water in the city centre in order to supply the large public buildings, fountains and nymphaea. We should again stress the odd number of exactly 120 reservoirs in the regiones VIII and IX, which could point to interference by the copyist. Baths Known from Ancient Authors and Inscriptions Several baths in Rome are only known through passages in ancient texts. In the SHA (see above, chapter 2), the thermae Septimianae and the thermae Severianae are mentioned, but their actual location within the city remains problematic.110 The thermae Falerianae are known from an inscription on a statue base, yet the name cannot be linked to archaeological remains.111 The thermae Olympiadis and the thermae Salusti or Salustianae are known from Early Medieval Passiones and itineraries.112 The smaller balnea known by text are numerous and none have been linked to archaeological remains (see the entry ‘balneum’ in the LTUR 1 for an overview). It is interesting to notice that the names of most of these baths seem to refer to the proprietor (or the initial patron), e.g. 109  Even if we assume that all the balnea of the notitia and the curiosum were open to the public (Foulché (2011) 609–610), there is no way of telling if these privately owned baths were stand-alone buildings or incorporated into the house. For this kind of semi-private, semi-public baths, see below, chapter 4, p. 207. 110  Tortorici (1994). 111   C IL 6.29806. See Di Stefano Manzella (1999) 58–59. 112  De Spirito (1999a) 62–63; De Spirito (1999b) 63–64; De Spirito (1999c) 63–64.

105

Archaeological Evidence

map 6

The number of balnea (B) and thermae (T) in each region of Rome according to the Curiosum / Notitia plan by the author after Fagan (1999a) 358, Map A3

the balneum Claudianum, the balneum Claudii Etrusci or the balneum L. Domiti Primig(eni), known from an inscription on a lead entrance-tessera.113 Baths in the Late Antique Urban Context The Forum Romanum and the Lower North Slope of the Palatine At the start of the 4th c., the monumental centre of the city was still restored and expanded. After a fire in AD 283, Diocletian undertook a large-scale renovation 113  B  alneum Claudianum: CIL 6.29767; balneum Claudii Etrusci: Mart. 6.42; Stat. Silv. 1.5; balneum M. Domiti Primigeni: Rostovtseff (1903–1905) 895.

program. On the forum Romanum, the basilica Iulia and the curia were restored to their former glory. A column was also erected in honour of the emperor.114 The portico south of the temple of Vespasian was restored in AD 367 by the praetorian prefect. East of the well of Iuturna, which was still in use, the statio aquarum (‘office of the aqueducts’) was embellished with statues, one dating from the reign of Constantine and one of AD 328.115 One of the largest buildings near the forum, the basilica nova, was constructed on the command of Maxentius, but finished under Constantine. The former also restored the 114  Giuliani and Verduchi (1995) 342. 115   C IL 6.37121; 36951.

106 temple of Venus and Rome and dedicated a small round temple to his prematurely deceased son Romulus.116 His successor and rival Constantine transformed it into a temple of Iupiter Stator.117 The house of the Vestal Virgins was in use until at least AD 380, the date of the last dedicatory inscription by the virgo vestalis maxima.118 After the ban on ‘pagan’ sacrifices by Theodosius (AD 394), the building was probably reused as offices for the imperial administration. The temple of Saturn was still repaired in the late 4th c.119 By the 5th c., we should assume that the classical temples had lost their original religious function. In fact, no attestations of restoration have been found. Nevertheless, they still very much shaped the visual landscape of the forum. The roof of the basilica Aemilia collapsed after a fire in the early 5th c., perhaps to be linked to the sack of AD 410, but there is no secure evidence that the building was restored afterwards.120 An inscription by an urban prefect records the restoration to the rostra after a battle against the Vandals in AD 470. At the eastern end of the Via Sacra, the so-called Baths of Elagabalus (C24) were constructed in the first half of the 5th c. The small baths (C26) located between the Temple of Antoninus and Faustina and the Temple of Romulus may also date from this period. These baths do not seem to have belonged to private houses. Both had street entrances, pointing to at least a partial public use. They may have belonged to a certain (mercantile) association that was active on the forum. The construction of a bathhouse near the forum had had a long tradition in the Roman empire.121 In the case of late antique Rome, these small baths seem to have been built on private initiative. During the 6th c., some Christian structures had found their way to the forum. The Church of Santa Maria Antiqua was built behind the Temple of Castor and Pollux.122 The rectangular structure behind the Temple of Romulus, actually belonging to the Temple of Peace, was transformed into the Church of San Cosmas e Damiano and incorporated the round temple as an entrance.123 At the end of the century, the ‘Baths of Elagabalus’ fell out of use. The House of the Vestals was still in use during the 6th c., although its specific function cannot be determined.124 116  Coarelli (1995c) 170–73; Ziemssen (2010). 117  Coarelli (2008) 107. 118  Coarelli (2008) 104. 119  Coarelli (1999) 235. 120  Lipps (2013) 115–16. 121  See the number of ‘Forum baths’ in Nielsen (1993b). 122  Zanotti (1996) 214–16. 123  Episcopo (1993) 324. 124  Johnson (2012) 110.

chapter 3

The last monument to be erected on the forum was a statue of Phocas (AD 608) placed upon on older column. The curia was transformed into the Church of Sant’Adriano.125 Similarly, part of the basilica Iulia was reused to construct the Church of Santa Maria in Cannapara. The Church of Santa Martina was commissioned by the archbishop Honorius I in AD 625. The Imperial Fora and the Temple of Peace Even if no bathhouses were found in this part of the city, the economic and political heart was still very much alive until at least the 5th c. The Forum of Caesar was restored by Diocletian and its tabernae again in the 5th c.126 The Forum of Trajan was still in pristine condition when Constantius II visited Rome and was still admired by Cassiodorus in the 6th c.127 The bibliotheca Ulpia was still used for recitals in the early 7th.128 The promulgation of laws continued until at least AD 451 and several statue bases dating from the first half of the 5th c. were also found.129 At the end of the 5th and certainly during the 6th c., the fora lost their original function, but still remained important symbolical centres.130 The Temple of Peace was out of use by the time of Procopius.131 The Palatine The Palatine remained the seat of power, even though Constantine moved the imperial court to Constantinople.132 His predecessor Maxentius had equipped the palace with a bathhouse overlooking the valley of the Circus Maximus.133 The small baths (C23) built inside a cryptoporticus along the Scala Caci reveal the continued activity in this area in the 4th c. From the 5th c. onwards, there are signs of abandonment.134 At a time when Rome was under threat, it is no surprise that an area which had mainly been used for the old classical religion and for the now absent imperial court was not a focus of restoration and reinvestment. The large sanctuaries fell out of use. The Temple of the Magna Mater and the Temple of Elagabalus were abandoned and reused as burial ground, while parts of the domus Tiberiana fell into decay.135 Theoderic, however, still restored parts 125  Coarelli (2008) 62. 126   C IL 6.1718; Morselli (1995) 301. 127  Cassiod. Var. 7.6.1. On the visit of Constantius II: Amm. Marc. 16.15.10. 128  Packer (1995) 349. 129  Coarelli (2008) 147. 130  Delogu (2000) 88; Coates-Stephens (2006) 300. 131  Procop. Bell. 6.21. 132  Papi (1999c) 35. 133  Manderscheid (2012). 134  Papi (1999c) 35. 135  Augenti (2000) 46.

Archaeological Evidence

of the imperial palace, while the Church claimed more control over the Palatine during the 6th and 7th c.136 An ecclesiastical complex from this period was probably equipped with a (private) bath as John VII took up residence on the Palatine and housed a part of the papal administration here during the 8th c. The Valley of the Colosseum and the Esquiline The Arch of Constantine near the Colosseum was erected by the Senate of Rome to commemorate the emperor’s victory over Maxentius at Milvian Bridge and his decennalia. It demonstrates the continued importance of the area, especially as a location to communicate imperial propaganda to the people.137 The Colosseum itself was restored in the 4th and 5th c. and venationes continued to be showed until at least AD 520.138 It is not surprising that the Baths of Titus and the Baths of Trajan were also kept in good order during the 4th and 5th c. During the 6th c. however, both bath complexes and the Colosseum fell out of use. At the end of the century, warehouses belonging to the Church had been constructed inside the amphitheatre.139 To the north of the Baths of Trajan, the rich domus on the Sette Sale reservoir had been equipped with a private bathhouse in the 4th c. From the time of Silvester (AD 314–335), the titulus Equiti (modern San Martino ai Monti) was located ‘iuxta termas Domitianas’ (sic).140 In the 5th c., the titulus Eudoxiae (modern San Pietro in Vincoli) was built west of the Baths of Trajan. East of the Colosseum, the titulus Sancti Clementi (modern San Clemente) is recorded in the literary sources from the late 4th c. onwards.141 It was built on top of a 3rd c. mithraeum. The titulus of Santi Marcellino e Pietro, towards the Lateran, is known from the reign of Siricius (AD 384–399). Under Symmachus (AD 498–514), the Church of Santa Lucia was built on the ruins of the porticus of Livia.142

107 on the forum Boarium under Constantine.144 The titulus Anastasiae already existed in the 4th c. and was restored in the 5th. The Ara Maxima, erroneously called the statio annonae, was restored in the 4th and 5th c., before being incorporated into the Church of Santa Maria in Cosmedin (6th c.). The spina of the Circus Maximus was equipped with a second obelisk under Constantius II (AD 357).145 In the 7th c., the Church of San Giorgio in Velabro was built under Leo II (AD 682–683).146 Regio I None of the 86 balnea or the two thermae mentioned in the Notitia Regionum have been found so far.147 For a regio with allegedly the highest number of baths, this is rather surprising. The fact that in the regio with the second highest number, Trastevere (XIV), there is the same discrepancy between archaeology and written sources could point to the modest size of these balnea and their limited chances of survival in the archaeological record. Just as Trastevere, regio I was an area of modest housing. The Baths of Caracalla, built on the fringes of this regio, were meant to fill in a thermae-void in a densely occupied quarter of the city. The titulus Crescentianae (modern San Sisto Vecchio) already existed by the end of the 4th c.148 The Church of San Giovanni near porta Latina was presumably built at the end of the 5th c.149

The Valley of the Circus Maximus and the Forum Boarium No bathhouses were found in regio XI, although the Notitia Regionum mentions 15 balnea.143 The area was still very popular in the 4th c., as a quadrifons was erected

Regio II The temple of Divus Claudius was still mentioned in the Notitia and probably continued to shape the summit of the Caelian hill for the centuries to come.150 The large domus in the neighbourhood continued to be inhabited in the 4th c.151 Some were even expanded and equipped with private baths (see above, p. 101–102). The titulus Pammachi (modern Santi Giovanni e Paolo) was built in the 4th c. upon a 2nd c. house with private baths. Under Constantine, the Castrum novum equitum singularium, housing the mounted guard of the emperor, was destroyed and replaced by the first ecclesiastical complex within the city of Rome. The location on the Lateran was not without symbolic significance, as the Basilica Salvatoris was built upon the barracks of the equites singulares who had sided with Maxentius. Other Christian

136  Augenti (1999) 39–40. 137  Coarelli (2008) 200–203. 138  E.g. CIL 6.1763; 32094; Cass. Var., 5.42. See Rea (1993) 31–32. 139  Rea (1993) 32. 140  L ib. Pont. 1.170. 141  Guidobaldi (1993) 278–79. 142  Panella (1999) 128. 143  This is the lowest number of bathhouses recorded for any of the 14 regiones in Rome. Note that we do not know the exact delimitations of the regiones described in the Notitia.

144  Coarelli (2008) 419. 145  According to Amm. Marc.17.4.12–16. 146  Coarelli (2008) 419. 147  The thermae mentioned in the Notitia must have been of a modest size. We should repeat that the term did not necessarily imply a large structure (see above, chapter 1). 148  Johnson (2012) 128–29 (with anterior bibliography). 149  Hansen (2015) 224–25. 150  A 13th c. papal bull still speaks of the clausuram Clodei (see Platner and Ashby (1929) 120–21). 151  Pavolini (2000) 147–48.

108 buildings such as martyria were built on the periphery, as the emperor probably did not want to upset the ‘pagan’ elite and plebs.152 During the construction works of the Basilica Salvatoris, a separate baptistery was constructed on top of an old bathhouse, reusing parts of the pools and building materials. Across the street, the Lateran baths may have been used throughout the 4th c. The titulus Aemilianae (now Quattro Santo Coronati) was created at the turn of the 4th and 5th c. The Church of Santo Stefano was erected in the second half of the 5th c.153 Regio V Constantine was credited with the construction of the basilica of Santa Croce, but the archaeological evidence seems to indicate a mid-4th c. date for the oldest remains.154 It is possible that the Constantinian phase only pertained to a small chapel that has not been identified.155 The imperial palace that perhaps once incorporated the amphitheatre was enlarged. Helena, mother of Constantine, restored the baths near the Porta Maggiore (hence the name Baths of Helena). Along the Aurelian Wall, north towards Porta Tiburtina, the so-called ‘Temple of Minerva Medica’ was possibly a palace built during the reign of Maxentius or Constantine. Just outside the walls, a late antique bathhouse was discovered under the modern porta San Lorenzo. A new bathhouse, perhaps belonging to a large house, was also constructed during the 4th c. in the area of the modern via Ariosto (C25). The baths under the piazza dei Cinquecento may have been out of use by the 4th c. Nevertheless, the Notitia still mentioned 75 balnea for regio V. The Christian influence in its urban fabric also became apparent with the Church of Santa Pudenziana (late 4th c.), built upon the remains of the Baths of Novatianus, the titulus Eusebii (5th c., modern Sant’Eusebio), Santa Bibiana (5th c.) and a church ‘iuxta Macellum Liviae’ under Liberius (AD 352–366), which would be enlarged to become the Basilica of Santa Maria Maggiore in the middle of the 5th c.156 Regio VI In AD 306, the largest thermae ever to be built in the empire, the Baths of Diocletian (C21), were inaugurated on a site formerly occupied by private residences and a temple precinct. The regio was obviously densely populated 152  Reekmans (1989) 865–66. 153  L ib. Pont. 1.249; Brandenburg (1999) 373–77. 154  L ib. Pont. 1.179. See Coarelli (2008) 272. 155  Episcopo (1996a) 28. 156  Cecchelli (1996) 217–18; Hansen (2015) 223.

chapter 3

during the 4th c., as the thermae of Constantine (C22) were built only some 750 m to the south of Diocletian’s Baths. The Notitia mentions 75 balnea in addition to these two thermae. The baths discovered along the modern via del Tritone may have been one of these, although it is unclear if they belonged to regio VI or VII (see above, p. 101). The Temple of Serapis was still mentioned in the Notitia. Close to the Baths of Diocletian, the titulus Cai (now Santa Susanna) was instated somewhere in the 4th c; the titulus Vestinae (modern San Vitale) also dates from the 4th c., but was restructured at the beginning of the 5th.157 After Alaric took the city in AD 410, his Gothic troops chose the Quirinal to set up camp and erected a small church, modern Sant’Agata dei Goti, the only Arian church foundation to have survived in Rome.158 Regio VII The Notitia mentions 75 balnea for this part of town, although possibly only one has been identified. The baths of the new Rinascente mall on the modern via del Tritone may have been in use in Late Antiquity (see above, p. 101). On the location of the modern Church of San Marcello al Corso, a titulus may have existed in the 5th c.159 The titulus Santi Apostolorum was inaugurated by Pelagius (AD 556–561) to commemorate the victory of Narses over the Goths.160 Regio IX The Notitia still mentioned the most important public buildings and temples. The Baths of Nero and the Baths of Agrippa were still in use, the latter probably until the late 5th c. (see above, p. 95–97). Excavations of the Crypta Balbi revealed how the public function of certain buildings gradually shifted towards a more industrial sphere.161 Except some tituli along the Via Lata, there seems to have been no important ecclesiastical buildings in the 4th and 5th c. The titulus Marci (modern San Marco Evangelista al Campidoglio) was presumable built at the foot of the Campidoglio hill in the early 4th c. The titulus Damasi (modern San Lorenzo in Damaso) was founded in the late 4th c. and expanded during the 5th. It was built over a mithraeum. In AD 609, Phocas gave the Pantheon to the Church.162

157  Milella (1996) 387–88; Zanotti (1995) 371. 158  Cartocci (1993) 24–25. 159  Episcopo (1996b) 211. 160  Lib. Pont. 1.303; Cecchelli (1999) 84–86. 161  Manacorda (1993) 328. 162  Titulus Marci: Reekmans (1989) 867; Titulus Damasi: Richardson (1992) 285.

Archaeological Evidence

Regio XII The mainly residential quarter was still known for its large domus in the 4th c. (see Notitia). One of these was transformed into a church during the second half of the 4th c. (modern Santa Balbina).163 The nearby Baths of Caracalla continued to service the area until the 6th c., when the aqueduct supplying it fell out of use and the building seems to have become ecclesiastical property (see above, p. 97). The site became a burial ground for a nearby xenodochium (modern Church of SS Nereus and Achileus). There are no traces of the 63 balnea mentioned in the Notitia. Regio XIII According to the Notitia, the most important temples on the Aventine still existed. The Baths of Decius were probably in use until the early 5th c., while the balneum Surae may have fallen out of use in the 4th c. The titulus of Aquila and Priscilla (modern Santa Prisca) dates from the 5th c. and was constructed on a mithraeum.164 The Church of Santa Sabina was constructed in the 5th c. over the ruins of a private house near the Temple of Iuno Regina.165 The significant deposition of amphorae on the Testaccio hill had ceased by the end of the 3rd c.166 Regio XIV and the Ager Vaticanus Little is known about the residential units and the commercial infrastructure such as horrea in the Trastevere regio during Late Antiquity. The titulus Callixtus (modern Santa Maria in Trastevere) was one of the oldest locations in Rome where Christians could celebrate Mass (4th c.).167 The titulus Caeciliae was built over private domus in the 5th c.168 The titulus Santi Chrysogoni (modern San Crisogono) also dated back to the 4th c.169 In the valley of the Vatican, the first basilica of St. Peter was constructed in the 4th c. on the location were the apostle himself was presumed to be buried.170 From the 86 balnea mentioned in the Notitia, not one is known through archaeological remains.

163  Episcopo (1993a) 155. 164  Zanotti (1999) 162–63. 165  Episcopo (1999) 221–23. 166   Rodríguez-Almeida (1984) 118. 167  Pronti (1996) 219–20. 168  Parmegiani and Pronti (1993) 206–207. The identification of a hypocaust and wall heating near the baptistery of the titulus has often been used to corroborate the story of Caeciliae who had supposedly died in her bathhouse (Mulryan (2014) 33). However, the absence of pools makes the interpretation as a bathhouse problematic, as many late antique houses had heated rooms. 169  Reekmans (1989) 871. 170  Coarelli (2008) 484–85.

109 Late Antique Baths and the Water Supply The water distribution in ancient Rome was more abundant than it is today. Frontinus, 1st c. curator aquarum of Rome (superintendent of aqueducts), mentions 9 large aqueducts, several of which had separate branches with separate names. The reconstruction of each aqueduct and its branches within the city is problematic, as the archaeological remains are dispersed and difficult to attribute to the right name. Furthermore, according to Frontinus, the water of different aqueducts often arrived in large reservoirs and cisterns, from which the water was distributed to various parts of the city.171 Most of the large aqueducts still seem to have been in use in the 4th c.172 The Aqua Alexandrina, supplying water to the Baths of Nero and to the Field of Mars, was restored in the 4th, 5th and even 6th c.173 The Aqua Claudia, supplying large parts of the city, was restored under Maxentius, while its caption points were reinforced under Arcadius and Honorius.174 The Aqua Virgo, supplying the Baths of Agrippa among others, was restored under Constantine and today still supplies water to some of Rome’s landmarks, including the Trevi fountain, the ‘Barcacia’ of the Spanish Steps and the Fountain of the Four Rivers on piazza Navona.175 The Aqua Alsietina (formerly the Aqua Augusta), supplying mainly Trastevere, was in use in the 4th and 5th c., as can be deduced from the passages in the Notitia Regionum and Polemius Silvius. The latter author also mentions several other aqueducts: the Aqua Anena (or Ania) and the Aqua Damnata with an unknown distribution, the Aqua Iulia with its wide distribution, the Aqua Marcia and the Aqua Traiana (mainly Trastvere). Procopius also recounts how the latter was cut by the Goths during the siege of AD 538, yet an inscription records the subsequent restoration by Belisarius.176 The restorations of the aqueducts under Byzantine rule is, however, problematic, as there are fewer literary sources and the repairs are more difficult to discern.177 Yet, parts of the Aqua Marcia were still restored under Adrian I (AD 772– 795). An Aqua Drusia, possibly part of the Anio Novus, was also mentioned in the 8th c. De montibus et aquis Urbis Romae.178 171  The Aqua Iulia supplied water to regiones II, III, V, VI, VIII, X and XII. Some 17 castella distributed the waters to the different regiones (Cattalini (1993) 67). 172   Coates-Stephens (2003a) 165. 173  Caruso (1993a) 60–61. 174   C IL 9.4051. See Mari (1993) 63–64; Aicher (1995) 42. C IL 6.31564. La Pera (1993) 72–73. 175   176  Procop. Bell. 5.19; CIL 11.3298. 177   Coates-Stephens (2001a); Coates-Stephens (2003a) 172; CoatesStephens (2003b) 419. 178  Palombi (1993a) 66.

110 The water supply was a fundamental concern for the rulers in Rome, whether it was the emperor, the Senate, the military commander or later the pope. However, the continued use of most aqueducts during Late Antiquity, at least until the Gothic siege of the mid 6th c., does not necessarily mean that all aqueducts functioned all the time. As studies on the Aqua Antoniniana have shown, there were periods during which the water supply was interrupted. If such a scenario unfolded, the baths depending on these aqueducts may have fallen out of use (as recorded for the Septimian Baths, see chapter 2, SHA). Alternatively, they may have closed down certain ‘water-consuming’ parts such as the frigidarium with its large piscinae and natatio, as argued by DeLaine for the Baths of Caracalla (see above, p. 97n59). New branches could also have been constructed to counter any water shortage. This is what may have happened when the Aqua Mercurii was built, possibly to supply a bath which was still known in the 8th c. as the balneum Mercuri.179 Alternatively, the baths may have been temporarily out of use, waiting for the aqueduct to be repaired (e.g. EI28). It is difficult to demonstrate the direct link between the abandonment of a bathhouse and the decline of a specific aqueduct. For the large (imperial) baths supplied by a separate branch (e.g. the Baths of Caracalla, the Baths of Diocletian, the Septimian Baths), the destruction of this branch could definitely spell the end, even if continued use on a much smaller scale cannot be excluded. For the smaller baths, the abandonment of a certain aqueduct may have been less dramatic, as the water supply often depended on several aqueducts supplying one large reservoir. The destruction of these smaller branches and additional pipes, which are even harder to trace in the archaeological record, was probably much more problematic for these baths. Ostia Concise History of the Urban Fabric The foundation of Ostia is much debated, as the location of the earliest settlement is not known. The literary sources such as Livy attributed its foundation to the fourth king of Rome, Ancus Marcius (642–617 BC).180 However, the oldest archaeological evidence dates back to the 4th c. BC.181 Some small finds and some architectural elements from the 6th and 5th c. BC were not related to any structures, but could point to an earlier settlement. The first structural remains pertain to the 179  Palombi (1993b) 69. 180  Livy 1.33.9. 181  Pavolini (2006) 20.

chapter 3

so-called castrum, the old city centre that resembled the layout of a Roman military camp. The oldest nucleus of the city was located at the location of its imperial forum.182 In other words, the forum was the centre of civic life from the earliest period onwards. Its construction date has been placed in the 5th or more probably the 4th c. BC. Therefore, it seems that Ostia was founded as a coastal colony to defend the mouth of the Tiber. Between the 4th and the 2nd c. BC, Ostia had an important military role as a naval basis. With the diminishing threat of Carthage after the Second Punic War (218–201 BC), the harbour transformed into a commercial hub, especially for stockpiling grain for Rome.183 Around 130–100 BC, Ostia had its own magistrates. In the power struggle between Octavian and Marc Antony, Ostia sided with the future emperor, probably because he protected the city’s commercial ties with the west.184 During the Principate, Ostia enjoyed a time of peace and prosperity. The theatre, with the adjacent ‘piazzale delle corporazioni’ (square of the corporations), was built during the reign of Augustus.185 Several horrea along the eastern part of the decumanus maximus, such as the horreum of Artemis (V, XI, 8) and the large horreum (II, IX, 7), also date from the 1st c. AD. Under Tiberius, the aqueduct bringing water from the Monti di Acilia was constructed. The enormous cisterns under the later Baths of Neptune date from the reign of the same emperor. They were partially fed by ground water and mainly by the aqueduct.186 The oldest known public baths (terme delle provincie, see below, p. 113) were constructed right next to it. Another bathhouse, in the insula of the Invidious (see below, p. 113), was constructed somewhat later and confirms that the water supply of the city was augmented. Furthermore, the urban expansion did not stop at the Sullan walls, with rich domus, such as the domus fulminata (III, VII, 3) of the 1st c. AD, found outside the Porta Marina.187 Several emperors undertook major building programs, the most important of which were carried out under Domitian, Trajan, Hadrian and the Severans. Under Domitian, large portions of the city were raised by almost 1 m to enable the construction of large multistorey apartment blocks.188 From the reign of Claudius onwards, a new harbour town called Portus was created directly to the north of Ostia, taking over much of its commercial and distributive functions. Ostia, however, 182  Calza (1953) 63–77 and fig. 19 (location castrum). 183  Meiggs (1960) 27. 184  Pavolini (2006) 24. 185  Calza (1953) 116. 186  Ricciardi and Scrinari (1996a) 248–49. 187  Calza (1953) 121. 188  Pavolini (2006) 32–34.

111

Archaeological Evidence

map 7

General plan of Ostia with the location of baths (blue) and main roads (red) plan by the author after Pianta generale di Ostia, version 2014

remained an important trade centre and probably served as a residential town for the citizens running Portus.189 The majority of the insulae along the via della Foce were built under Trajan. Large horrea (I, XIX, 4) were also constructed in this quarter close to the harbour.190 This was linked to the important investments in the harbour facilities of Ostia and Portus. The quarters outside Porta Marina were also developed. In the city centre, the new curia west of the forum was one of the most important additions. The acqua traianea mentioned in an inscription was probably a new branch of the existing aqueduct, reaching the newly built neighbourhoods.191 Several new baths were also constructed, mainly privately-owned establishments (map 7). However, the thermae of Porta Marina (see below, p. 117) are an example of imperial benefaction. Under Hadrian, the city within the Sullan walls witnessed its largest expansion. The forum was reorganized, with a new temple—the Capitolium—built at its northern end with spacious porticoes along its sides. North of the forum, along the cardo maximus heading towards the Tiber, an entire new neighbourhood was developed, including apartment blocks and horrea. The south-west part of town was transformed into a residential area, the ‘case a Giardino’ (‘houses with gardens’; III, IX) being an example of such a new housing project. In

all regiones, large multi-storey apartment blocks with tabernae on the ground floor were erected.192 The different collegia built their own assembly halls (scholae) and the rich elite invested in the construction of bathhouses (see below, p. 130). The construction of the city’s second imperial thermae, the Baths of Neptune (see below, p. 119), was also started. Under Antoninus Pius, the urban programme was brought to completion. Large apartment blocks, such as the so-called ‘palazzo imperiale’, were expanded and embellished and some new scholae were constructed.193 The most conspicuous addition to the city-centre’s public buildings were without doubt the Baths of the Forum, Ostia’s third thermae (see below, p. 124–125). The diffusion of oriental cult sanctuaries, especially mithraea, also started around AD 160. At the end of the 2nd c., Commodus seems to have paid special attention to Ostia. Large horrea were built near the Porta Romana and the theatre was restored. The city was even renamed Colonia Felix Commodiana for a short period of time.194 Under the Severan dynasty, its commercial infrastructure was kept in good use and even expanded. The piazzale delle corporazioni was reorganized and new scholae were erected. Several bathhouses, both publicly and privately owned, were

189  Meiggs (1960) 90. 190  Calza (1953) 123–24. 191   C IL 14.4326. Ricciardi and Scrinari (1996a) 253.

192  Pavolini (2006) 34–35. 193  Pavolini (2006) 36. 194  Pavolini (2006) 36.

112 restored (see below). The newly laid out Via Severiana points to the importance of the coastal zone and the connection with other harbour towns north and south of Ostia. During the 3rd c., building activities in the city centre seem to have come to a halt. The so-called ‘round temple’ was the last large public monument built within the Sullan nucleus of the city. During the second half of the 3rd c., some central horrea and large insulae were abandoned. The Baths of the Swimmer (see below, p. 115) and the caserma dei vigili (station of the city’s watch men) fell out of use and the piazzale delle corporazioni probably lost its commercial function.195 Even if these changes seem to point to an ‘urban crisis’ in the 3rd c., we must remember that large parts of Ostia’s suburban quarters remain unexplored. The decay of the old centre may have been the result of a shift in focus towards the periphery. At the beginning of the Tetrarchy, renewed building activity (or interest?) in the city centre put dilapidated buildings to new use, e.g. the Baths of the Philosopher (see below, p. 126–128).

chapter 3

Newly Built Late Antique Baths Most of the newly built late antique baths seem to date from the 4th c. (C7, C8?, C9, C10?, C11, C12?), a period of continued prosperity in Ostia, especially under Constantine and his successors. With the exception of the Byzantine Baths (C13), the new baths were rather small, with a total surface not exceeding 500 m2. With some of the larger baths in Ostia still in use until the 5th c. (see below) perhaps there was no real need for large and expensive thermae. Alternatively, a new large bathhouse may have been built in a new and as yet unexcavated part of town. It is also difficult to identify the patron who commissioned the construction. In the absence of building inscriptions, the stamps on lead pipes can sometimes name the patron. The water supply of the small baths west of the ‘palazzo imperiale’ (C7) seems to have been paid for by the praefectus urbi and his wife. However, we cannot conclude that the pipes belonged to the original construction phase, and thus if the bath itself was commissioned by the same people. Decorative elements such as mosaics can also hint at the ownership of the bath. The Baths of Musiciolus (C9), with its mosaic depicting athletes and their trainer, may have belonged to some type of organisation or collegium. Most of the new baths were constructed within a preexisting building, reusing the former walls in an obvious effort to save on construction materials and man hours (C8, C9, C10, C11, C12?, C13). The availability of water supply, e.g. the well near the Baths in the Horreum (C8) or in the Baths of Musiciolus (C9), could have been an extra

advantage. The insertion of baths within older buildings could point to a pressure on the intra-urban building space, with some quarters being especially attractive for commercial purposes, such as the tract along the Via Severiana (see below, p. 135–137). The limited building plot, defined by the pre-existing buildings, also limited the layout of the late antique baths (C8, C9, C10, C11, C12?, C13). All the basic bathing rooms had to fit within a predefined space, often prompting the architects to opt for simple linear or angular schemes with basic rectangular rooms. It should be noted, however, that there was a persisting preference for semicircular and round forms, such as apses and pools, even if this was not the most suited form to fit within a limited (and often rectangular) space (e.g. the east alveus of C8, the large piscina of C9). The heated rooms with adjacent service corridors would have been located near a street or alley to facilitate fuel supply (C7, C8, C9, C10, C11). It is important to note that there are always several heated rooms, even if the available building plot was limited. Most of the decorative schemes in the new baths made use of marble. Both in the cold and heated rooms, the floors, the inside of the pools and the lower parts of the walls were covered with marble slabs. Most of this veneer was lost in later phases, making it difficult to determine whether these slabs had already been reclaimed (spolia). Given the widespread practice of spoliation during antiquity, it is highly likely that a lot of the marble slabs used to embellish the late antique baths were spolia of older buildings. The mosaics were either geometric in design (C13) or depicted marine fauna and athletes (C9). For the former, large tesserae of precious stone (serpentine, marble) were used to create colourful patterns imitating opus sectile (see Gazetteer, fig. 63). The use of moulded stucco and paintings for the upper parts of the wall cannot be verified, as none of the late antique baths were preserved to a sufficient height and none of the available excavation reports mentions these forms of decoration. The heating systems of the new baths still relied on the traditional hypocaust system, consisting of square pillars made out of superimposed terracotta tiles, covered with large bipedales and topped with layers of hydraulic mortar. The walls were still heated by tubuli, although the dimensions of these seem to change. Whereas the tubuli of the High Empire measure (internally) 9–10 cm (width) by 5–6 cm (depth) by 33–34 cm (height), a new type of broader box tile was introduced somewhere in the 4th c.196 The broader type measures 11–13 × 7–8 × 37–38 cm and was used in baths dating to the 4th c. or later, and in baths that have a 4th or 5th c. restoration.

195  Pavolini (2006) 62.

196  Maréchal (forthcoming).

113

Archaeological Evidence

The number of furnaces in the caldarium is often equal to the number of alvei. Separate furnaces used only to heat the room are rare, the only case being the caldarium of the Baths at the Via Marciana-Via Severiana junction. The restricted size of the heated rooms only necessitated a single furnace working on a limited fuel supply. This observation brings us back to the hypothesis that the heated rooms may have been reduced in size to keep down fuel costs. As most of the newly built baths seem to have been privately owned, a small heated section with a low fuel cost may have been a deliberate choice to keep a lid on the expenses. This does not mean, however, that keeping down the fuel costs was a deliberate strategy to maximize profits. In contrast to our modern economic perception, the primary purpose of running a bathhouse was not to make profits, but acquiring the esteem of fellow citizens. The low admission fees attested in literature and inscriptions support this idea.197 Keeping the baths breaking-even would have made a bathhouse very ‘successful’ in the eyes of the proprietor. Some of the small alvei seem to have been heated by a type of testudo. The holes in their floors point to the phenomenon called ‘Beckenausbruch’ in modern research: the metal semi-cylindrical containers were torn out to be recuperated. However, for these small alvei, the small brick-built parallel walls that lined the furnace and supported the testudo are absent. Perhaps a type of semitestudo might have been used, consisting of only a metal plate in the bottom of the pool, lying directly above the furnace.198 The water management followed the same strategies as in previous centuries. Larger baths relied heavily on the aqueduct, often in combination with stored rainwater or ground water. In late antique baths, water reservoirs were not part of the complex and lay some distance away (e.g. C11, C13, C14).199 A decrease in the water supply does not seem to have been a problem for Ostia (see below, p. 130), as frigidaria were often equipped with large piscinae. The small single-person pools that are often found in the frigidaria of late antique baths must have had a special use, as they are constructed next to the larger piscinae for communal use (see below, chapter 4). If we look at the baths of the High Empire, we notice that such small piscinae were added from the 3rd or 4th c. onwards (e.g. the Baths of Neptune, the Baths of the Coachmen, the Baths of the Invidious, the Baths of the Trinacria, the Baths of Mithras). The size of the alvei is also restricted. This is in line with the general decrease in size of heated rooms. Alvei are often found in pairs in the 197  Nielsen (1993a) 131–35. 198  Maréchal (2017) 182–84. 199  Ricciardi and Scrinari (1996a) 144.

caldarium, with at least one of them being semicircular. Their narrow steps could not be used as a bench, suggesting they were used for quick dips rather than to linger in. Besides the decrease in the size of the hot rooms, the preference for curves and semicircular forms has often been mentioned as one of the most characteristic elements of late antique baths. The waste water disposal was ensured by the sewer network running underneath Ostia’s main streets. In most of the baths, the sewage channels of the baths seem to connect to these street sewers (e.g. C10, C11, C13?, C14). For some of the small pools in the heated rooms, no outlet could be identified (e.g. C9, C10, C11, C13, C14). The fact that the bottom of many of these pools were damaged when the testudo or semi-testudo was torn out probably masks the type of drain that once enabled the emptying of the pools. Perhaps a tap in the (semi-)testudo would have acted as an outlet, having the additional advantage that it could clean away the remaining ashes.200 Continued Use and Abandonment of Existing Baths The Baths of the Provinces The oldest bathhouse found in Ostia to this day dates to the reign of Claudius.201 The remains found under the via dei Vigili only comprised parts of the cold room. A splendid black and white mosaic with an emblema of 4 dolphins and with representations of 8 provinces gave the structure its name. The large reservoir (II, 4, 2) under the palaestra of the Baths of Neptune was probably constructed to supply water to these baths and other public and commercial facilities in the neighbourhood of the harbour.202 The baths were abandoned during the reorganisations of Domitian.203 The Baths of the Invidious (V, V, 2) The Baths of the Invidious (fig. 15) have been dated to the middle of the 1st c. on the basis of the opus reticulatum used to construct the walls, while tile stamps attest important restorations during the reign of Antoninus Pius. The mosaics are typical for the first half of the 3rd c.204 As no excavation report was ever published, it is impossible to discover when the baths fell out of use. However, the small piscina in the form of a quarter-circle cuts through the 3rd c. mosaic and hence seems a later addition (fig. 16). The reorganisation of the caldarium (3 on fig. 15), during which the eastern alveus

200  Maréchal (2017) 185. 201  Bloch (1953) 220. 202  Bukowiecki et al. (2008) 69. 203  Pavolini (2006) 61. 204  Pavolini (2006) 222.

114

chapter 3

figure 15 Plan of the Baths of the Invidious in Ostia (Italy) after Calza et al. (1953) pl. I

figure 16 Baths of the Invidious in Ostia (Italy), small pool in the frigidarium seen from the west photo: author, April 2014

115

Archaeological Evidence

figure 17 Plan of the Baths of the Swimmer in Ostia (Italy) after Medri (2013) 61, fig. 1.23

was sacrificed for an apsidal wall, may also belong to this last phase. The Baths of the Swimmer (‘Terme del Nuotatore’; V, X, 3) The Baths of the Swimmer (fig. 17) are the only baths in Ostia that have been excavated systematically. The majority of the work was carried out in the 1960s and 1970s and the results have been published in a number of Ostia-volumes and several articles.205 The baths were built in the late Flavian period (AD 80–90) and subsequently altered under Hadrian and Antoninus Pius. However, the baths fell out use in AD 230–250 when the entire neighbourhood seems to have been in decline.206 It is interesting to notice that during the second restructuration phase (AD 120–160), a small heated basin was added to the tepidarium (fig. 17). During the third rebuilding phase (AD 160–170), a small pool was added to the frigidarium (fig. 18).207

205  The final volume is Panella and Rizzo (2014), with anterior bibliography. 206  Medri (2013) 42–63, 66–68. 207  Di Cola (2013) 160 (tepidarium pool), 166–67 (frigidarium pool).

The Baths of Buticosus (I, XIV, 8) On the basis of tile stamps, the construction of the baths can be dated to the reign of Trajan, while a first restoration seems to have taken place in the middle of the 2nd c. Furthermore, the black and white mosaic representing a certain Epictetus Buticosus can be dated to the beginning of the 2nd c. Wall paintings with garden scenes are dated to the 3rd c., but no evidence of later use was found. However, the neighbourhood was still very much alive in the 4th c., as the famous domus of Amor and Psyche (I, XIV, 5) was built in this century and was still in use during the 5th.208 The Baths of the Six Columns (IV, V, 11) The baths (fig. 19) can be dated to the reign of Trajan by tile stamps and seem to have been built upon earlier houses. Some restorations took place under Antoninus Pius.209 The excavations of the 1930s and 1940s did not result in any publication, although no decorative evidence points to a late antique use of the building. In

208  Following the chronology in Pavolini (2006) 123–24. 209  Bloch (1953) 226; Mar (1990) 62.

116

chapter 3

figure 18 Baths of the Swimmer in Ostia, small pool of the frigidarium seen from the east, scale is 50 cm photo: author, March 2015

the hall with the 6 columns, a thin, poorly constructed wall delimited an area between the tepidarium and the western row of columns (fig. 20). In the north-west corner, a recess was hewn out of the wall to lodge 8 rows of tubuli belonging to the type (int. dim.: 7 × 12 × 38 cm) that can only be found in baths with a late antique phase of use.210 These tubuli could point to a 4th c. phase. Furthermore, the construction of the ‘heat lock’ in the hall is reminiscent of similar late antique additions in the Baths of Neptune and the Baths of the Lighthouse (see below, p.119 and p.126). It seems that the doorway between the heat lock and the tepidarium was also pierced at this time. The Baths of the Christian Basilica (III, I, 3) The baths are dated to the Trajanic period, when most of the neighbourhood was built in a recognizable opus testaceum. The building remained in use throughout the 2nd c. and was embellished with mosaics in the 3rd. At the end of the 4th c. or in the early 5th, the heated section was destroyed when a Christian building of unknown function, wrongfully called a Christian basilica in earlier research, was built.211 It is unlikely that the baths remained fully operational after the construction of the ‘basilica’, as the heated rooms were all destroyed.

figure 19 Plan of the Baths of the Six Columns in Ostia (Italy) with the small heat lock in (1) after Nielsen (1993b) 92, fig. 64

210  Maréchal (forthcoming). 211  Following the chronology in Heres (1980) 88, 94–95.

117

Archaeological Evidence

figure 20 Baths of the Six Columns in Ostia (Italy), heat lock seen from the north photo: author, March 2015

However, the continued use of the cold section of the baths cannot be ruled out. The Baths of Porta Marina (‘Terme della Marciana’; IV, X, 1) The large bathhouse built near the ancient shore line can easily be labelled as an imperial thermae (fig. 21). It was probably built on the initiative of Trajan, but was only finished under his successor Hadrian. The tile stamps, found out of context, are all of Trajanic and Hadrianic manufacture.212 An inscription recorded the restoration of the thermas maritimas, the original Latin name of the complex, in AD 375–378 (EI-18). The large thermae were still in use during the 6th c., as tile stamps dating from the reign of Theoderic were found.213 The nature of these interventions is difficult to interpret, especially as a thorough investigation of the remains is still a desideratum. The polychrome marble mosaic floor of the frigidarium dates to Late Antiquity, most probably to the renovations in the 4th c.214 It was made with large 212  Bloch (1953) 226. We should remember, however, that bricks and tiles may have been used long after their manufacture date, due to the large stocks that were sometimes acquired for large-scale building projects. 213  Pavolini (2006) 180. 214  Olevano and Russo (2001) 566; Pensabene (2007) 228–32.

tesserae of precious stone (marble, serpentine), just as in the Byzantine Baths (see C13). In the caldarium, the westernmost alveus changed its form, from a rectangular to an apsidal shape, possibly during the 4th c. restorations (in 3 on fig. 21). During the last phase of use of the caldarium, the tubuli were of the broad type. Several marble statues, including examples of Trajan’s sister Marciana and of Hadrian’s wife Sabina, were found in and around the baths.215 The Baths of the Coachmen (‘Terme dei Cisiarii’; II, II, 3) The Baths of the Coachmen (fig. 22) were built during the Hadrianic period on top of an abandoned commercial complex. Trenches made under the frigidarium confirmed the construction date that had been proposed on the basis of the mosaics. Tile stamps reveal that the complex was restored in the second half of the 3rd c.216 Small excavations in the 1970s revealed that there were at least 4 different phases of use, judging from the 4 floor levels in the frigidarium.217 In the last phase, the black and white mosaic that can still be seen today was repaired 215  Pensabene (2007) 226. 216  Carinci (1974) 563. For the tile stamps: Bloch (1953) 219. 217  Archivio Fotografico di Ostia, Inv. 12008, R1429.

118

figure 21 Plan of the Baths of Porta Marina in Ostia (Italy) after Nielsen (1993b) 67, fig. 94

chapter 3

119

Archaeological Evidence

figure 22 Plan of the Baths of the Coachmen in Ostia (Italy) after Calza et al. (1953) pl. I

with marble slabs. The small room (1) with an apsidal ending that connected the frigidarium (F) with the large tepidarium (2) was a later addition and is reminiscent of the similar room added to the hall of the columns in the Baths of the Six Columns (see above, p. 115–116). The small single-person pool in the north-east corner of the caldarium (3) was an even later element, as its walls were built in opus listatum (fig. 23). These baths were probably still in use during the 4th c. After the abandonment, two lime kilns were constructed within the building. The Baths of Neptune (II, IV, 2) The large thermae (fig. 24) were commissioned by Hadrian, but finished under his successor Antoninus Pius (AD 139). Apparently, the new emperor had to cough up an additional sum for the completion and the adornment (with marble) of the baths.218 Tile stamps confirm the general chronology and reveal that the caldarium was restored at the beginning of the 4th c.219 The excavations, carried out in the 1910s left no published records. Small-scale excavations in the 1960s and 1970s revealed 218   C IL 14.98. 219  Bloch (1953) 220–21.

the original layout of the cold rooms.220 The late antique modifications are difficult to discern. The mosaics of a large rest room north-east of the frigidarium depicted figures and symbols that may have had Christian connotations (fig. 25), perhaps proposing a late 3rd or early 4th c. date.221 Similarly, the abandonment of the original cruciform caldarium probably occurred during restoration works under Diocletian or Constantine. A smaller caldarium, with two new rectangular alvei, was constructed in the earlier sudatorium. The hypocaust and the tubuli were also renewed.222 The addition of a transit room between the frigidarium and the tepidarium is again reminiscent of similar rooms in the Baths of the Coachmen and the Baths of the Six Columns (see above, p. 117 and p. 115–116). A small oval basin lodged in the western doorway between the frigidarium and the hall with the mosaic of Neptune was also added at a late date (4th c.?).

220  Archivio Fotografico di Ostia, Inv. 11561, R982; Inv. 11562, R992. 221  Pavolini (2006) 60. 222  Heres (1979) 35–37.

120

chapter 3

figure 23 Baths of the Coachmen in Ostia (Italy), small basin in the caldarium seen from the south photo: author, March 2014

The Baths under the Forum of the Heroic Statue (I, XII, 2) The baths, of which little is known, were built in the Hadrianic period and reorganized during the Severan period.223 Recent excavations under the forum revealed that the baths were probably destroyed by fire at the end of the 3rd or the early 4th c.224 The bathhouse was then razed to the ground and filled in with the destruction debris. The layer was dated by ceramics and coins to the early 4th c.225 A second bathhouse, probably independent of the one described above, may have existed north of the Forum of the Heroic Statue. One of the rooms, which can still be identified in the cassegatio della cisterna (I, XII, 4), was octagonal in plan with 4 semicircular niches in its corners. The Baths of Mithras (I, XVII, 2) The baths (fig. 26), named after a statue of Mithras in the mithraeum found in the service corridors west of the complex, were built during the reign of Hadrian and 223  Lavan (2012) 668; quoting an oral communication by excavator A. Marinucci. 224  Gering (2014) §16. 225  Lavan (2012) 671.

restored in Severan times using a specific type of wall construction and Tuscan marble capitals.226 The plan was altered and the original caldarium fell out of use in the early 4th c.227 The original tepidarium was equipped with a large semicircular pool and a small single-person pool. The large piscina was probably filled in and replaced by a smaller square pool on the opposite side of the room. In the late 4th c., the northern part of the baths was transformed in a Christian building, the function of which is still debated.228 The northernmost room was equipped with a large apse in opus listatum and another semicircular construction, perhaps a presbyterium, in its centre. Christian monograms on the pillars have inspired researchers to interpret the building as a church (fig. 27), with the small unheated basins of the former baths used as baptisteries.229 However, the Christian decorative elements do not necessarily point to an ecclesiastical building. Symbols such as the Chi-Rho were often used as expressions of imperial power. The fact that 226  Bloch (1953) 219 (on the basis of tile stamps); Nielsen and Schiøler (1980) 151, esp. n. 17. 227  Heres (1982) 433. 228  Nielsen and Schiøler (1980) 152; Heres (1982) 433. 229  Nielsen and Schiøler (1980) 152; Mar (1990) 41; Pavolini (2006) 126.

121

Archaeological Evidence

figure 24 Plan of the Baths of Neptune in Ostia (Italy), with the 4th c. walls highlighted in red after Calza et al. (1953) pl. I

the northern apse cut off a public street may also point to an intervention (authorized) by civic authorities. The blocking off of (minor) public streets could also point to an insula in the hands of a rich owner and demarcated his ‘property’.230 The ‘Christianisation’ of part of the building does not imply a subsequent abandonment of the baths, as the Church often possessed bathhouses.231 The literary evidence from the Liber Pontificalis in Rome can be mentioned in this context (see above, p. 102).

The Baths of the Trinacria (III, XVI, 7) The baths have been dated by tile stamps to the early reign of Hadrian (AD 120), when the entire quarter of the city was built. During the Antonine period, the hypocaust system was restored and the mosaics of the heated part were laid.232 The neighbourhood was still inhabited in the 4th c., as the nearby Serapeum and the House of Bacchus and Arianna were reorganized at this time.233 Evidence for a 4th c. phase of the baths is difficult to identify. Some doorways were blocked up by a type of rubble concrete, while the mosaic floor was repaired

230  Gering (2010) 97. 231  DeLaine (2006) 340.

232  Bloch (1953) 225; Pavolini (2006) 136. 233  Pavolini (2006) 133.

122

chapter 3

figure 25 Drawings of the late antique mosaics in the Baths of Neptune in Ostia (Italy) after Becatti (1961) fig. 17

with large tesserae of the type used in the Byzantine Baths. The small square piscina in the frigidarium was constructed by blocking off a doorway or a niche with opus listatum (fig. 28). A niche above the western alveus of the caldarium also made use of opus listatum. A kiln was constructed at a late date in the southern water reservoir.

The Baths of the Seven Sages (III, X, 2) The structure in which the baths were lodged was constructed slightly later than the adjacent building of Serapis during the reign of Hadrian.234 The original plan of the baths in this period is unknown, except for the 234  Bloch (1953) 224; on the basis of tile stamps.

Archaeological Evidence

123 probably coincided with the transition to a new and powerful owner.236 The mosaics and the wall paintings of the bathing rooms confirm this chronology.237 Only a small latrine in opus listatum to the north-east of the round frigidarium seems to have been added at a late date (3rd c.?).238 Even if these baths were constructed within a commercial or residential building, the access may not have been restricted to inhabitants as the baths were accessible from public roads both to the north and south.239 The Baths on the Via di Iside (IV, V, 6) The complex was excavated in the first half of the 20th c., but the results still await publication. A re-examination of the remains in 2007 by Alfredo Marinucci was unfortunately not followed by a publication. The type of opus testaceum could point to an Antonine construction phase. Without any excavation data, it is impossible to discover when the baths fell out of use. The Baths by the Sullan Wall (Former ‘Terme Maritime’; III, VIII, 2) Small-scale excavations between 1968 and 1971 revised the Severan construction date proposed by the excavations at the beginning of the 20th c. The baths were probably built in Hadrianic times and remained in use until the 5th c. Tile stamps suggest that important restoration works took place around AD 210 in the western section of the baths.240 The heated rooms were also reorganized.241 Wall paintings suggest new restorations as late as the 4th c.242 Some opus listatum walls can be seen in the cold section. After the abandonment of the baths in the 5th or 6th c., two kilns were constructed inside the baths and several burials were also found in the praefurnium.243 The Baths of the Palazzo Imperiale The ‘palazzo imperiale’, a large apartment block that was mistakenly identified as an imperial palace by the excavators in the 1850s, dates from the Hadrianic period. The bathhouse was added in AD 145–150 and can be

figure 26 Plan of the Baths of Mithras in Ostia (Italy) after Heres (1982) 428, fig. 77

cruciform caldarium to the south. It was abandoned during the Severan period, when a new suite of bathing rooms was constructed.235 The expansion of the bathing complex into the commercial space of the building 235  Heres (1992) 99.

236  Heres (1992) 108. The circular room, which was equipped with a pool in a later phase, may have been part of a commercial complex such as a macellum (Heres (1992) 91). 237  Heres (1992) 85; Pavolini (2006) 140. 238  Heres (1992) 103. 239  Mar (1990) 45. 240  Bloch (1953) 222; Veloccia Rinaldi (1974) 563. 241  Pavolini (2006) 170. 242  Veloccia Rinaldi (1974) 654. 243  The kiln in a room east of the southern piscina is documented in the Archivio Fotografico di Ostia, Inv. 11174, R 613–616. The burials can be seen on Inv. 11505, R 926 and Inv. 1538, A 1538.

124

chapter 3

figure 27 Baths of Mithras in Ostia (Italy), alleged church in the northern part of the baths seen from the south. Stele with ChiRho sign on foreground photo: author, March 2015

compared to the Baths of the Seven Sages.244 The baths are still under investigation, so questions of chronology are still unresolved. It seems that some restorations were carried out during the 3rd c. and perhaps later. There is evidence for the blocking up of doors and the reuse of marble slabs in the caldarium.245 The Forum Baths (Also ‘Terme di Gavio Massimo’; I, XII, 6) The Baths of the Forum are the largest and most luxurious thermae in Ostia (fig. 29). Their construction was been dated to the reign of Antoninus Pius by an inscription, which also mentions repair works under Marcus Aurelius.246 Tile stamps have further specified the construction date to AD 160 and also revealed that the baths were restored in the Severan period and in the 4th c. Several dedicatory inscriptions, some of which are still displayed on site, confirm the 4th c. phase.247 An inscription on an architrave on the entrance mentioned the praefectus annonae of Rome restoring the marble decoration in AD 385–389.248 Late antique interventions

figure 28 Baths of the Trinacria in Ostia (Italy), small pool in the frigidarium seen from the north. Opus listatum blocking a pre-existing niche or door photo: author, March 2014

244  Bloch (1953) 225–226. 245  Spurza (1999) 135–38. 246   C IL 14.376. 247  Bloch (1953) 217; Pensabene (2007) 273. 248   C IL 14.4718.

125

Archaeological Evidence

figure 29 Plan of the Forum Baths in Ostia (Italy) after Nielsen (1993b) 95, fig. 69

concern both the cold and heated rooms. The large northern piscina in the frigidarium was given an apsidal ending in the 4th c. (dated by tile stamps), while the marble floors were renewed by making use of spoliated slabs.249 The principal entrance on the via della forica was closed off.250 The bathers now accessed the baths through a vestibule in the south-western corner. The decoration of the caldarium and tepidarium 1 can be dated to the early 4th c. The figurative capitals of the tepidarium and the Corinthian capitals of the caldarium are especially characteristic for this period (fig. 30).251 The walls of the caldarium, the sudatorium and exittepidarium were heated by the broad type of tubuli. In the latter, recesses were cut out of the southern wall to lodge these box tiles (fig. 31). The last evidence of activity inside the bathhouse is dated by a dedicatory inscription of the 5th c.252 A recent re-examination of the palaestra has revealed continued building activity during the 4th 249  Bloch (1953) 218 (stamps); Pensabene (2007) 273 (spolia). 250  Marinucci and Cicerchia (1992) 138. 251  Pensabene (2007) 275. 252   C IL 14.5387.

and possibly the 5th c. During the first excavations, several statues were found, including the head of a 4th c. prefect. The multitude of statues dating from different periods was perhaps the result of collecting statues from different locations in the baths. It seems that the palaestra square was put to new use, more as an annex of the forum than as a sporting ground for the bathers.253 The Baths of Silenus (IV, IX, 1) The bathhouse, located along the Via Severiana and facing the sea, is still under investigation by the University of Bologna. The preliminary results point to a construction date in the second half of the 2nd c. The piscina was filled in during Late Antiquity and covered with a new opus sectile floor, which was in turn spoliated in the 7th or 8th c. The ceramics and coins found in the filling of the pool date to the early 4th c.254 Perhaps the baths were transformed into a private residence somewhere in the second half of the 4th c.

253  Lavan (2012) 656–64, 687. 254  David (2013) 706–707; David et al. (2014) §13.

126

chapter 3

figure 30 Forum Baths in Ostia (Italy), late antique Corinthian columns from the fourth heated room seen from the south photo: author, April 2014

The Baths of the Lighthouse (‘Terme del Faro’; IV, II, 1) The baths were part of an insula and were probably privately owned (fig. 32). The first phases of the building, according to tile stamps and inscriptions on lead fistulae, seem to date from the reign of Marcus Aurelius.255 The emperor’s daughter Cornificia may even have owned the baths for a period of time. The mosaics and wall paintings, however, point to an early 3rd c. phase (reign of Caracalla). The names of the members of the senatorial aristocracy, such as Valerius Faltonius Adelfius or Anicia Italica, incised on the plumbing, point to a continued use of the baths in the early 5th c.256 The late antique phases can be recognized in the blocking up of doors with opus listatum and the use of a broad type of tubuli in the sudatorium (3 on fig. 32). The rows of box tiles are separated by vertical sherds, just as could be seen in the late antique repairs of the tubuli in the Baths of the Forum (fig. 31). A remarkable feature of the plan is the small frigidarium (1) with its own small piscina and the heat lock separating the tepidarium from the frigidarium (F). The main entrance was flanked by a taberna. The Baths of the Philosopher (V, II, 7) The Baths of the Philosopher (fig. 33) lie in a neighbourhood that was particularly active in Late Antiquity. The figure 31 Forum Baths in Ostia (Italy), recess with tubuli in the south wall of the exit-tepidarium photo: author, March 2014

255  Bloch (1953) 226; Pavolini (2006) 206. 256   A E (1954) 180; Pavolini (2006) 206.

Archaeological Evidence

figure 32 Plan of the Baths of the Lighthouse in Ostia (Italy) after Nielsen (1993b) 94, fig. 68

figure 33 Plan of the Baths of the Philosopher in Ostia (Italy) after Nielsen (1993b) 96, fig. 71

127

128

chapter 3

figure 34 Baths of the Philosopher in Ostia (Italy), small pool at the beginning of the heated section seen from the east, scale is 50 cm photo: author, March 2015

adjacent House of the Fortuna Anonaria (V, II, 8) to the north was expanded and embellished in the 4th c., while the House of Protirus (V, II, 4–5) to the south was enlarged in the late 3rd c. and continued to be inhabited in the 5th.257 The baths were constructed inside Trajanic warehouses and shops and a Severan temple.258 When the latter was destroyed in the second half of the 3rd c., the bathhouse was constructed. The western part, however, opening onto the Semita dei Cippi street, retained its commercial function. The baths may have served a specific group, perhaps a collegium. The hypothesis that the baths belonged to a philosophic school is based solely on the find of a Neo-Platonic portrait in the courtyard of the baths.259 The plan has already some of the elements that will be characteristic for late antique baths: a prominent frigidarium with two different piscinae, a single-person pool at the start of the heated section (fig. 34), a limited number of (small) heated rooms and a preference for apsidal or semicircular pools. The oval piscina shows resemblances to the circular pool of the domus of the Dioscuri (see below). The steps to enter the

pool, just as in the semicircular pool to the north of it, are too narrow to be used as a bench (between 10 and 13 cm, measurements by the author). Even if the excavations of 1939 did not leave us any hard evidence for a continued use in the 4th c., it is very probable that the baths were still in use, at least at the beginning of the century.

257  Pavolini (2006) 215–19. 258  Boersma (1985) 120–26. 259  Boersma (1985) 137.

260  Subias Pascual (1991) 111–21. 261  Pavolini (2006) 166. 262  Subias Pascual (1991) 71.

Private Baths The Baths of the House of the Dioscuri There has been much debate concerning the domus of the Dioscuri in Ostia. Some scholars believe that the complex is not a domus, but rather the seat of a collegium.260 Others maintain that the complex is a large residence, which paid exceptional attention to the reception of guests.261 The presence of a sizeable bathhouse has been mentioned in favour of a non-residential interpretation. However, the fact that the baths were only accessible from inside the building points to a restricted, non-public use. Important reorganisations of the building are dated by the mosaics to the 4th c.262 According

129

Archaeological Evidence

figure 35 Plan of the baths on the Isola Sacra (‘Terme di Matidia’) near Ostia after Veloccia Rinaldi (1974) tav. III

to Thea Heres, the construction technique of the walls of the baths point to an early 5th c. date.263 Even if a detailed description is not necessary in the framework of this study, it is worth noting that the frigidarium was equipped with both a large apsidal (4 × 3.5 m; 1.2 m deep) and a small oval piscina (1.8 × 1.5 m; 1.2 m deep; measurements by Subias Pascual).264 The latter could only accommodate one person at a time and must have had a specific function (see below, chapter 4). The internal steps were too narrow (averaging 10 cm) to be used as a bench, so the pool could not have been ideal for sitting and lingering.

not be confirmed that it still functioned as a bathhouse. The quarter of the pons Matidiae was still occupied in the 7th c.267

Baths on the Isola Sacra The Isola Sacra is the stretch of land between Ostia and Portus, enclosed by a branch of the Tiber (north), the Tiber itself (east and south) and the sea (west). Excava­ tions between 1972 and 1974 uncovered a commercial zone around a Roman bridge crossing the Tiber.265

Baths in the Late Antique Urban Context At the beginning of the 4th c., the most important public buildings and temples were still in use. The theatre was restored, the large thermae and the smaller baths still functioned and the city was embellished with several new fountains, nymphaea, and exedrae. Some regiones seem to have been more active in Late Antiquity than others. The area north of the decumanus maximus, especially the multi-storey apartment blocks and the horrea, were abandoned by the early 4th c. In the area south of the forum and towards the porta marina and porta Laurentina, several large houses were built. Several (secondary) streets were blocked, perhaps pointing to some sort of ‘ghetto-building’ in which important patrons fenced off a specific quarter of the city from the public streets and cargo transport.269 The decisive parameters for these different patterns of continuity were most likely the presence of the important traffic arteries, social hubs and economic activity centres. Bathhouses were often built near important traffic arteries, such as the decumanus or the cardo maximus, as these roads ensured a good visibility for the throngs

The Baths of Matidia A bathhouse was identified in one of the insula that was aligned with the road leading up to the bridge (fig. 35). It seems that the oldest phase of the building dates to the 2nd c. (opus mixtum), although it is unclear whether it already functioned as a bathhouse at this time.266 Late antique elements in the building can nevertheless be discerned: the large tesserae mosaic of the side entrances of the main hall and the small pools added to the original plan in the heated section. The multi-coloured glass tesserae, some of them gilded, probably belonged to wall mosaics of this last phase. Small finds such as coins suggest the building was in use in the 6th c., although it can263  Heres (1982) 477–85, esp. 485. 264  See Subias Pascual (1991) 53–56; de Haan (2010) 150–52. 265  Veloccia Rinaldi (1975). 266  Pavolini (2006) 279; Pensabene (2007) 464, 540 contra Veloccia Rinaldi (1975) 33.

The Baths under the Basilica of San Ippolito Some 100 m south-east of the Baths of Matidia, a basilica for the martyr Hippolytus was constructed at the end of the 4th c. Excavations in the 1970s revealed the existence of an older bathhouse on the location of the later basilica. However, the baths had already been abandoned before construction work on the basilica started, as early 4th-c. graves were found within its walls.268

267  Veloccia Rinaldi (1975) 30–35. 268  Pani Ermini (1979) 245–46. 269  Gering (2010) 94.

130

map 8

chapter 3

Public baths in Ostia in relation to the main aqueducts (green), public spaces (red) and new Christian buildings (yellow) at the end of the 4th c. plan by the author after Pianta Generale di Ostia, version 2014

of people passing by (map 7). This strategy was adopted in imperial times and was continued in Late Antiquity. The decumanus maximus, connecting the porta Romana with the forum, was still the centre of civic investment, with new fountains and plazas built along its course. Public monuments, such as the theatre and the Baths of Neptune were restored in an effort to keep some of Ostia’s prestige alive. The same goes for the forum and the Forum Baths, which were at the core of Ostia’s public life. A similar strategy can be discerned for the other important roads. The baths along the via della Foce (Baths of Mithras, Baths of the Trinacria, Baths of the Seven Sages) were possibly kept in good use and a new bathhouse (Baths on the via della Foce, C14) was built. This is even clearer along the Via Severiana, where several small (privately owned) baths were constructed (Baths on the Via Severiana , C11, Baths on the Via Marciana C10, Baths of edificio b, Baths of Musiciolus C9). In this case, the late antique baths act as an indicator for the growing importance of the road. Encroachment upon streets (e.g. the Baths of Mithras, the Baths of the Forum) can be interpreted as a reorganization of the road network, as a sign of a weakened civic authority regulating public space, or as the result of wealthy citizens ‘buying up’ parts of the public space. A second important element in the construction or repair of a bathhouse, seems to be the water supply (map

8, 9). The presence of a branch of the aqueduct was important, which in its turn was linked to the importance of the road. The baths along the decumanus maximus (e.g. the Baths of Neptune) definitely benefitted from the water supply channel running underneath it. The fact that Ostia’s aqueduct survived into Late Antiquity certainly accounts for the survival of the thermae and the larger balnea. At the same time, large roads were also equipped with sewers for disposing of the waste waters of the baths. The proximity of other social hubs, such as the forum or the theatre, certainly played an important role in restoring old or constructing new baths (map 8, 9). In areas where large crowds of people could be expected, a bathhouse was always near (Baths of the Forum, Baths of Neptune near the theatre). In Ostia, the old city centre continued its central role into Late Antiquity. It would be interesting to discover new late antique focal points (churches, fora) and examine if new baths were constructed in the immediate vicinity (e.g. the Baths of Musciolus near the synagogue?). Regio I The area of the forum was still the centre of the political and social life in the 4th c. (map 10a). The most important bath complex in the neighbourhood, the sumptuous Forum Baths (or of Gavius Maximus), were repaired

Archaeological Evidence

map 9

131

Public baths in Ostia in relation to the main aqueducts (green), public spaces (red) and Christian buildings (yellow) at the end of the 5th c. plan by the author after Pianta Generale di Ostia, version 2014

and decorated with new marble elements. North of these imperial thermae and along the decumanus maximus, the so-called Forum of the Heroic Statue—probably a macellum—was created at end of the 3rd c.270 For its construction, two different bathhouses, one to the south and one to the north, were razed and filled up.271 To the east of it, a large exedra closed off the Semita dei Cippi road (see below, regio V). On the other side of the decumanus, a new semicircular nymphaeum was constructed.272 Some of the apartment blocks east of the forum were still inhabited, although most were abandoned at this time.273 To the west of the forum, a large trapezoidal nymphaeum was built at the start of the via della foce.274 The Baths of Buticosus may already have fallen out of use by the 4th c., even if there was still some building activity in the neighbourhood, with the sumptuous domus of Amor and Psyche built just a few metres to the north-west. The Republican temple of Hercules was even adorned with a new statue by the praefectus annonae in AD 392–394.275 Restorations were undertaken in the Baths of Mithras, adding a layer of Christian veneer to existing structures. 270  Gering (2014) §13. 271  Lavan (2012) 668. 272  Ricciardi and Scrinari (1996a) 194–95; Lavan (2012) 680–81. 273  Pavolini (2006) 90–91. 274  Ricciardi and Scrinari (1996a) 195–96; Lavan (2012) 681–85. 275  Pavolini (2006) 120.

It is not known whether the baths still functioned or if the building was put to a new use. Some 70 m to the west, the new baths on the via della foce (C14) were built, perhaps to fill in the void left by the decay of one of the other baths. The entire part of the via della foce between the baths and the nymphaeum was closed off for traffic at the beginning of the 4th c., and the street was subsequently transformed into a promenade with lavishly decorated shops and fountains.276 The baths of the ‘palazzo imperiale’ may have been abandoned at the end of the 3rd c. A small bathhouse discovered nearby (C7), however, points to a continued occupation of the area. Until recently, the literature on Roman Ostia stopped at the end of the 4th c.277 Subsequent phases were summarized as an agonizing death struggle of a city in decline. Even the ‘barbarian’ invaders, such as Alaric, did not even bother to conquer the city, preferring nearby Portus instead.278 Thanks to recent investigations of the forum area by a British-German team (the Kent-Berlin Late Antique Ostia project, 2008–2013), combining noninvasive techniques (georadar, laser scanning, aerial 276  Gering (2010) 104. 277  Calza (1953) 155–63; Meiggs (1960) 89–90; Pavolini (2006) 38–39. 278  Heather (1996) 148.

132

MAP 10A

chapter 3

Regio I in Ostia during the 4th c., with location of baths and other buildings that were newly built or still in use plan by the author after Ricciardi and Scrinari (1996b) 14, general plan 1: Forum 2: Forum of the Heroic Statue 3: Exedra 4: Round Temple 5: Domus of Amor and Psyche 6: Forum Baths 7: Baths of Buticosus 8: Baths of Mithras

photography) with detailed stratigraphic excavations, late antique Ostia’s city centre has been reinterpreted. Test trenches underneath the paving of the forum revealed that it was kept in use until the end of the 5th c.279 The forum and the Forum of the Heroic Statue were restored and embellished as late as the second half of the 5th c. (map 10b). The economic and representational functions of the forum at this late date make it likely that other public buildings in the area, such as the Baths of the Forum, were also kept in use. The first signs of spoliation on the forum and of the construction of poorly built wooden structures date to the end of the 5th c.280 Regio II The Baths of the Coachmen, just north of the nymphaeum near porta Romana, may still have been in use during the 4th c. (map 11a). The Baths of Neptune were restored at the beginning of the 4th c., while the shops in front of them, lining the decumanus, were also reorganized.281 279  Lavan (2012); Gering (2014); both with anterior bibliography. 280  Lavan (2012) 688. 281  Heres (1992) 40.

A nymphaeum was built on the south-west corner of the insula.282 The theatre was renovated in the 4th c., but the adjacent piazzale delle corporazioni had probably fallen out of use.283 At the end of the 4th or the beginning of the 5th c., a Christian oratorium was built on top of the eastern semicircular nymphaeum of the theatre (map 11a, 11b).284 It was dedicated to Ciriacus, a martyr who died in the 3rd c. The shrine, which later became a small church, remained a pilgrimage site for Christians until the 12th c.285 Regio III The area delimited by the via della foce to the north and the decumanus maximus to the east was essentially a residential quarter planned during the Trajanic and Hadrianic period (see above p. 111; map 12a). During the 4th c., several houses and insulae reveal restorations (opus listatum walls, mural paintings, decorative elements). There is no evidence that the Baths of the Seven Sages 282  Ricciardi and Scinari (1996a) 199. 283  Pavolini (2006) 68–71. 284  Boin (2013) 229. 285  Pavolini (2006) 67.

Archaeological Evidence

MAP 10B

Regio I in Ostia during the 5th c., with location of baths and other buildings that were newly built or still in use plan by the author after Ricciardi and Scrinari (1996b) 14, general plan

MAP 11A

Regio II in Ostia during the 4th c., with location of baths and other buildings that were newly built or still in use plan by the author after Ricciardi and Scrinari (1996b) 57, general plan 1: Piazzale delle Corporazioni 2: Theatre 3: Christian oratory 4: Baths of the Coachmen 5: Baths of the Provinces 6: Baths of Neptune

133

134

MAP 11B

chapter 3

Regio II in Ostia during the 5th c., with location of baths and other buildings that were newly built or still in use plan by the author after Ricciardi and Scrinari (1996b) 57, general plan

MAP 12A Regio III in Ostia during the 4th c., with location of baths and other buildings that were newly built or still in use plan by the author after Ricciardi and Scrinari (1996b) 57, general plan 1: ‘Christian Basilica’ 2: House of the Serapeum 3: Baths of the ‘Christian Basilica’ 4: Baths of the Seven Sages 5: Baths of the Trinacria 6: Baths near the Sullan Wall 7: House of the Dioscuri

135

Archaeological Evidence

map 12b Regio III in Ostia during the 5th c., with location of baths and other buildings that were newly built or still in use plan by the author after Ricciardi and Scrinari (1996b) 57, general plan

were still in use after the 3rd c. Similarly, no late antique phase can be attested for the surrounding apartment blocks.286 In the Baths of the Trinacria, some opus listatum walls point to a late antique phase. The houses just west of the baths were restored during the 4th c.287 The continued occupation of the insula, as well as the water supply from a branch of the aqueduct, could explain why a small bathhouse (C8) was constructed in a horreum in the 4th c. Late antique building activity seems more prominent along the decumanus maximus (see also below, regio IV). The commercial potential of this important road and the availability of water related infrastructure (branch of the aqueduct, sewer) explain the continued popularity of this area. Several large domus were remodelled here during the 4th and 5th c., e.g. the domus on the decumanus, the House of the Nymphaeum and the House of the Dioscuri (map 12b).288 The Baths of the Basilica Cristiana probably functioned until the second half of the 4th or the beginning of the 5th c., only to make way for a new

Christian building. More or less contemporaneously, the House of the Dioscuri was equipped with a private bathhouse. The baths near the Sullan Wall, provided with water by the aqueduct running on top of the latter, were kept in good use during the 4th and 5th c.289 The aqueduct not only supplied those last two bathhouses with water, but also a 4th c. private nymphaeum.290

286  Pavolini (2006) 133–35. 287  Gering (2010) 101–102. 288  Pavolini (2006) resp. 150, 166 and 169.

289  Ricciardi and Scrinari (1996b) 169. 290  Ricciardi and Scrinari (1996a) 216–17. 291  Poccardi (2001) 169; DeLaine (2006) 342.

Regio IV Outside porta Marina, the neighbourhood along the Via Severiana was very active in Late Antiquity (map 13a, 13b). The Baths of Porta Marina benefitted from imperial patronage until at least the reign of Theoderic (early 6th c.), while several other small baths (C9, C10, C11 and the baths of edificio b), probably privately owned, were constructed in the 4th or 5th c. It is not a coincidence that several baths were built along this important road, connecting Ostia’s seaside neighbourhood with other harbour towns to the south.291 We could even imagine a competition between the bath complexes. The location

136

map 13a

chapter 3

Regio IV in Ostia during the 4th c., with location of baths and other buildings that were newly built or still in use plan by the author after Ricciardi and Scrinari (1996b) 131, general plan 1: Synagogue 2: House of the Columns 3: Tabernae of the Fish-sellers 4: Baths of the Lighthouse 5: Baths of the Via di Iside 6: Baths of the Six Columns 7: Baths of Porta Marina 8: Baths of Silenus 9: Baths of ‘Edificio b’

probably ensured a steady clientele and perhaps even profits for the owners.292 Other buildings along this road were restored in the 4th c., such as the synagogue. From the latter, a road headed north-east towards porta Laurentina, where the late antique Baths of Perseus (C12) were found. Returning to the porta Marina and the decumanus maximus, several residential and commercial structures were kept in good use throughout the 4th c. A part of the Baths of the Six Columns may have 292  See also DeLaine (2006) 342: “the string of small baths on the Via Severiana recalls the stazioni balneari of modern Lido di Ostia and the amoenissima civitas of Minucius Felix.”

been in use. The Baths on the via di Iside were probably abandoned by the 4th c. The eastern part of regio IV, along the cardo maximus, saw the construction of an important bathhouse (Byzantine Baths, C13), two nymphaea and the restoration of latrines. Luxurious houses, such as the domus delle colonne or the domus di via della Caupona, were still inhabited and even embellished in the 4th and 5th c.293 The senatorial elite of Rome still invested in the Baths of the Lighthouse at the start of the 5th c., 293  Heres (1982) resp. 488–493 and 501–504; Pavolini (2006) resp. 201 and 205.

137

Archaeological Evidence

map 13b

Regio IV in Ostia during the 5th c., with location of baths and other buildings that were newly built or still in use plan by the author after Ricciardi and Scrinari (1996b) 131, general plan

which may have based its ‘successful survival’ into Late Antiquity on its prime location along an important traffic artery near a city gate.294 Regio V From the porta Laurentina, we can take the Semita dei Cippi to re-join the decumanus. Along this important commercial axis, several late antique houses were built (see already regio I; map 14a, 14b). The domus of Protirus was restored in the 5th c., while the House of the fortuna annonaria was occupied until the 6th.295 The Baths of 294  Stöger (2011) 91–92. Stöger also concluded that the bath’s spatial organisation—stressing a very functional organisation and offering high levels of privacy—was ‘instrumental in sustaining the baths’ long period of use’ (Stöger (2011) 167). However, we should remark that the internal organisation of this bath was not in any way exceptional, but followed a standard design, lodging the alvei in recesses to create a sense of privacy. 295  Heres (1982) 542–48; Boersma (1985); Pavolini (2006) resp. 215 and 219–20.

the Philosopher, lodged between the two aforementioned houses, were built at the end of the 3rd c. and were probably still in use during the 4th. The Baths of the Invidious were restored in the 3rd c., but probably abandoned during the 4th. The Semita dei Cippi was cut off from the decumanus by a late antique exedra (see above, regio I). Further east along the decumanus, several houses had a late antique occupation. The Baths of the Swimmer, as well as the surrounding houses, fell out of use during the 3rd c. The horreum of Hortensius, one of the largest and oldest granaries in the city, was still operational during the 4th c.296 From the 5th c. onwards, the building may have changed function, as was the case for the other horrea along the decumanus. Some 285 m south of the decumanus maximus, along the via del sabazeo, geophysical prospection with subsequent test trenches discovered a Christian basilica, built on top of earlier Hadrianic remains.297 It can probably 296  Heres (1982) 559–65. 297  Bauer and Heinzelmann (2001).

138

map 14a

chapter 3

Regio V in Ostia during the 4th c., with location of baths and other buildings that were newly built or still in use plan by the author after Ricciardi and Scrinari (1996b) 166, general plan 1: Nymphaeum of Porta Romana 2: Basilica of Constantine 3: Horrea of Hortensius 4: House of the Fortuna Annonaria 5: House of Protirus 6: Baths of the Swimmer 7: Baths of the Invidious 8: Baths of the Philosopher

be identified with the first basilica of the city donated by Constantine and recorded in the Liber Pontificalis.298 Near the porta Romana, a sumptuous nymphaeum was built at the turn of the 3rd and 4th c.299 The castellum aquae was still in use during the 4th c. and possibly later, as a part of the aqueduct with 5th c. restorations was discovered near the Constantinian basilica.300

298  L ib. Pont. 1.183–184. 299  Heres (1982) 566–69; Ricciardi and Scrinari (1996) 235–36; Lavan (2012) 685. 300  Bukowiecki et al. (2008) 187–88, 190.

Cuicul Concise History of the Urban Fabric The site of ancient Cuicul (Djemila) on the border of the provinces Numidia and Mauretania Caesariensis is a perfect example of a well-preserved city that has been unearthed in period before scientific excavations. The late antique phases of the city, considered less important by these early archaeologists, are poorly understood. The small city was probably founded as a military stronghold under Nerva, around AD 96–97, on the crossroads of the Sitifis-Cirta with the Igilgili-Lambaesis highways. As Cuicul was located on a small plateau (850 m a.s.l.)

139

Archaeological Evidence

map 14b

Regio V in Ostia during the 5th c., with location of baths and other buildings that were newly built or still in use plan by the author after Ricciardi and Scrinari (1996b) 166, general plan

between two gorges, the shape of the city was necessarily triangular. The local economy was based primarily on agriculture and its strategic location made Cuicul an important trade centre.301 The mountainous surroundings supplied plenty of building material (limestone) and water sources. A first bloom of the city took place under the Antonine dynasty (AD 138–192), with the development of the northern part of the city. It was centred around the so-called ‘old forum’, the large square in front of the Capitolium, dedicated between AD 150 and AD 170. Shortly afterwards the curia was constructed. Other important public buildings dating to this first golden era included the basilica and the market. The temple of Venus Genetrix, just south of the basilica, must also have been built in this period.302 The basic infrastructure of a city 301  Allais (1938) 9. 302  Following the urban evolution of Février (1964) 9.

was not completed without a decent bath complex. The Baths of the Capitolium (‘Thermes du Capitole’) were probably built before the Capitolium itself (see below, p. 142), while the town’s largest baths, the Large South Baths (‘Grands Thermes Sud’), are dated to AD 183–184.303 It is also in this period that the population was expanded with the local inhabitants of the region, as is deduced from the names on numerous epitaphs.304 The second bloom came under Septimius Severus, again accompanied by significant building activity. The town was further expanded to the south, with the construction of a new forum.305 It was dominated by the town’s most impressive temple, that of the gens Septimia (dated AD 229). Several statues in honour of the imperial family and leading citizens were dotted around the main 303  Thébert (2003) 194. 304  Février (1976) 249. 305  Leschi (1953) 12; Février (1976) 250.

140 streets and squares.306 To cope with the town’s growing population, the city council invested in new grain storages (horrea) to the north-east of the new forum. Cuicul in the 3rd c., traditionally labelled a period of ‘crisis’, is poorly understood due to the lack of building inscriptions. However, restorations took place from the Tetrarchy onwards. During the 4th c., several churches were built in the southern part of town. Under Vandal rule, a persecution of orthodox Christians was ordered in AD 484, although after the Byzantine conquest of North Africa, a bishop of Cuicul is mentioned in a synod at Constantinople in AD 553.307 The area around the new forum must have been occupied until the Arab phase. Several wells were dug in the centre of town to ensure a water supply. Newly Built Late Antique Baths During the 4th c., a period of prosperity and growth, several new neighbourhood baths were built (C39, C40, C41). The baths of the baptistery were constructed at the end of this century, as part of a larger ecclesiastical complex. The small external baths of the House of Castorius (C42) may date to the 5th c. or even later. Most of the newly constructed baths were built inside pre-existing buildings (C39, C40, C41), which explains their modest size and simple linear lay out. The frigidarium is the most important room and is often as large as the entire heated section combined.308 It also offered the most possibilities to develop decorative schemes. The size of the cold room may have been the result of the wish to retain the function of the large vestibules that preceded the frigidaria of the thermae (see esp. the basilica thermarum of the large baths).309 Because of the restricted building space, the functions of this vestibule and the frigidarium had to be combined, resulting in one large cold room. The presence of two separate cold pools is also striking, even if the available space was limited (C41, C42). This may be reminiscent of the (lateral) piscinae of the large thermae commonly found in North Africa,310 or it could be that the pools had a different function. In fact, one pool is often smaller than the other. This is most clear in the Baths of the Baptistery (fig. 93), the Baths of the House of the Donkey (C41) and the external Baths of the House of Castorius (C42). The 306  Février (1976) 250. 307  Allais (1938) 31. 308  The importance of the cold rooms seems to have been a typical feature of baths in North Africa, already during the High Empire, see Nielsen (1993a) 91; Yegül (1992) 405. 309  Yegül (1992) 404. 310  Yegül (1992) 405; Nielsen (1993a) 89; Thébert (2003) 310 and 323.

chapter 3

pools—or rather tubs—were only large enough for one person. We could imagine that they were used for medical purposes or for a personal wash. The niches above the small cold pool in the House of the Donkey may have been used to store ointments, oils or additives for a special bathing / washing experience. The small pool in the first of the heated rooms in the Baths of the Baptistery may have had a similar function, located at the beginning or the end of the bathing itinerary. The high number of baths for a relatively small town may point to the baths attracting customers from outside the city. It is well attested for the 19th and 20th c. that the hammams in the cities of the Maghreb still attracted peasants and merchants from out of town.311 It is possible that the public baths of Roman cities attracted a similar crowd. The fact that the late antique semipublic baths were built along the main traffic arteries could support such a thesis (see also above, C30, baths in Herdonia; several baths along the Via Severiana in Ostia). Continued Use and Abandonment of Existing Baths The Large Baths (‘Grands Thermes Sud’) The Large Baths (map 15, nr. 10) were dedicated in AD 183–184 by the legatus Augusti.312 The southern palaestra was probably abandoned at the end of the 3rd c., although it seems that the baths were (partially) in use during the 4th c., when some pavements were restored.313 For a relatively small city such as Cuicul, these thermae are exceptionally large, covering around 3,000 m2. Its symmetrical layout and special attention to sport commodities (palaestra, natatio, basilica thermarum) are reminiscent of the imperial baths in the largest cities of the empire.314 The plan of several shifting axes centred on a dominant axis down the middle “display a mature and masterful handling of spatial relationships.”315 The presence of two small pools in the tepidarium is an interesting feature, as they allowed the bathers to take an additional dip after leaving the caldarium. One could imagine a special use of these pools, as the normal cold pools of the frigidarium lay just ahead. It may have provided a short cut, for those wanting to avoid the frigidarium. Alternatively, the small pools may have provided a more personal experience or may have had a medicinal use (see below, chapter 4).

311  Carlier (2000) 1311. 312   A E (1935) 45. 313   Blanchard-Lemée (1975) 233–34; Thébert (2003) 194. 314  Thébert (2003) 195. 315  Yegül (1992) 202.

Archaeological Evidence

141

map 15 Cuicul (Algeria) in the 4th c., with the location of public baths and the most important buildings after Leschi (1953) general plan 1: Macellum (market) 2: Old Forum 3: Capitolium 4: Basilica vestiaria 5: Severan Forum 6: Severan Temple 7: Theatre 8: Baths of the Capitolium 9: Baths of Terentius Donatus 10: Large Baths

142

chapter 3

figure 36 Plan of the Baths of the Capitolium in Cuicul (Algeria) in Thébert (2003) 626, pl. LXXVI.4; with permission by the publisher

The Baths of the Capitolium The Baths of the Capitolium (fig. 36; map 15, nr. 8) were probably the first large public baths built in Cuicul. Yvon Thébert convincingly argued that the baths were built before the Capitolium, probably shortly after the planning of the old forum (beginning of the 2nd c.).316 Little is known about the lifespan of these baths. It seems that at some date the west pool of the frigidarium was reduced in size.317 Judging from the plan, a small apsidal pool in the north-east corner of the room north of the frigidarium may have been added in a later phase, obstructing the original passage to the first of the heated rooms. The Baths of Terentius Donatus These small baths (26.2 × 23.5 m) were implanted in an insula opposite the curia (map 15, nr. 9).318 Excavated in the 1910s, they have never been the subject of a detailed study and were overlooked by Thébert (2003). A mosaic in the first room mentions how a certain Pumetius Longinianus restored the baths of Terentius Donatus.319 The frigidarium had a semicircular pool to the north-east and gave access to a larger rectangular pool preceded by a sort of vestibule. This cold pool seems to have been inserted in an existing room, as the walls are double. There were three rooms heated by a hypocaust in the southeast corner of the building. A long north-south oriented 316  Thébert (2003) 196–97. 317  Ballu (1917) 216. 318  Measurements from Ballu (1921) 226–27. 319  Ballu (1921) 228; Allais (1938) 44.

corridor gave access to 4 small rooms, while a door gave onto a small alley where a well was discovered (diam.: 1.5 m). No construction date has been proposed for these baths. Baths in the Late Antique Urban Context At the beginning of the 4th c., the inhabitants of Cuicul could still bathe in the large baths in the southern part of town. The restoration of castellum aquae on the new forum and the construction of a large public fountain some 20 m north of the large baths may have been part of a large ‘water program’ that also included restoration of the baths.320 A public latrine was also attached to the basilica vestiaria (dated AD 367).321 There is little information about the water supply network within the city and how the baths were linked to it. In the northern part of town, the Baths of the Capitolium may still have been in use. If the latter fell into decay, the newly built baths in the House of Europe, just across the street, may have compensated for this loss. Even if these were implanted in a private house, the double street access makes clear that they were at least accessible to a part of the larger public.322 It is unclear when the Baths of Terentius Donatus were built and if they were already or still in use in the 4th c. The proximity to the Baths of the Capitolium is remarkable. Maybe these small semi-public baths targeted a specific group of people? Too little is known about decoration and commodities 320  Lepelley (1981) 404. 321   I LS 5535. 322  Maréchal (2016) 128.

143

Archaeological Evidence

to understand the type of group this may have been (the elite, the poor, an association?). Alternatively, the owner of the house opened up private baths in an attempt to tie more citizens to his own familia and hence expand his base of political supporters. Near the end of the 4th c. (map 15), a bathhouse was inserted into the House of the Donkey, while a part of an insula near the public granaries was transformed into the East Baths. These (semi-) public neighbourhood baths were constructed in areas where there had not been baths before. Important traffic arteries such as the cardo maximus were prime locations to build such baths, not only due to the busy traffic, but also due to the presence of aqueducts and sewers running underneath these streets. The Large Baths obviously still fulfilled the ‘bathing needs’ of the southern part of town, as no new public baths seem to have been built there. The Baths of the Baptistery were probably for a selected group of (Christian) users. They were part of a large ecclesiastical complex comprising two churches and a chapel. It would be interesting to know who had paid for their construction and who was responsible for their exploitation. There is evidence from Rome and Ravenna that churches received the profits of a bathhouse to increase their revenues and attract more worshippers.323 The situation in the first half of the 5th c. may not have changed significantly (map 16). Our information about the 5th c. and later is too sketchy to reconstruct a clear picture of the urban fabric and the articulation of baths within it. The external baths of the House of Castorius were probably built at a late date. The presence of a sewer network and water supply in connection to the pre-existing ‘internal baths’ may have prompted their construction. During the Vandal and Byzantine period, few traces of building activity have been recognized. The Church of Cresconius, bishop in the early 5th c., is a rare example of such late building activity. In the 6th c., the northern part of town was partially deserted. A small church was constructed just within the southwestern corner of the old city walls at an unknown date, but seems to have been in use until the 7th c.324 The last occupation of the city seems to have been restricted to the area between the new forum and the Large Baths, perhaps not surprisingly on the highest parts of the town. The wells dug into the abandoned streets probably date to this last phase, meaning the water supply by aqueduct had disappeared. If a bathhouse did exist in this period, it must have been a very basic facility.

323  Sagui (1990b) 100; Stasolla (2002) 21–28. 324  Following the urban chronology by Sears (2007) 56–58.

Thamugadi Concise History of the Urban Fabric The city of Thamugadi was founded in AD 100 as a colony for veterans.325 The city lay only some 25 km southeast of Lambaesis, the station of Legio III Augusta, and on the road to Theveste (Tébessa). Furthermore, Thamugadi was built on the crossroads with a northsouth road leading from Cirta (Constantine) to the Aurès mountains in the south. The Colonia Marciana Trajana Thamugadi was conceived as a military camp, a large square (355 m on the side) crossed by a network of orthogonal streets creating square insulae (s: 20 m). Unfortunately, little is known about the history of the city, as there are few mentions of Thamugadi in the ancient literature and the epigraphic evidence mainly highlights the city’s economic aspects.326 The first bishop is mentioned in AD 256, while under Optatus (AD 388), Thamugadi became the most important centre of Donatism in North Africa. As an adviser to the comes Africae Gildo, Optatus had a strong influence on North African politics. The Vandal period is poorly documented in Thamugadi, just as in the rest of North Africa. By the turn of the 5th and 6th c., tribes from the Aurès mountains had invaded the city.327 However, the occupation of the site continued until at least the end of the Byzantine period and possibly into the Early Islamic period. The original nucleus of the city is a classic example of an orthogonal street grid with the forum and the most important buildings (curia, basilica) placed at the intersection of the cardo and decumanus maximus.328 Twelve rows of insula in the north-south axis and 11 rows in the east-west axis gave a total of 132 insulae. Around 70% of the city’s surface was made up of living space.329 From the first years of its existence, very few acts of euergetism are recorded in the inscriptions of the city.330 Large public infrastructure such as the theatre and the Large East Baths were constructed during the reign of Marcus Aurelius, when an aqueduct was also built.331 The walls of the city, forming a square of 11.5 hectares, were gradually dismantled after the 2nd c., as the town expanded. The expansion extra muros was most important to the west of the city centre, along the road leading to

325   I LS 6841; see Corbier (2009) for a foundation at the end of Trajan’s reign. 326  Sears (2007) 58–59; Corbier (2009). 327  Lassus (1976) 901. 328  Lassus (1976) 900. 329  Lohmann (1979) 167. 330  Corbier (2009) 184–85. 331   C IL 8.869; Courtois (1950) 38–44.

144

map 16

chapter 3

Cuicul (Algeria) in the 5th c., with the location of public baths and the most important buildings after Leschi (1953) general plan 1: Church IV 2: Church III 3: Church II 4: Church I 5: House of Bacchus

145

Archaeological Evidence

Lambaesis.332 The city’s largest temple, the Capitolium, was built some 100 m south-west of the western gate, possibly in the Severan era. The 3rd c. saw a further expansion west and south of the old city centre. Several large baths, including the Large North Baths and the Baths of the Filadelfes, were probably built in the late 2nd or 3rd c.333 The library along the cardo maximus may post-date the Severan era.334 The market of Sertius and an entirely new residential area developed around the western road leading to Lambaesis.335 Some 380 m south of the city, an important sanctuary with an enclosure wall dedicated to Dea Africa and Asclepius was built during the reign of Caracalla. The urban evolution after AD 250 is difficult to reconstruct due to the lack of building inscriptions and because early excavations did not see the importance of the late antique layers.336 Newly Built Late Antique Baths The excavations in the early 20th c. left us little reliable evidence for dating most of the newly constructed baths. It seems that only three new public baths were constructed after the 3rd c. (C57, C58, C59). This should not come as a surprise as Thamugadi was equipped with many large imperial thermae and smaller neighbourhood baths. Several of these complexes were still in use during Late Antiquity (see below). If we look at the plans of the new baths, we can see that little had changed. The baths still had a large cold room and at least three heated rooms, to be identified with tepidarium-sudatorium-caldarium (C57, C58, C59). The frigidarium of the Baths North of the Capitolium (C59) closely resembles the frigidarium of the Baths North-West of the Theatre in Bulla Regia (C36), while the shape of the northern piscina is also found in other late antique North African baths (e.g. C34, Sidi Ghrib). The North African preference for the ring type can be identified in the Small North Baths (C57) and the Baths North of the Capitolium (C59).337 The possible doubling of the bathing itinerary in the Small North-East Baths (C58) may have been linked to an advanced diversification of the bathing habit, with a separate wing for non-water related bathing activities (epilation, massage), or to a separation of sexes, with a separate wing for the female bathers. Such a gender division can also be found in the late antique baths of Egypt (e.g. C75B–C, C76, C87?, C89, C90, C91, C92, C95) and Palestine (C99, C116?). 332  Lassus (1976) 900. 333  Thébert (2003) 235–36; Corbier (2009) 187. 334  Gros (1996) 372. 335  Courtois (1950) 79. 336  Corbier (2009) 190. 337  Already remarked by Nielsen (1993a) 90–91 and Thébert (2003) 355–57.

The decorative schemes are poorly understood due to the bad state of preservation. It seems that most mosaics had geometric patterns, often in black and white, in the 4th and 5th c. (C57, C59). In later bathhouses, the use of basic building materials such as limestone or terracotta tesserae was more common (C58, Fort Baths). In the baths of the so-called ‘Donatist Cathedral’, the use of polychrome mosaics and marble revetment point to the care that was taken to embellish ecclesiastical complexes. The technology of the Thamugadi baths was often ignored in early excavation reports. It seems that the standard hypocaust pillars and wall heating techniques were used, even in Byzantine times (see the Fort baths). We can remark how the tubuli in the Fortress Baths did not cover the entire walls, but were lodged within recesses. Little is known about the water supply systems, but we can presume that the larger baths were linked to the aqueduct and / or had cisterns fed by rainwater. The sewage channels of the Small North Baths (C57) and the Small North-East Baths (C58) passed by the latrines before going into the street sewers. Continued Use and Abandonment of Existing Baths It is very difficult to reconstruct the number of bathhouses at the beginning of the 4th c. Few baths have been accurately dated, while the moment of decay has been pinpointed for even fewer. The large thermae were probably all built during the 2nd or 3rd c., the period of the most important expansion of the town. Several smaller baths have not been dated, but could very well belong to the late antique phases. We are certain that a few bathhouses fell out of use before the 4th c., as they were transformed into buildings with different functions. The high number of private baths in Thamugadi (approximately 14, i.e. 50% of the total number of baths) will not be discussed in detail here, as this falls beyond the scope of this research and have been discussed by Yvon Thébert.338 Imperial-Type Baths The Large East Baths were built in the 2nd c. and enlarged in AD 167.339 Some of the mosaics were probably added in the second half of the 3rd c., making it likely that these thermae were still in use in the 4th. At some point during the restoration works, the frigidarium was extensively altered.340

338  Thébert (2003) 243–52. 339  Tourrenc (1968) 215. 340  Germain (1969) 35; Thébert (2003) 231.

146 The Large South Baths were also built in the 2nd c. and enlarged in AD 198.341 Several statue bases dedicated to members of the imperial family have been found and one inscription even dates from the end of the 3rd c.342 A white mosaic in a small room near the frigidarium may even date to the 5th or 6th c.343 No dramatic changes to the plan seem to have occurred. The most important addition was an apsidal pool in the north-west heated room, blocking communication with another heated room.344 The Large North Baths were probably built during the Severan period, rather than in the first half of the 2nd c.345 The plan is of imperial type and does not reveal important structural changes. It is unknown when the baths fell out of use. The same is true for the Baths of the Filadelfes (fig. 37), located only 25 m west of the Large North Baths. The mosaics are attributable to the Severan period, but information on subsequent restorations is lacking. The mosaics in the portico of the courtyard are possibly late antique.346 The orientation of these baths is peculiar, as the heated section is pointed towards the north. An interesting feature is the heated room (6.5 × 3.7 m) to the north of the caldarium (1 on fig. 37).347 It was only moderately heated by a single furnace and had two small pools.348 On the threshold of this room is a mosaic inscription reading ‘SALVV LOTV’ (‘Have a nice bath’). The central panel of the floor mosaic gave the name to the baths, as it depicts an erotic Dionysian scene accompanied by the text ‘FILA.D.ELFIS VITA’ (‘Long live the Filadelfes’).349 In front of the western pool, the letters ‘B b’ were laid out in mosaics.350 This abbreviation remains unexplained. Judging from the mosaics, the room was obviously important. It has been interpreted as a gathering room for an association called the Filadelfes, although the exact nature of this group and the function of this room are not further explained.351 The message at the entrance in combination with the small one-person pools may point to a room used for special treatments, such as therapeutic baths.

341   A E (1894) 44. 342  Statue bases: e.g. CIL 8.2381–2383; late 3rd-century inscription: CIL 8.2393. 343  Germain (1969) 136. 344  Krencker et al. (1929) 230. 345  Thébert (2003) 235. 346  Germain (1969) 74, nr. 87, pl. Xxxi (portico), 77–79. 347  Measurements from Ballu (1905) 92–95. 348  Thébert (2003) 237. 349  Germain (1969) 77. 350  Thébert (2003) 516, nr. 170c. 351  Thébert (2003) 236.

chapter 3

figure 37 Plan of the Baths of the Filadelfes in Thamugadi (Algeria) in Thébert (2003) 654, pl. CIV.2; with permission by the publisher

Neighbourhood Baths Some baths that have not been accurately dated may have had a late antique phase. The Small East Baths, the Baths of the Capitolium and the Small South Baths belong to the modest neighbourhood baths that were popular in both the High Empire (e.g. the Small Central Baths) and in Late Antiquity (e.g. C57). The plan, consisting of three heated rooms and a large frigidarium with a single pool, was probably often repeated due to its simplicity and adaptability to the restricted space of an insula. The continued popularity of this type makes it impossible to assign it to a specific period. The Small East Baths protrude onto the streets, suggesting they were inserted in an existing insula (fig. 38). The complex is built primarily in brick and rubble concrete, contrasting with the opus africanum of the 2nd and 3rd-c. constructions. The terracotta mosaic found in the entrance corridor is reminiscent of the terracotta mosaic in the Baths of the House North of the Capitolium (C59).352 The plan is similar to the Small North Baths (C57): a frigidarium with a rather large pool and the 352  Germain (1969) 45.

147

Archaeological Evidence

figure 39 Plan of the Baths of the Capitolium in Thamugadi (Algeria) in Thébert (2003) 655, pl. CV.4; with permission by the publisher figure 38 Plan of the Small East Baths in Thamugadi (Algeria) in Thébert 2003, 649, pl. XCIX.6

succession of three heated rooms, ending in a caldarium (IV on fig. 38) with a rectangular and a semicircular alveus. An additional room with a small heated pool (exit-tepidarium, 6) ensured the bathers could follow a circular itinerary, while a small latrine was also incorporated. Thébert considers this small room to be part of a shortcut for athletes wishing to join the caldarium immediately after the exercises.353 However, this would imply that the oblong room (1) was used for exercises, a hypothesis that cannot be confirmed. The Baths of the Capitolium (fig. 39) were built in an area that developed quite late, after the erection of the Capitolium (first half of the 3rd c.). It flanks the road leading to the nearby ‘Donatist complex’ (end of the 4th c.). The plan resembles the one of the Small North Baths (C57): a large frigidarium with a large pool gave access to a set of three heated rooms. The caldarium (IV on fig. 39) was also the only heated room with pools, a rectangular one and a circular one.354 The frigidarium was in this case equipped with an additional small pool, a feature returning in many late antique baths.355 The entire surface covered by these baths does not exceed 500 m2.356 353  Thébert (2003) 232. 354  Germain (1969) 120. 355  We should remember that the frigidarium of the Small North Baths (C57) was badly damaged. There may well have been a small cold pool in the southern part of the room. 356  Thébert (2003) 240.

The Small South Baths fit within this same pattern. The modest building (ca. 350 m2) had a comparatively large cold room with a single piscina and a succession of three heated rooms.357 The caldarium had a rectangular and semicircular pool. The second heated room also had a pool. The latrine was located outside the bathing route proper and accessible from the outside. The mosaic floors were repaired with terracotta slabs, a late antique type of restoration.358 However, these repairs may have been carried out after the building had changed function. The Small Central Baths are dated to the 2nd c. The opus reticulatum panels would have been a late use of this construction technique. However, the mosaics of the frigidarium could be placed at the end of the 3rd c.359 The plan reveals the same type of rooms arranged in a retrograde itinerary. Just as in the Small East Baths, the pool in the first heated room could have been used at the end of the bathing itinerary. The North-West Baths (‘thermes de la porte de Lambèse’) may have been built during the development of the area in the second half of the 2nd c. (the nearby gate of Lambaesis is dated AD 167). However, the building history of the baths is difficult to reconstruct due

357  Thébert (2003) 243. 358  Germain (1969) 137. 359  Germain (1969) nr. 61; Dunbabin (1978) 275; Thébert (2003) 233.

148 to numerous restorations and changes.360 The preference for a high number of small—often semicircular or apsidal—single-person pools could point to a phase of late antique use. Only one bathhouse had certainly fallen out of use during Late Antiquity. The baths of insula 17 were private baths inserted in an existing domus. The frigidarium was transformed into a church by raising the floor level and adding an apse to the south (4th or 5th c.?). According to Ballu, the piscina was reused as a baptistery in the Byzantine period.361 Baths in the Late Antique Urban Context As so many baths (and other buildings) are difficult to date, the urban evolution, as presented in this study, necessarily shows an incomplete picture. It does seem, however, that the 4th c. was the period of greatest growth for Thamugadi (map 17). The quarter west of the Capitolium seems to have expanded, for example with a large new ecclesiastical complex (‘Donatist cathedral’) that was erected at the end of this century. The old city centre was certainly still bustling with activity. The Small North Baths lay near the northern gate of the city, along the cardo maximus. Such an important access road, leading towards the forum, ensured a steady passage of potential bathers, while the aqueduct and the most important sewer also ran underneath this traffic artery. The vitality of the old centre can also be deduced from the construction of the first and only intra muros church. It was built in the House of Januarius somewhere during the 4th c.362 The Large East Baths may still have been in use at this time. There is less certainty about the fate of the Small East Baths and the Small Central Baths. In the southern suburbs, the Large South Baths were still functioning, while there is a good possibility that the Small South Baths were active as well. The Temple of Mercurius was restored during the Tetrarchy, while the large House of Sertius was inhabited until the 4th or even 5th c.363 To the west of the ancient city, building activity flourished. The Capitolium was restored under Valentinian and Valens.364 A small church made out of spolia might date from the Donatist heyday of Thamugadi, rather than from the Byzantine period.365 The nearby Baths of the Capitolium might still have been in use during the 4th c. The large ecclesiastical complex, often labelled the ‘Donatist cathedral’, was built at the end of the 360  Ballu (1906) 204; Thébert (2003) 237. 361  Ballu (1905) 88. 362  Courtois (1950) 48. 363  Courtois (1950) 82; Lepelley (1981) 446. 364  Lepelley (1981) 447. 365  Sears (2007) 61 contra Lassus (1976) 901.

chapter 3

4th c. under the supervision of Optatus, perhaps to be identified with the Optatus known from Augustine and included a suite of bathrooms (fig. 40).366 Along a road leading north and eventually connecting with the road to Lambaesis, a large house with adjacent bathhouse (C59) was built at the turn of the 4th c. Together with the newly built ‘clothes market’ near the western gate of the old city,367 the new baths stress the continued investment in this part of town. Along the Lambaesis road, the North-West Baths may still have been in use. A small church was built further along the same road in an area used as burial ground. The Christian tombs found within its courtyard predate the 5th c.368 This points to a continuity between the ‘pagan’ and Christian burial ground. To the north-west and to the north of the old city centre, the Large North Baths and the Baths of the Filadelfes may still have been in use in the 4th c. A large basilica was also constructed here during the 4th c.369 The evidence for 5th c. Thamugadi is problematic. It seems that most of the public baths had already fallen out of use (map 18). Only for the Large South Baths is there evidence for a 5th and even 6th-c. phase. Other large secular and public buildings, such as the theatre and the Capitolium, were probably abandoned.370 The continued investment in urban infrastructure must have been primarily aimed at the churches and ecclesiastical complexes, although it is difficult to assess which buildings were still functioning. The Byzantine conquest of the city meant a limited return of epigraphic evidence. The fortress built some 375 m south of the old city was constructed in AD 539– 540 according to an inscription.371 It was built partially on top of the old sanctuary of the Aqua Septimiana (compare maps 17, 18) and equipped with a bathhouse (fig. 41). Some 300 m south-west of the fortress, a Christian graveyard was discovered.372 A small chapel built in the vicinity was dated by a dedicatory inscription to AD 641–647, mentioning the name of the dux of Tigisi.373 The larger church of the graveyard has not been dated. South of the old city, the Large South Baths may have been (partially) in use. In the old city centre, the Small North-East Baths (C58) may date to the Byzantine period, although this date only relies on a Byzantine type of capital found there. Around the north-west gate 366  See however the critical re-examination by Gui et al. (1992) 275–76; also Sears (2007) 62 with anterior bibliography. 367  Courtois (1950) 81. 368  Sears (2007) 61. 369  Sears (2007) 62–63. 370  Courtois (1951) 38. 371  Lassus (1981) 13. 372  Courtois (1950) 83. 373  Courtois (1950) 66.

Archaeological Evidence

map 17

Thamugadi (Algeria) at the end of the 4th c., with the location of public baths and the most important buildings after Thébert (2003) 667, pl. CXVII 12: Chapel of Ianuarius 1: Forum 13: Baths of the Filadelfes 2: Theatre 14: Large North Baths 3: Library 15: Large East Baths 4: Capitolium 16: Small East Baths 5: Market of Sertius 17: Small Central Baths 6: Basilica vestiaria 18: Large South Baths 7: Church V 19: Small South Baths 8: ‘Donatist cathedral’ 20: Northwest Baths 9: ‘Catholic’ complex 21: Baths of the Market of Sertius 10: Church I 22: Baths of the Capitolium 11: Sanctuary of Aqua Septimiana Felix

149

150

chapter 3

that were still in use during the 4th c. seems to confirm that the aqueduct system was still up and running. Carthage

figure 40 Plan of the Baths of the Donatist Cathedral in Thamugadi (Algeria) after Thébert (2003) 664, pl. CXIV.2

of the road to Lambaesis, some olive oil presses and Christian inscriptions were found, perhaps attesting a Byzantine occupation of this part of town.374 It is unknown when the water supply system of the city went out of use. Several branches of an aqueduct were found underneath the main streets of the old city (see map 17), probably originating from the Aïn-Morris source some 3 km south of the city. However, it seems that the north-easterly quarter was not served by the aqueduct and had to rely on wells.375 The extra muros expansions of the late 2nd and 3rd c. were probably supplied by new branches of the aqueduct, although only few traces of it have been identified. Nevertheless, large bath complexes such as the North-West Baths were probably supplied by an aqueduct. The high number of baths

374  Courtois (1950) 82. 375  Lohman (1979) 179.

Concise History of the Urban Fabric It would go beyond the scope of this research to summarize the entire history of Carthage, from its foundation by Phoenician tradesmen in the late 9th c. BC to the invasion of Arab armies in the 7th c. AD. A brief outline of the ‘Roman phase’ can be useful to contextualise Carthage’s position in Late Antiquity. After its defeat in the Third Punic War against Rome (146 BC), the city of Carthage lay in ruins.376 It was only in 40 BC, inspired by the late Julius Caesar, that the Roman state saw some potential in reconstructing the capital of its former enemy. Carthage was made the provincial capital of the new province of Africa Nova and was populated with colonists.377 The old Punic houses and public buildings were razed to the ground, “Après la mort de la cité, l’enterrement de ses vestiges” (“After the death of the city, the burial of its remains”)378 and from its ashes a new Roman city arose. With the foundation of ‘Roman Carthage’, a new urban layout was planned. The old citadel on the Byrsa hill was levelled to become the forum. This was the centre for laying out a regular street grid. Public monuments such as the theatre, odeum, amphitheatre and circus also date from this period. The so-called La Malga cisterns, a group of interconnected reservoirs with an estimated capacity of 51,000 m3, probably antedated the large aqueduct and supplied the 1st-c. city with water.379 In the Antonine period, the forum was restored after a fire had caused significant damage. The city’s largest bathhouse, the Antonine Baths, were dedicated in the period AD 145–161.380 The Zaghouan aqueduct, the longest in the Roman world (132 km in its last phase), enabled rather than followed the construction of these thermae.381 Christianity probably started to spread in Carthage in the 2nd c. The first martyrs, coming from a small town in Numidia called Scillium, were condemned by the proconsul in Carthage in AD 180.382 It is the oldest attestation of Christians in North Africa. In the 3rd c., Carthage became the episcopal seat of North Africa. The letters of bishop Cyprian (AD 200–258) give a good insight into the Christian community at that 376  Lancel (1992) 570–74. 377  Le Glay (1985). 378  Lancel (1992) 578; translation by the author. 379  Wilson (1998) 76–77. C IL 8.12513. 380   381  Wilson (1998) 92. 382  Passio Sanctorum Scillitanorum in Musurillo (1972) 86–89.

Archaeological Evidence

map 18

Thamugadi (Algeria) at the end of the 6th c., with the location of public baths and the most important buildings after Thébert (2003) 667, pl. CXVII

151

152

chapter 3

figure 41 Plan of the Baths of the Byzantine Fortress in Thamugadi (Algeria) after Lassus (1981) 124, fig. 87

time. Carthage was spared from the tribal incursions that threatened other regions in North Africa during the 3rd c.383 The political turmoil in Rome and a setback in trade during this century probably impeded important building works and public expenditure.384 Newly Built Late Antique Baths Unlike other cities discussed so far, the newly built baths in Carthage were more evenly distributed throughout the whole period of Late Antiquity, with baths being built in the 4th (C44, C47?), 5th (C43?, C45) and the 6th c. (C46, baths in the Baal Hammon Street). The average size of the baths is also atypical, due mainly to the large Byzantine Baths (C46) and the Douar Chott Baths (C47) having a surface easily exceeding 1000 m2. The neighbourhood baths probably fluctuated around 500 m2 (e.g. C43), but unfortunately most of these smaller baths are only partially known. The two large baths may have been built with public money, especially if we identify the so-called Thermae Theodorianae mentioned by Procopius (see above, chapter 2) with the Byzantine Baths. However, there is no direct evidence to support the public or private ownership of any of the baths. Several baths were constructed in or on top of preexisting buildings. For the Byzantine Baths (C46), the advantages of building new large public baths on the 383  Lepelley (1979) 84. 384  Thébert (1983) 102; Leone (2007) 23.

remains of its predecessor seems clear, considering the topographic continuity within the city and the availability of building materials, water supply and disposal systems. Smaller baths were built in all sorts of buildings (e.g. C44), including baths that were formerly private (C43). The transformation of a private into a public bathhouse was not uncommon in North Africa (see also below, chapter 4). The architecture of the new baths followed the trends already seen in the Severan period, with a lot of round, oval and semicircular shapes, especially for pools and apses (C45, C46, C47). The types of plan are difficult to reconstruct from the fragmentary archaeological data of most of the smaller baths. The symmetrical plan of the Byzantine Baths (C46) was reserved for large baths for which building space was not an issue. The appearance of small pools occasionally achieved by dividing larger pools in smaller units (C43) was an evolution that was also occurring in other regions in the Roman empire (see below, chapter 4). Again, we should note that this did not necessarily entail the disappearance of larger pools (e.g. C47). Continued Use or Abandonment of Existing Baths The Antonine Baths Before the Byzantine Baths were built on top of the Antonine Baths (fig. 42; map 19, nr. 10), the latter had been used throughout the 4th c. An inscription found on an entablature in the baths commemorates a restoration

Archaeological Evidence

153

figure 42 Plan of the Antonine Baths in Carthage (Tunisia) after Nielsen (1993b) 163, fig. 179

map 19 Carthage (Tunisia) at the end of the 4th c., with the location of public baths and the most important buildings and public areas after Leone (2007) 156, fig. 41 1: Amphitheatre 2: Forum (Byrsa Hill) 3: Amphitheatre 4: Theatre 5: Odeon 6: Circular harbour 7: Roman harbour 8: La Malga cisterns 9: Borj Djedid cistern 10: Antonine Baths 11: Baths on the Borj Djedid 12: Damous el-Karita Baths 13: Baths on the Juno Hill 14: Phenix Baths 15: Bir el Jebbana Baths 16: ‘Proconsular Baths’

154 of the building under Valentinian II, Theodosius and Arcadius (AD 388–390; see EA-31). The floors of the octagonal rooms were also embellished with new mosaics of horses and athletes. These can be dated to the late 4th c.385 The baths were in use even after the vault of the frigidarium had collapsed. Due to the bad state of preservation, it is not possible to reconstruct the changed plan of these reorganized baths. The ‘Proconsular Baths’ This bathhouse was discovered and unearthed in the 1920s (unpublished; map 19, nr. 16). It was identified as the ‘thermae proconsularis’ after an inscription mentioning baths with that name was found nearby (see EA-16). The text seems to mention a restoration under the proconsul Flavius Dardanius, and is dated AD 340–350. The Baths on the Juno Hill Between 1918 and 1921, a small excavation on the Juno Hill uncovered two large cisterns and some adjacent substructures.386 The cisterns belonged to a house with a peristyle, the ‘Maison de la chasse au Sanglier’, dated to the 3rd c.387 The vaulted structures at the foot of the cisterns seem to have been part of a large bathhouse, that may still have been in use during the 4th and 5th c. (map 19, nr. 13).388 The plan is uncertain. Two semicircular niches may have contained basins, but only one of the rooms had a hypocaust.389 In a later (Byzantine?) phase, a Christian chapel and accompanying cemetery was built on the locations of the baths.390 On the north slope of the hill, another bathhouse was unearthed in the 1880s. Some of its features, such as a series of rectangular basins and a round room with black and white mosaics,391 may point to a late antique phase. Baths on the Borj Djedid In 1908, part of a bathhouse was discovered on the Borj Djedid hill (map 19, nr. 11). A service corridor, covered by a vault made of tubi fittili, and a hypocaust with pillars made out of tufa were found. An inscription discovered on site commemorated the embellishment of the thermae (see EA-5).392 It is dated to the beginning of the 4th c.

385  Lézine et al. (1956) 426. For dating, see Dunbabin (1978) 252. 386  Lantier (1921). 387  Lantier (1931) 502; Dunbabin (1978) 252. 388  Leone (2007) 171. 389  Lantier (1921) 88. 390  Lantier (1921) 90–92. 391  Poinssot and Lantier (1922) LXI–LXII. 392  Delattre (1908) 593–94.

chapter 3

The Damous el-Karita Baths The remains of a bathhouse were discovered in the area of the Christian basilica at Damous el-Karita (map 19, nr. 12). Of the two rooms that were discovered, one had a rectangular pool. A brick stamp from the site has been dated to AD 155. The inscription found on the same site and dated to the Vandal period (see EA-48) may have belonged to a different bathhouse in the vicinity. In a late (Byzantine) phase, the pool of the bathhouse was covered with reused Christian epitaphs and inscriptions.393 The Phoenix Baths Although this bathhouse was described as ‘the Christian baths’ by some scholars (map 19, nr. 14), the construction date of this complex is unknown. The remains visible today do not reveal any datable elements.394 The Bir el Jebbana Baths (‘Bains de la villa de Scorpianus’) The area of these baths, situated some 100 m northwest of the amphitheatre along an important road (map 19, nr. 15), were first discovered and excavated in the 1880s by Father Alfred Delattre.395 The baths were interpreted as the private complex of a wealthy villa. Between 1994 and 1997, a Canadian team reopened the site and expanded the excavations. New research suggested that these baths were public, although no direct street access was found. A construction date between AD 115 and 144 was proposed on the basis of brick stamps and pottery found in the small test pits under the floors. At the turn of the 3rd and 4th c. a reorganization and repair changed the layout of the baths. The hypocaust of a heated room was replaced with a normal floor. The building was abandoned at the latest at the beginning of the 4th c., when tombs were dug in the northern rooms. A cemetery east of the baths reused several ornamental elements from the bathhouse, with burials dating between the 4th and 6th c.396 Private Baths on the Odeon Hill (‘Bains de la maison de la volière’) The bath of a large domus was discovered in 1903, but has since been severely damaged. Several rooms with a hypocaust and a cold section with pools were identified. The mosaics show several building phases, the most recent of which dated to the late 3rd or early 4th c.397 393  Delattre (1886) 235–36. 394  For the ‘Christian’ interpretation: Vaultrin (1932) 206. For the visible remains: Rossiter (2009) 186–187, fig. 4. 395  Delattre (1882); Delattre (1883). 396  Summarizing the results in Rossiter (1998) 110–13. 397  Carucci (2007) 129–31; de Haan (2010) 348.

155

Archaeological Evidence

Other Baths Several other bathhouses that were discovered in Carthage cannot be securely dated. Often a late antique phase could not be proven on the basis of the available evidence.398 The bathhouse mentioned by Augustine, the Baths of Gargilius, where the ecclesiastical congress of AD 411 was held (see above, chapter 2), has not yet been identified with any archaeological remains. Augustine mentions that the baths were located in the city centre399 and that its assembly hall (secretarium) could receive 500 bishops. With the present state of research, the Antonine Baths are the only ones to fit this description. It is possible that the Baths of Gargilius was just the late antique name of the imperial thermae.400 However, it is equally probable that the Baths of Gargilius simply have not been found yet. The Falernian Baths, the object of curse on a lead defixio (curse tablet, see EA-59), are also unknown from the archaeological record. As the word balneum is used to denote these baths, it probably indicated a modest neighbourhood bath. Baths in the Late Antique Urban Context During the 4th c., the urban fabric of the imperial period was still more or less intact (map 19). The harbours were still areas of important economic activity, while most public buildings such as the circus, the theatre, the odeum, the amphitheatre and the forum were still in use. The Temple of Cybele and Attis, the proconsular palace, the theatre, the odeum, the Antonine Baths and a fountain were restored.401 The Dermech region, located near the coast around the Antonine Baths, saw important building activity during the 4th c. The residential quarter north of the baths was enlarged and a church, the oldest known example in the city, was erected. In the area of the theatre, enlargement and refurbishment of the dense housing can be observed. A circular complex might have been a memorial site for a martyr.402 Only some small neighbourhood baths, such as the Bir el Jebbana baths, had fallen out of use. In the insula between the Byrsa and the Junon Hill, a small bathhouse—probably for private use—was added to the existing neighbourhood baths. The Zaghouan aqueduct still supplied water to the city and the large reservoirs such as those of La Malga and the Borj Djedid were still operational.403 According to

398  For an overview, see Rossiter (2009) 193–97. 399  August. Ad donatistos post conlationem 25.43. 400  Rossiter (2009) 178, n. 3. 401  Leone (2007) 124, table 8; Sears (2007) 38; both with anterior bibliography. 402  Leone (2007) 115–16, table 7. 403  Wilson (1998) 93.

the written sources, the first churches were built during this century.404 In the 5th c., large parts of the monumental town centre gradually fell into decline (map 20). The Vandals were, however, concerned with the preservation of the urban fabric. Several restorations in private houses can be dated to the second half of the 5th c.405 Some Christian churches, such as Dermech I or Damous el Karita, date from the end of the 4th or the beginning of the 5th c.406 Yet, it is impossible to pinpoint their construction to before or after the Vandal conquest.407 The city walls, which had been restored in AD 425, no longer had a defensive function, as houses were built directly against them.408 According to Victor of Vita, the Vandals destroyed the odeum, the theatre, the aedes memoriae (one of the oldest Christian martyria), and the basilica maiora, where St. Perpetua, St. Felicita and the Scilitan martyrs were buried.409 It is difficult to recognize these damages in the archaeological record. Only the data from the odeum confirm this literary evidence, as it was used as a burial place.410 Poor quality housing, presumably from the late 5th c., covered the ruins of the theatre. The circus and perhaps the amphitheatre may have been still in use, but more research is needed to confirm this.411 Two churches, one at Bir el Knissia and one at Bir Ftouha, seem to have been constructed after the Vandal conquest. On the forum, the large basilica seems to have been out of use, but the proconsular palace was still functional as it was used by the Vandal King for his court. Parts of the old circular harbour seem to have been abandoned.412 Even if the Zaghouan aqueduct was still functional, the La Malga and Borj Djedid reservoirs probably ceased functioning in the 5th c.413 The Antonine Baths were closed after the vaults of the frigidarium had collapsed. The bathhouses that were in use during this period lay on a north-south axis, corresponding to the cardo maximus. In the outskirts of town, burials were inserted in the residential quarters, pointing to a transformation of these domestic areas. Industrial quarters developed in the Sayda quarter.414 In the 6th c. (map 21), the remaining large public structures, such as the circus and the amphitheatre, fell out of use and were subsequently occupied by poor 404  Ennabli (1997) 15–44. 405  Leone (2007) 161–162 and table 12. 406  Ennabli (1997) 147–52; Bockmann (2013) 105–17. 407  Leone (2007) 157. 408  Sears (2007) 39. 409  Victor Vit. 1.8–9. 410  Leone (2002) 240. 411  Bockmann (2013) 62. 412  Leone (2007) 157–59; Bockmann (2013) 64–65. 413  Wilson (1998) 93–97. 414  Leone (2007) 164.

156

chapter 3

map 20 Carthage (Tunisia) at the end of the 5th c., with the location of public baths and the most important buildings and public areas after Leone (2007) 169, fig. 45 1: Damous el-Karita Church 2: Dermech I Church 3: Bir el Knissia Church

housing in the 7th c.415 Most building activity was focused on residential architecture and ecclesiastical buildings. Procopius recounts that Justinian ordered the walls to be restored and commanded the construction of several other buildings, such as a chapel for Mary Theotokos inside the proconsular palace and a new bathhouse. Furthermore, he built stoas on the Maritime Forum and a monastery on the shore.416 The archaeological evidence proves that the Byrsa hill was fortified. At least three churches were built in the Dermech area. In the urban periphery, the cemetery churches were enlarged. “Religious buildings played the principal role and transformed the focal centres of the urban and suburban reorganisation, completing a process that had already started from the end of the 4th c.”417 The new buildings were fitted within the ancient street grid of the city, although in some cases public and

private structures encroached on the street or the public space, foreshadowing the development that would eventually lead to the creation of the medieval souk.418 Sewers and drainage channels were kept in use and restored. The south-east quarter, especially in the harbour area, saw some revival with the construction of undefined public buildings and some modest housing. The Borj Djedid and Dermech area were still the most bustling parts of town, as new houses and churches were erected.419 The Byzantine Baths were constructed on the remains of the Antonine predecessor. Most of the other baths, such as the Odeon Hill baths or the Baths in the Baal Hammon Street fell out of use in the course of the 6th and 7th c. The aqueduct still functioned when Belisarius besieged the city in AD 533.420 Several smallscale reservoirs were constructed along the course of the Zaghouan aqueduct and in the Borj Djedid reservoirs.

415  Bockmann (2013) 62–63. 416  Procop. Aed. 6.5.8–11. 417  Leone (2007) 178.

418  Kennedy (1985). 419  Leone (2007) 172 (sewers, drains), 179 (houses, churches). 420  According to Procop. Bell. 4.1.2.

157

Archaeological Evidence

map 21 Carthage (Tunisia) at the end of the 6th c., with the location of public baths and the most important buildings and public areas after Leone (2007) 180, fig. 54 1: Baths in Rue Baal Hammon 2: Bir Messouda Church 3: Basilica of Carthagenna

However, the latter were never reconnected to the Byzantine Baths. It was not until AD 698, when Arab troops sacked the city, that the main aqueduct was destroyed.421 The distribution pattern of graves highlights the most ‘active’ part of town i.e. the north-eastern quarter.422 At the end of the 6th and the beginning of the 7th c., Carthage gradually shrank and monumental architectural heritage declined. The basilica of Cartagena fell out of use and seems to have been occupied by poorly built houses.423 Other churches survived as centres of pilgrimage.424 After the Islamic conquest, archaeological evidence points to a continued occupation in the Byzantine quarters, namely the north-east quarter and

421  Wilson (1998) 93–94, 98. 422  Sears (2007) 38. 423  Ennabli (1997) 68–69. 424  Leone (2007) 179.

the forum area.425 Production activities were centred on the harbour area, with smaller shops and houses pointing to production on a smaller scale, probably aimed at regional markets or local subsistence.426 Domestic and industrial structures encroached on public spaces, creating a dense network of small houses and intertwining alleys that can still be seen in today’s souks. Sufetula Concise History of the Urban Fabric The history of Sufetula is difficult to reconstruct due to the lack of literary sources. It seems that the first settlement was founded during the Flavian dynasty (second half of the 1st c.) by veterans of Legio III Augusta.427 425  Vitelli (1981) 5–39. 426  Leone (2007) 181. 427  Duval and Baratte (1973) 8.

158

chapter 3

The city was strategically located on a plateau, controlling an important road that connected the interior of Africa Proconsularis, especially the cities Ammaedra and Thelepete, with the coast. The city, which now is excavated over an area of around 20 hectares of a presumed total of 50 hectares, was delimited in the north and east by the Wadi Sbeitla. The economy was primarily based on olive cultivation, while ceramic production was also widespread in the region.428 The city followed the classic development of a Roman town, becoming first a municipium and later a colony (Severan period), although the exact dates of these ‘promotions’ is unknown. By the Severan era, Sufetula was a prosperous town governed by a city council and a curator.429 It further expanded throughout the 3rd and 4th c., benefitting from the trade in olive products. The first bishop is attested as early as the 3rd c. With Diocletian’s reforms, Sufetula became part of Byzacena. The city fell under Vandal rule and saw persecution campaigns against non-Arian Christians (AD 484).430 When the Byzantines conquered the city, it became an important military stronghold in holding off the raids of Amazigh tribes. The exarch Gregorius chose Sufetula as his capital when he cut ties with Constantinople (AD 647), but eventually the city was sacked by Muslim troops. This did not mean the end of all occupation, as evidence has been found of a continued urban activity.431 Although the site is relatively well preserved, the chronology is poorly understood. Most of the monuments were unearthed in the period 1917–22 and 1942– 47 without any archaeological methodology.432 Sufetula is a good example of a planned city (map 22), laid out following an orthogonal grid with oblong insulae (ca. 100 × 45 m). This was probably the result of a single project, dating to the foundation of the city or to an important reform in the first half of the 2nd c.433 The paved streets, all with sewers running underneath them, follow a north-east / south-west and north-west / southeast orientation. The hypothesis that Sufetula started as a castellum before evolving into a city is unconvincing.434 Possible structures that predated the street grid have been discovered under Basilica 1, yet there are no indications for a precise construction date. The Capitolium, named after the three temples standing next to each

other, was preceded by a large square, possibly the forum, and an arch dedicated to Antoninus Pius.435 The aqueduct-bridge may also have dated to this period, as an inscription of AD 145 was reused in a repair work.436 However, there is doubt whether this repair was carried out in antiquity or in the 19th c. The theatre and the amphitheatre have not been the subject of archaeological research. They probably dated from the great 2nd c. expansion, yet evidence is lacking. The Christian monuments were constructed at the beginning of the 4th c. Three basilicas (1, 3 and 4) were inserted in the city centre in an effort to ‘Christianize’ the public space and the most important social hubs.437 In an unusual act of conversion, a classic ‘open-courtyard’ temple was reshaped into basilica 1.438 An arch was built at the eastern end of town under the Tetrarchy.439 At the end of the 5th or beginning or the 6th c., renewed building activity upgraded the more peripheral areas of town (basilicas 2 and 5).440 By the 7th c., large parts of the town had been abandoned, while some fortified farms were built at the eastern edge of town. However, the city was never walled.441

428  Béjaoui (2004) 12–13. 429  Lepelley (1981) 308; Kuhoff (2010) 279. 430  Duval and Baratte (1973) 9; Duval (1982) 603; Béjaoui (2004) 7. 431  Duval (1982) 620–26; Kennedy (2007) 207; Kuhoff (2010) 293. 432  For the reports on these earliest interventions, see the bibliography in Duval (1982) 626–31. 433  Duval (1982) 602. 434  Duval (1990) 505–508.

435  Duval (1982) 602–606. 436   C IL 8.229. 437  Kuhoff (2010) 285–88. 438  Duval (1971a) 9–15. 439  Duval (1982) 612–13. 440  Duval (1971a) 153–62; Duval (1982) 625. 441  Kuhoff (2010) 292. 442  Dunbabin (1989) 29–30; Manderscheid (1994) 64.

Newly Built Late Antique Baths The dating of the new baths is again problematic due to the absence of solid datable evidence. All three new late antique baths (C55, C56, Bath of Basilica 5) seem to have been built in the late 5th or early 6th c. The plan of the baths follows the linear row type, with a large frigidarium and at least two piscinae, often with an apsidal end, and several heated rooms. In the first heated room—or last while backtracking—a small pool was present (C55, C56). The large frigidarium of the Bath nr. 26 and the Bath of Basilica 2 were preceded by a large multipurpose hall (C55, C56). The decoration of the baths seems rather low-key, with the absence of marble especially noteworthy. The floors of the Bath of Basilica 2 and Bath nr. 26 were paved with limestone slabs. The polychrome mosaics bordering the piscina in Bath nr. 26 depict marine scenes, a common theme in the decoration of baths.442 As in several other late antique baths, the bottoms of the pools were often finished with a plain white mosaic (C55, Baths of Basilica 5).

Archaeological Evidence

map 22

Sufetula (Tunisia) at the end of the 4th c., with the location of public baths and the most important buildings after Duval and Baratte (1973) fig. 2 1: Amphitheatre 2: Capitolium 3: Theatre 4: Large Reservoir 5: Nymphaeum 6: Basilica IV 7: Chapel with baptistery 8: Basilica III 9: Large Baths 10: Tabernae along the decumanus 11: Baths nr. 15 12: Possible baths (‘nr. 7’)

159

160 The technology of the late antique baths can only be investigated in the Baths of Basilica 2, as the brief descriptions of Baths nr. 26 and the Baths near Basilica 5 do not permit even a basic reconstruction. The heating system and water management of the Baths of Basilica 2 are standard, with pillars made of square tiles, tubuli for wall heating and sewers connecting to the latrines. The water for the pools was stored in two reservoirs, but it is unknown how these were fed. Continued Use and Abandonment of Existing Baths The Large Baths The largest bathhouse discovered so far in Sufetula is located along the decumanus maximus connecting the Capitolium to the theatre (fig. 43; map 22, nr. 9). It consisted of eastern and western wings, covering a total surface of around 3,500 m2. The construction date is problematic as the excavations of 1916, 1922, 1947 and the 1970s did not yield conclusive evidence. A statue base of the 3rd c. is incorporated into the wall of the vestibulum, but there is no evidence that this stone belonged to the first building phase, nor that it remained in situ.443 Two more inscriptions give a terminus post quem for restoration works. The first inscription, dated to AD 283, was discovered in one of the walls of a piscina.444 The second one is dated to the Tetrarchy on the basis of palaeography (see EA-1). There is no mention of the type of work that was commemorated, but it could be linked to restoration works on the (cellam?) piscinalem thermarum hiemalium, the ‘pool (room) of the winter baths’.445 If these inscriptions did belong to the baths, the two sections could be interpreted as summer and winter baths (resp. the eastern and western parts). The remains will not be discussed here at length.446 However, it is interesting to look at some of the structural changes that occurred during the lifespan of these baths. Even if it is impossible to date the interventions, we can assume that structural remains discovered during the excavations reflect the articulation of the rooms during their last phase of use. In the eastern frigidarium, an extra pool (ca. 5 × 6 m) was added and the route for visiting the heated parts was altered by the blocking up of doors. The western part was changed significantly.447 The original frigidarium was abandoned and a new cold room was created in 443  Thébert (2003) 153–54. 444  ILAfr 139. 445  ILAfr 141 = AE (1921) 30. 446  See Duval and Baratte (1973) 82–86. For concerns about attributing these inscriptions to the Large Baths, see Thébert (2003) 153–56. 447  Duval and Baratte (1973) 83, fig. 51. Compare Thébert (2003) 600, pl. L2 and L3.

chapter 3

what used to be a large vestibule next to the palaestra.448 Two pools were installed there: a square one (ca. 5 × 5.5 m) and a small apsidal one (ca. 3 × 3.5 m). As a result of abandoning the original frigidarium, the original entrance to the heated part was blocked up and a new one was pierced through the east wall, reaching the new frigidarium. Other doors inside the heated parts were also blocked to create a different route. The northwestern pool of the caldarium was reduced in size and changed to a semicircular shape.449 It is difficult to assess how many different phases these interventions represented. It is also unclear if both the eastern and western parts of the complex continued to be used simultaneously or if one part (the larger eastern baths?) was abandoned earlier. The most important observation about these interventions is the importance attributed to the pools, especially by adding small semicircular ones. The use of double-shafted and corner columns is typical for local late antique architecture.450 Bath Nr. 15 Bath nr. 15 was located in the north-east quarter of town, one insula north of the Large Baths and two insulae east of the Baths of Basilica 2 (map 22, nr. 11). As the baths were unearthed in 1956–58, the information about the chronology of this building was destroyed.451 Looking at the plan (fig. 44), we can recognize the same oblong vestibule at the beginning of the bathing route and the preference for apsidal pools as in the Large Baths. In a later phase, some doorways of the heated rooms were blocked.452 Baths in the Late Antique Urban Context At the beginning of the 4th c., the Large Baths were still in use. On the decumanus connecting the Large Baths with the Capitolium and the theatre, a nymphaeum was built in the first half of the same c., probably under Valentinian and Valens (map 22).453 Two other nymphaea were built, one on the cardo south of the Capitolium and one on the decumanus leading to the amphitheatre.454 It is unclear whether the amphitheatre was still in use, but the theatre was restored.455 Test pits in the area around the amphitheatre revealed considerable building activity, possibly several wealthy villas. 448  This change might well have been commemorated in the inscription, see above. 449  Thébert (2003) 156. 450  Kuhoff (2010) 282. 451  Duval and Baratte (1973) 74. 452  Thébert (2003) 156. A E (1958) 158/159; Duval and Baratte (1973) 30. 453   454  Duval (1982) 612. 455  ILAfr 116.

Archaeological Evidence

figure 43 Plan of the Large Baths in Sufetula (Tunisia) after Nielsen (1993b) 170, fig. 189

161

162

chapter 3

figure 44 Plan of Baths nr. 15 in Sufetula (Tunisia) after Thébert (2003) 602, pl. LII.1

A structure with what seems to be an apsidal pool may point to a public bathhouse in this neighbourhood.456 The oldest Christian monuments are also attributed to the first half of the 4th c., with Basilica 4, just north-west of the Capitolium, dedicated at the end of the 4th or the beginning of the 5th c. The same date can be proposed for Basilica 1 and the adjacent baptistery. Basilica 3 replaced a public building.457 There seems to be a hiatus in the building activity of baths in the excavated part of town. Only at the end of the 5th c. was a new large bathhouse near Basilica 2 (C55) constructed (map 23). It is not surprising that a bathhouse was built near such an important new social hub. In the same period, the quarter around the eastern exit road seems to have developed, with two new ‘neighbourhood baths’—Bath nr. 26 (C56) and the Bath near the new Basilica 5—newly erected there. Some 3 km south-east of the city, the remains of a 5th c. basilica (Basilica 7) have been discovered.458 The occupation in the north-west quarter continued throughout the 5th and even 6th c. (maps 23, 24). 456  Duval (1982) 614–15 and pl. XII, 18. 457  Duval (1971a) 85, 131 (basilica 1), 379 (basilica 4); Duval (1982) 617 (basilica 3). 458  Duval and Baratte (1973) 110.

During the 6th c., there were no radical transformations in the town’s layout (map 24). The large public buildings such as the theatre and the amphitheatre were probably abandoned at this time. The Capitolium had lost its original cult function, yet the open square in front of it was still used, as a Byzantine wall around it suggests.459 Basilicas 4 and 5 continued to serve the Christian communities. Nothing is known about the lifespan of the baths. However, the south-east quarter of town which had already been developed in the 5th c. saw some building activity after the Arab raid of AD 642.460 Fortified houses were built along decumanus entering the city in the east. This important traffic artery was kept in use by raising the street level. Both Bath nr. 26 and the Bath near Basilica 5 may well have continued to function. Other roads fell out of use, however, as olive presses were installed on them. Some parts of the town lay in ruins or were used as cemeteries, creating a picture of “scattered nuclei.”461

459  Béjaoui (2004) 44. 460  Duval (1982) 619–25. 461  Leone (2007) 184.

163

Archaeological Evidence

map 23

Sufetula (Tunisia) at the end of the 5th c., with the location of public baths and the most important buildings after Duval and Baratte (1973) fig. 2 1: Basilica II 2: Basilica I 3: Basilica V 4: Baths near Basilica V

Ptolemais Concise History of the Urban Fabric The city of Ptolemais was founded in the Hellenistic period, probably between 250 and 160 BC. It developed as a classical Greek polis, governed by a boulè. Located in the fertile plain between the coast and the northern escarpments of the Jebel Akhdar (‘the green mountains’), the city prospered thanks to trade in agricultural products

(wheat, olives), timber and wood.462 The location of the site was not a coincidence. Ptolemais was initially the harbour of the inland settlement of Barca (modern alMarj) and was possibly chosen for “the sure footing for building, the excellent opportunities for drainage, the relative degree of protection against the torrid Ghibli winds from the Sahara, and the tendency of the cooling 462  Kraeling (1962) 2–7.

164

map 24

chapter 3

Sufetula (Tunisia) at the end of the 6th c., with the location of public baths and the most important buildings after Duval and Baratte (1973) fig. 2

inshore breezes to counteract the heat of the summer, usually before midday”.463 In 96 BC, the Hellenistic king Ptolemy Apion bequeathed Cyrenaica to the Romans. However, it was not until the reign of Augustus that the whole province benefitted from integration into the Roman empire.464 During the Roman period, the original communication line with Barca was abandoned in favour of a coastal route connecting Ptolemais with Apollonia to the east and Taucheira to the west.465 463  Kraeling (1962) 33. 464  Reynolds and Lloyd (1996) 619–21. 465  Kraeling (1962) 14.

The city of Ptolemais flourished in the first 150 years of imperial rule. The city plan probably dates from the Hellenistic period, with a defensive wall stretching from the coast to the foot of the Jebel Akhdar. Inside these walls, the two north-south axes connected the harbour district to the city centre, while several smaller northsouth and east-west streets divided the remaining town into equal insulae.466 The archaeological evidence from this period is meagre: deep test pits in the ‘villa with a view’, carried out by a Polish team in recent years, have yielded 2nd c. BC remains, while the upper theatre and 466  Kraeling (1962) 37–39.

Archaeological Evidence

the stadium may also have a pre-Roman date.467 From the first decades of the imperial period, several private houses have been found underneath late antique remains, for example in ‘House G’, in the House of Paulus (see below), in the ‘Palazzo delle Colonne’ and in the Villa of the Four Seasons.468 The bridge over the eastern wadi and the aqueduct also date from the early Roman phase, but precise dates are lacking. The forum may have been situated just north of the odeum, itself perhaps a 1st or 2nd c. construction.469 Other structures that date from the High Empire include the ‘Byzantine theatre’ and the square of the cisterns. The latter consisted of a large square, possibly a Greek-style gymnasium, built upon a vast cistern of 17 interconnected vaulted chambers. Its presumed capacity exceeded 5 million litres.470 Recently, a geophysical survey by a Polish archaeological team discovered a structure that could be interpreted as a large cistern near the harbour.471 In contrast to the North African regions to the west, the period of the Severans does not seem to have been one of extraordinary expansion.472 Under Diocletian, however, the city became the capital of the province of Libya Pentapolis, surpassing Cyrene as most important city of the region (map 25). The presence of an administrative bureaucracy meant that Ptolemais became the focal point of trade and communications.473 During the 5th c., local tribes revolted and besieged the city on numerous occasions. Therefore, the imperial government decided to move the administrative capital to the more defensible Apollonia. City life still continued, albeit probably on a smaller scale, and urban activity was still mentioned by Procopius at the beginning of the 6th c.474 A renewal program under Justinian may have reorganized the shrinking town. In AD 643, the city fell under Arab rule and probably continued to serve as a harbour town until at least 14th c.475 Newly Built Late Antique Baths The examples of new baths in Ptolemais are fundamentally different from each other. The City Baths (C71) date from the 4th c. and show us how a new large public bathing facility was constructed along the most conspicuous street of the city. The bath of the House of Paulus (C72) 467  Kenrick (2013) 90. 468  Pesce (1950); Kraeling (1962) 119–39. 469  Arthur (1975); Goodchild and Kraeling (1962) 93. 470  Knudstad (1962) 62–67. 471  Misiewicz et al. (2010) 197. 472  Kraeling (1962) 18. 473  Kraeling (1962) 20. 474  Procop. Aed. 6.2.9–11. 475  Kraeling (1962) 29.

165 on the other hand show us how a former private bath was transformed into a small public bath in the late 5th or early 6th c. This difference can also be detected in the type of plan and the decoration. The City Baths had an impressive frigidarium centred on an octagonal pool, a tepid room and a cruciform caldarium (in the first phase). The bath of the House of Paulus had a very functional layout with the basic succession of rectangular cold, tepid and hot rooms. The reorganisation of the heated section of the City Baths at a later date shows some of the same characteristics that can also be found in the ‘Byzantine’ baths in Taucheira (C73) or Apollonia (C66), namely a caldarium in which two alvei flanked a boiler room. A similar arrangement can be seen in the baths of the House of Paulus, although in this case, there was only room for a single alveus next to the boiler room. The addition of several small single-person pools to the existing communal pools, both in the cold and heated rooms, is a development that we can also see in the 5th and 6th c. baths of Egypt (e.g. C75C, C78, C81, C87, C89, C95) and Palestine (e.g. C98, C99, C100, C114). The decoration of the City Baths was obviously supposed to impress. Reused slabs of marble, fluted columns with new Corinthian capitals and old statues brought in from other locations in town created an environment of luxury and tradition. However, in the second phase, these high standards could not be maintained and the new small pools were only coated in waterproof mortar. In the baths of the House of Paulus, the transformation of the private into a public facility did not entail an upgrade of the decoration. The small pools were simple tubs with a waterproof coating. The hypocaust pillars of the second-phase City Baths were constructed with standard square tiles, although later restorations made use of round tiles. Such round tiles were also used in the baths of the House of Paulus. This could suggest that both baths were in use at the same time, although only a close comparison of both sets of tiles could confirm if these were made at the same production site. The water supply of the City Baths probably relied on a combination of supply systems. The large reservoirs may have been fed by rainwater catchment from the rooftops or by an aqueduct, such as the branch found coming from the street and running along the niches of the frigidarium. The small well in the praefurnium of the original caldarium was also reused, possibly for cleaning rather than supplying the pools. The drainage channel of the octagonal piscina did not pass the latrines, which had a separate sewer heading west. In the baths of the House of Paulus, the large reservoir in the courtyard may also have been fed by rainwater and / or a branch of an aqueduct. The waste waters of

166

map 25

chapter 3

Ptolemais (Libya), plan of the 4th c. city with the location of public baths and the most important buildings after Blas de Roblès (1999) 103, unnumbered fig. 1: Arch of Constantine 2: West Central Church 3: Odeon 4: ‘Palazzo delle Colonne’ 5: Temple 6: Large cisterns 7: Large cisterns 8: Fountain of the Maenads 9: Amphitheatre 10: ‘Byzantine theatre’ 11: Stadium 12: Upper theatre

167

Archaeological Evidence

figure 45 Plan of the Baths of House T in Ptolemais (Libya) after Ward-Perkins et al. (1986) 141, figs. 15 and 19

the pools were collected in a sewer which flushed the latrines on its way to the street sewer. Continued Use and Abandonment of Existing Baths For an important city such as Ptolemais, it is remarkable that no bathhouse dating from the High Empire has been found. It has been argued that the City Baths were built over an imperial predecessor, although archaeological evidence to support this hypothesis is lacking.476 Two private baths dating from the imperial period have been identified. The Baths of the House of Paulus were transformed into a public bathhouse in Late Antiquity (see above p. 165). The Baths of the ‘Colonnaded Palace’ were destroyed at the same time as the luxurious house.477 The violent collapse of heavy architectural elements suggests that the house was abandoned after an earthquake, possibly the one of AD 365.478 Private Baths The Baths of House T (Triapsidal Hall) The excavations of this large domus were interrupted when the excavator passed away and much of the excavation data was lost. Re-examination of the structures and the surviving records were only published at a later date.479 The house was originally built in the High Empire, but was reorganized profoundly in 476  Kraeling (1962) 175. 477  Pesce (1950) 49. 478  Kenrick (2013) 82. 479   Ward-Perkins et al. (1986).

Late Antiquity.480 A large triapsidal hall was the most important element of the new layout. The peristyle was equipped with a large fishpond and a small bath was added in the south-west corner of the building, opposite the original main entrance (fig. 45). The bathhouse was constructed just opposite the main entrance to the domus. This could point to a public use of the baths. However, this door was blocked at a certain point, perhaps when the baths were constructed.481 If this was the case, the use of the baths was obviously restricted to the inhabitants of the house. The bath consisted of a small vestibule, a cold room and a heated room. Both the cold and warm rooms had a pool. It seems that the baths were never finished and that the furnace was never fired. Several cisterns or possibly large wells have been discovered in the house. When the baths were constructed, a water supply pipe entered the house near the north-western corner of the baths and possibly fed the fish tank and the garden.482 Baths in the Late Antique Urban Context When Ptolemais became the provincial capital under Diocletian, the city was embellished with monuments fit for its status (map 25). The so-called ‘Street of Monuments’ was adorned with a triumphal arch dedicated to Constantine and Licinius (AD 311–313). Furthermore, it was equipped with a colonnade under 480   Ward-Perkins et al. (1986) 126. 481  Kenrick (2013) 78. 482   Ward-Perkins et al. (1986) 140–42.

168 Valentinian I (AD 367–375) and subsequently expanded or restored under Arcadius and Honorius (AD 395–408).483 Statues—the bases of which can still be seen—were erected alongside this street. Not surprisingly, the City Baths were a part of the ‘monumentalisation’ of the city centre. A Hellenistic puteal (decorated kerb-stone of a well), carved out of Pentelic marble, was relocated to a fountain on the same street. The fact that a new public fountain was some 40 m east of the baths could point to a restored or enhanced water distribution in this part of town.484 Around the same time, a project for a private bath was started in the House of the Triapsidal Hall, while the odeum was used as a venue for swimming displays.485 The large cisterns in the southern part of town were still in use. The city’s water storage capacity was even enhanced by the construction of two new large reservoirs near the old cisterns.486 The largest of them seems to have been an open square that was reused by coating the walls with a waterproof mortar.487 Perhaps the expansion of the city, boosted by its new status of a provincial capital, encouraged the authorities to anticipate possible periods of drought, a recurring problem in the region. Few building activities can be dated to the 5th c. The testimonies by Synesius of Cyrene, bishop of Ptolemais from AD 410 onwards, mention a period of turmoil and tribal raids.488 The Villa of Orpheus, named after a mosaic in the triclinium, has been attributed to the early 5th c.489 Most of the public buildings probably fell out of use at the end of the 4th or the beginning of the 5th c.490 The aqueduct may have suffered the same fate, although the City Baths were still functioning. One could imagine that their water management was adapted to the new circumstances, for example by relying more on rainwater storage. The Byzantine period saw increased building activity (map 26). Several fortified ‘blockhouses’ were built throughout the city; however, none of these has been excavated.491 Their ‘Byzantine’ construction date has only been deduced from the type of building (see above, e.g. Sufetula) and by the fact that their position diverges 483  Resp. SEG IX 71, nr. 364 and SEG IX 71, nr. 365. 484  Kraeling (1962) 81. 485  Kenrick (2013) 92. 486  Kraeling (1962) 68–73. 487  The interpretation of this structure as an open-air reservoir has rightfully been challenged in modern research. Some scholars tend to identify it as a defensive structure (Misiewicz et al. (2010) 202). 488  Synesius, Opera quae extant omnia. 489  Harrison (1962). 490  Kraeling (1962) 73. 491  Kraeling (1962) 100–107.

chapter 3

slightly from the city grid.492 The Street of Monuments remained an important axis within the city, as its eastern end was adorned with a tetrapylon. The private baths of the House of Paulus were turned into a public facility (C72), perhaps because the City Baths had temporarily closed or could not function at full capacity. In one of the later phases, probably near the end of the 6th c., the water management of these baths also changed. At this point, the aqueduct and the large reservoirs in the southern part of town had perhaps ceased to function. In the prolongation of the street, the ‘Fortress of the Dux’ was constructed as a fortified stronghold. An edict of Anastasius (AD 491–518) once stood in front of it.493 It has been proposed, therefore, that the building was a residence of the military governor of the province.494 Byzantine building phases have also been discovered inside some of the earlier houses, such as in ‘House G’ or the House of the Triapsidal Hall. From the three churches that have been identified in Ptolemais, not one has been studied.495 It is believed that they were built during the Byzantine (Justinianic) ‘revival’ of the city, but an earlier date cannot be ruled out. The Central West Church still awaits excavation. The West Church has been labelled as a ‘fortress church’ due to its unusually solid construction.496 The building was reused in the Early Islamic period, as some Arab graffiti reveal. Other traces of early Arab occupation are found throughout the city, often modifications of existing houses.497 Even if the Arab conquest did not mean the abandonment of Ptolemais, the last recorded evidence of activity inside the City Baths and the Baths in the House of Paulus date from the first half of the 7th c. No baths from the Early Islamic period have been discovered yet. Scythopolis Concise History of the Urban Fabric The city of Scythopolis (modern Baysan or Beth She’an) was located in the fertile Jezreel valley, on the crossroads of a route connecting Syria and the Far East to the Mediterranean. A settlement was already on this location in the 2nd millennium BC. The city steadily grew under the Hellenistic kingdoms, first under the Ptolemies and then under the Seleucids.498 It was 492  Misiewicz et al. (2010) 202. 493   S EG IX 68–70, nr. 356. 494  Kraeling (1962) 101. 495  Kraeling (1962) 97–100. 496  Kraeling (1962) 97. 497  Kenrick (2013) 69. 498  Liwak (2006).

Archaeological Evidence

map 26

Ptolemais (Libya), plan of the 6th c. city with the location of public baths and the most important buildings after Blas de Roblès (1999) 103, unnumbered fig. 1: Tetrapylon 2: ‘Palace of the Dux’ 3: House T

169

170 Pompey (63 BC) who took the city and established Roman control. Little is known about the further development of Scythopolis in the Roman period, although according to Josephus the city was sacked during the First Jewish Revolt (AD 66–70).499 The Roman city developed around the foot of the acropolis (the tell), upon which the Temple of Zeus Akraios was built. The residential area was established on the slopes of the hill without following a regular grid.500 Only a small part of the public infrastructure has been dated to this first phase, including the basilica, theatre, the temple with the round cella and a bathhouse. The city was monumentalized during the reigns of Antoninus Pius and Marcus Aurelius, adding other temples, an odeum, a monumental nymphaeum and a ‘reflecting pool’ to the city centre.501 To the south of the city, a hippodrome was constructed in an area that had been uninhabited at that time. No forum or macellum has been identified to this day. A large square surrounded by a portico that has sometimes been labelled the forum should rather be interpreted as a large temple complex. The street network of this High Imperial phase is difficult to reconstruct, as the urban fabric was heavily altered during Late Antiquity. Several buildings were also dismantled to make way for new structures (see below).502 In Late Antiquity, the city reached its largest expansion after a period of relative peace (maps 27 to 29). The first Christians in the city are attested by written sources in the early 4th c., yet classical culture was still very much present in the city until at least the early 6th c.503 In AD 363, an earthquake hit the entire Palestinian region. Even if the catastrophe is documented by the sources, archaeological traces of this event are hard to find. The impact on the city was gradually obliterated as the municipal authorities were able to restore most of the damage. The city continued to grow until it reached its zenith in the first half of the 6th c. A revolt by the Samaritans (AD 529) and the Justinianic plague (AD 542 and after) disrupted this period of prosperity. It seems that the city council (curia or βουλή) had lost much of its administrative power, leaving the initiative with the local elite. These preferred to invest in ecclesiastical property rather than the public facilities. When the Sasanians took the city in AD 614, the ruling class was 499  Josephus, BJ II.458. 500  Tsafrir and Förster (1994) 99. 501  Tsafrir and Förster (1997) 92. The ‘reflecting pool’ was a shallow pond that acted as a mirror to reflect the surrounding architecture. 502  Tsafrir and Förster (1997) 89–97. 503  Christians: e.g. Eus. De Mart. Palest. 1.1–2.; Tsafrir and Förster (1997) 127–32.

chapter 3

cut off from the main imperial administration and had to rely solely on local resources to survive. The city fell into a gradual decline, making it an easy target for the Arabs in AD 635. Under Muslim rule, the city continued to be inhabited by the Christian and Jewish communities that had lived there for centuries. According to later Arab sources, the conquered cities had to hand over half of the houses to the new rulers and had to pay an annual tribute. Scythopolis, now called Baysan, continued to function as a bustling market town, be it on a smaller scale, until another earthquake in AD 749 destroyed most of the remaining infrastructure.504 Newly Built Late Antique Baths Both the Southern and Western Baths (C114, C115) seem to date from the 4th-c. reorganisation of the city. The large Western Baths were a real prestige project, offering a new large public facility on the most important street of the city. The high number of pools, especially the natatio and piscinae, could point to the large capacity of the baths, but also stresses the intended opulence of the building. Similar large pools can be identified in the smaller Southern Baths. Both baths also had several heated rooms, which seems to rule out problems of fuel supply. The apsidal room that was added to the peristyle in the Western Baths could have played a role in the non-bathing activities that are known to have taken place in baths, such as religious meetings and juridical processes (see above, chapter 3, esp. Augustine, Socrates Scholasticus and John Malalas). The decoration is quite basic, with plain white mosaic floors (for example in the pools) or black-and-white mosaics with geometric patterns. There is evidence for wall mosaics with glass tesserae in the Western Baths. Local limestone and shale was also used to pave floors. The columns with their Corinthian capitals gave the Western Baths a monumental outlook. The hypocaust pillars of the Western Baths were somewhat particular, as they consisted of a conical base on which round tiles were piled. In the Southern Baths, the hypocaust pillars were made with rectangular and round pillars. The tubuli in both baths were rectangular in section. The water supply of both complexes is difficult to reconstruct. In the Southern Baths, there must have been a rainwater collection on the rooftops, as several gutters were identified.

504  Following the urban evolution as reconstrcuted by Tsafrir and Förster (1994) and Tsafrir and Förster (1997).

Archaeological Evidence

Continued Use and Abandonment of Existing Baths The Baths under Valley Street The remains of a bathhouse, probably dating from the 1st c. AD, were discovered underneath Valley Street (map 27, nr. 8). The baths were out of use by the time the street was laid out in the 2nd c.505 The Eastern Baths The 2nd c. baths underwent important changes during the 4th c. and were used until the end of the late antique period (map 27, nr. 9). In AD 515–516, the socalled Silvanus Hall was built partially on top of the now abandoned baths. The late antique changes included the addition of the so-called ‘pillared hall’, possibly with a fountain in an apse, and the blocking up of several doorways. In one of these doorways, a small basin for ablutions was installed at a late date.506 The collection of statues portraying gods of the classical pantheon was probably assembled during this late phase. Several of these statues seem to have been dumped in the ravaged hypocaust once the baths had fallen out of use. The excavators interpreted the deposition as an intentional act by the Christian inhabitants of the city, although it cannot be ruled out that the statues were damaged during one of the numerous earthquakes that hit the city and then subsequently dumped in the ruined hypocaust as part of the levelling layer.507 Baths in the Late Antique Urban Context The late antique phase of Scythopolis is well-documented through the excavations.508 The single-most important event for the urban fabric in the 4th c. (map 27), was the earthquake that hit the Palestinian region in AD 363. Even if it is risky to connect archaeological data to precise historical events, the way in which some buildings were damaged, in addition to 4th-c. occupation layers, point to an earthquake around this time. The basilica was destroyed and never rebuilt. The theatre, the reflecting pool and the nearby colonnaded street, on the other hand, were restored to former glory. The Eastern Baths were not only restored, they were expanded. The most important new construction involved the creation of the monumental colonnaded Palladius Street connecting the southern foot of the Tell with the theatre. Along its western side, the erection of the Western Baths underlined the continued importance of buildings of leisure 505  Tsafrir and Förster (1997) 98. 506  Tsafrir and Förster (1997) 123; Mazor and Bar-Nathan (1998) 12–13. 507  Tsafrir and Förster (1997) 129. 508  The articles by Tsafrir and Förster (1994) and Tsafrir and Förster (1997) will act as are key sources in summarizing the late antique urban evolution hereunder.

171 and hygiene. At least 4 aqueducts have been identified for late antique Scythopolis, although their route within the city could not be traced. In the area between the theatre and the hippodrome, a new residential quarter was developed, which would continue to expand in the 5th c. It is around this time that the Southern Baths were constructed to serve one of these new neighbourhoods. During the second half of the 4th c., the hippodrome was transformed into an amphitheatre. At the start of the 5th c., the importance of the area between the Tell and the theatre was confirmed by the construction of the so-called Byzantine Forum (map 28). North of it, the monumental nymphaeum was restored. A new water supply channel ran through the temple with the round cella, demonstrating that the latter had been abandoned before this renovation. Similarly, the Temple of Zeus on the Tell had been abandoned by the late 5th c., as a church was built partially over it. The wall surrounding the city and incorporating the Imperial era gates was probably built in the 4th or 5th c. In the first half of the 6th c., Palladius Street was restored and a semicircular plaza, the so-called Sigma Plaza, was constructed partially on top of the newly ruined odeum. A new colonnaded street started from the nymphaeum and ran east to a new large hall. Two small churches were built north of the city, around Tell Iztaba, and a monastery dedicated to Lady Mary and another to Abba Justinus were constructed to the north-west. A synagogue north of the Byzantine wall also belonged to the period of the 5th to 7th c. (map 29). In the second half of the 6th c., building activity slowed down. The walls were renovated under Justinian and the Baths of the Lepers, known by an inscription (EP-1), were restored in AD 558. However, several public buildings were abandoned, probably as the result of poor maintenance and the inability to carry out necessary restorations. The Eastern Baths, the Southern Baths, the theatre and the amphitheatre all fell out of use. At the end of the 6th c., Palladius Street had been covered by a layer of colluvium. The Western Baths persisted until the arrival of the Arabs. At the turn of the 6th and 7th c., an earthquake heavily damaged the Silvanus Hall, the Byzantine Forum and the Sigma Plaza. These buildings would never recover. By the second half of the 7th c., most of the public monuments were transformed into industrial activity centres. Furnaces and workshops were discovered inside the Western Baths and on the Byzantine Forum. A row of shops was constructed before the nymphaeum and underneath the portico of Silvanus Street, giving the street a souk-like appearance. Churches and the synagogue were still in use when the earthquake of AD 749 put an end to the remaining monumental infrastructure.

172

map 27

chapter 3

Scythopolis (Israel), plan of the 4th c. city with the location of public baths and the most important buildings after Tsafrir and Förster (1997) fig. C 1: Temple of Zeus on the acropolis 2: Large temple complex 3: Odeon 4: Temple with round cella, nymphaeum 5: Reflecting pool 6: Theatre 7: Amphitheatre (ancient circus) 8: Baths under Valley Street 9: East Baths

Archaeological Evidence

map 28

Scythopolis (Israel), plan of the 5th c. city with the location of public baths and the most important buildings after Tsafrir and Förster (1997) fig. C 1: Market 2: Palladius Street 3: ‘Byzantine’ Forum 4: Church

173

174

map 29

chapter 3

Scythopolis (Israel), plan of the 6th c. city with the location of public baths and the most important buildings after Tsafrir and Förster (1997) fig. C 1: Sigma plaza 2: Church of the Martyr 3: Monastery of the Lady Mary 4: Andreas’ Church

175

Archaeological Evidence

Archaeological Evidence of Early Hammams North Africa and Cyrenaica The archaeological evidence for public baths in 7th c. North Africa is rather flimsy (see Gazetteer). In Carthage, the Byzantine Baths and the Baths in the Baal Hammon Street may still have been in use until the late 7th c. (see appendix 3B), but there is no evidence for any type of architectural transformation. Similarly, the Baths of the 5th Century in Sitifis were still standing in the 7th c., although the building may not have been used as a bath at this time. In Cyrenaica, the Byzantine Baths at Taucheira were still in use in the Early Islamic period. Unfortunately, a large part of the excavation data has been lost, making it difficult to discern possibly Early Islamic alterations to the original design. Evidence for the new construction of baths is even scarcer. In Thugga, a small bathhouse at the foot of the forum may date to the Early Islamic period.509 The only other confirmed example of a hammam is found in Volubilis, renamed Walila (Morocco). The so-called extra-muros baths (fig. 46) were dated to the late 8th c. by coins and ceramics found underneath the pavement of the cold and heated rooms. The plan was of an angular row type, with a large cold room, two tepid rooms and a heated room on hypocaust. The heated room of these baths strongly resembled those of the Byzantine baths found in Taucheira and Apollonia (Fortress baths, C66, C73). The room had a hypocaust, but no wall heating and was heated by a central furnace flanked by two single-person pools.510 We could even assume that the architects of the extra-muros baths were inspired by the last form of Byzantine bath types in North Africa. It is also important to note that both construction techniques and building materials do not differ from Roman period materials. In fact, the bricks are reused from Roman buildings.511 This continuity in construction techniques has also been identified in other water-related structures, such as aqueducts and cisterns, using the same type of hydraulic mortar as Roman predecessors.512 Without the stratigraphic information from the trenches, these baths would have been difficult to assign to the Early Islamic period. There is no archaeological evidence for early 509  The baths, as yet unpublished, were presented by C. Touihri during the conference ‘Byzantine and Medieval Islamic Baths and their use 600–1200 AD’ in Rome (5–6 June 2014). They have been preliminarily dated to the Early Islamic period, but could also date from the Late Byzantine period. It is unclear whether the baths described by C. Touihri are the same as the ‘private baths’ described by de Haan (2010) 258–60 (K.35). 510  Khayari (1994) 302–305; Fentress and Limane (2010) 110–111. 511  Khayari (1994) 306–307. 512  Wilson (2003) 129; also Ugurlu and Böke (2009).

figure 46 Plan of the Extra-muros Baths in Volubilis (Morocco) after El Khayari (1994) 303, fig. 2

hammams in the newly-founded cities of Kairouan and Tunis. The written evidence for this period is limited to the work of the 9th c. historian Ibn Abd-al Hakam. Egypt Both the archaeological (appendix 3D) and the papyrological evidence (appendix 2) reveal that several Romano-Byzantine bathhouses were kept in use after the Arab conquest. There is no evidence for radical transformations of the architecture of these existing baths, which is understandable, as the same people as before were using them.513 Indeed, the Arab colonists that settled in Egypt mainly lived in Fustat and Alexandria. The excavations at Fustat have not yielded remains of early hammams, even if the water supply of the city was an important element of the urban infrastructure from its earliest phases.514 The 14th-c. historian Ibn Duqmaq mentions a small hammam—hammam al-Far or Baths of the Rat—being erected during the foundation of the city.515 To this day, no remains of 7th or 8th-century hammams have been discovered in Egypt.516 Palestina Several public baths in Palestina survived the Arab conquest (appendix 3E). Some larger complexes such as in Tiberias (until the 11th c.) or the thermal baths of Hamat Gader (8th c.) were used and adapted until well after 513  Denoix (2009) 24. 514  Gayraud (1998); Sheehan (2010). 515  Ibn Duqmaq IV.104–105; Denoix (1992) 73–80. 516  Boussac et al. (2014) 18–20.

176

map 30

chapter 3

Umayyad baths in the Middle and Near East plan by the author after Boussac et al. (2014) 38–39, maps 3–4

the period that is central to this study. Some smaller baths may even have dated from the Early Islamic period, rather than to the late antique phase (Ha’on, Nahf, the Ophel baths and the possible baths in Beit Hanina in Jerusalem). The construction material and building techniques certainly do not permit an identification of these baths as hammams. There is some evidence of newly constructed hammams in Palestina, although their construction date lies in the 8th rather than the 7th c. (map 30). The baths near the Umayyad palace of Khirbet al-Minya, now disappeared, could be one such example.517 A well-studied example, is the (semi-public?) bath of Khirbet al-Mafjar (fig. 47).518 The baths were attached to a large audience hall, which has sometimes been interpreted as a banquet hall used to entertain political allies.519 In this hall, there was a large pool (19.5 × 3.4 m) made of local bituminous 517  Grabar (1993) 1051. 518  Hamilton (1959) 45–105. 519  Ettlinghausen (1972) 17–65; Tohme (2011) 72–73.

limestone and coated on the inside with a type of hydraulic mortar. To the north of the hall, a small bathing suite with two unheated rooms ended in a heated rectangular room with a semicircular niche and an octagonal warm room with a horseshoe-shaped niche on each side, presumably used as individual cells to relax. Both rooms were heated by a hypocaust—the brick-built pillars ending in arches—and tubuli. The rectangular room was heated by a furnace to the east, surmounted by boilers—possibly in metal—in the alcove, which must have generated steam (fig. 48). A water reservoir above the furnace of the octagonal room served a similar purpose, while simultaneously supplying hot water for the basins lodged in the niches flanking the furnace.520 The architecture, the heating technology as well as the decoration—figurative and geometric polychrome mosaics—show clear influences of Romano-Byzantine and Sasanian culture.521 520  Hamilton (1959) 49–57. 521  Arce (2007) 497–506.

Archaeological Evidence

177

figure 47 Plan of the palace at Khirbet al-Mafjar (Palestine) with location of the baths after Yegül (1992) 345, fig. 435 based on a plan by Hamilton (1959)

The Larger Middle Eastern Context Although Lebanon, Syria and Jordan do not fall within the geographic limits of this study, it is nevertheless important to have a quick look at some of the baths that were still in use during the Umayyad period (map 30). Archaeological research in Syria in particular over the past decade has revealed priceless new information about some of the mechanisms of transition between Romano-Byzantine and Islamic baths. In Lebanon, two early hammams have been found in ‘Anjar. In what was described as the first real Umayyad city, a palace, a mosque, an administrative neighbourhood, commercial areas and private residences were laid out along an orthogonal street grid with two main traffic arteries that met at a tetrapylon in the centre of the city.522 A wall enclosed the entire settlement with 4 gates oriented towards the cardinal points and semi-cylindrical defence towers. A small and a larger bathhouse were found near the north gate. The largest bath had a significant hall, its vault supported by 4 columns. There were three warm rooms, heated by a single furnace servicing a hypocaust. The second room had a semicircular apse to the east, while the third room had similar apses to the east and to the west.523 The smaller baths—which were never published—were built against the northern city figure 48 Baths of Khirbet al-Mafjar (Palestine), plan of the hypocaust system, furnace with reservoir after Hamilton (1959) 58, fig. 21

522  Hillenbrand (1999) 59–61. There has been some discussion about the Umayyad dating of the site, but recent scholars have convincingly argued that Anjar was built at the start of the 8th c. (Hillenbrand (1999) 62; for the discussion about the date, see esp. 59, n. 3). 523  Chehab (1963) 21–24.

178

chapter 3

figure 49 Plan of the Baths of al-Bara (Syria) in the Roman period after Charpentier (2014) 486, fig. 7

figure 50 Plan of the Baths of al-Bara (Syria) in the Islamic period after Charpentier (2014) 490, fig. 16

wall, near the mosque. It was similar in layout, with a large hall preceding three heated rooms. Several Romano-Byzantine baths in Syria were kept in use after the Arab conquest. The evolution of a Roman-style bathhouse to an Islamic one can be seen in the baths of al-Bara. The first phase of the building (fig. 49) was dated to the 5th c.524 The plan follows a specific type of layout found in the villages of the northern Syrian plateau.525 A large rectangular hall—possibly used for different types of social gatherings—had a small singleperson cold pool and was flanked on the southern long side by 4 bath rooms, two of which were heated by a hypocaust system. The last heated room probably had a warm water pool.526 The baths then continued to function without any important transformations until the end of the 8th c. (fig. 50), when the caldarium was transformed into a furnace room and a new hot room, without pool and wall heating but with three warm water labra, was created in the second heated room.527 The

new hypocaust system was constructed using monolith pillars. The furnace that heated it was surmounted by a small water reservoir (fig. 51). In the bottom of this reservoir, a round opening was covered by a metal sheet (diam.: 1.15 m), which stood in direct contact with the furnace underneath, enabling the water of the reservoir to turn into steam. This steam was then channelled from the top of the reservoir into the new heated room by a chimney opening.528 This transformation of the heating system shows an interesting mix of Roman elements and the new heating by steam technique that became characteristic for hammams. A similar mixed heating system was found in the baths of Halabiyya-Zenobia (fig. 52) and Qasr al-Hayr East (or Qasr al-Hayr al-Shaqri, fig. 53).529 Gérard Charpentier suspects that the introduction of the steam furnace should be identified as an Umayyad transformation of the original Roman heating system.530 The baths of Sura also started out as a Romano-Byzantine style bath (5th or 6th c.), but were equipped with a steam system during the Early Islamic period. The heated rooms, however,

524  Charpentier (2014) 469. 525  Tchalenko (1953) 25–28; Charpentier and Denoix (2011) 73–87. 526  Charpentier (2014) 470–75. 527  Charpentier (2014) 478–79. The cold-water pool received a new internal coating, blocking up the outlet at the bottom. Without any apparent form of drainage, this pool may have been used as a shower or as ornamental basin. A cylindrical basin, fed by a conduit coming out of the wall, was added next

to the pool to be used for simple ablutions (Charpentier (2014) 475–76). 528  Measurements and description by Charpentier (2014) 476–77. 529  Resp. Lauffray (1991) 113–30; Grabar (1978) 90–97. 530  Charpentier (2014) 480.

179

Archaeological Evidence

figure 51 Baths of al-Bara (Syria), cross-section of the furnace with superimposed boiler during the Islamic phase after Charpentier (2014) 491, fig. 19

figure 52 Plan of the Baths of Halabiyya-Zenobia (Syria) after Charpentier and Denoix (2011) 90, fig. 10

still had wall heating (tubuli) and three single-person pools.531 In Syria and Jordan, several original Umayyad hammams have been identified.532 The most famous example is the hammam of Qusair Amra (Jordan, early 8th c.), probably part of a larger rural estate of the Umayyad rulers.533 The baths consisted of a large three-aisled hall, with the side aisles ending in apsidal chambers (fig. 54). In the northern end of the eastern 531  Othman (2014) 499–502. 532  Boussac et al. (2014) 20–22. 533  Musil (1907); Almagro et al. (1975); Fowden (2004).

aisle, a shallow basin may have served for quick ablutions or simply as decoration. From this hall, a doorway led into a first room without any noticeable features (dressing room?). The following room was heated by a hypocaust and had a small rectangular cold pool. The last room, also with a hypocaust, had two heated semicircular pools. Above the furnace, water was converted into steam in a reservoir and then channelled into the heated room through a vent. The water was supplied by a saqiyah with adjacent reservoir to the north of the baths. The concrete domes and vaults of the building were decorated with polychrome wall paintings depicting hunting, bathing, athletics, dancing girls and figures

180

figure 53 Plan of the Baths of Qasr al-Hayr East (Syria) after Yegül (1992) 341, fig. 427

of philosophy and poetry. The references to classical themes, such as a woman bathing, wearing only a loin cloth, is reminiscent of the many Venus depictions in Roman baths.534 The complex, which stands in isolation in the desert, has therefore been interpreted as a ‘hunting lodge’ for the prince and his entourage. However, a recent survey of the immediate surroundings of the baths identified some small dwellings and a mosque, so “the bathhouse was originally less isolated than it now appears”.535 The baths of Hammam as-Sarah (Jordan, early 8th c.) have a very similar plan and water and heating technology.536 The bath of the Caliph’s Palace in Amman (Jordan) was accessible from the street as well as from the palace itself, meaning this hammam had a semipublic use.537 The plan (fig. 55)—with a large dressing room flanked by the triad of cold, tepid and hot rooms— showed obvious similarities to late antique baths such as al-Bara (see above). The baths of Qinnasrin (Syria) probably belonged to a larger building, perhaps a palace. The excavations, which have yielded successive phases from the turn of the 6th and 7th c. to the 9th c., made clear that a Byzantine bath 534  Fowden (2004) 38–42, 229–30 (paintings). 535  Fowden (2004) 46. For the survey, see Genequand (2002). 536  Butler (1919) 77–80; Arce (2015) 142–59. 537  Arce (2015) 160.

chapter 3

was replaced by a new bathhouse in the Early Islamic phase. The heated section was transformed into a cold section with a single-person pool.538 In the Umayyad fortress of Khirbet al-Maqsurah (Syria), a small bathhouse was found against the eastern defensive wall. The excavations so far revealed at least three rooms heated by a hypocaust, one of which had a rectangular and a semicircular heated single-person pool. The hypocaust pillars were made of tiles and ended in arches. The walls of the second and third warm rooms were heated by tubuli. The furnace may have had an overlying reservoir to generate steam, but the walls are too damaged at this location to allow for any such conclusions. To the south of the heated rooms, a cold room and a larger hall were discovered, reminiscent of the baths of the Umayyad desert castles.539 The baths at Gabal Says can be dated to the reign of Caliph al-Walid (AD 705–715) and lay some 120 m east of a fortress.540 The plan was less monumental than the baths at Qusair Amra or Hammam as-Sarah. The bathers entered a hall with an apsidal ending to the north and two side rooms to the south. Some limestone basins and an underground sewage channel in this hall point to the use of water. A first small room with a water basin to the west was followed by two rooms with hypocaust heating. The second heated room possibly had two rectangular heated basins. The furnace was surmounted by the reservoir-steam system that has already been described for the baths at Qusair Amra. The water was supplied by a well some 170 m north of the baths, possibly operated by a saqiyah.541 Summarizing the Archaeological Evidence of Early Hammams In the first phase of the Islamic conquest of former Romano-Byzantine territory, Roman-style bathhouses continued to be used, without any apparent transformations of the plan or the technology. Except for a single inscription in Taucheira, there is little evidence that the Arab rulers used or maintained these baths. The first hammams that can be dated to Umayyad rule seem to have been built under Caliph al-Walid (AD 705–715). These hammams were very similar to late antique baths, with a traditional pillar hypocaust and even small pools for individual use, but without a space for physical exercises.542 The hammams of the Syrian ‘desert castles’ or ‘hunting lodges’ had a large hall as their most characteristic element. The decoration, clearly influ538  Rousset and Rochette (2014) 520–25. 539  Omeri (2014) 510–12. 540  Schmidt (2012). 541  Bloch (2014) 586–88. 542  Fontana (2011).

181

Archaeological Evidence

figure 54 Plan of the Umayyad baths of Qusair Amra (Jordan) after Almagro et al. (1975) fig. 4

figure 55 Plan of the Umayyad baths attached to the Palace of the Caliph in Amman (Jordan) in Piraud-Fournet (2014) 708, fig. 17 after the plan by I. Arce; with permission

enced by Byzantine and Sasanian art, point to a social function of the baths, in a larger context of leisure and pleasure.543 Some scholars have even suggested that parts of the palaces and most likely the baths were accessible to the elite and entourage of the caliph residing on the estates.544 The baths of the Palace of the Caliph in Amman indeed point to a semi-public use of a private hammam. In the cities, existing Roman baths were often kept in use. The newly built baths can be difficult to attribute to Byzantine or Islamic rule simply because it was still the same people building and using the baths. The so-called ‘characteristic’ elements of early hammams, i.e. a large multipurpose cold room and a small 543  Dow (1996) 32–38; Tohme (2011) 72–73. 544  Genequand (2009) §28.

but steamy heated section, are in fact the continuation of an architectural evolution that had already begun in the small public baths of the Byzantine Near East.545 The transformation of Romano-Byzantine baths in Syria during the later 8th c. mainly concern the disappearance of the pools in favour of simple washbasins (ĝurn, pl. aĝrân), the disappearance of wall heating and the introduction of a steam-boiler (qadr).546 The change from pools to basins has been linked to the Islamic prescriptions that one should wash in flowing water.547 The disappearance of the wall heating may have been 545  Fournet (2012b) 230. 546  However, immersion pools (hawd, pl. ahawad) were maintained in Egyptian hammams (De Miranda (2010) 70). 547  Benkheira (2003) 400; De Miranda (2010) 69.

182 linked to the disappearance of production centres of ceramic building material or to the invention of different types of heating techniques.548 The introduction of a steam-boiler may have been one of these. It has been identified as an original Islamic invention, even if the caldaria of the smaller Romano-Byzantine baths may already have been a little ‘steamier’ than in the imperial period.549 In some Roman-style baths, there seem to have been experiments with introducing steam directly into the caldarium. The baths in Karanis (Egypt, see C85) or the small baths in the Cyrenaican hinterland (see C68, C69, C74) are good examples. Not surprisingly, experiments with new heating techniques and new architectural forms could be more easily achieved in small neighbourhood baths than in large (pre-existing) thermae.550 Several centuries before, technological and architectural experiments in small private baths led to the full development of the Roman-style bathhouse.551 548  Grotzfeld (1970) 57. 549  Charpentier (1995) 233, n. 61; Fournet (2012a) 332; Fournet (2012b) 230; Charpentier (2014) 481. 550  Fournet (2012b) 230. 551  de Haan (2010) 37.

chapter 3

The evolution from the Romano-Byzantine baths to hammams is less clear in North Africa and Egypt. However, it seems that there were no equivalents of the luxurious baths of the caliphs as found in Palestina and Syria. Indeed, in the following centuries, the hammams of the Maghreb and those of the Mashreq followed different architectural traditions.552 It is interesting to notice that some of the Islamic baths found in Andalusia (Spain) still follow the same architectural design found in the Maghreb (Walila), which in turn had been influenced by the ‘Byzantine’ type of baths found in Cyrenaica (Taucheira, Apollonia).553

552  Dow (1996) 14–16; Denoix (2009) 25, 30; De Miranda (2010) 69–70; Lamei (2012). 553  For a recent overview of the Islamic baths in Andalusia, see Espinar Moreno (2014) esp. 215–78 and Fournier (2016). Most of the surviving remains were constructed after the 12th c., but the basic scheme of rectangular rooms and hot room with one or two hot pools flanking the furnace-cum-boiler still stands in sharp contrast to contemporary hammams in Egypt and the Mashreq.

chapter 4

Continuity and Change in Late Antique Public Baths and Bathing Habits Architecture

Imperial Type The so-called ‘imperial type’ bathhouse, not to be confused with imperially funded baths (which in theory could have any type of layout), was only built in the largest cities of the empire. For the Italian Peninsula, their construction was also restricted to the early decades of the 4th c., when the emperors still resided there. It is hardly surprising that cities of imperial residence such as Rome (C21, C22), Aquileia (C1) and Milan (C5) were equipped with new ‘palaces of the people’. One can speculate if such baths were also constructed in Ravenna once it became the new administrative capital in the early 5th c., or if an existing bathhouse was restored or perhaps enlarged. The same question can be asked for the reign of Theoderic in the first half of the 6th c. The construction of several modest bathhouses (C18, C19, C20) demonstrates that the bathing habit was far from extinguished at this time, yet it remains doubtful whether the construction of a new imperial bath was financially possible or even advisable. Considering the modest size of the population and the continued existence of several earlier baths, a restoration and embellishment programme may have sufficed to gain the favour of the people. Such a programme seems to have been enacted in other large cities such as Rome and Ostia. This is confirmed not only by the archaeological remains, but also by the literary and epigraphic evidence

(chapter 2). Authors such as Ammianus Marcellinus, the Scriptores Historia Augusta, Ausonius (all 4th c.) or Olympiodorus (5th c.) mention the limited construction in the Urbs at the beginning of 4th c. and the subsequent restorations of existing buildings later that century. The inscriptions pertaining to imperial type baths (e.g. EI-2; 45; 46) also commemorate restorations, and not a fundamentis building for that matter. The situation is very similar for other regions. There are no new imperial type baths built in North Africa. The so-called Unfinished Baths at Leptis Magna may have been an exceptional attempt, but as their name indicates, the project was never completed. In Carthage, the Antonine Baths were kept in use (EA-31). The Thermae Maximianae mentioned by Jerome,2 inaugurated in the same year as the Baths of Diocletian in Rome (AD 302), might have pertained to restoration works on the Antonine Baths. If this passage refers to a new bathhouse, the fact that they were commissioned by Maximianus, an Augustus during the Tetrarchy, does not necessarily imply that the layout was of the imperial type. The same can be said for the several attestations of imperially funded baths found in the written sources (literature, laws and inscriptions). The numerous baths built by Justinian in the reconquered territories—as recounted by Procopius (see chapter 2)—could hardly have been of the imperial type. In none of the archaeological sites mentioned by Procopius have imperial type baths of the 6th c. been found. In Cyrenaica, there are no examples of imperial type bathhouses, not even for the High Empire. Even in larger cities such as Cyrene and Ptolemais, the late antique bloom did not result in the construction of imperial type baths. In Palestina, the late antique ‘monumentalisation’ of Caesarea Maritima did not entail the construction of imperial type baths. However, with the construction of the Western Baths (C115) in Scythopolis, at least one new imperial type bathhouse was built in this prosperous region. We should nevertheless note that with its 3,200 m2, the Western Baths could not compete with the sumptuous double symmetrical thermae found in Rome or Carthage. It is also good to remember that the imperial type was the exception rather than the rule, even during the High Empire. The historical trajectory of Cyrenaica, Egypt and Palestina, in particular the impact of Roman urban concepts, may partially explain the small number

1  See for example Brödner (1983), Nielsen (1993a) or Yegül (1992).

2  Jer. Ab Abr. 2317.

Plans of the Baths Even if there existed certain common characteristics in the layout, architecture and technical aspects of Roman bathhouses all over the empire (see chapter 1), modern scholars have discerned certain regional characteristics of baths.1 There is no reason to assume that for Late Antiquity, with its divergent historical trajectories for each region, these characteristic features disappeared in favour of a standard type of ‘late antique bathhouse’. If we want to study the architecture—and indeed all other aspects—of late antique baths, it is first necessary to look at the remains on a regional level, before attempting to distil any conclusions on an empire-wide level. Furthermore, we should not forget the chronological evolution within each regional dataset when comparing the remains.

© koninklijke brill nv, leiden, ��20 | doi:��.��63/9789004419421_006

184 of imperial type thermae in these rather prosperous regions.3 The long urban history that preceded Roman rule may have made it more difficult for Roman urban concepts, and especially baths as a Roman flagship, to be embraced.4 However, it is possible that an imperial type bath may still lie buried underneath the streets of modern Jerusalem or Benghazi. The texts of John Malalas (chapter 2, p. 61–63) remind us that emperors still saw the upkeep and restoration of (large) baths as an important aspect of their benevolence, especially in the larger cities such as Antioch and Alexandria. In Egypt, the only imperial type bath built in Late Antiquity was located in Alexandria (C77). This prosperous harbour town was the central hub for trade in Egyptian grain. The Kom al-Dikka baths were possibly not the only imperial type baths in the city (C78?), but so far the examples of the High Empire remain to be found. Two more imperial type baths were found in Egypt, one in Thebes (near the temple of Karnak) and one in Antinopolis (now vanished). The baths in Thebes were only abandoned in the late 4th c., when the entire neighbourhood was in decline. Other cities in Egypt may also have had imperial types, although, in the so-called extinct cities where much of the city fabric has been mapped, such as Hermopolis Magna, no large structures that may be interpreted as imperial type thermae have come to light. Papyrological evidence mentions several baths named after an emperor in Oxyrhynchus (see P-2; 10) and possibly Panopolis (see P-18), yet there is no way of knowing whether these had an imperial type layout. Even if the construction of imperial type baths withered after the 4th c., it would be wrong to assume that inhabitants of the late antique city did not bathe in these luxurious thermae any more. On the contrary, in important cities (administrative capitals, regional centres), the imperial baths, often prestige projects financed by the emperors themselves, had a better chance of being restored by the grace of the emperors than the more modest neighbourhood baths (see the epigraphic evidence). Furthermore, large cities could also afford to send envoys to the emperor asking to repair or built baths. The construction and subsequent restoration of the imperial type can be seen in Aquileia, Milan, Ostia, Rome, several North African cities, Alexandria, Scythopolis and even further east in Syria, Lebanon5 and of course Constantinople (see the evidence of the Zeuxippos Baths in chapter 2). In each of these cases, the eventual 3  Only three in Egypt (see below) and only two in Palestina (Western and Eastern Baths in Scythopolis). 4  For Egypt see Redon (2017b); for Palestina see Hoss (2005) 92–99; for the Middle East in general see Fournet (2012b). 5  See Fournet (2012b).

chapter 4

abandonment of the imperial baths followed a more general decline of the urban fabric. Row Type and Ring Type The row type was the most common choice in each of the discussed regions. The popularity for this linear succession of the main bathing rooms lay in its simplicity and its adaptability to the spatial constraint imposed by the surrounding urban fabric. The application of this type goes back to the oldest Republican examples.6 Indeed, for many baths that were inserted in an existing insula, the linear (or axial) layout of the rooms offered an obvious solution for fitting all of the rooms into a predefined space (e.g. C3, C11, C23, C35, C57, C67). The parallel row type was especially suited for rectangular spaces (e.g. C16, C41, C86). It had the advantage over the ring type that it did not need an exit-tepidarium between the caldarium and the frigidarium. In the eastern regions, the linear arrangement of the rooms was combined with a square or rectangular courtyard to form a specific ‘eastern’ layout (e.g. C65, C73, C89, C98, C109). The fact that the row types could be heated by a single furnace, placed in the axis of the rooms, was an additional advantage. This was especially useful in regions with a limited fuel supply and for small and fortress baths (e.g. C74, C100, C101, C106, fortress baths in Taucheira and Yotvata). The ring type became popular from the second half of the 1st c. AD onwards, probably to improve the circulation within the baths, by adding an exit-tepidarium between the caldarium and the frigidarium.7 It was often used when there was a large building plot available. This type of layout was especially popular in North Africa, where they were often constructed in new neighbourhoods created during the Roman ‘monumentalisation’ of a city.8 The type was rather uncommon in the rest of the Roman West.9 This trend continued in Late Antiquity. The number of ring types in the Italian Peninsula is very low (C32?), even if some baths from the High Empire might still have been in use (e.g. Curinga, Misterbianco). In North Africa, Cyrenaica and Egypt, the type is more common (C36, C37, C49, C59). Sometimes this layout was even chosen when the available building space was rather restricted and a row type might have been the more sensible choice (e.g. C61, C64). In Egypt, several baths were transformed from a ring or row type in a double (angular) row type (C75A to C75B–C, C91, C92). Such double baths, which could be double from the start (C76, C89, C90, C95), seems to be an original 6  Staccioli (1958) 273–74; Yegül (1981) 109–110. 7  Thébert (2003) 121. 8  Nielsen (1993a) 90–91; Thébert (2003) 355–57. 9  Thébert (2003) 358.

Continuity and Change in Late Antique Public Baths and Bathing Habits

Egyptian feature, possibly linked to a separation of sexes in two separate (yet often identical) wings.10 The fact that some of these baths lie en route to or near Christian pilgrimage sites (e.g. Marea, Abu Mena) has been seen as a possible reason why these baths in particular had such an obvious and strict separation.11 In Palestina on the other hand, the ring type was not popular.12 When the necessary building space was available, the (parallel) row type was preferred over a ring type (e.g. C111, C112). As a general remark, we can say that the choice for a specific type of plan was still mainly inspired by the size of the baths. This, in turn, could depend on a variety of factors, including available space, available funds for construction or the number of potential bathers in any given city or neighbourhood that was targeted. The most obvious choice for smaller baths was the row type, be it linear or angular depending on specific circumstances. For example, the double baths in Egypt often used ‘interlocking’ angular row types around a central furnace. However, the popularity of the ring type in North Africa, which had developed during the High Empire, was a regional preference that continued to influence the choice of plans in Late Antiquity. From Cyrenaica to Palestine—and further east to Syria and Cilicia—a ‘block type’ consisting of a courtyard attached to a row type, became popular in Late Antiquity. This new typology fitted within a larger bath-related architectural evolution in the east, which saw an increased importance of the cold section (see below, p. 187–188). Size of the Baths If we group late antique baths into three categories— namely baths larger than 1,000 m2, baths between 500 and 1,000 m2 and baths smaller than 500 m2—we get similar patterns for the different regions (all periods mixed) and a clear difference between the three categories.13 The number of baths with a surface larger than 1,000 m2 is limited for every region. In the Italian Peninsula, imperial type baths influence the slightly higher number. Besides these imperial types, few bathhouses in these regions exceeded 2,000 m2. In the High Empire, such large baths were not uncommon, even in smaller cities.14 Especially in North Africa, baths with a surface of over 2,000 m2 were frequently constructed in the 2nd and 10  Fournet and Redon (2017a) 307–309. 11  Fournet and Redon (2017a) 309. 12  Hoss (2005) 64. 13  The graph is based on the surface of the catalogued baths (under 6). If the size of the baths is unknown, it was not taken into account. However, the size of the rooms may give an indication of the maximum size of the building (in brackets under 6 of the individual catalogued baths). 14  See Nielsen (1993b) 2–47.

185

3rd c.15 The new late antique examples were mainly built in large towns with an important late antique phase (Rome, Milan, Ostia, Carthage, Alexandria, Scythopolis). Most were built in the 4th c., some perhaps at the beginning of the 5th and only one in the 6th (the Byzantine baths of Carthage).16 Similar patterns can be identified in other western regions of the empire. The only baths exceeding 1,000 m2 were built in the (early) 4th c. in the most important cities, e.g. the Kaiserthermen in Trier (Germany)17 or Constantine’s Baths in Arles (France).18 In Hispania and Lusitania, no new large baths dating to Late Antiquity have been discovered yet.19 In the eastern provinces, Late Antiquity was a rather prosperous period, which was reflected in the investment in new urban infrastructure. In Asia Minor and the Middle East, most larger baths were built in the 4th c., e.g. the complete reconstruction of the Zeuxippos Baths in Constantinople under Constantine, the Bath E in Antioch and possibly the Large Baths at Tyr (Lebanon).20 The construction of this category can be linked to the interest of the Roman emperor in a specific city (Rome, Arles, Trier, Constantinople), to the wealth of the local city council (Volaterrae, Gadara, Scythopolis) or a private investor (Piazza Armerina, Oued Athmenia). The category of baths with a surface between 500 and 1,000 m2 seems to disappear in late Antiquity in the Italian Peninsula. This was the type of modest neighbourhood baths that was frequently found in Ostia (Baths of the Trinacria, Baths of the Coachmen, Maritime Baths, Baths of Buticosus), Rome (Baths under the Piazza dei Cinquecento), Thamugadi (Small East Baths, Baths of the Capitole) and other cities all over the empire.21 In North Africa, baths of this size were still built in important cities such as Thamugadi (C57, C58) or Sufetula (C55, C56). The dataset of Cyrenaica, Egypt and Palestina points to a similar evolution, with baths in Cyrene (C67), Alexandria (C81) and Caesarea Maritima (C100). However, these figures are biased by the inclusion of double baths (C58, C75, C76, C87, C89, C90, C91, C92, C99, C116), which actually consist of two separate baths with a surface under 500 m2. If we consider the surface available to the bathers, these baths actually fall within the category of small baths. If, however, 15  Thébert (2003) 305–308, 321–22, 341–42. 16  We should keep in mind that several of the large imperial period baths were kept in use. The important point, however, is that large baths were not built any more. 17  Krencker et al. (1929). 18  Bouet (2003b) 41–44 with anterior bibliography. 19  Reis (2004) 43–45; Peréx et al. (2014) 69–70. 20  Constantinople: Yegül (1992) 324; Antioch: Levi (1947) 260; Tyr: Charpentier and Duvette (2014) 385. 21  See Nielsen (1993b) 2–47.

186 we consider the space allotted to a bathhouse within the urban fabric and the resources and manpower needed for their construction, they belong to the modest category. The majority of baths built in Late Antiquity, in all of the regions discussed above, belong to the category of small baths, with a surface that did not exceed 500 m2. The smallest examples, not larger than 100 m2, often date to the later centuries (C27, C42?, C68, C69, C74, C85?, C101, C104?, C110, C113?, Thamugadi Donatist Cathedral, Tipasa, Taucheira Fortress baths, Jerusalem Third Wall?). They often belong to a fort or an ecclesiastical complex. In North Africa, several baths were part of a larger domus, even if they had separate street entrances. One of the most deeply rooted ideas about late antique baths is the theory of the disappearance of public baths in favour of private baths (see above chapter 1, p. 33). The archaeological evidence that has come forward in the present research shows a more nuanced picture. We should also keep in mind that the small size of a bathhouse does not necessarily imply that it was just a basic facility without any luxury. The decoration itself could give some small baths a very ‘boutique-chic’-like atmosphere (see below, p. 196–198). Furthermore, studies on contemporary hammams in the Maghreb and Mashreq have shown that the service provided by the staff contributes to the quality of the bathhouse, even more than the aesthetic and architectural features.22 Rooms of the Baths Cold Rooms In the imperial type baths, the cold section still consisted of a large rectangular frigidarium—the centrepiece of the building—with cold pools for communal use in its corners. The presence of a natatio seems guaranteed in the imperial baths of Rome (inside the main bathing block) and Aquileia (in the palaestra?). The cold section of the imperial baths in Milan (C5) and Alexandria (C77) are too damaged to make any definitive conclusions about the presence or absence of a natatio. In the Western Baths of Scythopolis (C115), there were several cold pools surrounding the bathing block, although none had the size of the natationes in Rome. In the frigidarium of the Kom al-Dikka baths in Alexandria (C77), post-dating the ‘western’ examples by over half a century, we find small pools with heated water. The apodyteria in Rome’s imperial baths, as far as can be deduced from the incomplete plans and Renaissance drawings, stood in connection to the palaestra or basilica thermarum. 22  Kolb and Dumreicher (2008) 23.

chapter 4

In the other types of baths, some architectural trends that had developed during the 3rd c. continued during the entire late antique period (in all the different regions). The frigidarium was the largest and most important room, often rectangular in shape, with one or two piscinae. In North Africa, the plan of the frigidarium could have an alternative shape, such as a hexagon with semicircular apses on its sides (C36, C59). This form was also found in the Villa Baths of Piazza Armerina and a similar shape (decagonal) in the possible baths of Minerva Medica in Rome. The use of curvilinear forms in the North African architecture had been very popular since the 2nd c., especially in bathhouses (e.g. the Hunting Baths in Leptis Magna, the Large Baths in Thenae). In Cyrenaica, Egypt and Palestina, an open courtyard or large room with benches often combined the function of cloak room, waiting room and frigidarium (see below p. 187). Some public gatherings such as assemblies and law courts described in the literary sources (see above, chapter 2, esp. Augustine, Socrates Scholasticus, Malalas) may have taken place here. The peristylium in the baths of Piazza Armerina (C15), Volaterrae (C32) or the villa baths in Galeata (fig. 75) may also have been used for such assemblies. The piscina often had a semicircular shape or ended in an apse (e.g. C13, C16, C24, C37, C54, C62, C99, C112). The older rectangular shape also remained popular. Often, these piscinae lay on opposite ends of the room (e.g. C13, C14, C25, C29, C31, C32, C33, C36, C50, C54, C59, C88, C95, C96, C114). In North Africa, the pools could take more unusual shapes, such as a trefoil (C36) or quatrefoil clover (C45), an oval (C47), a hexagon (Sidi Ghrib), a U-shape (C52) or a rectangle with an apse on each short end (C34, C59, Sidi Ghrib). The villa bath of Piazza Armerina, again, shows remarkable similarities with these North African examples, which is not surprising considering the strong commercial and cultural ties between Sicily and North Africa. In the Italian Peninsula and Palestina, a round cold pool made its appearance (C6, C12, C13, C14, C26, C27, C98, C100, C115). The form in itself was certainly not new, as it is reminiscent of the earliest frigidaria pools in the Vesuvian cities. The natationes were uncommon, as these were mainly restricted to imperial type baths or very large neighbourhood baths (C1, C13, C15?, C21, C22, C114, C115), both rare during Late Antiquity in the regions investigated. A common feature of the frigidaria was the appearance of small single person pools in addition to the large communal pool (see below, p. 190–194). These smaller pools were sometimes added to the original layout of imperial period baths during Late Antiquity (e.g. Ostia, Baths of Neptune; Sufetula, Large Baths; Thuburbo Maius, Winter Baths, Volubilis, North

Continuity and Change in Late Antique Public Baths and Bathing Habits

187

figure 56 Plan of the baths in Brad (Syria) after Charpentier and Denoix (2011) 88, fig. 9

Baths). In some exceptional cases, a pediluvium (foot bath) was added to the frigidarium, often before the doorway connecting to the heated rooms (C94, Fortress Baths in Legio, baths of Nagazza?). Multipurpose Cold Room In Cyrenaica, Egypt and Palestina, a special type of cold room developed (C65, C70, C71, C73, C75, C76, C80?, C86, C89, C90, C95, C98, C109, C116).23 An open courtyard or a large hall served as a multifunctional cold room. The introduction of this lounge-type frigidarium, combining the functions of dressing room, cold room and hall for social gatherings, can also be detected in the late antique baths of Syria and Cilicia.24 A courtyard, such as in Brad (Syria, 3rd c.; fig. 56) and Télanissos (Syria, 5th c.?), or a large rectangular hall, such as in Serdjilla (Syria; fig. 57) was the main feature of the building and probably served as a meeting place for different types of social events.25 In Syria, in the village baths of the socalled Apamaean type, this large cold hall even became 23  See also Fournet and Redon (2017a) 283–90, who identify this type in Egypt and link it to examples in the Near East, but overlook the evidence in Cyrenaica and Cilicia. 24  Syria: Charpentier and Denoix (2011); Fournet (2012a) 332; Cilicia: Yegül (2003) 59. 25  Biscop and Blanc (2014) 423, fig. 5.; Charpentier and Denoix (2011) 93.

the main feature of the building, while the bathing block was of secondary importance.26 In this context, it is interesting to recall the passages in the late antique literature about rites de passages (see Gregory of Nazianzus), religious activities (see Eunapius), Christian celebrations (see Augustine, Socrates Scholasticus, Sozomen, Palladius, Gregory the Great) and secular affairs such as law courts (see Augustine) that took place inside bathhouses. The large halls may have been the ideal locations for such non-bathing activities. The platform in the courtyard of the City Baths in Ptolemais (C71) and the balcony in the hall of the baths in Serdjilla could be evidence of the social function of such halls. The gradual loss of the exercise yards (palaestra, basilica thermarum) with its exedrae and colonnades, which also served as a location for social encounters, may have been partially made up for by these halls.27 The geographical spread of this cold hall, from Cyrenaica to Cilicia, does not mean that there was a single ‘type’ of baths that was adopted in the east during Late Antiquity. Furthermore, the courtyard preceding the bathing block also appeared in the Italian Peninsula (C15, C32, Galeata), pointing to a general trend of increasing the cold section during Late Antiquity. However, 26  Charpentier (1995) 228. 27  Fournet and Redon (2017a) 302.

188

chapter 4

figure 57 Plan of the baths in Serdjilla (Syria) after Charpentier (1994) 121, fig. 7

some regional trends can be discerned. In Cyrenaica and Egypt, a courtyard, often with colonnades on two, three or four sides and possibly unroofed, preceded the actual bathing block (C65, C71, C73, C75, C76, C84, C89, C90, C91, C92, C94, C95). Sometimes the colonnade in front of the main bathing block formed an oblong corridorlike space that acted as the actual frigidarium, with two larger cold pools on the small ends (e.g. C73, C89, C92). Small single-person pools (rectangular or semicircular) flanked the doorway to the heated section. From here, the bather entered a small room without a hypocaust (tepidarium?), in which a small basin or labrum may be found, followed by two or three small heated rooms laid out in a linear or axial row. A similar plan can be found in Syria and Jordan (Placcus Baths in Gerasa).28 However, the heated section lies parallel to instead of in the axis of the cold hall (baths of Brad, see fig. 56). The popularity of this layout is confirmed by its continuing into the Umayyad period (see the Baths of Halabiyya-Zenobia or Qasr al-Hayr East, figs. 52, 53). In northern Syria and Cilicia, the bathing bloc is also parallel to the large hall, but for these buildings, the hall actually becomes the most important room and the bath suites are downgraded to a secondary function. Palaestra Few newly built baths had a palaestra. The absence of these large exercise courts is most notable in the small neighbourhood baths, although this was an evolution that had already started in the High Empire (see for example the baths of the Piazza dei Cinquecento in Rome). 28  Lepaon (2015).

This did not mean that bathers were no longer familiar with palaestra. The new 4th-c. imperial type baths in the Italian Peninsula still had large palaestra (C1, C5, C21, C25). The mosaics depicting athletes in these baths also demonstrate that physical exercises were still part of the bathing routine. In the 4th-c. villa baths of Piazza Armerina, a mosaic also depicts women playing a ball game. The Kom al-Dikka Baths in Alexandria (C77) did not seem to have a palaestra, although the cold section of this building is difficult to reconstruct due to the bad state of preservation. The Western Baths in Scythopolis (C115) had an open courtyard with several pools. The fact that this area was paved with flagstone seems to indicate that it was not used as an exercise yard. As we have seen in Rome, Ostia and other important cities in the empire, the existing imperial type baths were often still in use in Late Antiquity, meaning a palaestra was still available even if the smaller neighbourhood baths no longer offered a place to exercise. The literature (see Augustine, Socrates Scholasticus, John Malalas) and recent excavations—for instance in the palaestra of the Forum Baths in Ostia—show a remarkable change in function. It seems as if existing palaestra became multipurpose halls or squares where all sorts of meetings could take place.29 We can recall the passages in Augustine, Socrates Scholasticus or John Malalas, in which the large halls of the baths were used for hearings, 29  These squares could be adorned with statues brought in from other locations, as was the case in the Forum Baths in Ostia (see also below p. 198). A similar evolution of frigidaria into multipurpose halls can also be discerned in smaller baths, especially in the East (see above).

Continuity and Change in Late Antique Public Baths and Bathing Habits

legal matters and public affairs (see chapter 2). We should probably interpret the large halls that preceded the cold rooms as such multi-purpose areas rather than as exercise courts. A similar evolution of frigidaria evolving into multi-purpose halls can also be discerned in the smaller baths, especially in the east (see above). Late antique medical texts still mention exercises in combination with bathing (see Oribasius, Caelius Aurelianus), yet it is not clear if these should be performed inside the bathhouse in an adapted room (palaestra). In the late antique literature mentioning baths and bathing by both Christian and non-Christian authors, exercise is not mentioned in combination with bathing.30 Furthermore, there are no inscriptions commemorating the construction or restoration of rooms for physical exercise in bathhouses. Consequently, the archaeological remains, the literary passages and the inscriptions seem to indicate that a specific room for physical exercises (excluding swimming) disappeared from the layout of bathhouses. Number of Heated Rooms Most of the late antique baths in the Italian Peninsula and North Africa had a minimum of three heated rooms (tepidarium, sudatorium and caldarium). In the imperial type baths, the heated rooms were still large and equipped with several large alvei (C1, C5, C21, C22), resembling the layout of the imperial types of the High Empire. Only in the small baths of the 5th and 6th c. was the heated part sometimes restricted to one or two heated rooms (C27, C33, C42, C43 C53, C54). For some baths, the small heated part fitted within the modest ambitions of the architect, providing only the essential bathing rooms (e.g. C27, C42, Thamugadi Donatist Cathedral). In other cases, the small heated section may have been related to the lack of fuel, as the cold section was large and lavishly decorated (C53, C54). The growing importance of the cold rooms could point to a changing bathing habit, in which sweating and relaxation in warm water became less important. This evolution is also visible in some earlier baths in which one of the heated rooms was closed in a later phase (C23, C39, Sidi Ghrib, Vicus Augustanus baths; Sufetula, Large Baths; Thuburbo Maius, Capitolium Baths; Madauros¸ Large Baths). Again, a lack of fuel may have been the reason why some of these heated rooms were abandoned. Alternatively, the owners were not capable or willing to restore heated rooms that had fallen out of use, perhaps due to a lack of funds to pay for the repairs or for the 30  See, however, the passage by Sidonius Apollinaris (Ep. 5.17) in which the author watches a ball game before attending Mass. This game was played outside and not in a bathhouse.

189

fuel once the rooms had been restored. Or perhaps the skilled labour and necessary building materials were not available any more. This seems to have been the case when the hypocaust pillars were repaired with stone monoliths (e.g. C30, Mesange, Malvindi-Campofreddo baths; Volubilis, baths of the House of the works of Hercules; Thuburbo Maius, Winter Baths). We must remember that the heated section in North African baths had always been quite small.31 In Cyrenaica, Egypt and Palestine, the imperial baths followed the Roman standard, with several heated rooms (C77, C115).32 Smaller late antique baths sometimes had only one or two rooms in their original layout. In arid regions such as the Negev or the Libyan desert, a lack of fuel may well have been the most important reason (e.g. Mechili, C109, C111, C113, Abu Sha’ar Fortress baths?). In other cases, the limited heated section may have been the result of a small construction budget (e.g. C68, C104, C108) or simply modest ambitions (C74, C85, C96, C100, C101, C117), especially in fortresses (Taucheira, Ayn Gharandal, Bir Madhkur, Yotvata) and monasteries (Jerusalem Mount Scopus, and Third Wall; Ma’ale Adumin baths). For soldiers and monks, a basic facility to maintain hygiene may have sufficed. In larger baths that obviously had the space and the funds to construct several heated rooms (e.g. C71, C89, C99), the choice for a limited heated section may have been inspired by a change in bathing habits, limiting the time spent in the heated rooms in favour of the social gatherings in the larger cold sections. When only one or two heated rooms are present, it is likely that the function of caldarium and sudatorium merged (i.e. there was no separate room for sweating). In Egypt, some baths had two caldaria instead of one caldarium and one sudatorium (e.g. C75B/C, C76, C89). The Kom al-Dikka Baths in Alexandria (C77) even had different types of pools in almost all of the heated rooms. The popularity of small pools can have several explanations (see below, p. 190–194). Whatever the number of heated rooms, it is important to point out that the late antique baths still had at least one room with a hypocaust. In other words, a visit to a public bathhouse still implied a moment of sweating in a heated room. The idea behind the bathing habit still seems to have been inspired by the medical theories of the time. These still followed the same basic Hippocratic ideas (see Chapter 2, e.g. Theodorus Priscianus, Cassius Felix, Alexander of Tralles): the 31  Lézine and Elouard (1961) 9; Yegül (1992) 409; Thébert (2003) 326. 32  This was also the case in Syria, Lebanon and Jordan, where several imperial type baths were still in use and even attained their largest expansion in Late Antiquity; see Fournet (2012b).

190 balance of bodily fluids could be maintained or—in case of disease—restored by exposing the body to cold, hot, moist or dry circumstances.33 The baths still offered all these environments. Form and Features of Heated Rooms The caldarium often had a rectangular or cruciform shape. Both forms were not new, but had already existed during the High Empire. The rectangular caldarium often had two pools, which could be rectangular, semicircular or apsidal in shape. These pools stood opposite or perpendicular to one another (e.g. C9, C24, C30, C53, C55, C63, C67, C83, C98, C115). The cruciform caldaria seem to have been particularly popular in Palestina (C99, C105, C111, C114, C116, C117). In Egypt, a rectangular caldarium with pools on two or three sides (C75C, C92, C95) or two small ‘twin’ pools on one side and a larger round or semicircular pool on an adjoining side (C81, C84, C86, C91), are common. In Cyrenaica, a rectangular caldarium with a furnace flanked on each side by a single-person pool seems to have been popular in the Byzantine period (esp. C66, C73, Taucheira Fortress baths). This simple form may have been a quick solution to construct a bathhouse in ashlar in a short amount of time. Especially in a military context, such an uncomplicated scheme could have been welcome. However, this type of caldarium was uncommon in other regions.34 The Baths of the Byzantine Fortress in Thamugadi (fig. 41) did not have such a caldarium. The heated room of the Islamic extra muros baths in Volubilis (8th c., fig. 46) had a very similar form. Perhaps the early Islamic settlers modelled their baths after the ‘Byzantine types’ which they had encountered during their conquest, especially in their operating base in Cyrenaica. In the Italian Peninsula, the other heated rooms (tepidaria and sudatoria) were only rarely equipped with pools (C4?, C12?, C14, C15, C16, C17?). In the Kom al-Dikka baths in Alexandria (C77), all the heated rooms were equipped with several small pools, sometimes in addition to larger communal pools. In North Africa, a small single-person pool was often found at the beginning of the heated section (see below). As most of the plans had a reversing track, this pool could also be used at the end of the bathing itinerary. The fact that some pools in the caldarium were apparently also heated by a testudo, while others were supplied by a boiler and still others simply by an underlying furnace (e.g. C14, C29), 33  See also Duffy (1984). 34  The late antique baths in Dharih (Jordan) have a similar design (C101), although the two rectangular pools of the caldarium are adjoining instead of separated by a furnace-cum-boiler; see Sartori (2015).

chapter 4

points to a difference in the temperature of each of these pool (also below, furnaces). The small size of some of the heated rooms could be the result of new expectations of a visit to the baths, rather than of a lack of fuel or a lack of funding. In baths with large cold sections, but comparatively small heated sections (e.g. C9, C13, C15, C32, C34, C39, C54, C56, C61, C71, C75, C82, C99, C100), there obviously were enough funds, but a deliberate choice was made to invest in the cold rather than in the heated section. Most of these baths also had several heated rooms, which would be odd if fuel was really an issue. Hence, there seems to be a preference for smaller rooms, with smaller pools, creating a more intimate bathing experience. In an exceptional case, the heated room may not have had a pool. In the Baths of the Fifth Century in Sitifis (C54), the heated section only consists of one small rectangular room with no apparent pool. The reasons for this omission are unclear. The two large piscinae of the frigidarium exclude a scarcity of water. The most likely explanation according to Thébert is that the alveus has simply not been recognized during the excavations.35 If not, we should consider the possibility that the small heated section (merely 21m2) was of secondary importance, used only as a sweating room. The prevalence of the cold bath over the heated bath could not be clearer. Small Pools A recurring element of late antique baths that has often been labelled ‘characteristic’ (see above, chapter 1) are the small single-person pools. These pools are sometimes called ‘sitz-baths’ in excavation reports, making the link to the hip bathtubs found in Greek-style bathhouses.36 However, there is no evidence that all these small pools were used in the same way as their Greek counterparts, i.e. scooping water over the sitting bather. In contrast, the Roman type pools were often fed by a water conduit. Several scholars have proposed different explanations for their introduction. Before analysing these different explanations, we should first point out that the pools seem to have been introduced as early as the second half of the 2nd c. The frigidarium and the small tepidarium in the Baths of the Swimmer in Ostia, abandoned in AD 230–250, were equipped with such small single person pools in its last restoration phase (dated AD 160–170; figs. 17, 18).37 Similarly, in the North Baths of Volubilis, the small pools in both the frigidarium and the caldarium were later additions to the original layout (fig. 58). Their construction 35  Thébert (2003) 224. 36  Ginouvès (1962) 187. 37  Medri (2013) 66–68.

Continuity and Change in Late Antique Public Baths and Bathing Habits

figure 58 Plan of the North Baths in Volubilis (Morocco) after Nielsen (1993b) 128, fig. 132

was dated in the 3rd c. These baths had fallen out of use by the end of this century, giving a terminus ante quem of the late 3rd c. for the introduction of these small pools in Volubilis.38 Even if the introduction of such small pools arrived at an earlier date than previously assumed, their wide distribution started a century later and reached its peak during Late Antiquity. Not only were these pools incorporated into the original layouts of the newly built baths, they were also added to the existing imperial period buildings. In the Italian Peninsula, this is especially noticeable in the archaeological evidence from Ostia (Baths of Neptune, Baths of the Coachmen, Baths of the Invidious, Baths of the Philosopher, Baths of Mithras, Baths of the Trinacria, Baths of the Swimmer), Rome (Baths under the piazza dei Cinquecento; baths under Palazzo Valentini), Fiesole (undated intervention), Canusium (Terme Ferrara; undated intervention) and 38  On the dating of these baths, see Thouvenot (1945) 160–61; Thébert (2003) 273.

191

Venusia.39 In North Africa, we can mention the examples of Belalis Maior (Forum Baths), Madauros (Large Baths), Thuburbo Maius (Winter Baths, Baths of the Labyrinth), Sufetula (Large Baths) and Leptis Magna (Schola Baths). Inge Nielsen considered the small pools to be the result of a break-down of the water supply systems.40 A concern for water may have influenced the size and number of pools in more arid regions such as Egypt or Palestina. In small bathhouses, such as those in Cyrene’s hinterland (C68, C69, C70) or the baths built during the Byzantine military occupation (C73, fig. 41), the size of the pools seems the result of a limited water supply, which relied on the catchment of rainwater. Filling a large natatio with rainwater in a region where the average rain fall is 500 mm per year does not seem the most efficient use of water. Furthermore, these baths probably served small communities, which would make large baths and accompanying pools unnecessary.41 All in all, the modest size of these baths, and of their pools, seems to be the result of local conditions. And yet, these baths often had more than one small pool (C68, C69?, C70 and possibly the baths in Siret Ain Relles). The bathconstructors, in other words, chose several small pools above one communal pool. In contrast, the water supply for large cities such as Rome and Ostia was still in use at the time these pools were built, so water shortage was probably not the prime instigator for constructing small pools. A number of small pools were added to the total number of pools rather than replacing the larger communal pools, especially in cases of renovation. This shows us that the small pools actually added an extra volume of water to the total pool capacity, necessitating a higher water supply. Hence the pools seem to have been built for a specific use, rather than as a measure to economise on water. Another popular theory claims that the small pools were the result of Christian morals, which frowned upon close bodily contact and (partial) nudity.42 This view is largely based on decontextualized excerpts of Early Christian authors, such as Jerome (see chapter 2, p. 38– 40). Christian critique on baths was mainly pointed towards mixed bathing (male with female) and intended for ascetics. Furthermore, the introduction of this type of smaller pools predates the rise of Christianity (see above) and did not occur alongside the removal of the communal pools. What would be the point of taking your heated bath in an individual pool out of piety, if 39  For the baths in Fiesole, see Bellini delle Stelle et al. (1984) 36. 40  Nielsen (1993a) 57, 116. 41  The absence of efficient fuel such as firewood also seems to have limited the size of the heated sections of these baths. 42  Yegül (1992) 461, n. 88; but also Nielsen (1993b) 57.

192 you had to take the cold one in the communal piscina afterwards, or vice versa? As the alternating between a hot and cold pool formed the core of the bathing habit (see chapter 1), it is difficult to believe that one of the two would be skipped. Puritan Christian ideas about baths may well have influenced the architecture of some bathhouses, especially those near a church (C19, C27, C75–77, Thamugadi Donatist Cathedral), in a monastery (Jerusalem Third Wall, Ma’ale Adumin, Bawit) or near a baptistery (Cuicul, Thamugadi, Tipasa), but they cannot be held responsible for a general reduction in size of the pools.43 In the 1950s, the French scholar René Ginouvès proposed an alternative explanation, based on the late antique transformation of three Roman-style bathhouses in Greece.44 According to Ginouvès, the small pools were the result of an evolution in bathing habits. Smaller baths facilitated the use of different types of water or water of a different temperature.45 We know from textual evidence that the Romans distinguished different types of water and attributed different properties to them.46 These different types of water were also used in Greco-Roman medicine.47 Several medical authors prescribed different types of water for bathing or drinking (see chapter 2).48 There is also evidence that seawater was used in bathhouses (see chapter 2). Ginouvès connected the use of the small pools in public baths with those found in thermal baths.49 The water in small pools would also have been easier to refresh after each bather. As these small individual pools seem to have been introduced in the 3rd c., we could speculate that the impact of the ‘Antonine plague’, which hit North Africa particularly hard, had anything to do with a more personal bathing habit. The high number of medical texts written in North Africa during Late Antiquity (see chapter 2) may also point to the increased importance of medicine in this region.50 Alexandria in particular became the most important centre for late antique medicinal theory.51 We must, however, admit that references to such ‘medical prevention in bathing habits’ are difficult 43  From surviving monastery rules, we know that the monks living at the pilgrimage sites could bathe, even in public baths, but not together, out of fear of sexual temptations (rule of Shenoute, canon 3; Pachomius, Praecepta 92); see Crislip (2005) 30. 44  Ginouvès (1955). 45  Ginouvès (1955) 151. 46  Already Plin. NH 31.2.2; see above, chapter 2, Oribasius. 47  Jackson (1999). 48  See also Garzya (1994) 109–110. 49  Ginouvès (1955) 150; more recently also Bouet (2003a) 292 for the small pools in Gallia Narbonensis. 50  Benseddik (1989). 51  Nutton (1984) 5.

chapter 4

to find in ancient literature. We are reminded of the passage in Ammianus Marcellinus in which members of the elite demanded of their slaves that they bathed after they had visited sick friends.52 Or of the bathers in a passage by Theodoret who refused to submerge in the ‘contaminated water’ of a heretic.53 There is little proof of baths being used for containing a disease outbreak.54 The ‘Baths of the Lepers’ in Scythopolis (EP-1) is insufficient evidence for a category of baths ‘reserved for the sick’. The medical texts also prescribe different temperatures of bathing water for different ailments (see chapter 2).55 This theory playing out in baths has recently been supported by archaeological evidence. Chemical analysis of the calcareous deposits found in the pools of the thermal baths at Jebel Oust (Tunisia) confirmed that the different pools were filled with water of different temperatures.56 By a system of cooling tanks and direct pipes from the hot spring, or by combining the waters of both, the water of the pools could be made cold, tepid or hot (fig. 59). Such a diversification may also explain the high number of pools in the ‘Asclepium’ baths in Lambaesis (C50 and fig. 84). This idea of offering different bathing experiences may well have been repeated in public baths. In fact, the chemical analysis of the scale that was scraped of the bottom of the pools in the Kom al-Dikka baths in Alexandria confirmed that different pools were fed with water of different temperatures.57 The fact that some baths offered a choice in water temperature finds confirmation in a text by Libanius (see above, chapter 2).58 The same passage in Theodoret about bathers demanding the water of a pool be refreshed after a heretic had bathed in them further to points to such a service.59 Ginouvès explained the tendency to offer different types of bath water in small individual pools as the result of an increasing ‘individualisation’ of society.60 This fits within the context of a changing world view, which also saw the rise of eschatological religions such as Christianity, Manichaeism and the Mithras cult. Besides temperature or the type of water, the different pools may also have offered different ‘acts’ of bathing. Looking at the location of the small pools within the 52  Amm. Marc. 14.6.23. 53  Theod. Hist. eccl. 4.13. 54  Morley (2005) 199. 55  See also Fontanille (1985) 15–16; Villard (1994) 52; Nutton (2004) 202–203. 56  Broise and Curie (2014) 574; Broise (2015). 57  Koɫątaj (1992) 75. 58  Lib. Or. 2.34. 59  Theod. Hist. eccl. 4.13. 60  Ginouvès (1955) 135.

193

Continuity and Change in Late Antique Public Baths and Bathing Habits

figure 59 Plan of the thermal baths at Jebel Oust (Tunisia) with pools of different temperatures in Broise and Curie (2014) 581, fig. 6; with permission

bathhouse, it can be observed that they are often found in the caldarium (e.g. C39?, C52, C55, C71, C73, C86, C88, C89, C94, C114), in the frigidarium (e.g. C9, C12, C13, C36, C58, C60, C64, C71, C89, C114) or, particularly in North Africa, at the beginning of the heated section (C14, C15, C16, C25, C34, C37, C55, C56, C62, C64, C65, C75A–C, C82, C94, C99, C100). The availability of such a wash basin for personal ablutions (e.g. C82, C101) is reminiscent of the labra in Republican and Early Imperial bathhouses.61 The Baths of the Forum in Herculaneum had a labrum in the apodyterium and in the caldarium in the men’s section, and a labrum in the caldarium of the women’s section.62 The Central Baths in Pompeii on the other hand, still under construction when the Vesuvius erupted in AD 79, did not have the characteristic apse for the labrum (schola labri), but instead included a small heated basin in the caldarium (fig. 60).63 This basin does not seem to have had any internal steps, even if 19th-c. restorations blur its original appearance. The heated basin might have replaced the labrum as a feature for personal ablutions. In a further evolution, such a basin may have been replaced by a fully-fledged pool. Around the time that the schola labri disappears from bathhouse architecture (end of the 1st c. AD), the tepidaria were 61  Maréchal (2016) 135–36. 62  Pesando and Guidobaldi (2006) 360–62. 63  Pesando and Guidobaldi (2006) 85–87.

equipped with pools. According to Nielsen, these pools took over the function of the labra in the caldarium.64 As most of the late antique baths were of the row type with a reversing route (see above, p.184), the small pool at the beginning of the bathing route could also be used at the end. If these pools were used after a visit to the caldarium, their function matched that of the exit-tepidarium pools. It is interesting that in some North African examples, the pools at the beginning of the heated section or even in the frigidarium were supplied with warm water (C60, C62), just as some labra had been in early baths. Some of the small pools did not have a drain, e.g. the semicircular pool in the caldarium of the Karm Kandara Baths (C86) or the small pool at the beginning of the heated section in the Baths of the Philosopher in Ostia. This suggests a different use of water inside this pool, perhaps as a type of shower. The absence of a drainage system inside a pool is indeed evidence for a continuous flow of water.65 Such a different water use could perhaps explain the presence of small pools in the frigidaria (C12, C58, C71, C72, C87), as a difference in the temperature of the water seems unlikely here (except for C60). A special use of small pools may also be found in the small rooms immediately following the open courtyards in the baths in Cyrenaica, Egypt and Palestina (see above, p.188). 64  Nielsen (1993a), 158. 65  Manderscheid (2000 a) 508.

194

chapter 4

figure 60 Central Baths in Pompeii (Italy), detail of the small basin in the caldarium seen from the northwest photo: N. de Haan and K. Wallat, April 2007, with permission

These small rooms may have functioned as winter frigidaria. Alternatively, such pools are also reminiscent of the exit-tepidarium pools encountered in North Africa (see above, p.184) and could have had a similar function. It is striking that in some of these rooms, there was no small pool, but only a basin in the wall (C73, phase 3; C83, C101) or a labrum (C87?, C88?, C89, C91?, C95). From the options above, we can conclude that there is not one single explanation for the small pools in late antique baths. It was an architectural feature that could be used for different purposes, in different parts of the bathhouse. But whatever its function, we cannot deny that the use of this pool must have been a more private bathing experience, certainly in contrast to the communal pools. New Rooms? There do not seem to have been any new rooms added to the basic design of Roman-style baths—at least on a technical level. It is difficult to reconstruct what went on inside the different rooms, especially if no special ‘fixed’

features are present. In some cases, we can observe a combination of functions, such as the courtyard and halls of the cold section of the baths from Cyrenaica to Cilicia. In some late antique baths or during late antique restoration phases of earlier baths, a small heat lock was built between the frigidarium and the first heated room to reduce heat loss (Ostia: Baths of Neptune, Baths of the Six Columns, Baths of the Lighthouse; Banasa, Baths of the Fresco; Piazza Armerina Villa Baths, Sidi Ghrib, C57, C64, C85, C95, C111). These rooms were not large enough to permit any type of activity other than waiting. Such rooms were not a late antique ‘invention’; they already existed in earlier baths (e.g. Ostia, Baths of the Swimmer; Sabratha, Baths of Regio VIII). In baths with a circular track, the heat lock often took the form of an exit-tepidarium with a small pool, e.g. in the Baths of Caracalla in Rome (fig. 7), the Baths of Licinius in Thugga or the Baths of the Hunters in Lambaesis (C49). The terminology to describe the different rooms in the written sources does not change radically with the transition to Late Antiquity. The word lavacrum

Continuity and Change in Late Antique Public Baths and Bathing Habits

is sometimes encountered to denote a bathhouse or maybe a pool (EI-28, EA-13, EA-17, EA-32), but it does not seem to point to a new specific feature.66 Other terms such as oceanum (EA-25), stagnum or lacus seem to point to a larger pool, but are more likely to be literary ‘Spielerei’ than common terms.67 The word colymbus (EI-37) or baptisterium (EI-29, Sid. Apol) also denote pools, but have already been used by respectively Pliny the Younger and Galen.68 More interesting in connection to the archaeological evidence is the word atrium (EA-20, EA-26), which could perhaps be linked to the growing importance of the cold section. However, the baths in these inscriptions were funded by the emperor and therefore likely had a variety of unheated auxiliary rooms, which sometimes may have been called atrium. The xustos mentioned in a papyrus (P-10) could also have been a standard palaestra, or could point to the large multi-purpose halls (with colonnades) that became popular at this time in Egypt. And how should we imagine the matronikon (P-39)? Did this term denote a special room exclusively for women or rather the separate women’s wing of the double baths? A passage in Gregory of Tours mentions a room with thick vapour.69 Although Gregory is most likely talking about a sudatorium or a caldarium, the explicit mention of the ‘arduos vapores’ may be linked to the new, smaller yet steamier rooms that were common in Late Antiquity. Absence of Certain Rooms The fact that it is very difficult to attribute a specific function—not to mention a Latin term—to the archaeological remains, also makes it difficult to understand if and why some rooms disappeared. The palaestra is the most obvious feature that is absent in most late antique baths (see above, p. 188–189). Only the imperial type bathhouses still had large exercise courts in Late Antiquity, although these probably had a different function. Other rooms that could disappear were the sudatorium, which merged its function with the caldarium (see above), or the apodyterium, which merged with the frigidarium (see above, p. 188). In the smaller baths, we see an absence of ‘auxiliary’ rooms such as an unctorium or districtarium, although we should restate that these did not have a recognizable form or characteristic features. Besides, these rooms were not part of the standard layout of baths and were mainly an optional feature of the larger baths.

66  On lavacrum, see Maréchal (2015). 67  S tagnum: Sid. Apoll. Carm. 19; lacus: Auson. Mos. 335–348. 68  Plin. Ep. 2.17.11 and Ep. 5.6.25; Gal. De meth. med 7.6 and 11.10. 69  Greg. Tur. Hist. 3.31.

195

Construction Techniques and Building Materials The construction techniques and building materials that were common during the High Empire in the different regions were still being used during Late Antiquity. In the Italian Peninsula, opus testaceum was used, especially for the large imperial type baths in Rome. Smaller baths in and around Rome were often constructed in opus listatum. Sometimes, bricks of older structures were reused in the testaceum or listatum walls (e.g. C14).70 Architectural elements and statues from earlier periods were also salvaged as building material, often in the foundations (e.g. C4, Rome Minerva Medica, C101). The practise of reusing material, especially the more expensive materials such as marble, was not new.71 However, the contraction of the urban fabric and the subsequent decay of (public) buildings put a large quantity of building materials on the market. This is not to say that the production and import of new materials disappeared, as most imperially funded construction still made use of new materials.72 The opus caementicium technique was also very popular, especially in regions were brick production seems to have been rare. In North Africa, a ‘framework’ of orthostats in local stone often provided additional stability. The so-called opus africanum was a popular construction technique for all types of large buildings and had been in use since the Punic period. In this part of the empire, bricks were only used for the parts of baths that stood in direct contact with the heat of the furnace (furnace mouth, pillars, wall heating, suspensura). The scarcity of bricks does not need to be interpreted as a lack of clay or an inability to produce ceramic building materials. A type of bottle-like terracotta building element called tubi fittili were produced on a large scale to be used in vault construction, especially in North Africa.73 They were also found in Sicily, southern Italy and as far north as Ravenna by the 6th c. No examples have been found in Cyrenaica, Egypt and Palestina. The tubi had existed since the 3rd c. BC, but became more popular from the 2nd c. AD onwards.74 In Cyrenaica, Egypt and Palestina the influence of the ancient Greek building tradition in ashlar continued into the Roman period (opus quadratum). The late antique baths were also made out of locally quarried stone. In Cyrenaica, the tradition of hewing buildings out of the rock dates back to Hellenistic times, as 70  See also DeLaine (2001) 241. 71   Ward-Perkins (1984) 213; Christie (2006) 209; Russel (2013) 234–39. 72   Ward-Perkins (1984) 215. 73  Storz (1994). 74  Wilson (1992) 110.

196 demonstrated by the Greek baths of Cyrene.75 Only the vaults were made of opus caementicium or—for flat roofs—light building materials such as palm branches, reed and mud (as found in the 4th c. phase of the baths of Trimithis). In Egypt, the ancient tradition of mud bricks was also continued, although bricks—especially in the Nile Delta—were more common. Sometimes, only the heated section of the baths was made with fired bricks (C94, Trimithis) or even only the parts of the building that stood in direct contact with the heat of the furnace (C85). It is remarkable that the hypocaust pillars of late antique baths in Egypt were often built with standard bricks, instead of special tiles, hence avoiding a specialisation (or import) of ceramic building materials. Water-related building material such as pipes for supply and discharge were also made of ceramic. Vaulting techniques, as far as can be deduced from surviving walls, essentially stayed the same. Rooms with a hexagonal, octagonal or other polygonal shape must have been covered with domes, rather than vaults. Such domes are described by Basil of Caesarea and later Cosmas Indicopleustes (see chapter 2). Decoration Mosaics Mosaics were still an important type of decoration, especially in the Italian Peninsula and North Africa, but far less so in Egypt. They could be found in almost every type of room, both on the floor and on the inside of vaults.76 The floor of the frigidarium, the largest room of a bathhouse, was often embellished with a large figurative scene. Auxiliary rooms (apodyteria, unctoria, corridors) could be decorated with geometric mosaics (meanders, Salomon’s knot), both black and white and polychrome. The topics of the figurative scenes included marine landscapes (e.g. C9, C52, C56, Sidi Ghrib, Tipasa), mythological characters such as Venus and Neptune, even until a late date in North Africa (C1, C23, C53, C54, C60, C61, Sidi Ghrib, baths in Ouled Hafouz) and otium-related scenes such as hunting (C1, C52, Piazza Armerina Villa Baths). All of these topics had been popular in the High Empire as well.77 The depiction of athletes decreased in comparison to the High Empire (e.g. C1, C9, restorations of the Antonine Baths in Carthage; Manderscheid (1994), 65). Floral designs, birds and vineyards became increasingly popular (e.g. C9, C16, C29, C33, C34, C39, C41, C61, C64, 75  Wright (1957) 307–309. 76  The evolution of the style of mosaics in Late Antiquity falls beyond the scope of this study and has been addressed in detail elsewhere; see for example Dunbabin (1999). 77  Dunbabin (1989) 24–32; Manderscheid (1994) 64.

chapter 4

Sidi Ghrib, Tipasa, Jerusalem Mamilla baths). Obvious Christian themes and symbols were rather exceptional (C41, restoration of the Baths of Neptune in Ostia, fig. 25). The pools, both the piscinae and the alvei, frequently had simple white mosaics (e.g. C30, C36, C45, C55, C99, C112, C115), perhaps to imitate the costly marble slabs found in other baths (see below). The idea of imitating marble decoration can also be seen in the large tesserae mosaics imitating opus sectile floors (e.g. C13, C54, the restored frigidarium floor of the Porta Marina Baths in Ostia). Already in the Baths of Caracalla in Rome, the geometric mosaics in the basilicae thermarum seem to imitate the opus sectile floors of the main bathing rooms. Such mosaics, it has been suggested, were made with the waste of the cutting of marble slabs.78 A special type of mosaic decoration in North Africa and Palestina was a text laid out in tesserae (e.g. EA-2; 12; 13; 35; 37; 48; 63; EP2–5; in Italy in C15). These were often addressed directly to the bathers, wishing them a good bath (C33), praising the virtues of the baths, and of the water in particular. The comparison to Baiae, the thermal resort in the Bay of Naples, was a popular theme in North Africa (e.g. EA-2; 35; 49; 60).79 These mosaics were probably laid by specialized teams, sometimes coming from out of town (see EA-24). The mosaics on the vaults and the intrados of arches have rarely been preserved. Only the tesserae, often in glass and with silver or gold leaf, have been found during the excavations (e.g. C1, C20, C21, C29, C30, C98, C99, C115). Such decoration may have been a special luxury— in contrast to a simple plaster coating, painted or not— and became popular in the Byzantine cultural sphere (e.g. the churches in Ravenna). Marble Slabs and Other Stone Building Materials Marble slabs were mainly used for the floors and the facing of walls and often the pools themselves. The effect of sunlight hitting white marble through glittering water was especially appreciated, as can be deduced from inscriptions (EA-52; 55) and passages in Ausonius and Venantius Fortunatus (chapter 2).80 The floors of the large halls, such as the frigidarium or the basilica thermarum, could be paved with simple rectangular slabs or with a polychrome opus sectile floor (e.g. C1, C18, C21, C72, C81, C100, C110, C116). The prestige of marble elements in a bathhouse is also reflected in several passages of ancient authors, specifically mentioning the marble 78  DeLaine (1997) 31. 79  See also Allen (2009). 80  Auson. Mos. 347; Venantius Fortunatus Carm. 1.21–22; see also de Haan (2006) 351.

Continuity and Change in Late Antique Public Baths and Bathing Habits

decoration (see above, chapter 2: Ausonius, Sidonius Apollinaris, Olympiodorus, Palladius, John Malalas). When marble was either not available or not affordable, the rooms could be paved with locally quarried stone. Sometimes, the main rooms such as the frigidarium were paved with marble slabs, while corridors and auxiliary rooms had to make do with local flagstones (e.g. Galeata, C86). For most cities that had a construction boom during the High Empire (e.g. Rome, Ostia, Thamugadi, Alexandria), the availability of spolia from abandoned public buildings ensured that the baths could be embellished with different kinds of high-end building materials, even when the trade in these materials had diminished or stopped (e.g. C13). Columns were often salvaged from dilapidated buildings to enjoy a second life in a new or a restored building (e.g. C77, C116; see also P-27). Construction teams in more remote areas (e.g. San Giusto, Mechili, Karanis, Mampsis, Oboda, Rehovot; maps 31–34) only had access to locally quarried building materials and did not import more luxurious stone. Fortress baths were not built to impress and consequently were not embellished with luxurious materials (Thamugadi, Taucheira, Abu Sha’ar, Nag-el-Hagar, Ayn Gharandal, Legio, Yotvata). Wall Paintings and Stuccos The poor state of any wall paintings or moulded stuccos surviving in the archaeological record makes it challenging to trace the evolution of these popular forms of decoration. In the Italian Peninsula, the information is scarce. The walls in the baths in Ostia and Rome were mainly faced with marble, possibly until the start of the vaults. We must nevertheless keep in mind that wall paintings of previous periods could still be admired in the imperial period baths that were kept in use, e.g. the Severan paintings of the rape of Europa by Jupiter and the birth of Venus above the small frigidarium pool in the Baths of the Lighthouse in Ostia.81 The restoration of a wall painting was an intervention worth commemorating with an inscription (EI-37; also in North Africa EA-37). In North Africa and Sicily, examples of painted marble imitation were found on the lower parts of the walls (Piazza Armerina Villa Baths, C54, C61). A piece of stucco moulded in the form of a seashell is the only surviving example of this decorative form in the baths of Tubactis Municipium (C63) in North Africa. Fragments of painted geometric and vegetal patterns were found in Cyrenaica and Egypt (C74, C76, C83, C85, C89, C95), but most have been popular in other regions as well (unknown themes in C33 and C63). Figurative scenes must 81  Pavolini (2006) 207.

197

still have been painted (C60, C75C, C76, C95). The lower part of the wall could also be painted in a uniform red colour, while the top part was simply left white (C83, C88). For regions with no access to luxurious building materials such as marble, wall paintings were probably the best alternative to embellish the walls (e.g. C85, see P-10 for wall paintings in Oxyrhynchus). The evidence of wall paintings in the baths in late antique Palestina is scarce. Examples of 5th and 6th c. tombs point to the popularity of geometric and vegetal patterns, e.g. the Cave of the Birds on the slopes of the Mount of Olives in Jerusalem (4th c.).82 The vegetal and animal motifs found in bath mosaics suggest that these themes were not restricted to funerary contexts. The patterns and themes of the wall paintings seem to match those of the mosaics. The absence of figurative scenes and the preference of vegetal motifs could be a mere coincidence, if we keep the poor state of preservation in mind. The number of published wall paintings is indeed limited. However, we can cite the description of a private bathhouse by Sidonius Apollinaris (see chapter 2, p. 51), who was pleased that no naked figures, actors and wrestlers were depicted on the walls.83 Of course, we cannot generalize Sidonius’ personal taste to an empire-wide, Christian-inspired aversion for certain figurative art, but the passage does show that, in certain circles, the traditional themes of athletes, deities and otium were considered indecent and not fit for a bathhouse. The Christian critique on statues (e.g. Ambrose) fits within this same line of thought. However, we should not conclude from this that figurative art disappeared completely from the walls of bathhouses, as the Early Islamic baths of the Umayyad elite still displayed naked figures on their walls, pointing to a direct influence by contemporary Romano-Byzantine bath architecture. Statues Most of the statues that once adorned the baths have vanished into medieval lime kilns, which were often conveniently located inside the abandoned bathhouse (e.g. the examples in Ostia, chapter 3).84 The imperial baths in the Italian Peninsula were adorned with statues of the emperor, members of the imperial family and deities, just as had been the case in the Baths of Caracalla. The niches in the walls of both large and more modest baths also point to the presence of statues (e.g. C11, C45, C71, C112). Passages in literature (e.g. Ambrose, Evagrius 82  Kloner (2000). 83  Sid. Apoll. Ep. 2.2.4–9. 84  These lime kilns create an additional problem for modern archaeology, as the statues found in the vicinity may have been transported from other locations to be burnt in the kiln.

198 Scholasticus, Malalas) and inscriptions (EI-37; EA-25; 45) also point to the continued importance of statues as part of the decorative scheme. These could sometimes be brought over from other dilapidated buildings (e.g. EI-16).85 The statues were probably deities and mythological figures or local benefactors, such as the Amor riding a dolphin and draped figures (found in the Laberii Baths in Uthina). As long as the statues were no longer ‘polluted by pagan sacrifice’, there was no harm in displaying these among a Christian crowd of bathers.86 The epigrams of Christodorus of Coptus (6th c., see chapter 2) describe the busts of important statesmen, philosophers and classical gods displayed at the Zeuxippos Baths in Constantinople.87 Statues and busts of ‘pagan’ deities could be reused for their aesthetic value, even if some Christian bathers may still have been offended.88 The names of certain bathhouses were probably inspired by a famous statue displayed there. The most famous example is the Zeuxippos Bath at Constantinople, named after a bronze statue of Apollo Zeuxippos (‘the horse-yoking Apollo’) according to John Malalas (see chapter 2).89 Other baths also owed their popular (nick) name to (bronze) statues, such as the Medeia Baths in Antioch, the Icarus Baths in Tripolis (see Malalas) and the Baths of Horse in Alexandria (Johannes Grammaticus in the Ant. Pal, see chapter 2). We should mention, however, that these baths were built (and named) in the High Empire. We cannot verify whether the statues in question were still present in the bathhouses during the lifetime of the late antique authors or if only the name had survived. One can ask the same question for the balneum Veneris (‘baths of Venus’) mentioned in a 5th c. inscription (EI-49), if the bath was named after a statue at all. The practise of moving imperial period statues to baths has been attested in inscriptions (EI-16, 37; see also the Large Baths in Iol Caesarea), as well as in the archaeological record (see the palaestra of the Forum baths in Ostia).

chapter 4

Decorative friezes were only found in large thermae, such as the Baths of Diocletian or of Constantine in Rome, although few fragments have survived. Bronze water spouts in the form of animal heads (as mentioned in P-71; in bronze or stone in the private baths of Sidonius Apollinaris, see chapter 2) were recuperated and melted once the baths fell out of use. Similarly, precious materials such as gold and silver (see Evagrius Scholasticus, chapter 2), possibly used in the wall mosaics, but perhaps also in taps or on doors, were reused after the abandonment of the baths. No examples have come out of late antique baths so far. Architectural elements made of wood (as mentioned in P-58?) have decayed, although the extraordinary dry conditions in Egypt have given us some negative imprints (e.g. Trimithis). The same can be said for moveable wooden furniture, curtains and cushions, which must have embellished the interior of the baths (as mentioned in the Talmud Yerushalmi, Bata Batra 4, 6). The stucco claustra of the late antique baths in Marea (C89) and possibly Clysma (C82), pierced with round and heart-shaped oculi and fitted with green glass panes, are the predecessors of the characteristic oculi of the Islamic baths. Bath Technology

Other Decorative Elements The capitals that have been discovered in late antique baths in the different regions mainly belong to the Corinthian order and sometimes to the Composite order. It is remarkable that even in Italy (C1, restorations in the Forum Baths in Ostia) and North Africa (C58) some of these capitals seem to show an eastern workmanship.

Hypocaust In most baths of the regions discussed in this book, square or—less frequently—round tiles were used to construct the hypocaust pillars. Stone monoliths or other alternative types of pillars, such as tubuli filled with concrete (e.g. C30), were used to carry out restorations, presumably when tiles were hard to come by.90 The use of monoliths (e.g. C29) or other types of pillars was not new and had been adopted since the earliest Roman-style baths as a way of dealing with local conditions (lack of ceramic building material, availability of fire resistant stone such as basalt).91 Different types of hypocausts, such as parallel walls instead of pillars, are exceptional. They seem to have been used for small rooms (e.g. C2, corridor in C61), probably to ensure additional stability. This technique had already been adopted for the hypocausts underneath heated pools in earlier periods.92 In some late antique baths in Egypt and Palestina, such parallel walls—pierced by arcades—were used for all of the heated rooms (e.g. C81, C83, C85, C89, C110, C111, Abu

85  See also Strong (1994) 20; Lepelley (1994). For the imperial baths in Rome, see DeForest (2013) 60–63. 86  Lepelley (1994) 6–7. 87  A  nt. Pal. 2. 88  Just as Ambrose had been in Amb. Ep. 18.31. 89  Malalas 12.20; 291–292.

90  In exceptional cases, the pillars were made of stone monoliths from the start (Fortress Baths in Taucheira). 91  Adam (1984) 290; Degbomont (1984) 104–106; Lehar (2012) 57–60. 92  Adam (1984) 290; Lehar (2012) 62.

Continuity and Change in Late Antique Public Baths and Bathing Habits

Sha’ar). The reasons for building parallel walls instead of simple pillars can only be speculated upon. Besides the extra stability they provided for the suspensura, such walls may also have lasted longer than individual pillars (see EA-24, 42 for restorations of suspensurae). An alternative solution was building arches between the different pillars (C77, C88, C101, C106, C108, C109, Jerusalem Mount Scopus). This system has been found throughout the Roman provinces, especially in the Iberian Peninsula.93 However, the system does not seem to have been popular in the Italian Peninsula and North Africa. The regional preference for this type of pillar is difficult to explain, as other systems were also found in the same regions. For the baths in Egypt and Palestine, the lack of bath-related building materials—such as the square and round tiles for the pillars or tubuli—possibly necessitated alternative ways to construct the hypocausts.94 In the baths of Alexandria (C77, C81), Marea (C89) and Karanis (C85), the only type of ceramic building material available for construction was a rectangular brick. The architects had to improvise a stable support for the suspensura, which also ensured the air flow underneath the entire floor. The different ways of supporting the suspensura have to be linked to the availability of building materials in a certain region. The geographical context, rather than the chronological context (faltering construction techniques in Late Antiquity), seems to be a determining factor here. However, the import of (special) building materials to more remote locations may have stopped during Late Antiquity, due to shifting trade networks. One approach to this problem could be to study the change in construction materials on a case-by-case basis, linking the disappearance of certain building materials to the prosperity and connectivity of the city, also in comparison to other sites in the region. Wall Heating In the Italian Peninsula, the walls were mainly heated by box-like tubuli, even in small baths of a late date (C27). Sites such as Ostia and Portus revealed that a specialized production of tubuli started in the early 4th c. (C11, C13, C14, C17), possibly initiated by renewed building activity under Diocletian and Maxentius.95 In the warm regions of North Africa, Cyrenaica, Egypt and Palestina, the wall heating was sometimes reduced to a minimum or absent altogether (C36, C42, C43, C48, C54, C73, C106, C111,

93  Nielsen (1993a) 14, esp. n. 9. 94  See also Fournet and Redon (2017a) 288. 95  Maréchal (forthcoming).

199

Legio baths).96 The fact that tubuli actively contributed to the heating of the room is not only supported by the numerous modern experiments in recreated hypocaust systems (see above, chapter 1), it is also confirmed by authors such as Ausonius and Caelius Aurelianus and the legal sources.97 Reducing the number of walls that were heated by tubuli could thus reduce the average temperature in a room. In Egypt, only parts of the walls were heated by creating recesses that lodged a limited number of tubuli rows (e.g. C77, C85, C88, C89).98 In other baths, the absence of a production centre for tubuli might also explain the absence of a system of wall heating (e.g. C68, C69? and the fortress baths in Taucheira and Legio). The use of portable brazier may have offered a solution.99 In the absence of tubuli, the recesses were sometimes only covered by tiles or bricks (e.g. C85). The practice of controlling the temperature inside a room by changing the wall heating can also be noted when the caldarium—heated by its own furnaces—was deprived of wall heating, while the indirectly heated tepidarium did have such a system (e.g. C100, C102, C111). The examples of imperial period baths in Egypt and Palestine point to similar problems and inventive solutions (see above). Just as with the type of hypocaust, the type of wall heating was connected to the geographical context. Besides the presence of production centres of ceramic building materials (itself linked to the presence of natural resources), a regional preference for small steamy rooms may also have influenced the type of wall heating (see below, p. 200). Furnaces The furnaces were still located underneath the warm water pools to ensure a maximum use of the heat. In some cases, the use of a testudo, or possibly a semitestudo, can be deduced from the circular opening at the bottom of the pool and the arch above the furnace (e.g. C10, C23).100 In the baths of the Italian Peninsula, almost every room with a hypocaust had its own furnace (see C1 to C32). Only the room connecting the heated section to the frigidarium was sometimes indirectly heated by the furnace of another room (and hence often called tepidarium in modern literature). The use of boilers 96  The number of baths could be even higher, as several older excavation reports did not bother to describe the heating techniques. The remains that can be seen today are often too damaged to draw any conclusions (e.g. Thébert (2003) 200–201 about the heating system of C41). 97  Auson. Mos. 340; Cael. Aur. Morb. chr. 3.40; Dig. 8.2.13.pr. 98  Some of these recesses lodged only a single row of flues, acting more as chimney flues than as heating elements (e.g. C98). 99  See the remarks by Cael. Aur. Morb. chr. 3.40. 100  For the possibility of a semi-testudo, see Maréchal (2017) 182–84.

200 above the furnaces must have been common, even if the vessels themselves have been broken out and melted for reuse after the abandonment of the baths. A platform above the furnace, often with a circular imprint still visible in the mortar (e.g. C29, C85), points to the use of a boiler. In all of the regions discussed above and in all centuries, such boilers were used.101 It is only in certain arid regions in Palestina and in Cyrene’s hinterland, that the evidence is scarcer, perhaps due to the lack of metal ores to produce these boilers. The installation of large boilers could indeed be the source of great pride (see EA-20). In the absence of evidence for different types of water heating devices, the water may have been heated in the pool itself, with just the heat of the furnace tunnel under it (e.g. C2, C10, C24, C105, C113?). The absence of a boiler may in some cases been counterbalanced by the use of a testudo (e.g. C10).102 In some caldaria with several alvei, the way in which the latter were heated could point to a different water temperature. An alveus with a direct warm water supply from a boiler, and often also lying directly above a furnace, must have contained warmer water than an alveus that was only heated by the hot air from the furnace tunnel under it (e.g. C29).103 Smaller baths often had a single furnace. In combination with a linear row type, this furnace could heat two rooms, hence ensuring the standard frigidariumtepidarium-caldarium sequence. The lack of fuel, especially in arid regions (the Negev desert, Egypt’s Eastern Desert, the Sinai), may have inhibited the construction of larger baths with several heated rooms (C85, C101, C109, C113, baths in Mechili). Alternatively, the baths were only meant to serve a small group of people and therefore there was no need to build large complexes, as for example in fortresses (Taucheira, Abu Sha’ar, Yotvata). Similar sites dating from the High Empire had equally small baths, e.g. the fortress baths in the eastern Egyptian desert, the 2nd c. fortress baths of Tamuda or the baths in the pre-desert zone of Tripolitania (at Gheriat al-Gharbia, 2nd c.).104 In Egypt, the furnaces were sometimes located in corridors running underneath the bathrooms (C76, C77, C81, C84, C89, C93, C96). This was especially useful to heat centrally located rooms (C76, C77, C93), which 101  See Gazetteer; also mentioned by Malalas 11.10 and repeated by John of Nikiu, see chapter 2, p. 63 and p. 67. 102  See also Maréchal (2017). 103  We should note that an alveus could nevertheless be filled by water from a boiler which did not lie directly above its furnace. There are several examples of large boilers standing independently from the pools they supplied, feeding the alvei through a network of metal supply pipes, e.g. C82. 104  Egypt: Reddé (2009); Tamuda: Campos Carrasco et al. (2012); Tripolitania: Barker and Mattingly (1996) 104–105.

chapter 4

otherwise had to be heated from praefurnia inside the baths (e.g. C21, C56). In double baths, such a central furnace often heated a boiler (or boilers) that supplied the warm pools of both wings (C76, C84, C91, C92?). There is scarce evidence for such underground service corridors in preceding periods, e.g. the Hellenistic baths in Taposiris Magna.105 Also in Egypt, a special type of furnace was sometimes used. Instead of constructing the furnace just in front of the hypocaust and then channelling the heat by means of a furnace tunnel, the furnace was constructed as a pit in the hypocaust floor, which was supplied through a stoke hole connected to the praefurnium (C77, C87).106 This system might have been used because of the high flames that were produced by the burning of reed, straw and flax.107 The limited supply of firewood in arid regions might have incited the bath-operators to seek out new types of fuel, including agricultural waste products such as chaff or olive pits, dried cow dung or even animal bones.108 An interesting system for introducing steam directly into the caldarium was discovered in Karanis (C85). The steam of the boiling water in the boiler was channelled directly into the caldarium. As the caldarium was rather small in size, the steam created a sauna-like atmosphere, which would also explain the limited use of wall heating in this bath. This demonstrates a shift in bathing preferences that also became particularly popular in the Near East and would eventually develop further in the Early Islamic period. Water Supply The water supply of late antique baths did not differ from the system that had already been used in previous periods in each region. As we have seen above, the quantity of water that was needed did not diminish drastically because of fewer or smaller pools (see above). Aqueducts were maintained as long as possible, and replaced by cisterns and wells if necessary.109 Besides the archaeological evidence for branches of an aqueduct running towards a late antique bath or late antique restorations of aqueducts (e.g. Ostia, Rome, Carthage, Caesarea Maritima), both literature (e.g. John Malalas, Zacharias of Mytilene, Cassiodorus, Procopius; see chapter 2) and inscriptions (EI-18; EA-15, 17, 23, 32) point to the continued link of aqueducts and baths. The mainte105  Fournet and Redon (2009). 106  Compare to the most common furnace, see Degbomont (1984) 118–31. 107  Koɫątaj (1992) 174. 108  Olive pits: Meyer (1989); Cow dung: Bouchaud (2014); Bone: Fournet and Lepetz (2014). 109  Squatriti (1998) 11–26; Christie (2006) 246–47.

Continuity and Change in Late Antique Public Baths and Bathing Habits

nance of the aqueducts was one of the main concerns of local administrations and was important enough to be included in juridical sources.110 Alternatively, the water for a bathhouse could be tapped directly from a nearby stream or a local source (C18, C52, C112, C117; also mentioned by Sidonius Apollinaris and in EA-38). The use of a well was attested in sites where the water table was easily reached (C8, C12, C30, C35, C43, C84, C87, C93, C97). This type of water supply does not necessarily point to the decay of an aqueduct system. It was merely an alternative water supply system that could co-exist with an aqueduct (e.g. in Ostia).111 In Egypt and Palestine, the water from wells was brought to the surface by a mechanical waterwheel (saqiyah) turned by oxen, horses or asses (e.g. C75A–C, C76, C84?, C89, C95?, C113, el-Burj and Horbat Tarbenet). The fact that the saqiyah—a local invention of the region dating back to the 3rd c. BC—was preferred over a Roman-style waterwheel (noria) is evidence that local architects held on to local technologies, even when they were used in a Roman type of building. An important system of water supply that—due to the bad preservation of the upper parts of the buildings— has often been overlooked is the catchment of rainwater. Especially in regions with few natural sources and water tables lying deep beneath the surface, the collection of rainwater was probably the most important type of water supply (e.g. C63, C113). From some well-preserved imperial period baths in North Africa, we know that rainwater was also used as an additional form of water supply, complementing a noria (e.g. the Eastern Baths in Leptis Magna), a well (C43) or even a branch of the aqueduct (Hunting baths in Leptis Magna). In complexes that aimed to be self-reliant, such as forts or monasteries, the rainwater may well have been the only water supply for the—necessarily modest—bathhouse (e.g. baths at Ma’ale Adumin, Fortress Baths in Taucheira). For privately owned baths, tapping in to the ground water table and relying on rainwater could also be cheaper than paying the vectigal for aqueduct water.112 For privately used baths, it was also a way of bypassing possible limitations when using aqueduct water.113 Whichever type of water supply system was used, the water had to be collected in reservoirs before being distributed to the different pools, basins, fountains and taps. These reservoirs could be underground cisterns— sometimes cut out in the bedrock (e.g. Nahf, Jerusalem Third Wall)—or reservoirs (e.g. C13, C62, C71), often on 110  See Cassiod. Var. 5.38.2.; Cod. Iust. 11.43.6.2. 111  Manderscheid (2000a) 487. 112  Nielsen (1993a) 124; Vladu (2017). 113  See Cod. Theod. 15.2.3.

201

an elevated level as to make use of gravity for easy distribution within the building (e.g. C54, C73, C85; also mentioned by Palladius, see chapter 2). A combination of both types of water storage was also possible (C43). The distribution of the water from the reservoirs or the boilers to the pools was still ensured by terracotta or lead piping (e.g. C7, C52, C93, restoration of the Baths of the Lighthouse in Ostia, Elusa; for restoration works, see EI-22, 24; EA-36; P-12, 49, 50, 58), the latter withstanding more pressure than the former. As the pipework was often located at a certain height in the walls, their exact distribution pattern has disappeared along with the top parts of these walls. Waste Water Disposal Inside the baths, there were several manners of draining the water from the various pools. Most of the pools were drained by an outlet at the bottom. This was often a lead pipe that could be closed with some type of plug (e.g. the bottom of an amphora found in C88). The different openings then led the waste water into a network of sewage channels running underneath the baths (for possible repairs, see EA-7). Separate sewage systems for the warm and cold pools could exist, depending on the topography of the site (C13, C31, C43, C88, C96).114 If available, the sewage channels were made of brick and coated with a waterproof mortar. In regions were brick was less readily available, the sewers could be made of opus caementicium with a waterproof coating. The sewage network of the baths was connected to a larger urban sewer network or directly into the sea (possibly the Ostian baths along the Via Severiana) or a wadi (e.g. C52). In some cases, the waste water of the baths was used to flush an adjacent latrine (e.g. C30, C52, C56, C75C, C78, C99, C100). In a small minority of baths, a pool was drained by an outlet directly onto the bathroom floor (C30, C85, C109, Sidi Ghrib). This system, which had been used since the Republican period, was an easy solution to empty pools whose bottom lay on the same level as the room (e.g. C85, C101, piscinae of C112), for example, when a pool was added at a later date and simply placed on top of an existing floor.115 Some heated pools with no apparent drain (e.g. C10) might simply have been emptied by an opening in the testudo, having the additional advantage of flushing away the ashes of the underlying furnace.116 There were some small pools, however, without an apparent outlet and without a testudo (e.g. C86, C88). These pools must have been scooped out manually. In 114  See also Maréchal (2017). 115  Manderscheid (2000a) 502; Maréchal (2017). 116  Maréchal (2017) 185.

202 the case of the pools of the alleged Qalandia baths, the bottom slopped gently towards a hole in one of the corners, presumably to facilitate emptying it. This is reminiscent of the Greek hip baths with a similar hole in the bottom, serving this exact purpose.117 Alternatively, the overflow mechanisms found in some pools (C83, C88) could point to a continuous flow renewing the bath water.118 These pools would have necessarily been filled with cold water as the boilers could not supply such large volumes of warm water. They might also have served as a type of shower to fulfil a basic personal ablution, much as the labra once did.119

chapter 4

Urban Contexts The wider context in which a bathhouse was erected, in the centre of a Roman city or on an extra-urban estate for instance, has often been difficult to reconstruct. This is especially difficult for baths known through salvage excavations—see the numerous cases in Palestina— and excavations in modern cities—particularly in Italy for both cases— as the area just outside the bathhouse is often unknown. Consequently, it is often difficult to make assumptions about the incorporation of a bath into the urban fabric or the surrounding countryside (e.g. the villa baths in Sicily and North Africa). However, there are some patterns that can be deduced from the few sites that offer us a larger context. In all of the regions discussed in this book, the presence of a road was, unsurprisingly, an important factor in the location of baths.120 Not only were urban baths often built along one of the main traffic arteries (e.g. Ostia, Cuicul, Sufetula, Thamugadi, Ptolemais, Caesarea Maritima), baths in smaller settlements were also situated along an important road that ensured the link of this small community with the wider region (e.g. Valesio, Ain En-Ngila, Kerkouane?, Mechili, Tell Kanaïs, Mampsis). In addition to the advantages of accessibility and the guarantee of passers-by, which must have been especially important in the many linear settlements along roads, the main traffic arteries were often equipped with a branch of the aqueduct and / or a sewer. A building plot close to such a road could not only mean access to customers, but also access to water and waste water disposal. Furthermore, the road also ensured easy accessibility for

carts delivering fuel from out of town. The baths along the Via Severiana in Ostia even had their praefurnia and service corridors partially protruding onto the public street.121 We should note that this was not a late antique development and that the baths of the High Empire were constructed with the same pragmatic logistics in mind. But this arrangement shows which traffic arteries were still used or were restored or upgraded. The Via Severiana in Ostia was evidently still a very important traffic artery and may have serviced one of the most important commercial districts, linking the Ostian shoreline with other coastal towns. Similarly, the late antique restoration of the baths of Herdonia confirms that the local Via Traiana was still important in the late antique city. The long period of use of the Large and Small South Baths in Thamugadi point to the importance of the southern access road to the city, coming from the Aurès Mountains. The construction of the Byzantine fortress along this road in the Justinianic period confirms this. The baths along the eastern access road to Sufetula show a similar picture. The construction or restoration of baths along a traffic artery can also point to continued use of the branch of the aqueduct and the sewer. Linked to the roads, the presence of public spaces and buildings often seems an advantage for the erection or restoration of a bathhouse. Fora were the bustling centres of the antique city and attracted bathhouses, especially along the roads leading up to them (e.g. in Ostia, Rome, Cuicul, Leptis Magna, Taposiris Magna). Roman cities were often conceived as a succession of impressive monuments, reflecting the wealth and power of the local community. It should not be surprising that bath buildings, flagships of Roman technology and essential in maintaining good health and appearance amongst citizens, were part of this display. The Street of Monuments in Ptolemais with its two baths is a case in point. Baths were also found in combination with other ‘social hubs’ such as theatres (e.g. Volaterrae, Bulla Regia, Sabratha, Sufetula, Cyrene, Alexandria), amphitheatres (Rome), or churches (Ravenna, San Giusto, Cuicul, Madauros, Sufetula, Tipasa). Some of these baths were built to take advantage of the crowds that still frequented these social hubs, such as the small baths on or near the Forum Romanum in Rome (C24, C26) or those north of the Capitolium (C59) in Thamugadi. There is also evidence that a bathhouse itself attracted social and economic activity. The larger bathhouses, especially those

117  Ginouvès (1962) 187–88. 118  Manderscheid (2000a) 508. 119  Maréchal (2015) 157. 120  Saliou (2014) 667–68 comes to similar conclusions for the baths of Antioch.

121  For most of the baths for which the urban context is known, the service corridors had a direct entrance from the street. The most obvious examples in the gazetteer are C8, C9, C10, C11, C16, C24, C29, C34, C36, C39, C40, C41, C42, C44, C47, C55, C56, C57, C58, C72, C73, C77, C81, C99.

Public Baths and Their Contexts

Continuity and Change in Late Antique Public Baths and Bathing Habits

of the imperial type, were themselves social hubs and could even attract the building of other baths, e.g. in Thamugadi (map 17) or in Ostia, where the Baths of Porta Marina (in use until the 6th c.) did not discourage the construction of several—possibly privately owned— small baths along the Via Severiana (in the 4th c., map 13a). Market stalls that hoped to cash in on the bathing crowd can be seen around the Central Baths in Turris Libisonis (C29), Marea (C89) or Abu Mena (C75C), and are known through the written sources (see the Codex Theodosianus about the Zeuxippos Baths in Constantinople). The Church also saw the opportunity to benefit from the attraction of the baths. Building a church near an existing bath meant locating it in a long-standing social hotspot. In larger cities such as Rome, baths could also function as known anchor-points within a crisscross of streets (see the papyrological evidence in chapter 2). If one knew how to get to the baths, one also knew how to get to the church. The 4th-c. titulus just in front of the Baths of Caracalla could be a case in point. On the other hand, the Church could deliberately build a bathhouse as part of an ecclesiastical complex (C55 in Sufetula, the baths in Tebessa Khalia, possibly C19 in Ravenna and C37 in Bulla Regia). Providing a public bath may have boosted the popularity of the ecclesiastical authority, but may also have been seen as a responsibility of the Church, to provide this indispensable facility to its followers (see the passage in Theodoret). The same logic probably applied to baths for pilgrims (see the evidence from the Book of Pontiffs in Rome, the baths in Umm el-Amr, Jerusalem—Mount Scopus, Jerusalem—Third Wall, Ma’ale Adumin). The combination of bath and baptistery could have been a practical choice, as catechumens could bathe before being baptized. It also had a pragmatic reason: the baptistery could use the bath’s water supply and sewers (the possible small bathroom in the Lateran baptistery in Rome, C27, baptistery baths in Cuicul, Tipasa, Thamugadi). These water issues in combination with the topographical importance of baths as social hubs were probably the main reasons why abandoned bathhouses were transformed into churches (Lateran baptistery and Santa Pudenziana in Rome, Baths of Mithras in Ostia?; ‘Terme della Rotonda’ in Catania, church in the Small Baths of Madauros, church in the West Baths of Mactaris).122 When churches became the most important social hubs within a city or pilgrimage centres, they in turn may have started to attract bath buildings (C75–76 near the sanctuary of Abu Menas; C116 near the sanctuary of St. Hilarion). 122  For similar evolutions in Spain, see Sánchez and Carbonell (2004).

203

The absence of zoning within the antique city—such as truly commercial zones or exclusively residential areas—make it difficult to link the construction of bathhouses to a specific target group of bathers. The Baths of Musiciolus in Ostia (C9) may well have been used by inhabitants of the surrounding neighbourhood, Jews visiting the nearby synagogue and merchants on their way in or out of town. Some baths may have been used by members of a collegium, yet it is difficult to assess whether these baths were for an exclusive use, or if nonmembers, perhaps at different times, were allowed to bathe here.123 Wealthy proprietors of a public bathhouse possibly wanted to gain popular support from their fellow citizens and hence probably built baths where they could acquire building space, even if this was ‘next’ to the baths of a political rival (see below, p. 208). This question of building plots and ownership brings us to semi-public baths that were built inside a house, but also had a street entrance (see below, p. 207). In this case, the location of the bath within the urban fabric was subject to the location of the house. Consequently, there were streets in which several rich patrons had had the idea of building a semi-public bathhouse, e.g. in Volubilis or Cuicul (map 39 and 15). In both cities, the private baths may have tried to fill a void left by the closure of the large public baths. Alternatively, rich patrons financing the construction of a bath may have tried to tie a neighbourhood closer to their house, thus ensuring political support from fellow citizens. The location of some baths must also have depended on some pragmatic issues, such as the availability of building space within a city and of building material. The site of a temple or public building that was abandoned after an earthquake may have offered readily available building space and large quantities of (luxurious) building materials (e.g. C8 in an ancient horreum, C4 near the dilapidated Capitolium, C46 in the Antonine Baths, C67 in the Trajanic Baths, C73 in the old gymnasium). After having discussed the various elements that may have influenced the location of a bathhouse, we can also invert this logic and ask if, in some cases, the baths were used to reinvigorate parts of the city. Inscriptions teach us that such building programs were often initiated after catastrophes—such as wars and earthquakes— or after a period of abandonment. In the archaeological dataset, such planning is known from the imperial baths in Rome, e.g. the Baths of Caracalla or the Baths of Diocletian (C21), which were implanted in a region without imperial type thermae. The imperial type baths in Aquileia (C1) and Mediolanum (C5) may well have been similar projects. We can recall the passage by Ausonius, 123  Thébert (1991).

204 who mentioned that an entire neighbourhood in Mediolanum was named after the Baths of Herculeus. In Alexandria, the Kom al-Dikka Baths (C77) seem to have been part of a larger urban renewal program, also including new houses and auditoria. In Ptolemais, the newly embellished Street of Monuments included the City Baths (C71) and the formerly private Baths of the House of Paulus (C72), as well as a new nymphaeum. The same can be seen in Palladius Street in Scythopolis, which was created after an earthquake and included the Sigma Plaza and the Western Baths. Such planning efforts could be ascribed to the local city councils—who sometimes had some financial support from the emperor after an official plea for help (see John Malalas, chapter 2)—but perhaps also to private initiatives, in which a rich patron might have tried to tighten the grip on a certain part of town. Furthermore, we should not forget that baths, especially of the large imperial type, helped to define the public space and the outlook of a city.124 Investing in such eye-catching buildings was important to maintain the self-esteem and heritage of a city. Non-Urban Contexts Some baths stood in apparent isolation in a suburban or rural environment. Some bathhouses attached to large villa estates (e.g. Galeata, Piazza Armerina, Oued Athmenia?, Sidi Ghrib, Palazzolo near Ravenna) were freestanding buildings or had at least an entrance from outside the villa. This probably means that these baths were accessible to non-residents. We can imagine that the master of the house invited friends, family and colleagues to his suburban estate and boasted with his luxurious baths. There is no literary evidence that these villa baths were also accessible to the inhabitants of the surrounding countryside. However, the discovery of a small bathhouse (C15) in the pars fructuaria of the opulent Villa del Casale in Piazza Armerina suggests that the master of the house did provide baths for the people working his land. The site of Tébessa Khelia might have been the initiative of a local landlord, offering churches, olive presses and a bathhouse to a small rural community. Other ‘isolated’ baths were often labelled ‘private baths’, even if no residence was found in the vicinity (e.g. C33, C51, C62, C100, Nahf). However, we can imagine that several of these baths may have belonged to small villages, linear settlements along roads or rural communities (as was the case for the baths in Curinga). In regions where perishable building materials such as mud brick were used, the surrounding settlement may have been overlooked, especially in the absence of surveys. 124  For Rome, see DeForest (2013).

chapter 4

The small bathhouses in Cyrene’s hinterland remind us that baths (and churches) were sometimes the only buildings that were erected in stone. The baths of fortresses were either located outside or inside of the fortress walls. The reasons for the location are not always clear. We might expect that ‘external’ baths were also accessible to non-military personnel living in the surrounding countryside (e.g. Abu Sha’ar, Pelusium, Yotvata, Be’er Sheva, Be’er Shema) or that the baths were built at a later date, when there was no more space inside the fortress. The ‘internal’ baths may have been planned in the original design of the fortress, or were only for the use of the troops (Thamugadi, Taucheira, Nag el-Hagar, Legio). Alternatively, the baths were built inside the fortress for security reasons, in regions that were often under threat (Taucheira in the period of the Arab conquest, Thamugadi under threat from tribal raids, Legio under Sasanian threat). A last category of non-urban contexts is the monastic complex. Two types of baths can be discerned here. The baths that were intended for the faithful and pilgrims were located in the ‘public’ part of the complex (Umm el-Amr, Jerusalem Mount Scopus?). The baths for the monks and the nuns, also known through literature (e.g. Zacharias, Cassiodorus, Gregory of Tours), were located in the inner parts of the monastery, near the monks’ cells (Pelusium, Bawit, Suhag, Ma’ale Adumin, Jerusalem-Third Wall, Thamugadi Donatist Cathedral).125 As these were baths to clean the body and nothing more (see for example Augustine), these were small and very basic. Secondary Use of Baths When the baths fell out of use, the solid shell of the building was often reused for a new function. In a few examples, one of the large halls of the baths, such as the frigidarium, was modified and reused as a church (Rome, baths of Novatian; Mactaris, West Baths; Madauros, Small Baths and perhaps the Baths of Mithras in Ostia).126 Not only could the large halls of a bathhouse make a perfect nave, the presence of luxurious building materials and of water supply systems for a possible baptistery were additional advantages. If the structure of the baths was not really suited for transformation into a Christian place of worship, the church could simply be constructed in the vicinity or on top of the demolished baths, reusing both the water supply and the building material (e.g. the Baths under the Lateran Baptistery in Rome, 125  In literature, e.g. Cassiod. Inst. 1.29.1. 126   We can also remember the frigidarium of the Baths of Diocletian in Rome turning into Santa Maria Degli Angelli in the 16th c.

Continuity and Change in Late Antique Public Baths and Bathing Habits

Baths of Erythron, Baths behind the Water Company in Alexandria). The transformation of the baths into a church could also have the advantage of a prime location within the urban fabric. Basically, one social hub was transformed into another. Even if the sample is very small, we can observe that aspects of ownership might have decided whether a bath was transformed or not. The Small Baths in Madauros, which must have been the Winter Baths, were obviously the property of the city. With the increasing influence of Christianity in 4th c. Africa, the (already dilapidated?) baths perhaps came into the hands of the Church, which decided to build a place of worship here. The Baths of Novatian in Rome, on the other hand, might have been privately owned. According to the Book of Pontiffs,127 the baths came into Church hands in the 2nd c., and the owner may have donated some real estate to the Church too. The fact that the larger baths were often city property restored as long as necessary (or possible) could have impeded a widespread transformation of bathhouses into other types of buildings. Bathhouses were sometimes transformed into dwellings and / or small productive units for all sorts of arts and crafts.128 Especially frequent are pottery workshops (C46, C64, Galeata, Sidi Ghrib, Khirbet Lassan), while bakeries are less common (Eastern Baths in Leptis Magna) and, in later medieval and Early Islamic phases, lime kilns (e.g. in Ostia; Volubilis, Palace of Gordian; Leptis Magna, Eastern Baths; Ptolemais, City Baths; Khirbet Lassan) were constructed inside abandoned bathhouses. The sturdy shell of the baths, good accessibility and the presence of fire-resistant construction materials must have been a welcome benefit for the construction of ovens and kilns. In the larger cities, the site of the bathhouse was sometimes used as a burial ground (e.g. the imperial-type baths in Rome; Maritime Baths in Ostia; Baths of Pollena Trocchia; Baths of Venusia; C44; C47; C77; C115). In some rare cases, the baths were even reused as shrine for a holy man (C110, Sidi Ghrib and possibly in the West Baths of Mactaris). Burying the dead inside the decaying urban fabric was an empire-wide phenomenon, not restricted to the Christian West, but also common in Early Islamic North Africa and the Middle East. It should be stressed that

127  L ib. Pont. I.132. 128  Poor houses / squatter occupation: C1, C18, C22, C27, C36, C47, C65, C71, C73, C77, C81, C87, C88, C94, C101, C102, C104, C105, Sidi Ghrib, Legio. For arts and crafts: C66, C76, C88, Sidi Ghrib, Schola Baths in Leptis Magna, Erythron. We can remark that the production units were often linked to a domestic function, as had also been the case in previous centuries.

205

bathhouses were not the only buildings that were reused as burial grounds.129 It is difficult to find out if the secondary use of the baths can be linked to ownership, for example if abandoned privately owned baths were more easily transformed into private dwellings or production units. The evidence presented in this book does not enable us to make definite conclusions. As it is often impossible to pinpoint the exact date of abandonment, we cannot know if the baths had been out of use for a long period before they were transformed. The Baths in the Via D’Azeglio in Ravenna were probably privately owned and may hence have been sold, or transformed by the owner himself, to build a luxurious domus. However, the houses in the imperial baths of Aquileia were possibly constructed long after the building’s abandonment, when the ownership of the building was no longer relevant. We can also imagine that the burials inside the gardens of Caracalla’s Baths in Rome were ‘authorized’ by the Church, which seem to have been the new proprietor of the building. For burials inside other baths, such as the Maritime Baths in Ostia, it seems that this was more haphazard, inside the ruins of a building that had been abandoned for a long time and was possibly nobody’s property anymore. Similarly, a bathhouse in ruins, just as any other building, was ideal to dump all sorts of waste. The same is probably true for the Islamic burials in the baths of Carthage. The link between secondary use and ownership needs to be studied in a caseby-case manner, with special attention to the time lapse between the date of abandonment and start of the reuse. The transformation of a bathhouse into a different type of building, or the abandonment and subsequent use for squatter occupation or rubbish dump, can tell us something about the economic and demographic trajectory of a region. The transformation of a bathhouse into a church is a sign of continued urban development and points to economic prosperity of a city or region. Unsurprisingly, the ‘baths turned into churches’ were found in prosperous late antique cities such as Rome, Ostia, Mactar or Madauros. Similarly, it is no coincidence that a sumptuous house with mosaics was built on top of the Via D’Azeglio baths in 5th c. Ravenna. As of AD 402, Honorius had transferred the imperial capital to Ravenna, starting a golden age for the city.130 However, when baths were transformed into poorly built houses, this was usually during a period of economic decline. Civic authorities or wealthy patrons were no longer able to maintain public infrastructure. The imperial baths in 129  See the contributions in Brogiolo and Cantino Wataghin (1998), esp. Cantino Wataghin and Lamber (1998). 130  Gelichi (2000); Manzelli (2000); Cirelli (2008).

206 Aquileia (C1) were inhabited during the 6th c., when the Lombard invasion had swept through northern Italy and Aquileia had shrunk to a small settlement. Burials inside dilapidated baths also point to the abandonment of certain quarters. Several tombs dating from the turn of the 5th and 6th c. were found in the service corridors of the ‘Terme Maritime’ in Ostia. The burials in the Odeon Hill baths in Carthage date from the 6th or 7th c. In the Western Baths in Scythopolis, the burials occurred in the Islamic period, after the baths had fallen out of use and had been dismantled for building materials. The construction of workshops points to some economic vitality, even if the city (or region) was not wealthy enough to restore the public baths. In Leptis Magna, the bakery in the Eastern Baths was installed during the 4th c., when Tripolitania suffered raids from Amazigh tribes.131 In Uthina, the Baths of the Laberii (C64) were transformed into a pottery workshop in the 5th or 6th c. The city only survived on a small scale during Late Antiquity and no evidence for new buildings such as churches have been discovered yet. The Byzantine Baths in Carthage had fallen out of use in the Islamic period, but a pottery workshop was installed in the service area. Carthage remained an important city, even under Islamic rule. To conclude, the baths of Khirbet Lassan (Palestina) illustrate how the fate of a bathhouse was linked to fate of the larger regional context: the baths were in use throughout the prosperous 4th and 5th c. By the start of the 6th c., the baths had fallen out of use, but were transformed into a pottery workshop. This coincides with a more ‘agrarian’ phase of the settlement, also including wine presses and kilns. The life-cycle ended when the building was used as a rubbish dump, as the settlement only survived on a small scale during the Early Islamic and Mamluk period. Privatisation of the Bathing Habit? The ‘privatisation’ theory is mainly based on two paradigms: the literary evidence of Early Christian authors, reflecting a presumed Christian aversion to baths, and a changed role of the city councils which discouraged its members from investing in public infrastructure. This may have seen more of the wealthy elite building their own small private baths rather than building or maintaining public bathhouses. Evidence for this theory was allegedly found in literature, epigraphy and archaeology. Ausonius, Sidonius Apollinaris and Venantius Fortunatus describe such rich private baths. The dwindling number of bath-related inscriptions, especially those mentioning thermae, have also been used 131  Brett and Fentress (1996) 76.

chapter 4

to prove the downfall of public baths. Furthermore, the villa baths of Sicily (Piazza Armerina), Italy (Galeata) and North Africa (Oued Athmenia?, Sidi Ghrib) have been recognized as the archaeological confirmation of this theory. A closer look at these three categories of evidence, however, raises issues with this ‘privatisation’ theory. Literary Evidence As has been remarked for the Christian literature in chapter 2 and will be further elaborated below, the Christian aversion to baths and bathing must be put into perspective. Even if there was a plea by certain Christian authors to bathe with moderation and avoid mixed bathing, there are no passages in which wealthy Christian are advised to bathe in private baths rather than in public facilities. Only in some monastic rules were monks advised not to bathe in groups but on their own. There are authors such as Ausonius, Sidonius Apollinaris and Venantius Fortunatus who describe luxurious private baths, but such passages are certainly not restricted to Late Antiquity, nor does their number seem to increase.132 In fact, Ausonius’ Mosella poem shows strong similarities to a poem by Statius, making the text perhaps more of an exercise in style, consequently not even referring to an existing bath. The numerous attestations of public baths in the literary evidence as presented in chapter 2 clearly show that these facilities were still being built, restored and used until a very late date. Archaeological Evidence A recent comparative study of Roman villas in Lazio and Umbria revealed that the majority of private baths were actually built in the 2nd and 3rd c.133 For Lazio, only one private bath suite of the 384 studied villa sites was built in Late Antiquity.134 In a survey of villa baths in Calabria, Carmelo Malacrino came to similar results. Only one of the 47 baths found in the region was added to the villa after the 3rd c., but several were kept in use or enlarged during Late Antiquity.135 Furthermore, the baths of exceptionally large villas in southern Italy, Sicily and North Africa are now believed to have had a semi-public 132  de Haan (2010) 119–22. 133  Marzano (2007). De Haan (2010) 137–38 came to similar conclusion in her study on private baths in Italy and the western provinces. 134  Marzano (2007) 435. 135   Malacrino (2014) 297–99. Only the baths of the villa in Nicoterra, località San Teodoro seem to date from the 4th to 8th c. The baths of the villa di Casignana are a good example of a 1st c. AD bath, which was subsequently enlarged in the 2nd and especially the 3rd c. The 4th-c. changes concerned the addition of semicircular pools (Malacrino (2014) 294–96, fig. 4).

Continuity and Change in Late Antique Public Baths and Bathing Habits

function for the surrounding settlements.136 Not only the size, but also their independence from the villa itself seem to indicate that they were accessible to nonresidents. The literary sources tell us of wealthy patrons inviting friends and colleagues to their country-estates to enjoy a peaceful time dedicated to otium (see chapter 2, esp. Ausonius). The decorative schemes seem to corroborate this idea: both the Baths at Oued Athmenia (C52) and Sidi Ghrib had splendid mosaics depicting scenes related to otium (visit to the baths, hunting, horse breeding) and classical mythology, referring to the elite paideia (Neptune and the Nereids, Venus riding sea-creatures). However, the baths may also have served the local population of the countryside who possibly worked on the estate. This would match the type of patron-client relationship that existed in urban contexts. The only argument used to identify ‘private’ baths in an urban context as actually being private has often been their size. This is the case even if only part of the plan has been uncovered and no direct link to a house was found (e.g. the baths on the Celio in Rome, C3, C63). In North Africa, urban private baths with a separate street entrance seem to have been popular in Late Antiquity.137 These baths were not part of the original layout, but were added at a later date. In Cuicul and Althiburos, the construction of such baths is dated to the end of the 3rd c. or during the 4th. Only the external baths of the House of Castorius (C42) may have a later date. The examples in Volubilis date from an earlier period, possibly the 3rd c.138 If we look at the plans of all of these baths, two elements stand out. Firstly, the baths took up a considerable space within each house: about one third of the ground floor. Secondly, the baths often had a rather eye-catching street entrance, accentuated by a vestibule or columns (C34, C40, C41; Volubilis, House of Venus; Volubilis, House of Orpheus). These must have attracted the attention of the potential bathers who were strolling along the important traffic arteries on which these baths were located (e.g. map 15). Who could use these baths and why did the owners of the house or villa made such an important investment? It seems likely that access was restricted to a select target group of which the owner obviously was a member. We could think of baths for collegia or simply for the 136  E.g. the Villa del Casale in Piazza Armerina, see Pensabene (2010) 11. 137  Maréchal (2016). 138  We must remember that the end of the 3rd c. is often used as terminus ante quem for all Roman construction in Volubilis, as Diocletian withdrew the troops and the administration from southern Mauretania Tingitana. However, recent research has proved that Volubilis was inhabited until at least the Early Islamic conquest, see Lenoir (1985); Es-Sadra (2012).

207

friends or the clientes of a patron.139 The existence of specific baths for collegia has been proposed, but is still debated.140 The custom of inviting clients, colleagues and friends to one’s private bathhouse was a tradition that existed since Republican times and is well attested.141 If the baths were also accessible to a larger public, probably on payment of a (small) entrance fee, the baths could also have a direct financial benefit. Already in the Late Republican and imperial period, operating a public bathhouse had been a popular and sometimes lucrative business for the elite.142 For example, it has been argued that basically all of the 856 balnea mentioned in the 4th c. regional catalogue of Rome were privately owned baths, accessible to the general public (see above). Even if these baths could turn out to be financially rewarding investments, we should remember that the operating costs of baths, including water supply, fuel consumption and staff fees, must have been quite high.143 The main reason for building and operating a bathhouse was perhaps more about social prestige and charming fellow citizens than making money. The baths could hence be used to tie a certain neighbourhood to a patron. Such acts of benevolence towards the community can be found in Justinian’s Digests, in which a wealthy citizen of Tibur (Tivoli, Italy) granted his fellow citizens by will free access to the baths adjoining his house (see above, chapter 2).144 A different legal case study judged that if the baths of a house had been opened to the public, there should be a direct access between the baths and the houses in order for the baths to be recognized as an integral part of that house.145 As was the case for the large baths of villa estates, the private baths in urban context also served a wider part of the population. This type of semi-public baths therefore continued the tradition of the elite offering baths to the community. It also confirms recent theories that balance between private and public in houses of the elite did not necessarily shift to the former in Late Antiquity.146

139  de Haan (2010) 130. 140  Thébert (1991). 141  E.g. Vitr. De arch. 6.5.1; de Haan (2010) 119–40; Trümper (2014) 210. 142  Nielsen (1993a) 119–22; Fagan (1999a) 123–27, 142–54. 143  Nielsen (1993a) 122–24; Blyth (1999) 87–90. Even if the baths were operated by unpaid slaves, these would represent an important financial investment, as these slaves could not be used for other tasks while working at the baths. However, some sources confirm that running a bathhouse was considered to be a safe investment, see the Talmud Yerushalmi discussed in Jacobs (1998). 144  Dig. 32.35.3. 145  Dig. 32.91.4; see above chapter 2. 146  See Bowes (2010) esp. 46.

208 Competing with the large imperial thermae in size, facilities and luxury must have been an impossible undertaking. The smaller, privately owned baths had to rely on providing a more intimate bathing experience, special amenities and the quality of the services provided. One only has to remember some of the advertisements and graffiti in private baths in cities like Pompeii to get a glimpse of their strong points: the choice between sea water and fresh water, the wide range of bathing facilities or the quality of the food and the prostitutes.147 If we look at the location of the late antique semi-public baths in the urban fabric, we can see how they were placed along important traffic arteries, but also that they often avoided the catchment area of the larger baths, which were still in use (see map 15 or 17). The Codex Justinianus specifically mentions that it is forbidden for balneatores, contractors and builders of houses to exercise a monopoly.148 The law is obviously aimed at proprietors of balnea meritoria and could again point to the financial gain of operating a bath.

chapter 4

important sums were used to visit the baths (P13 to P-15). It is clear that Theophanes did not pay his host to make use of his baths, but visited the local public baths. The only direct evidence for private baths—or better, baths attached to a house, which may even have been semipublic in use—dates from the late 6th and early 7th c. (P-51; P-65). However, at this late date, there are still papyri mentioning public baths as well (P-59; 62; 66; 67). It is remarkable that only two papyri mention ‘private’ baths (2.4% of the sample in appendix 2) as the papyrological evidence includes many ‘personal’ letters and domestic archives. Private baths are highly likely to be mentioned here.

Epigraphic and Papyrological Evidence The alleged privatisation of the bathing habit has sometimes been deduced from the fact that the new construction of a bathhouse was hardly ever recorded in the epigraphic evidence of Late Antiquity (see above, chapter 2). As the construction or restoration of private baths was not commemorated with inscriptions—the goal of the inscription mainly stressing the benevolence of the patron towards the people—it is impossible to detect a ‘rise’ of ‘private bath’ inscriptions. On the other hand, the drop in inscriptions concerning public baths should be seen in the context of a general decline of epigraphy from the late 4th c. onwards. The construction of a semi-public, semi-private bath attached to or in late antique houses, mainly in North Africa, was probably not commemorated with inscriptions and neither were the baths attached to ecclesiastical complexes, so an argumentum ex silentio should be approached with extreme caution. Papyrological evidence mainly refers to public baths. Even if the specification ‘public’ (δημόσιος) is not often mentioned, it is clear that the receipts addressed to the officials (logistès, stratègos) concern expenses of public baths, both of construction (e.g. P-21, P-37) and restoration (e.g. P-19, P58). The archives of Theophanes, who kept a list of expenses from his travels, reveal that

The Church and the Baths Most scholars now agree that the Church did not actively oppose baths and bathing.149 Indeed, bath-related archaeological evidence from ecclesiastical complexes (e.g. C27, C55, Cuicul, Thamugadi, Tipasa, Suhag, Ma’ale Adumin) and Christian pilgrimage sites (Abu Mena, Umm-el-Amr) all over the empire makes clear that the Church not only allowed public bathing, but was involved in offering the possibility to bathe. Literary sources such as the Book of Pontiffs, Theodoret and Zacharias confirm this. The staunch criticism on bathing by pious ‘church fathers’ such as Jerome or Augustine should be framed in their context: these letters and texts were meant for a special group of exemplary Christian— ascetics and virgins—on how to follow an extraordinary lifestyle in order to come closer to God. Several literary attestations from all over the empire and from a timespan covering the 4th to 7th c. recount how the ‘average Christian’ and even members of the clergy bathed in public. Literary attestations and inscriptions account how pious emperors such as Valens, Theodosius and Justinian sponsored the construction and restoration of baths. There is even evidence that the sacrament of baptism, a founding pillar of Christian belief, was administered in public baths (John of Nikiu, Ananias of Shirak). Indeed, the technique to build baptismal fonts was the same as building bath pools.150 Several thermal sources, which had been dedicated to nymphs or local deities, were reinterpreted and attributed to Christian saints or martyrs (Itinerarium Burdigalense, Sophronius).151 Nevertheless, there does seem to have been a shift in the perception of baths, at least according to the writings of Early Church Fathers. The special relationship

147  Sea water and fresh water: Pompeii, CIL 10.1063; bathing facilities: Bononia, CIL 11.721; quality of food and prostitution: Herculaneum, CIL 4.10677. 148  C  od. Iust. 4.59.2.1.

149   Already Dumaine (1910); Stommel (1959); more recently Berger (2011) 49; Bady and Foschia (2014) 992–95. 150  Stommel (1959) 6; Contiero (1987). 151  See also Dvorjetski (2007) 393–423.

Continuity and Change in Late Antique Public Baths and Bathing Habits

between the Christians and water—the washing away of the sins by baptism—had to be carefully redefined, so that a simple visit to the bathhouse would not free the believers of all sins. The baths, in other words, could only provide a superficial cleansing of the body. Furthermore, the criticism of baths that had been vented by conservative minds such as Seneca, concerning the weakening effect of hot baths or the promiscuity and debauchery that took place inside bathhouses, were repeated and emphasised by Christians. A sober bathing habit fitted the way of life by which early Christians tried to address the lower classes. There is, however, one pitfall which scholars should avoid. If the Christian authors advocated a sober way of bathing, this did not imply a purely hygienic form of bathing.152 The modern concept of bodily hygiene did not exist in antiquity. Bathing always implied the basic elements of sweating out bad bodily fluids, washing and scrubbing the skin, relaxing the body in hot water and then reinvigorate it by plunging into a cold pool. This concept remained unchanged, even for Christians, and can be recognized in the architecture of ‘Christian’ baths by the presence of several heated rooms, hot and cold pools (e.g. C55, C75, C76, Cuicul, Tipasa, C116). Rather, a sober way of bathing did not last hours on end (as heat was the fuel for human passions), did not involve excessive eating and drinking or, even worse, mixed bathing, and did not involve the display of wealth. The appearance of double baths in Egypt, possibly on settlements linked to pilgrimage itineraries, does seem to indicate a Christian influence on bath design: two (nearly) identical wings, one for male and one for female bathers, centred around a communal furnace with boilers. The impact of this sober Christian way of bathing has been recognized in the architecture of late antique baths: small baths that could accommodate a smaller number of bathers, with smaller heated sections and single-person pools. The fact that the inner steps of the small pools also acted as benches, positioning the bather with his back to the room, also points to a more individual bathing experience: one did not look at his fellow bathers when sitting in the pool. Conversely, smaller rooms and smaller pools have also been interpreted as an evolution towards a more intimate form of bathing, favouring close contact between bathers.153 Some pools could indeed accommodate only two or three bathers at a time, resulting in a reduction of personal space in comparison to larger pools. And even if a reduction in size was chosen to impede ‘social bathing’ in the baths

152  On the sober Christian baths, see already Zellinger (1928) 10. 153  Eger (2007) 149.

209

of ecclesiastical complexes, we cannot accept this influence as a general trend. As mentioned above, the reason for building smaller heated section and especially smaller pools were diverse, sometimes linked to a scarcity of fuel or water, sometimes out of a deliberate choice to offer a different bathing experience. Large communal pools were still built in several late antique baths. The decoration of public baths seems to have evolved towards a non-figurative scheme, preferring floral and animal designs and abandoning the mythological themes of the old pantheon. Even if we can detect such a preference in some regions (Egypt, Palestina), we should stress that the classical decorative schemes of imperial period baths still in use by late antiquity were not replaced by a new Christian scheme. In North Africa, bath mosaics still depicted scenes of classical mythology as late as the 5th and 6th c. (C53, C54, Ouled Hafouz). As we have already mentioned (above, p. 203), local ecclesiastical authorities acknowledged the role of baths in the social fabric of the city. Indeed, popular baths may well have attracted titular churches. The siting of baths and churches certainly shared the same attraction to main traffic arteries (e.g. in Ravenna, Sufetula, Thamugadi). There is no evidence of baths being deliberately destroyed by the Church, even if some churches were built inside the shell of former public baths. This, however, was more likely linked to pragmatic considerations, such as the reuse of sturdy buildings and the connection to water supply and sewers (for the baptistery). On the contrary, there is ample evidence that the Church also built baths, often in combination with a baptistery (e.g. in Rome, Ravenna, Cuicul, Thamugadi, Sufetula, Jerusalem), possibly assuming the responsibility once carried by the city council. It would have boosted the Church’s image as ‘new city patron’ as well. By the end of Late Antiquity, the Church controlled most of the public baths in the cities of the Italian Peninsula. In Rome, the patriarch of the Church, later called the Pope, exercised increasing control of the city and therefore responsibility for public facilities such as water supply and bathhouses. In the other regions under investigation, this link between Church and baths is not as straightforward. In North Africa and Cyrenaica, the ‘last’ baths were rather the result of Byzantine military occupation. In Palestine, the baths related to ecclesiastical structures from the 5th c. still existed, but the last new constructions do not show an obvious link to the Church. In Egypt, the baths that survived well into the 7th and sometimes even 8th c. were located in sites that can be linked to Christian pilgrimage itineraries. Even if these baths were not financed by the Church (which they might have been), the continuous flow of Christian worshippers created a demand for baths.

210

chapter 4

Decay or Continuity of Baths and Bathing Habits

In North Africa, a similar picture emerges. Archaeology and epigraphy show that councils maintained and in some cases expanded the bathing infrastructure of their cities (e.g. Cuicul, Thamugadi; map 32, graphs 1 and 5). We can note the high number of newly built baths in comparison to the baths that were still in use (27 versus 29 on graph 3). Despite the alternative career paths that bypassed curial duties, some members of the elite continued to provide civic services such as opening and operating bathhouses, all be it under a slightly changed form. Indeed, a special type of semi-public, semi-private bathhouse became popular in some cities in the 4th c. It probably enabled the important families to make the link between the familia and civic services more explicit. Fellow citizens were now invited into the domus of the elite in a bid to gain their political support and thus overshadow rivalling families. The fact that the city councils in North Africa were still the main breeding ground for the recruitment of the imperial bureaucracy, meant that civic euergetism continued to play an important role for the North African up-and-coming families. And yet, not every region of North Africa enjoyed a prosperous 4th c. The withdrawal of the imperial bureaucracy in Mauretania Tingitana under Diocletian left important cities such as Volubilis and Banasa behind. There is, however, reason to believe that some maintenance of the urban infrastructure continued. The raids by nomadic tribes into Tripolitania in the second half of the century may have redirected building efforts to more defensible structures.

Continuity in the 4th c. If we consider the evidence of the written (chapter 2) and archaeological (chapter 3) evidence for public baths in Late Antiquity, we get a picture that evolves through time and space. For the 4th c., archaeology (appendix 3, graph 3), ancient literature and inscriptions (appendix 1, graph 1) show a strong continuity of the bathing habit in all of the regions. Especially in the first half of the century, both new construction and restoration works demonstrate a vivid interest in baths by both emperors and city councils (see graph 4). With the high number of baths built during the High Empire, it is unsurprising to find a considerable proportion of them still in use in the 4th c. (51 to 24 newly built baths, see graph 4). 4th-c. authors confirm continued imperial involvement in bath construction and restoration (e.g. Ammianus Marcellinus, Eusebius of Caesarea). On the Italian Peninsula, the quantitative aspect of the data set is, however, heavily influenced by the evidence from Rome and Ostia. Large imperial constructions in Aquileia (C1) and Milan (C5) confirm that the most conspicuous construction activity took place in the most important cities (map 31, graph 4). In smaller cities, more modest baths were constructed or the larger imperial period baths were kept in use. Traffic arteries were not only important for the survival of the cities, they also played an important role in the survival and in the new construction of bathhouses (C9, C10, C11, C24, C30).

4th c. Baths 80 70

Number of Baths

60 50 40

Continued

30

New

20 10 0

Italy

N. Africa

Cyrenaica

Egypt

Palestina

Continued

51

29

1

4

12

New

24

27

3

8

15

graph 3

Public baths in the different regions during the 4th c., based on the data from Appendix 3 (possible baths not included)

Continuity and Change in Late Antique Public Baths and Bathing Habits

211

map 31 Public baths in Late Antique Italy not shown: C28–29 on Sardinia, Roman numerals refer to Augustan regions plan by the author

212

chapter 4

Public Baths in Italy

90 80 70 60 Number of Baths

50 40

Possible

30

Continued use

20

New

10 0

4th c.

5th c.

6th c.

7th c.

Possible

5

1

2

0

Continued use

51

40

16

4

New

24

8

1

0

Unknown

6

graph 4

Public baths still in use in late antique Italy, based on the data from Appendix 3A. It should be noted that some baths were in use for several centuries, and were hence counted as ‘continued use’ for each century following their construction.

map 32

Public baths in Late Antique North Africa the Roman numerals refer to the provinces: I. Mauretania Tingitana, II. Mauretania Caesarensis, III. Mauretania Sitfensis; IV. Africa Proconsularis and Byzacena, V. Tripolitana plan by the author

213

Continuity and Change in Late Antique Public Baths and Bathing Habits

Public Baths in North Africa

70 60 Number of Baths

50 40 30

Possible

20

Continued use New

10 0

4th c.

5th c.

6th c.

7th c.

Possible

3

0

0

0

Continued use

29

17

7

2

New

27

5

2

0

graph 5

Unknown

4

Public baths still in use in late antique North Africa, based on the data from Appendix 3B. Possible baths are only counted once, even if a continued use of the building could be identified.

Public Baths in Cyrenaica

12

Number of Baths

10 8 6

Continued use

4

New

2 0

4th c.

5th c.

6th c.

7th c.

Continued use

1

3

3

3

New

3

1

2

0

graph 6

Unknown 10

Public baths in late antique Cyrenaica, based on the data from Appendix 3C

In Cyrenaica, the mainly archaeological evidence shows continued building activity in the 4th c. (e.g. C65, C67, C71; map 33; graphs 3 and 6). The sample is small (graph 6) and the majority of bathhouses could not be dated due to the lack of good excavation data (10 unspecified versus 6 securely dated, see graph 6). We must also take into account that far fewer archaeological (rescue) excavations are carried out in Cyrenaica than for example in Israel. In Egypt, archaeological evidence mainly comes from fortresses (Abu Sha’ar, Nag el-Hagar, Pelusium)

and Alexandria (C77, C78, C79; map 33, graph 7). Papy­ rological evidence shows that construction and restoration continued (graph 2) mainly in provincial capitals such as Oxyrhynchus and Hermopolis Magna. Again, we should note that some baths could only be dated to Late Antiquity, rather than a specific century (13 new baths, not including 8 possible baths, see graph 7). In Palestina, the 4th c. is a period of urban expansion, with several new baths constructed in both large and small cities (map 34; graph 3 and 8). For such a small strip of land, the number of newly built baths

214

chapter 4

map 33

Late antique public baths in Cyrenaica and Egypt, not shown: Siret Ain Relles, Zawiet el Argub, Gasr Bandis, Gasr Sherbin plan by the author



Public Baths in Egypt

25

Number of Baths

20 15 Possible

10

Continued use New

5 0

4th c.

5th c.

Possible

1

1

Continued use

4

5

12

12

New

8

7

3

0

graph 7

6th c.

7th c.

Unknown 8

Public baths in late antique Egypt, based on the data from Appendix 3D

13

Continuity and Change in Late Antique Public Baths and Bathing Habits

map 34

Late antique public baths in Palestina plan by the author

215

216

chapter 4

Public Baths in Palestine

35 30 Number of Baths

25 20 15

Possible

10

Continued use New

5 0

4th c.

5th c.

6th c.

7th c.

Unknown

Possible

2

2

7

2

15

Continued use

12

17

15

9

New

15

3

3

0

graph 8

8

Public baths still in use in late antique Palestina, based on the data from Appendix 3E

is remarkable (15, not including the 4 possible baths, see graph 8), especially in comparison to Egypt or Cyrenaica.154 Furthermore, for at least 8 and possibly another 17 baths, the century in which they were constructed could not be confirmed. The baths that were still in use (at least 8) also point to a continued investment in public infrastructure. The construction boom in 4th c. Palestine was not restricted to bathhouses and has been linked to an economic growth of the region.155 The construction and restoration (often enlargement) of public baths in Syria shows the same trend.156 To conclude, the archaeological remains of baths point to continuity in both architecture and technical aspects during the 4th c. We can find the same type of rooms, water management and heating systems. Some regional characteristics, such as the modest heated sections in the North African baths, also persisted in Late Antiquity. In the Italian Peninsula—especially in postConstantinian phases—and in North Africa, newly built baths were of a modest size. In Egypt and Palestina, where the ‘monumentalisation’ of cities had started later, several towns were equipped with larger bath complexes (Alexandria, Caesarea Maritima, Scythopolis).157 The decoration of the 4th c. baths was very important, even 154  We should note that in recent decades, the number of rescue excavations in Israel has increased spectacularly, resulting in the discovery of many small late antique baths that would otherwise have been demolished. 155  Tchalenko (1953); Tchalenko (1958); Shereshevski (1991); Tate (1992); Foss (1995). 156  Fournet (2012b). 157  See Broise (2009); Hoss (2012).

when the reuse of building materials was widespread. An urban bathhouse still offered the citizens a certain luxury. As far as we can tell, the continuity in architectural forms—both in new and restored baths—points to a continuity in bathing habits. The literary evidence, in particular the medical texts (e.g. Oribasius), seem to confirm this. Strong Regional Trends in the 5th c. In the Italian Peninsula and in North Africa, the turbulent 5th c. saw an important regression in the number of newly built baths (graphs 4 and 5). On the Italian Peninsula, only 8 baths were constructed during the 5th c., versus 24 in the century before (graph 4). The new constructions were restricted to Ostia, Brixia, Ravenna and Rome and were very small (appendix 3A). The total number of baths in use also decreased (from 75 to 48, see graph 4). The loss of North Africa to the Vandals in AD 430 was a severe blow to the food supply of Rome, which entailed a sharp demographic decline, and was a severe blow to the wealth of the Roman elite owning large rural estates in Africa.158 In combination with invasions of the Goths (sack of Rome in AD 410) and later the Vandals (second sack of Rome in AD 455), the unstable political situation and economic climate made important civic investments less appealing. The impact of invading ‘barbarians’ sacking towns and plundering the Italian countryside must have had serious local consequences.159 Yet it is important to remember that 158  Wickham (2005) 34. 159  Wickham (1998) 290.

Continuity and Change in Late Antique Public Baths and Bathing Habits

such raids also happened during the High Empire, especially in frontier regions.160 The long-term danger lay in damage to state infrastructure such as roads and bridges hindering the circulation of goods.161 In North Africa, the difference between the 4th and 5th c. is even more striking: from 27 to only 5 new baths (graph 5). The drop in the number of baths still in use is also significant. This would mean that more and more cities shrunk and struggled to invest in public infrastructure. The new constructions in North Africa were mainly restricted to ecclesiastical complexes and were also small in size (appendix 3B). There are no remains of bathhouses that can be ascribed with certainty to Vandal North Africa. It is only the literary evidence that is proof that the Vandals did indeed bathe in the Roman style (EA-48; 52–56; Procopius). The references to thermal site at Baiae in the mosaic inscriptions and poems of the Anthologia Latina (EA-49; 53; 57) evoke the healing power and opulence that was associated with this resort, and prove that there was still some enthusiasm for bathing. However, when the Vandals seized power, there does not seem to be any regression in the economic activity.162 The large rural estates in Proconsularis and Numidia were probably confiscated by the Vandal elite, but the production changed little.163 Archaeological surveys of the rural landscapes in modern Tunisia seem to confirm this continuity; even recording a peak of sites during the 6th c.164 Furthermore, the system of municipal authorities continued.165 The reasons for this drop in building activity should not be ascribed to a general economic depression under Vandal rule. The difficulty in identifying ‘Vandal’ construction and restoration in archaeology and epigraphy could very well bias our view of this period. For most North African baths, the date of abandonment is unknown (see appendix 3B). Detailed stratigraphic excavations in Sabratha have demonstrated that at least two baths which were thought to have fallen out of use in Late Antiquity, were actually still use in the 6th and even 7th c. (see appendix 3B). For Cyrenaica, the archaeological evidence is problematic, as the small baths that have been identified in a number of sites in Cyrene’s hinterland cannot be securely dated, even if a 5th c. construction is possible (see chapter 3; graph 6). However, the difference with the 4th c., even if the sample is limited, is not as dramatic as in the Italian Peninsula and North Africa. The apparent break in the archaeological and epigraphic evidence 160  Heather (2010) 18–111. 161  Wickham (1998) 284; Christie (2011) 194–95. 162  Courtois (1955) 157. 163  Merrills and Miles (2010) 156–59. 164  Leone and Mattingly (2004) 138–42. 165  Lepelley (1992); Leone (2007) 141.

217

for western North Africa, probably linked to the difficult identification of the Vandal phases, is not apparent in Cyrenaica, which escaped Vandal annexation. In Egypt, there is a certain continuity in building activity compared to the previous century (appendix 3D, graph 7). At least 5 baths were constructed during the 5th c. (C75, C80, C81, C89, C95), while another 7 could be added (C77, C78, C84, C88, C90, C91, Xois). The total number of baths in use in this century is also the highest. This picture fits with a period of political and economic stability in the region. The newly constructed baths of Abu Mena—South (C75) and Marea (C89)—can be linked to pilgrimage sites or itineraries. The increase in the number of baths at these pilgrimage sites may be linked to the growing importance of the Church in civic matters.166 The decrease in the number bath-related papyri fits within a general decrease of papyri production in the 5th c. and—considering the archaeological evidence—does not reflect a decrease of public baths. The regression of building activity in Palestina is remarkable, especially as the region was still quite prosperous.167 The number of new baths drops from 15 to just 2 in the 5th c. (graph 8). The high number of baths that were still in use seems a direct result of the building boom of the previous century: many of the 4th c. baths were still in use in the 5th, pointing to the continued popularity of bathing in public baths (see appendix 3E). This could also help explain the small number of new constructions: most cities and settlements were already equipped with baths. In general, we can see a decrease of public baths in all regions during the 5th c. (compare graph 3 and graph 9). The drop in newly built baths is most remarkable, except in Egypt. Concerning the architecture of the baths, we can note that some smaller baths (C53, C54), especially those attached to ecclesiastical buildings (C27, Thamugadi Donatist Cathedral, Ma’ale Adumin) no longer had all the basic facilities of a Roman-style bathhouse. In some cases, the heated section was restricted to a single room, with or without a single-person pool. These bathing facilities could only have offered a very basic version of the Roman bathing routine. The decorative scheme of these modest baths was equally sober. It seems as if the impact of the austere Christian attitude towards the ‘sinful’ body was reflected in the architecture and decoration of these baths. Baths that were located near pilgrimage sites, but open to pilgrims instead of the monks, did have a more complex plan with often several heated rooms and large communal pools (C75, C76, C89, C90, C116). In Egypt, the double 166  Wipszycka (2007) 332–39. 167  Walsmley (1996); Kennedy (2006) 602.

218

chapter 4

5th c. Baths 60 50 Number of Baths

40 30

Continued

20

New

10 0

Italy

N. Africa

Cyrenaica

Egypt

Palestina

Continued

40

17

3

5

17

New

8

5

1

7

3

graph 9

Public baths in the different regions during the 5th c., based on the data from Appendix 3 (possible baths not included)

baths may have been an answer to the Christian critique on mixed bathing. However, the ancient authors, and again the medical writers in particular (Cassius Felix, Caelius Aurelianus), inform us that the bathing routine itself and the underlying theories about the human body changed little. Difficulties in the 6th c. In the Italian Peninsula it seems that very few baths were constructed ex novo (only one confirmed example), whereas several (larger) baths in important cities remained in use (graph 4 and 10). The few inscriptions record restorations of balinea (EI-51; 52) by civic authority or the clergy. Thanks to tile stamps mentioning Theoderic, we know that some baths were still active in Ravenna, Ostia and Rome, although the situation in the latter must have deteriorated after the Ostrogothic siege of AD 537–538 and the destruction of the aqueducts. Extensive building programs would not have been sensible during the Gothic and Lombard wars of the 6th c. Literary passages reveal that the new ‘foreign’ rulers such as Goths and Lombards were enjoying Roman-style baths (see chapter 2, esp. Cassiodorus, Jordanes, Gregory of Tours, Anonymous Valesianus). The time of large-scale (and archaeologically identifiable) restorations, constructions and commemorations of such works had passed. Similarly, there is only limited evidence of new construction in North Africa (graph 5). The inscriptions are actually epigrams by a poet named Felix, who praised the baths built by Vandal kings (graph 1). As has already

been mentioned, the difficulty in identifying the late antique phases in the archaeological record may bias our view. Detailed stratigraphic excavations in two baths in Sabratha pushed the date of abandonment into the 6th and even 7th c. After the Byzantine reconquest of North Africa, cities often shrunk to small fortified settlements, characterized by intermixed and interactive civilian and military areas.168 Baths that fell outside the new nucleus may have been definitively abandoned. According to Procopius, Byzantine authorities restored or constructed baths inside the fortified centres (see chapter 2, p. 64–65). The remains of such ‘Byzantine’ baths have been identified in Thamugadi and possibly Thugga (see appendix 3B), and in one inscription (EA-58). In Cyrenaica, there seems to have been limited building activity in some cities that fell under Byzantine control (Apollonia, Taucheira; see appendix 3C and graph 6). The basic components of a Roman-style bath were still present, even if the layout was simple and the finishing of the interior rather basic. However, the large public baths in Cyrene might still have been in use, while the City Baths in Ptolemais were probably abandoned after the aqueducts were cut. The number of baths in use in 6th c. Cyrenaica, however small the sample, shows no decrease in comparison to the 4th and 5th c. (graph 6). In Egypt, several baths were still in use in the 6th c., although only three were constructed ex novo (C76, C83, C96; appendix 3D and graph 7). The impressive 168  Leone (2007) 188–89.

219

Continuity and Change in Late Antique Public Baths and Bathing Habits

6th c. Baths

Number of Baths

20 18 16 14 12 10 8 6 4 2 0

Continued New

Italy

N. Africa

Cyrenaica

Egypt

Palestina

Continued

16

7

3

12

15

New

1

2

2

3

3

graph 10 Public baths in the different regions during the 6th c., based on the data from Appendix 3 (possible baths not included)

expansion of pilgrimage sites during the 5th c. had left most of them with large (double) baths, which now could be enlarged (e.g. C75 and the construction of an additional bath C76). This evolution is clearly visible in graph 7: the high number of new baths of the 5th c. is followed by a high number of baths in continued use during the 6th (graph 10). The continued popularity of baths can also be seen in the high number of papyri (see graph 2). In Palestina, several baths were still in use during the 6th c. (graph 10) and some smaller buildings were constructed (appendix 3D, 3E). The technology in these regions also shows us how alternative solutions were invented to cope with the local scarcity of ceramic building materials, water and fuel. By the second half of the 6th c., several baths fell out of use. It seems that the (partial) abandonment of public baths fitted within a larger process of urban decay, also affecting other types of public infrastructure such as fora, nymphaea and even roads.169 Tension had been building up in the east in the second half of the 6th c., with the Sasanians sacking Antioch in AD 540 and Apamaea in AD 572. Furthermore, plague swept through the area, wiping out 30 to 50 percent of the local population according to recent estimates.170 Failed crops and earthquakes added to the misery, possibly resulting in decreased investment in public infrastructure. 169  A similar process has been recorded for the baths in Syria, see Fournet (2012b) 220. 170  Mitchell (2007) 373–74, esp. n. 10.

Towards Medieval and Islamic Baths in the 7th c. The evidence for bathing in public baths in the Italian Peninsula is restricted to ecclesiastical structures (appendix 3A). The Bagni del Clero in Ravenna, the continued use of the church baths of San Giusto, the baths on the Palatine in Rome and the evidence of the Liber Pontificalis reveals that the Church constructed bathhouses, often for the poor or pilgrims. The archaeological visibility of these baths or the restoration of existing baths is poor (graph 11). The first medieval examples— the Crypta Balbi baths in Rome and the baths of Ravello and Amalfi—date only from the 12th c.171 The existence of these baths and the written evidence for baths in the 7th to 10th c. stress that bathing itself did not disappear and was still essential to maintaining a healthy body.172 However, the building in which this cleaning of the body took place had changed over time. Only the hypocaust system was a faint reminder of Roman-style baths. Bathhouses also did not disappear overnight in North Africa (appendix 3B). In Carthage and even in some smaller settlements such as Sitifis, baths seem to have survived the Arab conquest. The 8th c. extra muros Baths in Volubilis (Morocco) are still built following Roman principles (cold room, tepid room, hot room and with hypocaust), which points to a continuity of bathdesign and technology. In Cyrenaica, the city baths of Taucheira seem to have been used in the Islamic period. 171  Sagui (1990a); Caskey (1999). 172  Sagui (1990b) 104–106.

220

chapter 4

7th c. Baths 14

Number of Baths

12 10 8 Continued

6

New

4 2 0

Italy

N. Africa

Cyrenaica

Egypt

Palestina

Continued

4

2

3

12

11

New

0

0

0

0

0

graph 11

Public baths in the different regions during the 7th c., based on the data from Appendix 3 (possible baths not included)

The Byzantine fortress bath that was constructed in the early 7th c. reminds us that, even at this late date, small basic bath facilities were built. The lack of information about continued bathing activity may well be the result of the poor archaeological visibility of repair works. The 7th c. and the succeeding Early Islamic rule is a blind spot in our knowledge on baths and bathing habits in North Africa and Cyrenaica. The lack of contemporary written sources adds to the problem. It is, however, doubtful that the new rulers introduced a radically new type of bathhouse. If anything, the evidence (e.g. in Volubilis), or the lack thereof, points to continued use of the old Roman-type baths. The rulers may have changed, but the people using and building baths stayed the same in this early phase of Islamic conquest. In Egypt, 12 baths remained in use in the 7th c. (see graph 7 and 11 and appendix 3D), and 5 of these may have functioned into the 8th c.173 Despite the Sasanian occupation between AD 617 and AD 629 and the Muslim conquest from AD 639 onwards, Egyptians seem to have bathed as usual. The Muslim presence in Egypt restricted itself mainly to the newly founded capital Fustat.174 Papyri dating to this period show that daily life in the rest of Egypt changed little in the first 50 years 173  However, it is sometimes difficult to distinguish the last ‘bath’ phase from the first phase of secondary use. The actual abandonment of the baths in Abu Mena is a case in point. See also Lepaon (2015) 112–13 about the 8th c. secondary use of the Placcus Baths in Gerasa (Jordan). 174  Sijpesteijn (2007) 442–43.

of Muslim rule. The same officials remained in power and Greek remained the administrative language.175 As long as wealthy people were prepared to invest money in operating costs and maintenance, baths survived. Not surprisingly, the South Baths in Abu Mena, the second bathhouse in Pelusium, the Byzantine Baths in Marea and possibly the baths of Tell el-Makzan, lay on profitable pilgrimage routes. The 7th c. was a period of turmoil in Palestine. The first Muslim raids into Syria began two years before the Prophet’s death through the incorporation of the Arabian Bedouin tribes from the fringes of the cultivated lands.176 The Arabs must have known the area well, as they were renowned traders. Amr bin al-As, the commander who conquered Egypt, was himself a former merchant. Between AD 634 and AD 644, the Middle East fell into Muslim hands. Considering that Arab rule consisted mainly of a small military presence in conquered cities, without the immediate settlement of Arab colonists, the absence of new urban infrastructure should not come as a surprise.177 As was the case in Egypt and North Africa, the Arab conquerors left the administration of the Middle Eastern cities to the traditional institutions (councils, Church). In most cases, only taxes were imposed, collected by the same Christian officials, but ultimately handed over to Arabic-speaking rulers. 175  Kennedy (2007) 159–61; Sijpesteijn (2007) 444–51. 176  Hillenbrand (2005) 334; Kennedy (2007) 71–72. 177  Kennedy (2007) 214–15.

221

Continuity and Change in Late Antique Public Baths and Bathing Habits

Number of Baths in Use

Late Antique Baths 80 70 60 50 40 30 20 10 0

4th c.

5th c.

6th c.

7th c.

Italy

75

48

17

4

North Africa

56

22

9

2

Cyrenaica

4

4

5

3

Egypt

12

12

15

12

Palestina

29

20

18

11

graph 12 Total number of all late antique baths in use (new and continued use) per region and per century (possible baths and baths with an unknown construction date not included)

Greek or Syriac remained both the every-day and official language. When Arab settlers did decide to move in, they often gathered according to their old tribal configuration in separate quarters of the town, sometimes even in the suburbs.178 The competition that developed between these groups only strengthened the transformation of the Roman town that was already under way. The colonnaded streets were taken in by shops and stalls, while the forum was overbuilt by houses.179 Islamic laws do not seem to have been strict on the encroachment of former public space as long as neighbours or the commercial life of the city were not disturbed.180 The unfolding of this souk-like fabric was not necessarily a sign of decline. Rather, it could be interpreted as a renewed focus on trade and commercial activities.181 In Palestina, several bathhouses survived the Arab conquest (graph 11; appendix 3E). It seems that the baths were only abandoned when their part of town fell into decline (see the case of Scythopolis). It was only in the 8th c. that gradual change to bathing habits, which had started in Late Antiquity, resulted in the widespread adoption of ‘steam’ technology and the disappearance of the pool, two elements labelled as main characteristics of the hammam.

178  Carver (1996) 207–208. 179  Kennedy (1985) 12–13. 180  Galikowsky (1997) 343. 181  Kennedy (1985) 21–26; Galikowski (1997) 344–49.

A Habit in Decline? In conclusion, if we look at the evolution of the number of public baths in use from the 4th to the 7th c. in all regions under consideration, we can remark on a general diminution (see graph 12). However, a closer look at the data for each individual region shows a more complex picture. For the Italian Peninsula, we see a gradual decrease, which could be linked to the growing socio-economic problems that challenged the region’s prosperity, especially from the 5th c. onwards (loss of North Africa, incursions by Goths, Huns and Lombards). The drop in numbers is also quite clear in North Africa, and again it is sharpest from the 4th to the 5th c. It is less clear how the Vandal conquest affected the public baths, but there is evidence of continuity. In fact, the poor archaeological visibility of 5th c. phases may be the main problem here. The situation is quite different for the eastern regions. In Cyrenaica, spared of Vandal conquest, the sample is too small for making sweeping statements, but there is a continuity in the number of baths. The same goes for Egypt, where the larger sample actually shows a status quo, and even a small rise during the 6th c. Moreover, if we compare the number of public baths from the Greek-Hellenistic period to Late Antiquity in Egypt, we can note how most baths were actually built during the Greek-Hellenistic period (see graph 13). The slow penetration of Roman culture and urban concepts in Egypt has been put forward as an important reason for the low number during the High Empire. However, political stability and prosperity,

222

chapter 4

Baths in Egypt 50

46

45

Number of Baths

40 35

31

30 25 20

17

15 10 5 0 Greek-Hellenistic (3rd - 1st c. High Empire (1st - 3rd c. A.D.) B.C.)

Late Antiquity (4th-7th c.)

graph 13 Newly built public baths in Egypt during the Greek-Hellenistic period, High Empire and Late Antiquity. The numbers derive from the most recent catalogue of public baths in Egypt (Redon (2017a)).

linked to strong agricultural production that few invaders dared to disrupt, resulted in a rise in numbers during Late Antiquity. In Palestine, there is a gradual decrease, but not nearly as marked as in the Italian Peninsula or North Africa. The region’s prosperity during Late Antiquity is well-attested. The sharpest drop in numbers is not coincidentally in the 7th c., when Sasanian and Arab invasions temporarily disrupted Palestine’s

affluence. With a renewed stability under Arab rule, the number of public baths rose again (see map 30). The Egyptian and Palestinian examples demonstrate that the equation ‘High Empire equals boom in bath construction’ is not regionally all-encompassing. It reminds us how the longevity of a specific type of public building not only depends on a favourable economic context (funds), but also needs cultural support (demand).182 182  Maréchal (2018).

Epilogue—The Transition to the Hammams The adoption of the public bathhouse by Early Islamic societies is difficult to reconstruct through the textual evidence, as no contemporary sources have survived. The archaeological evidence is concentrated on the Middle East, especially Jordan and Syria, and dates mainly from the 8th c. There is no textual or archaeological evidence for a type of public bathhouse in the Arabian Peninsula before the start of the Islamic conquest. It seems then, that the Arabs first came into contact with bathhouses when they visited (as merchants) and later conquered the Romano-Byzantine cities of the Middle East. Existing Roman-style baths were kept in use during the Early Islamic period (see appendix 3A–3E), with few transformations in the original plan (see chapter 3). A change in ruler had not seen a change in the way people bathed. It is unknown to what degree the new Arab rulers participated in the bathing habit or if they cared about the fate of the bath buildings. In some cities at least, the bathhouse was restored with an Islamic touch (e.g. the inscription in Taucheira, the Byzantine Baths). The Umayyad caliphs were charmed by Roman-style baths. Several ‘hunting lodges’ with a luxurious suite of bathing rooms were discovered in the Umayyad heartland (see chapter 3 and map 30). These baths were built in the Romano-Byzantine style, including the standard succession of cold, tepid and warm rooms, the presence of pools and a hypocaust heating system. Even the decoration showed important influences of Romano-Byzantine, but also of Sasanian art. There was no ban yet on representing the human figure. The most important novelty of the Umayyad baths seems to have been the steam boiler that helped in heating the caldarium. The full development of this technique took place under the Umayyads, even if there are already early examples of such a system (e.g. Karanis) in Late Antiquity. Furthermore, the small caldaria with several small pools often found in late antique baths must have been quite steamy and could hence be seen as a prelude to the development of steam rooms (dakhli, harâra, or maghtas in Egypt). The main problem in pinpointing the introduction of the steam-boiler (qadr) is that the remains are rarely preserved to a sufficient height to identify this system. In the Middle East, where the preservation of the buildings is often excellent, the steam-boiler seems to have been introduced in the course of the 8th c., possibly contemporaneous with the disappearance of wall heating. The disappearance of pools in favour of simple washbasins (ĝurn, pl. aĝrân) is also characteristic for the later hammams, although this development is regional, as the hammams the Maghreb (Volubilis, extra muros © koninklijke brill nv, leiden, ��20 | doi:��.��63/9789004419421_007

baths) and in Egypt (hawd, pl. ahawad) kept the immersion pools. The disappearance of the pools in Syria might be linked to stricter religious teachings that also start to appear in the contemporary texts. The separation of the sexes and the obligatory loincloth are just two of the measures that appear in legal texts. From the 9th c. onwards, there is more Islamic textual evidence to inform us about baths and bathing habits. Religious teachings about bodily hygiene, especially ablutions preceding prayers and important events and the habit of epilation, must have facilitated the acceptance of bathhouses in Early Islamic society. Furthermore, medical authors still followed the classical medical ideas and included baths in their prescriptions. Secular texts mention baths in various contexts, whether the collection of rents for the operating of a bathhouse or complaints by citizens about the smoke of the furnaces. It seems as if the baths had become an integral part of the Islamic city. Rich patrons and caliphs were building hammams to gain the support of the people. Public bathhouses were indeed still important tools for the selfpromotion of the upper class. And the baths were still important social hubs, especially for women. But in this social aspect also lurked a possible danger, as the baths were not only a “lieu de sociabilité” but also a “lieu social du corps”.1 Just as in the Roman period, the aspects of nudity and sexuality associated with these buildings ignited religious debates about acceptable and unacceptable habits and behaviour. These debates were often written down in religious-legal texts, laws and codes, which mainly had a local action-radius. Archaeological evidence convincingly demonstrates how Early Islamic society incorporated the existing Romano-Byzantine style bathhouses and gradually adapted both plan and technology according to its own preferences and needs. The textual evidence of the later periods reveals how public baths became important social hubs within the Islamic city and how building and using these baths became an essential part of Islamic society. In these aspects, the hammams are undoubtedly the heirs of the Roman baths. With the transformation of the bathing habit and of the bathhouse itself, a proper Islamic-style bathhouse developed that can be regarded as fundamentally different from the Roman baths, depending on what one considers to be the key element of the Roman bathing habit. If we consider the communal immersion pools as the most important characteristic— especially in comparison to Greek-style baths—we can 1  Carlier (2000) 1305.

224 state that Islamic baths were completely different (but, then again, so would most of the Byzantine type of baths with their individual pools). If we narrow it down to relaxation in hot water—which could also be enjoyed in single-person pools—the Islamic baths were content with simple hot water ablutions. If we consider the gradation of heat in the different rooms as typically Roman, we must conclude that the Islamic baths were actually not that different, even in the manner of achieving this

Epilogue—The Transition to the Hammams

effect (hypocausts). And if we consider the idea behind the Roman bathing habit—taking care of one’s body while participating in a set of shared cultural values— we can remark that the hammams provided a similar environment for its bathers. Going to the hammam, as the text confirm, was not only about cleaning one’s body, it was about sharing a cultural experience, in a building that reflected the shared values in both its functionality and its decoration.

General Concluding Remarks and Prospects for Future Research We can conclude this book with some general remarks about the evolution of public baths and bathing habits during Late Antiquity. The first and possibly most important conclusion is that public baths did not suddenly disappear after the reign of Constantine, the builder of the last large imperial type baths in Rome. In all of the regions discussed in this study, public baths continued to be built and restored, both by the emperor and by local authorities—including the Church—as well as by private benefactors. And what is equally important is that these bathhouses were still built in the same manner as in previous centuries. If we follow the theories of architectural sociology, the fundamental expectations of the people using these buildings had changed little over time. Baths still comprised a heated section with warm water pools for immersion and a cold section with cold water pools in order to counterbalance the weakening effects of the heat. The fact that the earliest Islamic-style baths still functioned in the same manner confirms that the core of the bathing habit had remained unchanged. Of course, there were several changes to the layout, and sometimes even in the technology, changing the ‘look’ of the bathhouse, and undoubtedly also changing the bathing experience. The best examples of this are the disappearance of the area for physical exercises or the smaller—and hence more intimate—bathing rooms (and especially pools). Yet, whether one considers these changes to be fundamental enough to speak of ‘a new type of late antique—or Byzantine—bathhouse’ depends on one’s definition of a Roman bathhouse. If one considers the technology (hypocausts, tubuli, testudines) to be the key element, and hence also the gradation of heat between the different rooms, one should admit that little changed in Late Antiquity. If one considers the collective immersion pools as the most characteristic element, especially in comparison to the Greek-style baths (however, see the remarks by Monica Trümper presented in chapter 1), one will conclude that late antique baths had turned towards a more ‘personal’ form of bathing, in a collective space. However, evidence presented in this book points to a continuation of the existing bathing habit, itself closely linked to medical theories and the classical world view. From this approach, a classical ‘mental legacy’ still survived in Late Antiquity, despite the important impact of Christianity. Lodged between the rise of public baths in the Greek-Hellenistic period and the success story of the hammam from the late 7th c. until this day, Roman-style collective baths can be seen as a phase—yet a very important and influential phase—of Mediterranean © koninklijke brill nv, leiden, ��20 | doi:��.��63/9789004419421_008

bathing habits. That what makes the Roman era stand out from the other periods is the importance that was given to the public bathhouse in the urban fabric. Not only was it a social hub offered by those in power to the people, it was also an architectural and technological ambassador of Roman culture. Never before, and arguably never again, would public baths be constructed on such a large scale, in such a homogenous form, and spread over such a vast region. The popularity of the building, and the underlying ideas about taking care of the body and presenting oneself to the community, is reflected in its longevity. Long after other types of Roman public buildings such as theatres or circuses had fallen out of use, baths were still being built and restored. This shows that public bathhouses remained a defining element in the urban layout of a city. Moreover, baths often took over the civic functions of other types of building that disappeared. So if we consider the role of public baths within the urban fabric as the most important characteristic of Roman-style baths, then there most definitely is a continuity into Late Antiquity. If we pass from this general observation of the late antique bathing habit to a more detailed analysis of the bathhouse, we should immediately state that the regional and even local contexts were the decisive force for change and continuity. As has best been explained for the small pools, there is no one-size-fits-all explanation for the general trends that have been identified for late antique baths. The size and separate architectural features of late antique baths were mainly the result of regional or local circumstances and developments, whether it was the availability of certain resources (water, fuel, building materials) or the wealth of the people in charge. These ‘external factors’ were often already present in previous periods, only to become increasingly prominent in Late Antiquity due to changing socio-political circumstances (e.g. a city being cut off from a trade network). We can state that the Middle East and Egypt, and to some extent North Africa, benefitted until a late date from favourable circumstances. In these regions, Roman-style baths were still being constructed in a remarkably similar basic layout well into very Late Antiquity. No book can end without the obligatory plea for more research. As this work has attempted to provide a first general summary of the late antique public baths and bathing habits, much work still needs to be done. Archaeological evidence again and again proved the importance of solid excavation reports. The chronology of the bathhouse is indeed one of the most important

226

General Concluding Remarks and Prospects for Future Research

obstacles to presenting a reliable evolution of this important building. A re-examination of previously excavated sites through published literature and especially through unpublished excavation reports—the so-called grey literature—can undoubtedly improve our understanding of late antique baths, especially in times when subsidies for archaeological field work are limited and certain areas of the Mediterranean become inaccessible due to turbulent political situations. The focus on the baths of a specific city, which has been attempted herein, can also yield important information regarding the evolution of the baths in a changing urban fabric and can even provide pointers for an intra-site chronology (see the tubuli of Ostia). In times when national authorities are reluctant to open yet another excavation site that has to be maintained afterwards, small-scale test pits in previously excavated sites can prove their worth, as a meticulous stratigraphic sounding can provide invaluable chronologic information. The research projects into the baths of Sabratha and the Forum Baths in Ostia are good examples of how important results can be obtained with only very limited intervention. In this context, the potential of surveys and geophysical prospection should also be mentioned. Surface finds in the surroundings of an excavated building or one

that has been identified on the results from geophysics can at least give an idea of the general chronology of a site. Small test trenches can then confirm or reject the proposed chronology (or identification as bathhouse in the case of geophysics). The baths west of the Palazzo Imperiale in Ostia are a case in point. The presence of (small) semicircular or apsidal forms (pools) and cruciform rooms (often caldaria) can act as ‘markers’ in identifying baths in the results of geophysical survey. The geophysical survey of the area immediately surrounding a (presumed) bathhouse can also provide important data about the (urban) context in which the building was constructed (e.g. the extent of an insula, the presence of sewers, the presence of other buildings / structures, etc.). Such methods have successfully been tested in Ammaia (Portugal, IAAP 2009 Radio-Past project) and Ptolemais (Libya, Polish archaeological mission). In the hope of having paved the way for future scholars on late antique baths, archaeologists and historians alike, the challenge now is to further elaborate, confirm or reject the results of this study. If anything, this research might act as a ‘lifeguard’ for starting researchers, helping them to avoid drowning in the depths of bathrelated literature. Salvom lavisse!

appendix 1

Late Antique Inscriptions Mentioning Baths Late antique inscriptions pertaining to baths and bathing habits have been catalogued by region. The epigraphic evidence found in Italy has been given the code EI, the data from North Africa EA and from Palestina EP. Within each group, the inscriptions are ordered chronologically. For each inscription, the following information is given: Reference to the epigraphic corpus, journal or monograph in which the inscription has been published. The abbreviations of the corpora can be found at the beginning this book (p. xxi–xxii). Loc. The location where the inscription has been found. The ancient find-spot is mentioned in italics, followed in brackets by the ancient region / province and the name of the (closest) modern city / village in the vicinity. Cont. The context in which the inscription was found. Date Proposed date of the inscription with information upon which this date has been based in brackets. The inscriptions included all date between AD 300 and AD 700. Text The text of the inscription, as interpreted by the publication under Ref. The transcription follows the Leiden System of epigraphic conventions. Transl. Translation of the text (only for a selected number of inscriptions). Note Additional information about the inscription (only for a selected number of inscriptions). Ref.

Inscriptions from the Italian Peninsula EI-1 Ref. CIL 6.1130 = CIL 6.31242 = ILS 646 Loc. Rome (Reg. I) Cont. Large slab found in two parts in the neighbourhood of the Baths of Diocletian. Date AD 305–306 (imperial titles) Text [Dd(omini) nn(ostri) D]iocletia[nus et ⟦Maximian]us⟧ Invicti / Seni[ores Augg(usti) patres Impp(eratorum) et Ca]ess(arum) et / Dd(omini) nn(ostri) Cons[tantius et Maximianus Invicti Augg(usti) et] / [Severus et] Ṃax[iminus nobilissimi Caesares] / [ther]mas Feli[ces Diocletianas quas] / [Maxi]mianus Aug(ustus) [rediens ex Africa sub] / [praesen]tia mai[estate disposuit ac] / [fieri iussit e]ṭ [Diocletiani Aug(usti) fratris sui] / [nomine consecrav]ịṭ [coemptis aedific]ịịs / [pro] ta[nti ope]ṛịs magnị[tudine omni c]ultu / perfec[tas] Romanis sui[s] dedic̣[ave]ru[nt]. © koninklijke brill nv, leiden, ��20 | doi:��.��63/9789004419421_009

EI-2 Ref. Loc. Cont. Date Text

CIL 6.1173b Rome (Reg. I) Found in the Baths of Caracalla. AD 305–306 (imperial titles) Maxim(o) Aug(usto) / n(ostro) Diocletiano II co(n) s(ulis) / sub felice proc(urator) aug(ustus) n(ostrus) / therm(ae) antoniniarum

EI-3 Ref. AE (1984) 151 Loc. Lavinium (Reg. I, Pomezia) Cont. Marble architrave in several fragments, found in situ in a bathhouse excavated in 1962. Date AD 313–324 (imperial titles) Text [Dd(omini) nn(ostri) Flavius Valerius Consta]ntinus Maximus et Valerius Licinianus Licinius Pii Felices Inv[i]cti semp[er Augusti] / [thermas ---]i temporis deformatas Laurentibus suis addito cultu restituerunt curante Camilio Aspro v(iro) c(larissimo) cu[ratore ---] EI-4 Ref. CIL 6.1136 = CIL 6.31244 Loc. Rome (Reg. I) Cont. Marble slab found in fragments at different locations (in gardens near S. Croce in Gerusalemme). Date AD 317–324 (imperial titles) Text D(omina) n(ostra) He[lena venerabilis do]mini [n(ostri) Constantini A]ug(usti) mater e[t] / avia beatis[simor(um) et floren]tis[simor(um) Caesarum nostr]oru[m] / therm[as incendio d]estru[ctas restituit] EI-5 Ref. AE (1937) 119/120 = SupIt-09 (1992), 85–90 (nr. 34) Loc. Amiternum (Reg. IV, L’Aquila) Cont. Bronze tabula ansata in honour of the citizen C. Salius Pompeianus Sofronius. Date AD 324 (imperial title; prosopography) Text (..) thermas quas iam olim disperier{e}ant antiquitus inpendiis et sua pecunia cum porticis novis factis et omni ornamento at pulcridinem restauravit / statuisque decoravit et nomine d(omini) n(ostri) Constantibeatiss(imi) Caes(aris) natalle idibus Nob(embribus) dedicavit, quarum dedicatio{b}ne biduum teatrum et dena iuve/naliorum spectaculis exsibuit sub {v} presentia Cl(audi) Urani, v(iro) p(erfectissimo), corr(ectoris) n(ostri) (..) Note For Claudius Uranius, see PLRE 1.982, s.v. “Uranius 4”.

228

appendix 1

EI-6 Ref. ZPE (2001) 272–273 nr. 3 = AE (2001)1009 Loc. Aquileia (Reg. X) Cont. Marble base found during works on the Piazza Capitolo. Date AD 324–330 (imperial titles) Text [Imp(eratori) Ca]es(ari) Flavi[o] / [Constant]ino Maxim[o] / [Victori s]emper Au[g(usto)] / [---therm] as indulgent[ia] / [eius vetusta]te conla[psas] / -----

EI-11 Ref. CIL 9.2338 = ILS 5732 Loc. Allifae (Reg. I, Alife) Cont. Reused in the stairs of the altar in the church of Santa Maria della Neve. Date AD 352–357 (prosopography) Text Fabius Maximus v(ir) c(larissimus) rect(or) prov (inciae) / thermas herculis vi terrae mo/tus eversas restituit a fundamentis

EI-7 Ref. Loc. Cont. Date Text

EI-12 Ref. CIL 9.2447 Loc. Saepinum (Reg. IV, Sepino) Cont. Unknown. Date AD 352–357 (prosopography) Text Fabius Maximus v(ir) c(larissimus) / re[c]tor provinciae / thermas Silvani vetustat(e) / conlasas restituit / curante Neratio Const[ante] / patrono sum(p)tu proprio [.

AE (1972) 202 Asola (Reg. X) Rectangular stone found reused in a wall. AD 336 (prosopography) Valentius B[ae]bianus Iunior / balneum a so/lo fecit Nepoti/ano et Facu/ndo cons(ulibus).

EI-8 Ref.

AE (1996) 694 = ZPE (2001) 271–272 nr. 2 = AE (2001) 1008 Loc. Aquileia (Reg. X) Cont. Large block in two fragments found during excavations of the baths. Date AD 306–337 (imperial titles) Text [Restitutori operum publi]/corum [D(omino) n(ostro) Fl(avio) Constantino] Maximo [Pio Felici Victori] / semper Augusto / Septimius Aelianus v(ir) c(larissimus) et F[l(avius)] / Mucianus v(ir) p(erfectissimus) p(rae)p(ositi) operis / [f]elicium thermarum / [Co]stantiniarum pieta[ti] / [eius] semper dicatissi[mi].

EI-9 Ref. CIL 14.135 Loc. Ostia (Reg. I) Cont. Unknown. Date AD 337–350 (imperial titles) Text [Cons]tantius et Const[ans] / [therm]as incuria longi temporis destitua[s] / [---]orum ad pristinum statum refoman[das sumptibus] / [s]uis rediderunt / [---]o v(ir) c(larissimo) praefecto annonae cum iure gladii EI-10 Ref. CIL 14.3594 = InscrIt-04–01, 00151 Loc. Tibur (Reg. I, Tivoli) Cont. Marble statue base found during works in the basement of a modern house. Date AD 350 (prosopography) Text Furius Maecius / Gracchus v(ir) c(larissimus) / corrector Fla/miniae et Pice/ni ornatui / thermarum dedicavit // dedicata XI Kal(endas) Maia[s] / Gallo et Flacco co(n)s(ulibus)

EI-13 Ref. CIL 9.2212 = ILS 5690 Loc. Telesia (Reg. IV, Telese) Cont. Unknown. Date AD 352–357 (prosopography) Text Fabius Maximus v(ir) c(larissimus) / rect(or) prov(inciae) thermas / Sabinias restituit curante ordine splendidissimo Telesinorum. EI-14 Ref. Loc. Cont. Date Text

CIL 6.31916c Rome (Reg. I) Marble base found near S. Maria Maggiore. AD 358 (prosopography) Naeratius / Cerealis / v(ir) c(larissimus) / cons( ) ord( ) / conditor / balnearum / censuit

EI-15 Ref. CIL 11.4781 Loc. Spoletium (Reg. IV, Spoleto) Cont. Marble slab found near the church of S. Salvator in Piscina. Date AD 355–361 (imperial titles) Text Reparatores orbis adque urbium resti/tutores dd(omini) nn(ostri) Fl(avius) Iul(ius) Constantius P(ius) F(elix) semper Aug(ustus) / et Iulianus nobilissimus ac victoriosissimus Caes(ar) / ad aeternam divi nominis propagationem / thermas Spoletinis in praeteritum igne consump/tas sua largitate restituerunt EI-16 Ref. CIL 10.3714 = ILS 5478 Loc. Puteoli (Reg. I, Pozzuoli) Cont. Unknown.

229

Late Antique Inscriptions Mentioning Baths Date AD 364–378 (prosopography) Text Signa translata ex abditis / locis ad celebritatem / thermarum Severianarum Audentius Aemilanus, v(ir) c(larissimus), cons(ularis) / Camp(aniae), constituit dedicarique precepit, / curante Tannonio Crysantio, v(iro) p(erfectissimo) EI-17 Ref. AE (1998) 453 Loc. Rusellae (Reg. VII, Roselle) Cont. Marble slab found in 9 fragments at the eastern end of the decumanus (reused). Date AD 367–369 (prosopography) Text [Cond]itor antiquus taceat sit muta vetustas / [h]oc quod cernis opus rudibus consurgere tectis / sedula rectoris sollers instantia fecit / has rector thermas dedicat Arzygius EI-18 Ref. Loc. Cont. Date Text

CIL 14.137 Ostia (Reg. I) Marble panel found in three fragments. AD 367–375 (imperial titles) Thermas maritimas intresecus refectione cellarum fori[---] / [---]s soli adiectione DDD(omini) NNN(ostri) Valens Gratianus et Valentinianus Victo[---] / [victo] r(es) ac triumf(atores) semper au[g(usti) Fl(avvio)] / Proculo Gregorio v(ir) c(larissimo) praefecto annon(ae) urbis Romae curante decorarunt

EI-19 Ref. SupIt-05-RI, 00006 = AE (1913) 227 Loc. Regium Iulium (Reg. III, Reggio di Calabria) Cont. Unknown. Date AD 367–375 (imperial titles) Text Imperator Caesar Flavius Valentinianus Pius Felix Victor / ac triumfator semper Augustus et Flavius Valens Pius Felix / Victor ac triumfator semper Augustus et Flavius Gratianus / Pius Felix Victor ac triumfator semper Augustus Reginis suis / [t]hermas vetustate et terrae motu conlasas in meliorem / cultum formamque auspiciis felicioribus reddiderunt / reddita basilica marmorum quas numquam habuerat / pulchritudine decorata nova etiam porticu adiecta curante / Pontio Attico v(iro) c(larissimo) correctore Lucaniae et Brittiorum dedecatas(!) / quartum Kalendarum Iuliarum die domino nostro Gratiano / Augusto tertio et Flavio Equitio v(iro) c(larissimo) consulibus EI-20 Ref. SupIt-05-RI, 00007 Loc. Regium Iulium (Reg. III, Reggio di Calabria) Cont. Unknown. Date AD 367–375 (imperial titles)

Text Suffragan[tibus ddd(ominis) nnn(ostris) Valentiniano Valente et Gratiano] / ppp(rincipibus) mmm(aximis) ther[mae vetustate 3 et] / ruin[a conlapsae EI-21 Ref. CIL 10.6312 Loc. Tarracina (Reg. I, Terracina) Cont. Unknown. Date AD 378 (prosopography) Text [Avianius] / Vindicianus / v(ir) c(larissimus) cons(ularis) Camp(aniae) / [t]hermas vi / [ignis consumptas] / [restituit]. EI-22 Ref. Loc. Cont. Date Text

EI-23 Ref. Loc. Cont. Date Text

CIL 5.7250 = ILS5701 = AE (2007) 0890 Segusio (Alpes Cottiae, Susa) Found in the Baths of Gratian. AD 375–378 (imperial titles) Salvis d(ominis) n(ostris) Valente Gratiano et Valentiniano Imp(eratoribus) [perp(etuis) Aug(ustis) bon(o) r(ei) p(ublicae) nat(is)?] / thermas Gratianas dudum coeptas et omissas Mag(nus)(?) Aput[--- praeses] / Alp(ium) Cott(iarum) extruxit ornavit et usui Segusinae reddidit civit[ati --- re]/formavit(?) fistulas dedit aquam deduxit ne quid vel utilitati vel us[ibus deesset].

CIL 10.6656 Antium (Reg. I, Anzio) Found on a private estate. AD 379–383 (imperial titles) Florente imperio ddd(ominorum) aaa(u)ggg(ustorum) q(ue) nnn(ostrorum) Grattiani / Valentiniani et Theodosi principum maximorum / thermarum speciem ruinae deformitate sordentem / et periculosis ponderibus inminentem quae laantem / populum metu sollicitudinis deterrebat exclusa totius / scarie vetustatis ad firmam stabilitate usumq(ue) tectorum / Anicius Auchenius Bassus v(ir) c(larissimus) pro consule campaniae / vice sacra iudicans r(e)paravi in meliorem civitatis effigiem.

EI-24 Ref. AE (1979) 323 Loc. Cornus (Reg. I, Cuglieri) Cont. White marble slab found in the basilica of Cuglieri; later medieval inscription on reverse. Date AD 379–383 (imperial titles) Text Salvis d(ominis) n(ostris) Flaviis Gratiano V[alen­ tiniano et] /[Th]ẹọḍosio Invictissimis princip[ibus thermae] / Ạẹṣtivae quae olim squalorẹ [et magna] / ṛụina fuerant conlabsae a [fundamentis] / coṇstitutạ[e] ṇunc de fonte du[ctae sunt ---]

230 EI-25 Ref. Loc. Cont. Date Text

appendix 1

AE (1984) 10 Ostia (Reg. I) Inscription on a marble architrave. AD 331–400 Maximus has olim therm[as ---] / divinae mentis ductu cum O[------ // λουτρὸν ἀλεξίπον[ον ---]ιξεν βίκτωρ ἀρχὸς ἐὼν κύδιμος Αὐσονίης

EI-26 Ref. CIL 10.4559 = IATrebula 00009 Loc. Trebulla Balliensis (Reg. I, Treglia) Cont. Unknown. Date AD 340–400 (prosopography) Text L(ucio) Alfio Fannio Primo So[---] / quaest(ori) curatori frumento / du(u)mviro omnib(us) {h}onerib(us?) et / honoribus functo sacerd(otali) / viro patrono et curatori / [---] A(uli) Pisoni aurum atque / argentum obraetium se/rio il(l)ustravit thermas {a}e/tiam Constantinianas [l]on[g]a / vetustate corrupta(s) ex virib(us) suo / quam etiam e Sabinianeus ord/[---] rem filio viaceni/re statuam bene merenti patrono / [p]a[estan]tissimo [---] / duoviro senatus populusque Trebu(lanorum) / me[---] statuam decreverunt / d(ecreto) d(ecurionum) EI-27 Ref. CIL 10.5200 Loc. Casinum (Reg. I, Cassino) Cont. Unknown. Date Second half of the 4th c. (paleography) Text C. Pacci Felicis / C. Paccio Felici pat(ri) omnib(us) honor(ibus et honeri/bus pe[r]functo filio C. Pacci Felicis patron(i) / coloniae Casin(atium) cuius inmensis bene/ficiis patria cognoscitur cumulata cur(atore) / r(ei) p(ublicae) nost(rae) cuius provisione semper feliciter / guvernati (sumus) cur(atori) et instauratori aedium / publicorum cuius opera et sollicitu/ dine{m} inpendiisque propriis post / seriem annorum therme Noviani / nobis in usu sunt restitutae ob his / omnibus laboribus eius quos cir/ca patriam civesque suos exibuit / digno patrono universus populus coloniae Casinati/um una cum liberis nostris / statuam marmoream erigen/dam digne censuimus. EI-28 Ref. ILS 5700 = AE (1896) 112 Loc. Tarentum (Reg. II, Taranto) Cont. Unknown. Date 2nd half of the 4th c. (prosopography) Text Pentascinensibus thermis quae longo temporis / tractu intercepto aquae meatu lavacris fre/[que]ntari desierant undis largioribus afluen/[tibus ny]mphalem aquam in meliores usus sua / [liberalitate C(aius)]

Furius C(ai) l(ibertus) Togius Quintilius / induxit / [---] lio Petrio v(iro) p(erfectissimo) Transl. ‘Furius Togius Quintillus, freedman of Gaius, brought flowing nymphal water in more copious surges to the Thermae Pentascinenses at his own expense and for their better use. For a long time the baths had ceased to be visited in their lavacra, owing to the interrution of the water supply by an earthquake. [Aure?] lius Petrius, vir perfectissimus, saw to the work.’ (Fagan (1999a) 312) EI-29 Ref. CIL 11.4974 = ILS 5709 Loc. Cures Sabini (Reg. IV, Passo Corese) Cont. Marble slab reused as building material in a modern mansion. Date 4th c.? (paleography) Text Maria Anthusa et P(ublius) M[---] / baptisterium et cella[m] / de sua pecunia ma[ EI-30 Ref. Picus (2007) 254 = AE (2007) 535 Loc. Fanum Fortunae (Reg. VI, Fano) Cont. Limestone slab reused in a medieval and Renaissance cemetery. Date 4th c. or later (paleography and decoration of the slab) Text Splendore civitatis ad mon[---]/mis hiemis et aestatis lavac[rum --- patro]/nus civitatis Setorius Aux[---]/nis subtraxisset fatale decret[um --- hu]/iusce civitatis cum fatr[---] / summam manum pro[---]. EI-31 Ref. Loc. Cont. Date Text

CIL 11.4094 Ocriculum (Reg. VI, Otricoli) Marble cippus found in the Terme Iemali of Otricoli. 4th c. (prosopography) C(aio) Volusio Victor/i qu(a)estori r(ei) p(ublicae) Oc/riculanorum IIII / vir[o] aedil[i] III vir[o] iur/e d(i)c(undo) IIII (viro) quinq(uennali) cives / plebei hocemerito stat/uam huic posuerunt q/ui t(h) ermas iemalis ad pristi/nam dig(nitatem) restauravit et d(e)d(i)c(a)v(it) // P(o)s(i)t(a) di(e) III nonas No(vembres).

EI-32 Ref. CIL 11.4095 Loc. Ocriculum (Reg. VI, Otricoli) Cont. Unknown. Date 4th c. (prosopography) Text Provocati temporis beatitudinem dd(ominorum) / impp(eratorum) Constani [et Constantis] augustorumque nn(ostrorum) / voluptatem thermarum hiemalium Sex(tus) Cluuius Mar/tinus et M(arcus) Caesolius Saturninus omnibus honori/bus functi de sua pecunia

231

Late Antique Inscriptions Mentioning Baths oridini sev(?) civibus ocri/colanis ad meliorem pulcritudinem pro civi(tate) ad/fectione cum augmento operi novi exercientes / asicnaverunt et dedicaverunt EI-33 Ref. Loc. Cont. Date Text

EI-34 Ref. Loc. Cont. Date Text

EI-35 Ref. Loc. Cont. Date Text

CIL 11.4096 Ocriculum (Reg. VI, Otricoli) Large marble statue base of unknown provenance. 4th c. (prosopography) Bonae originis suboli et sin/ceritate praecipua prae/ dito Sex(to) Cluuio Martino / omnibus honoribus functo / laudabili viro restauratori / thermarum hiemalium cum M(arco) / Caesolio Saturnino fratre suo / pro tantis meritis erga se / ordo et cives splendidis/ simae civitatis Ocricola/nae statuam marmore/am patrono dignissimo / ad Perenne testimonium / publice censuerunt / feliciter // Dedicata XVI Kal(endas) / Decembres / Marcellino et / Probino Conss(ulibus)

CIL 11.4097 Ocriculum (Reg. VI, Otricoli) Large marble statue base of unknown provenance. 4th c. (prosopography) Bonae originis suboli et / sinceritate praecipua praedi / to M(arco) Caesolio Saturnino om / nibus honoribus functo lauda/bili viro restauratori therma/rum hiemalium cum Sex(to) Cluuio / Martino fratre suo pro tantis / meritis erga se ordo et cives splen/didissimae civitatis Ocricolanae / statuam marmoream patro/ no dignissimo ad Perenne tes/timonium publice / censuerunt / feliciter // Dedicata XVI Kal(endas) / Decembres / Marcelli / no et Probi / no Conss(ulibus)

CIL 6.1131 Rome (Reg. I) Found in the Baths of Diocletian. 4th c. (event, paleography) thermas Diocletianas a veteribus principib(us) institutas omn[es]

EI-36 Ref. CIL 6.29764 Loc. Rome (Reg. I) Cont. Unknown. Date 4th c.? (paleography) Text Balineum / Iuliorum / Akariorum EI-37 Ref. Loc. Cont. Date

CIL 6.1179 = ILS 5732 Rome (Reg. I) Marble slab of unknown provenance. 4th c. (imperial titles)

Text [fortissimis] / invictissimisque princ(ipes) / ddd(ominis) nnn(ostribus) Valentiniano / Valenti et Gratiano Aug(ustis) / Fl(avius) Antigonus v(ir) p(erfectissimus) p(rae)p(ositus) colymbum nemus vetustate lap/sum testaccio picturis ac statuis / cum omni cultu ador(navit) EI-38 Ref. Loc. Cont. Date Text

CIL 14.5387 Ostia (Reg. I) Tabula found in the vicinity of the Forum Baths. 4th c.? (paleography) [Quod thermarum species ita h]abebatur ut lava[ri] n[emo posset ---] / [--- s plen]dore excultam ad usum popu[li ---] / [---] annonae praefec[to]

EI-39 Ref. CIL 10.1707 = ILS 5692 Loc. Puteoli (Reg. I, Pozzuoli) Cont. Unknown provenance (possibly Pozzuoli, now in Naples). Date 4th or 5th c. (prosopography) Text Septimio Rusti/co v(ir) c(larissimo) cons(ularis) Camp(aniae) / provisori ordi/nis restaurato/ri thermarum / ob insignem amo/rem splendi/dissimus ordo / et honestissi/mus populus / patrono praes/tantissimo. Note For Rusticus, see PLRE 1.787, s.v. “Rusticus 3”. EI-40 Ref. CIL 10.212 Loc. Grumentum (Reg. III, Grumento) Cont. Unknown. Date 4th or 5th c. Text Rullus Festus / v(ir) c(larisssimus) corr(ector) Luc(aniae) et / Brit(tiorum) ad ornatum / thermarum / collocavit. Note For Festus, see PLRE 1.337, s.v. “Festus 13”. EI-41 Ref. Loc. Cont. Date Text

CIL 10.5349 Interamna Lirenas (Reg. I, Pignataro Interamna) Base found during excavations (NSc 1879, 187). AD 408 (prosopography) Industriae ac sapientiae praeclaro / M(arco) Sentio Redempto v(iro) l(audabili) primario civitatis / nostrae omnibus honoribus et honeribus cu/riae suae perfuncto ex origine patronatus / veniente qu populum suum pro sua be/nivolentia ab indictione auri argentiq(ue) pop(u)/lum suum liberum reddidit termas (a)etivas / in sordentibus ac ruina conlabsas ex prop(rio) / ad summam manum operis revocavit cuiuc tantis / be(ne)ficiis ca. nos comulatis ad perpetuam fa/mam statuam eidem erigendam universus / populus interamnatium censuerunt

232

appendix 1

// D(edicata) prid Non(ius) Iulias / Basso et Filippo vv(iris) c(larissimis) cons(ulis) Transl. ‘To industry and wisdom. To the outstanding M. Sentius Redemptus, vir laudabilis, the principal man of our community, who held all the municipal magistracies of his curia, a man of patronal descent. He freed his people from the impositio, of a tax on gold and silver, when his people came to him on account of his kindness; he restored at his own expense the summer baths down to their last detail, after they had fallen into ugliness and ruin. For such great benefactions heaped up around us, the entire people of Interamna voted that such a statue be erected to him for his everlasting fame. (On the side) On July 6 when Bassus and Filippus, of senatorial rank, were consuls (i.e. 408).’ (Fagan (1999a) 292) EI-42 Ref. CIL 6.1703 = ILS 5715 Loc. Rome (Reg. I) Cont. Marble tabula found in the garden of a private mansion on the Aventine. Date AD 414 Text Salvis ac Florentibus DD(ominibus) NN(ostribus) Honorio et Theodos[io] / perpetuis semper augg(ustis) Caecina Decius Acinatius / Albinus v(ir) c(larrissimus) praef(ectus) urbi vice sacra iudicans / cellam tepidariam inclinator omni pariete labent[em] / de qua cellarum ruina pendebat erectorum a fu[n]/damentis arcuum duplici munitione fulcivit / D N M Q eorum. Note The restoration may possibly pertain to the Thermae Decianae, just as the restoration of an unknown public building in CIL 6.1659 (La Folette (1999) 51). For the urban prefect Caecina Decius Aginatius Albinus, see PLRE II, Albinus (7). EI-43 Ref. SEG 42–931 Loc. Rome (Reg. I) Cont. Unknown. Date AD 425 Text Εἰσοράᾳς με, φίλος, Φαῦ[στον, σοφὸς ὅσπερ ἔπαρχος] Ῥώμης ξεινοδόκον θῆκ̣ [ε λοετρὸν ˘ - ·] Εἰσέτι δ’ ἀγγελίης τε θρασυπτόλεμ[οι κλύον ἀοτοὶ] χῶρος τ’ Εὐφροσύνης ἐρχομένο[υς δέχεται ·] Ἑζομένους δ’ ἵπποισιν ἰδὼν ἑκάτερθ[εν Ἀνάκτας,] Φαῦστον ἀνυμνήσεις καὶ πρὶν ἵ[κῃ πρόδομα] EI-44 Ref. Loc. Cont. Date

CIL 10.4865 Venafrum (Reg. I, Venafro) Found in a private garden. AD 370–450

Text

[Quintil]iani / [pro eius me]ritis et obse/[quiis colle]gia urbis Vena/[franae ---] Quintiliano / [rectori S]amnitico pa/[trono optim]o et examina/[tori aequis]simo statuam / [loco publi]co positam ob / [--- atq]ue therm[as.

EI-45 Ref. CIL 6.1670 = CIL 6.31889 = ILS 5716 = IG 14.1062 = IGUR I.36 Loc. Rome (Reg. I) Cont. Base found in the Baths of Titus or the Baths of Trajan. Date AD 465–467 (prosopography) Text D(is) M(anibus) / Iulius Felix Campanianus / v(ir) c(larissimus) praefectus urbis / ad augendam t(h)ermarum / Traianarum gratiam collocavit EI-46 Ref. CIL 6.1750 = CIL 6. 31920 = ILS 5703 Loc. Rome (Reg. I) Cont. Square stone found in the vicinity of the Baths of Constantine. Date AD 430–500 (443?) Text Petronius Perpenna Magnus Quadratianus v(ir) c(larissimus) et inl(ustris) praef(ectus) urb(i) / Constantin­ianas thermas longa incuria et abolendae civilis vel / potius feralis cladis vastatione vehementer adflictas ita ut agni/tionem sui ex omni parte perdita desperationem cunctis repa/rationis adferrent deputato ab amplissimo ordine parvo / sumptu quantum publicae patiebantur angustiae ab extremo / vindicavit occasu et provisione largissima in pristinam / faciem splendoremque restituit EI-47 Ref. CIL 9.1596 = ILS 5511 = AE (2013) 331 Loc. Beneventum (Reg. II, Benevento) Cont. Base that was reused in a palazzo wall. Date AD 420–555 (historical event) Text ] SPECT / [------] / iustitia [admir]abili castita/ te conspic[uo repar]atori fori pro / magna [parte co]nlapsi in ruin(as) / cotidie [auctus] restitutori ba/silicae [--- cu]m porticibus Sagit/ta[riorum et] regionis viae novae / repar[atori] thermarum Commodiana/rum re[para]tori collegiorum repa/ ratori [porti]cus Dianae reparatori / basilicae [L]ongini ac totius prope civi/tatis [post h]ostile incendium condi/tori [ob] insignia eius in omnem pro/ [v]in[ciam] praecipuaque in se et patriam / [urbem m]erita populus Beneven/[tanu]s ad aeternam memoriam / [---] statuam conlocavit. EI-48 Ref. CIL 5.4500 Loc. Anagnia (Anagni, Reg. I)

233

Late Antique Inscriptions Mentioning Baths Cont. Mosaic in the remains of a presumed bathhouse. Date 5th c.? (on the basis of the mosaics) Text Bene / lava Salv(m) / lotu(m) Peripsu/ma su EI-49 Ref. Loc. Cont. Date Text

ILS 5693 = AE (2003) 339 = AE (2010) 318 Liternum (Reg. I, Literno) Marble slab of unknown provenance. 5th c.? Balneum Veneris lon[gi tempo]/ris vetustate corruptum / Domitius Severianus v(ir) c(larissimus) con[s(ularis)] / Campaniae ad pristinam faciem [revo]/cavit curante hac(!) dedican[te] / Sentio Marso v(iro) c(larissimo) comite divinor[um] / curatore Capuensium Liternin[orum] / et Cumanorum Transl. ‘The Baths of Venus, damaged by the aging of a long period of time, Domitus Severianus, of senatorial rank, governor of Campania, built to its original appearance. Sentius Marsus, of senatorial rank, companion of the divine rulers, curator of Capua, Liternum, and Cumae, saw to the work and dedicated it.’ (Fagan (1999a) 245)

EI-50 Ref. CIL 11.7298 = ILCV 00364 (add) Loc. Volsinii (Reg. VII, Bolsena) Cont. Marble slab found reused in the pavement of the crypt of S. Christianae. Date 5th or 6th c. (paleography) Text ] / [is]tius ordinis sive civitatis [et] rec[t]ori omniu[m] / [c]ommeantium Maecio Paterno curatori et pa[trono] / [h]uiusce civitatis iudicio omnium conprob[ato] / [r]estauratori thermarum Tusciani qui vix[it] / [a]nnis LVII menses VIIII d(iebus) XX et fecit cum uxore su[a] / [vi]rginia ann(is) XXXV Apra uxor sed et Paterna / [et M]arcellus fili(i) patri piissimo depositio VI[--K(alendas?)] / [Se]ptembr(es) pax tibi cum sanctis. Transl. ‘[–] of that ordo or the city, and to the guide of all travellers, Maecius Paternus, curator and patron of this city; he earned the respect of all, restorer of the baths of Tuscianus. He lived 67 years, 9 months, and 20 days and spent 35 years with his wife Virginia. Apra his [second?] wife, and also Paterna and Marcellus, his children, (offer?) a trust to their most holy father, on this (date lost) of September. Peace to you with the saints.’ (Fagan (1999a) 281) EI-51 Ref. AE (2008) 524b Loc. Tarquinia (Reg. VII, Tarquinii) Cont. Marble bloc reused in the rural church of Santa Restitua.

Date AD 504 (consul) Text Aurelius Gloriosus l(audabilis) p(uer) mag(is)t(er) et c[u]/rat(or) ind(ictione) XII balneum ordenis / civitatis Tarquiniensium iuris / pribati longa incuria et temp[o] / ris vetustate consumtum / procurante Proiecticio v(iro) c(larissimo) pa/tre suo {consumtum}adqu(e) inminente Serban/do v(iro) h(onesto) suffecto / d(e) s(ua) p(ecunia) r(estituit) / usuique publico reddedit / Cethego v(iro) c(larissimo) cons(ule). EI-52 Ref. Loc. Cont. Date Text

ICUR n.s. II.4794 = AE (2009) 137b Rome (Reg. I) Marble slab found in two fragments in the Via Ostiense. 6th or 7th c. (paleography) [In] nomine dei patris omnipot[ent]is et domini nostri Iesu Chr(isti) fil[ii et] / [s]ancti paracleti Eusebius in fa(cti)s [re](n)ovavit cymeteriu(m) totu(m) [et] / columnas in portic(cib)s pictura[s] qua in ruinis era(n)t tota et [te]/[c]tu(m) cum teguls et tab(u)l[am]en(tis) et acuts et materi[at]/[is] tot(s) balineu(m) marmo[ra] qu(a)e minus (h)abuit et scamna [fenes]/[s]tras spec(u)lar(i)a item in s[up]erior marmoravit al[teu(m)] / [c]l(a)stra ncinos et claes p[o]suit ut potuit usque d[um] / [es]set in s(a)eculo fecit reliqu[a(m) f]abrica(m) quando exivit de [hoc] / [s](a)eculo remisit alumnis sui[s pe]cunia(m) et ipsi fabrica(m) vel[ave]/runt introitu(m) a martyres [qu]od est in public a funda[m(entis) f[a]/bricait me(n)sas a martyr[es t]otas fecit aquam in aline[u(m)] / per mangana fecit a con[ch]a(m) cubiculu(m) et cancellu(m) fec[it] / causa fur fecit (u)ia mul[ta m]ala fac(i)nt item sar(ta) t[ect]/a suscepit sigilla (u)inque in por[ticu p]osuit cm podiol // Aeterna b(onae) [m(emaoriae)] quies(cit) pacem in [Chr(isto) sperans] / in futuro primae resurrec[tionis die] / Palatio neps ei[u]s fecit d(e)p(osita) XIIII K(a)l(endas) I[---]

EI-53 Ref. ICUR n.s. I.1485 = ICLV 1901 Loc. Rome (Reg. I) Cont. Marble slab in two fragments. Mainly hypothetical reconstruction. Date 7th c.? (paleography) Text Balnea quae fragilis suspendunt corporis aestum / et reparant vires quas labor afficerit / quae constricta gelu validis aut solibus usta / admixto latici membra [liquore] levant, / [ut]amur causa propri[ae suadente] salutis / [at cave ne mors sit me]dicina homini / lubrica ne sensus rapiat turpetque [v]oluptas / effera

234 ne mentem luxuries stimulet / ebria neu vino dapibus neu viscera crud[a] / dissol[v]at fluxo corde lab[ante gula] / sobria sed casto foveant [tibi membra lavacro] / et quaesi[ta salus sit sine damno animae]. // Haec [tibi si quis amor vitae] te tangit h[onestae] / [quicumque es homi]num dicta fuisse [putes] / [tu tam] en ista magis cautus servare me[mento] / grex sacrate d(e)o corpore men[te fide] / cui bellum cum carne subest quae et vic[ta resurgit] / quam cohibere iu[v]at si refo[v]ere p[aras] / clau[--- ---s]aluti [-- -] / vulnere [ne doleas ---] quod medeare iterum / [--- in]veni bene parta remedia carn[is] / [--- - --] / [non] nostris nocet officiis nec culpa la[v]acri / quod sibimet generat lubrica vita malum est. Transl. ‘Baths, that alleviate the rage of the fragile bodies and repair the virtues, that were weakened by labour; baths, that reinvigorate the limbs which were immobilised by frost or burned by the sun, when the water is mixed, these are the baths we want to use for our good health. (Yet be aware that) the medicine (does not become the doom) of man! That slippery lust does not carry of and defiles the senses, that unrestrained luxury does not stimulate the mind, that the water of the initiated, when they are drunk from wine or recently fed by hearty meals, does not dissolve, when the vulnerable heart loses all control. Sober, the body will refresh itself (in a purifying bath) and the dead (health will not harm the soul). This is said to you (if you love a life of glory, wherever you are). But be aware that you remember these precautions even better, you herd by God blessed in body, mind and faith, who battle the flesh permanently. That you will bring it down, when it rises again while you get ready to refresh yourself (..) if you are not sick from a wound that you have to heal again (..) I have found a good remedy for the flesh (..) (Bathing) does not harm our duties and bathing is not guilty for what it brings forth: a slippery lifestyle is the evil.’ (after Bush (1999) 537–38) EI-54 Ref. CIL 6.1165 Loc. Rome (Reg. I) Cont. Found in the vicinity of the church S. Maria de Monterone. Date AD 344–345? (emperors and prosopography) Text [Dd(omini) nn(ostri) Constantius / ac Cons]tans triumphatores Augusti / termas vetustate labefactas restauraverunt / Q. Rustico v(iro) [c(larrissimo) praef(ecto) urbi - - Note We could assume that these baths were quite large and important (thermae) if the emperors paid for the restoration. The inscription might refer to the Baths of Agrippa, which were located in the vicinity of the church.

appendix 1

Inscriptions from North Africa EA-1 Ref. Duval (1989) nr. 33 Loc. Sufetula (Sbeitla, Tunisia) Cont. Large entablature block in two fragments, found in the Large Baths. Date Beginning of the 4th c. (paleography) Text ---]s semper [Augustorum? -- invic]tissimorum Caes[arum --EA-2 Ref. Loc. Cont. Date Text

AE (1968) 610 Sullecthum (Salakta, Tunisia) Mosaic inscription in the frigidarium. Beginning of the 4th c. (paleography) En perfecta cito Baiaru(m) grata voluptas / undantesque fluunt aq(uae) saxi de rupe sub ima / nisibus hic nostris prostratus libor anhelat / quisquis amat fratrum veniat mecumq(ue) laetetur.

EA-3 Ref. CIL 8.9183 = AE (1937) 31 = CLE 577 = ILCV 275 = Busch (1999) 223–24. Loc. Auzia (Aïn Bessem, Algeria) Cont. Slab found in two fragments. Date Beginning of the 4th c. (paleography) Text Balnea, rura, domus fec[it C]onstantius auctor et fecit ut memore[nt c]ari de sanguine nati nam ut plene, lotor, [ [dis]cas quid sit perferre laborem ]m est memoranda per annos est novi exempli et q[ lateri iuncta viro [prim]ma data coniugi tali adque suum docu[ ]ia tum imitare priora ut nomen ut quae[ ]s ut honor ampliet in aeuum Auzias qia poten[s ]s his per saecula vita Tuque dabis civibu[s ?] et tibi patria laudes et dabis ut supere[s ]um linguas inanes. Note Verses found on two slabs. Note how the first letters of the left-hand text vertically spells out the phrase Bene lavate (Thébert (2003) 500, nr. 75; see also Busch (1999) 223–24. EA-4 Ref. CIL 8.10766/16812 = ILAlg I, 1187 Loc. Naragga (Saqiyat Sidi Yusuf, Tunisia) Cont. Unknown. Date Beginning of the 4th c. (emperor) Text [Pro salute imp(eratore)] Caes(are) M(arco) Aureli Valeri ⟦Maximiani⟧ [- - -]iridi / coeptas tantum et per longam annoru seriem [- - -] / onus partim sua propria et civium liberalitate [- -] / patriae futurum et [- - -] earumdem therma[rum - - -] / [ded]icavit

235

Late Antique Inscriptions Mentioning Baths EA-5 Ref. Loc. Cont. Date Text

CIL 8.24582 Carthage (Tunis, Tunisia) Marble slab found in a bathhouse on the Borj Djedid. AD 317–318 (proconsul) s[---] / ther[m ---] / exquisitis a solo [---] / decoro cult[u dedicante / A]cone Catullino c(larissimo) v(iro) [---

EA-6 Ref. CIL 8.4878 = ILS 2943 = AE (1903) 97 Loc. Thubursicum Numidarum (Khamissa, Algeria) Cont. Unknown. Date AD 324 Text Beatissimo sa[eculo d(omini) n(ostri)] / Constantini ma[ximi victoris?] semper Aug(usti) et C[onstantini] / et Constanti(i) no[bbilissimorum Caess(arum)], / plateam v(e)terem S[---] / lapide spoliatam [---] / Nonius Marcellu[s ---] / Herculius, qui[nquennalis? ---] / constravit P[--- ther]/mas(?) et cel[las? ---] / [ruin]a(?) dilap[sas -----EA-7 Ref. CIL 8.22853 = ILTun 106 Loc. Thysdrus (El-Djem, Tunisia) Cont. Found in the cisterns of the palaestra of the Large Baths. Date AD 324–333 (imperial titles) Text a]ntea cultu / [---]rectu formatas / [---] pro beatitudine saeculi [ddd(ominorum) nnn(ostrorum) Imp(eratoris)] / [Flavi] Constantini maximi victoris [ac triumf(atoris) semper Aug(usti)] / [Const]antini et Constanti nobilissimorum [---] / [---]ne Caesarum in meliorem faciem mac/[ //]ris in mult[---] / [---]vati accessum [---] / [--- s]peciem decor[---] / [--- curant]e(?) Val(erio) I[---] / [---]gaio III[---] / [---]D[ Note As there is no mention of a bathhouse in this inscription, it is not sure if the works (restoration or decoration?) commemorated here, concerned the Large Baths (Thébert (2003) 497). EA-8 Ref. Loc. Cont. Date Text

AE (1978) 864 Belalis Maior (Henchir el-Faouar, Tunisia) Lintel found in the vestibule of the Baths of the Forum. AD 326–333 (imperial titles) Beatissim[o saeculo invictorum principum Fl(avi) Valeri Constantini maximi victoris semper Aug(usti) et Constantini Iun(ioris) et Constanti glori]/ osissimor[um Caes(arum) ---] / parva solum [---] / quo cla(u)sa(?) clu[aca? --- pro]/[con]sul(atu) [M(arci) C]ae(i)on[i Iuliani c(larissimi) v(iri) ---] / [---]a vetusta[te conlapsa ---] / [---]L[---]C[------

EA-9 Ref. CIL 8.28065 = ILAlg I, 1033 Loc. Thagura (Taoura, Algeria) Cont. Long fragment of 1.2 m with two inscriptions of a different date. Date AD 326–333 (imperial titles) Text Constanti]nus et Constantius nobb(ilissimi) ac beatissimi Caes(ares) // post(?) missione]m honestam ex centurione Plu[---] / [---]ONE thermas ex HS CCCC mil(ibus) nummu[m Note The inscription consists of two parts. The upper part was added at a later date and is written in a smaller script. It probably commemorated the restoration of baths, whose dedication was recorded in the lower part of the text (Thébert (2003) 516). EA-10 Ref. AE (1992) 1776 Loc. Ammaedra (Haïdra, Tunisia) Cont. Inscription on a large apex stone, found in a bathhouse. Date AD 336 (consuls) Text Nepotia[no et Facund]o co(n)s(ulibus) / P(ublius) Rutilius V[--- flam(en)] / perp(etuus) curator [r(ei) p(ublicae)] / absidam a solo [in ther]mis / hiemalibus sua pecunia addidit. EA-11 Ref. Loc. Cont. Date Text

ILTun 814 = AE (1925) 72 Tubernuc (Aïn Tebornok, Tunisia) Found in the bathhouse of Tubernuc. AD 337 (imperial titles) Virtute clementia m[emor]ando pie/tate omnes a[ntecedenti?] d(omino) n(ostro) Fl(avio) Clau/ dio Consta[n]t[ino iu]niori / Aug(usto) / L(ucius) Pap(inius) Pacatianus Fl(avius) Ablabius [[[---]]] / [[[---]]] C(aius) Annius Tiberianus Nes/[to]ri[u]s Timonianus viri cla/[rissimi p]raefecti pr(a)etorio

EA-12 Ref. CIL 8.23131 Loc. El Haouria (Tunisia) Cont. Mosaic inscription on the threshold of a cold room in a possible private bathhouse. Date First half of the 4th c. (mosaics) Text Invide livide titula ta/nta quem adseveraba/s fieri non posse perfecte / sunt dd(ominis) nn(o)ss(tris) mi/ nima ne contemnas. EA-13 Ref. CIL 8.21518 Loc. Castellum Tingitanum (El Asnam, Algeria) Cont. Mosaic inscription of a bathhouse depicting a hunting scene.

236

appendix 1

Date Middle of the 4th c. (mosaic) Text Si/liqua frequens foveas mea membra / lavacro. EA-14 Ref. Loc. Cont. Date Text Note

CIL 8.7024 = ILAlg II.593 Cirta (Constantine, Algeria) Fragment found in the Large Baths. 4th c.? (paleography) [thermas Consta]ntinian[as] The reconstruction of the text is uncertain.

EA-15 Ref. ILTun 622 = AE (1934) 133 Loc. Henchir Haouli (Tunisia) Cont. Unknown. Date AD 340–350 (imperial titles) Text Saeculo felic[issimo ---] / dd(ominorum) nn(ostrorum) Fl(avii) Constanti et Fl(avii) Con[stantis ---] / administrantibus etiam / Fl(avio) Dardanio amp(lissimo) proc(onsole) c(larissimo) v(iro) et lgatis [---] / [p]ulcerrimum factum cum porticus [---] / [--- ]turpia foedabantur ad statum [---] / [pisc]inalis ad restaurationem d[---] / [-]s colitumque nitent soliar[em cellam ---] / [-] oleum a fundamentis perc[---] / [-] proneum aquiducti a fu[ndamentis ---] / [-]ium solium vero inst[auravit ? --] / [-]t C(aius) Aurelius Stat[ianus ---] / [-] cur(ator) r(ei) p(ublicae) una cum omn[ibus decurionibus? ---] / [--]sinsensium perfecit et [dedicavit]. EA-16 Ref. Loc. Cont. Date Text

ILTun 1093 Carthage (Tunis, Tunisia) Found on the Byrsa hill (out of context). AD 340–350 (proconsul) --- Fl(avius) Dar?]danius thermis procon[sul ---

EA-17 Ref. AE (1916) 87–88 Loc. Thuburbo Maius (El Fahs/Henchir Kasbat, Tunisia) Cont. Marble slab found in several fragments in the Summer Baths. Date AD 361 (imperial titles) Text que hu[more superfluo [m]agis f[l]eban[t qu]am prol[ue]bant vigili cura sollertique / [lab]ore inpens[o i]ntra septimummensem adiecti[s] omnibus perfectisque cunctis / quib[u]s lavacra indigebant Ann[i] us Namptoius fl(a)m(en) p(erpetuus) iurisconsultus [magister] / studi[or]um cur(ator) rei p(ublicae) Thu[burbitan]ae urbis florentissimo sena[tu] / c[u] n[ct]aque eius p[l]e[be / per]feci[t e]xcoluit dedicavit. Beatissimo saeculo dd(ominorum)] nn(ostrorum) C[onstanti Pii Fel]ici[s Maxim]i / [et invictissimi Augusti] et Iuli[ani no]bil[issimi] Cae[saris / pro] co[ns]ulatu Clo[di H]ermogenian]i v(iri) c(larissimi)

p]roc(onsulis) [p(rovinciae) a(fricae) et le]gatione [Crepe]rei / Optatiani v(iri) c(larissimi) leg(ati) Karthag(inis) [t]hermas [aes]tivales po[s]t ann[os solidos] octo / i[n]tra septimum mensem a[d]iectis omnibus perfectisq[ue] cuncti[s / qu]ibus lavacra ind[i]gebant Ann[i]us Namptoivius flam(en) [p(er)] p(etuus) / iuris consultus magister st[udiorum] cur(ator) rei p(ublicae) cum Thub[ur]bi[t]anae / [u]rbis ordine amplissim[o c]unct[a]que eius plebe / [per]fecit excoluit dedicavit. Note Both texts were written on different sides of the same slab. It is interesting to note the time span ‘post annos solidos octo intra septimum mensem’ (‘after 8 whole years within the 7th month’), probably referring to the 8 years that the baths had been left unfinished. The restorations were then completed after the first 7 months of the benefactors’ year in office. The term lavacra seems to have been a part of the thermae mentioned earlier in the text. The first lines of the first part (‘humore superfluo magis flebant quam proluebant’ translated by G. Fagan as ‘with the liquid overflowing they wept rather than washed’) indicate that these were probably the pools (Fagan (1999a) 276, nr. 273a and 273b). EA-18 Ref. CIL 8.20267 Loc. Satafis (Aïn Kebira, Algeria) Cont. Unknown. Date AD 361–363 (converted calendar year) Text Post fla]mmas cinere[squ]e suos nova surgere foenix / [scit nu]nc ut pulc(h)ra r[e]novetur fabrica mole / [tu Quinti]ne facis cui la[u]dem nobile corpus / [ac vir]es peperere suae quibus omnia polles / [regia] Romuleo genitum quem stem[m]ate p[atru]m / [praefectu] m Mauris claro [p]ermisit honore / [---]turis thermis I[---]onos iste resurget / [que]m gaudet sibimet [civitas n]utrisse Satafis / pr(ovinciae) CCCXX[---]II EA-19 Ref. ILAlg I. 2100 Loc. Madaurus (M’Daourouch, Algeria) Cont. Found in two fragments, one in the forum and one north of the theatre. Date AD 361–363 (imperial titles) Text B[eatissimis temporibus --- sa]eculo d(omini) n(ostri) [[Iul[ia]ni]] / perpet[ui Aug(usti) thermae --- in] curia paene ad in/teritum [--- ca]meris omnibus / et soliis e[---] non tantum in/feriorum [---]tis voragini / [--- sup]erior[um //]s qui picturae grati[am] / [---] melioribus ornam[entis] / [---]sunt proconsula[tu ---] / [---] sum(p)tu publico [// Ma]rciani ducena/rii [--splend]id[i] e[t] laudab[i]l[i]s v[i]ri cur(atoris) re[i] p(ublicae) perfectae sunt et cum ordine [---] / eiusdem col[oniae no]st(rae) et populo ded[ic]atae.

237

Late Antique Inscriptions Mentioning Baths EA-20 Ref. Loc. Cont. Date Text

ILAlg I. 2101 Madaurus (M’Daourouch, Algeria) Large block reused in a later wall in the Large Baths. AD 364 (imperial titles and proconsul) Pro tanta securi[tate temporum] / dd(ominorum) nn(ostrorum) Valentiniani [et Valentis perpetuo] rum Au[gg(ustorum)] / [therm]as aestivas olim splen[did(issimae)] coloni[ae --- ]a[---] / [---]tro annis ruinarum labe deformes pa[rietibusque omni]um soli/ orum ita corruptis ut gravibus damnis adficerent [nun]c omni idonitate con/structas et cultu splendido decoratas sed et patinas ampliato aeris pondere / omni idonitate firmissimas proconsulatu Publi Ampeli v(iri) c(larissimi) Octavio Privatia/no v(iro) c(larissimo) legato Numidiae C(a)ec(ilius) Pontilius Paulinus f{f} (lamen) p(er)p(etuus) p(atronus) c(oloniae) curat(o) r rei p(ublicae) pecunia / publica perfecit porticum quo[q]ue ingredientibus ab atrio sed et pronaum / eidem coh(a)erentem commeantibus per viam trabibus tignis [---] ceterisque / [--- Pont]ilius Pauli/[nus ---] ordine.

EA-21 Ref. AE (1911) 217 = AE (1987) 1082 Loc. Mascula (Kenchela, Algeria) Cont. Large slab found in several fragments in a probable bath building. Date AD 364–367 (imperial titles and consuls) Text Aureis ubique temporibus dd(ominorum) nn(ostrorum) Valentiniani et Valen/tis perpetuorum (Au)gg(ustorum) statum desperata recipiunt ami/ ssa renovantur ruinarum deformitatem decor novit/ atis excludit iamdudum igitur thermarum aestival/ ium fabulam factam depellens faciemque restituens / Publi(li)us Ceionius C(a)ecina Albinus v(ir) c(larissimus) consularis / [s(ex)f(ascalis) p(rovinciae) N(umidiae) C(onstantinae)] / ad splendorem tam patriae quam provinciae restituit / perfecit dedicavit[que?] omni nisu [cur]antib[us] / Aemilio Flaviano Fabio Praetexta[to F]lav[i]o / Innocentio Mario Secundino [---io E]xcu[s]antio / fll(aminibus) pp(erpetuis) / b(onis) b(ene). EA-22 Ref. ILAlg II.595 Loc. Cirta (Constantine, Algeria) Cont. Fragment of red marble (sic) found in a school (during construction works?). Date AD 364–367 (imperial titles) Text Felicissimi[s temp(oribus) d(ominorum) n(ostrorum) invic(tissimorum?) Valentiniani] / et Valentis [semper Aug(ustorum) thermas Constan]/tinias v[etustate conlapsas --- an]/nis tot re[tro

EA-23 Ref. Loc. Cont. Date Text

CIL 8.5335 = ILS 5730 = ILAlg 256 Calama (Guelma, Algeria) Inscription on an architrave (findspot unknown). AD 366–368 (proconsul) Beatissimis temporibus dd(ominorum) nn(ostrorum) Valentiniani et Valentis perpetuorum Augg(ustorum) procons[ulatu v(iri) c(larissimi) Iuli Festi Hymeti(?) legatio]ne v(iri) c(larissimi) Fabi Fabiani piscinam quae antea tenuis aqu(a)e pigra fluenta capiebat nunc ve[---] / [--- unda]rum intonantium motibus redundantem Q(uintus) Basilius Flaccianus fl(amen) p(er) p(etuus) augur et cur(ator) [---] / [---] et excepto[rio --- ex]tructo adq(ue) perfecto cum [Bas]ilio Maximo Auf[id]iano [filio suo dedicavit].

EA-24 Ref. ILAlg 2102 Loc. Madaurus (M’Daourouch, Algeria) Cont. Found in several fragments reused in a wall in the Large Baths. Date AD 366–368 (imperial titles and proconsul) Text Pro tanta felicitate tempor[um invictissi]/morum principum dd(ominorum) nn(ostrorum) perp[p(etuorum) Augg(ustorum) Valenti]/niani et Valentis piscinalem istam [---] / et soliarem cellam lacuniis densis ita foed[atas ut ima pavi]/menti monstrarent atque ita retentione[m caloris prohi]/berent compellente religione sanctae P[--- et commodo Roma]norum(?) civium exquisitis diversorum co[lorum marmoribus] / artificibus quoque peregrinis adductis et [adhibitis? splen]/dentes novoque omnino opere tes(s) ellatas pr[oconsulatu Iuli Festi v(iri) c(larissimi)] / Fabio Fabiano v(iro) c(larissimo) et inlustre legato Numidi[ae --- cur(ator)] / rei publicae inter cetera in quibus ia(m)du[dum ---] / cum ordine splendido et universo popul[o restituit et dedicavit?] / felicit[er]. EA-25 Ref. AE (1975) 873 Loc. Abbir Maius (Henchir en Naam, Tunisia) Cont. A lintel broken into three fragments (findspot unknown). Date AD 368–70 (imperial titles and proconsul) Text Salvis ddd(ominis) nnn(ostris) Valentiniano Valente Gratiano perpetuis Auggg(ustis) proconsul[a]tu Petroni Claudi c(larissimi) v(iri) et Mari Victoriani l[e] gati Kart(h)aginis c(larissimi) v(iri) / oceanum a fundamentis coeptum et soliarem ruina conlapsum ad perfectionem cultumque perductos ingressus novos signis adpositis decoravit / Flavianus Leontius alm(a)e Kart(haginis) principalis curator rei p(ublicae) ordinis splendidissimi conlatione cum amore populi inco(h) av[i]t perfecit dedicavit.

238

appendix 1 et virtutibus fovit suble/vabit erexit etiam ob ea quae sibi / specialiter conlata sunt civitas / Sabrathensis exsultans quod po/st ruinam et abnegatum therma/ rum populo exercitium citra ullius / dispendium ornamentis patriae / revocauit ordo populusque / concinentibus omnibus vo/tis statuam patrono pr(a)estan/ tissimo gratanti studio / conlocavit.

Note Y. Thébert ((2003) 486) wrongfully identified the city with modern Henchir al Khandaq (Roman Abbir Germaniciana). The clarissimi viri that were responsible for the restauration came from Carthage. They possibly had land in the vicinity of Abbir Maius. EA-26 Ref. AE (1925) 31 = ILTun 1500 Loc. Thugga (Douga, Tunisia) Cont. Found in three fragments, two near the Capitolium and one in the Licinian Baths. Date AD 367–383 (prosopography) Text Atrium thermar[um ---]inianarum ab antiquis c[oe] ptum excep/toriis in eodem loco su[biectis] quod inperfecto opere corruptum adque / ruderibus foedatum [erat ---]dius Honorati(a)nus fl(amen) p(erpetuus) cur(ator) rei p(ublicae) II [cu]m statua / signoq(ue) Felicissimi Fl(avi) Gr[atian]i CCCRATV opere perfecit itemq[ue dedica]vit. EA-27 Ref. CIL 8.16400 Loc. Musti (Henchir Mestis/El Krib, Tunisia) Cont. Unknown. Date AD 371–373 (imperial titles and proconsuls) Text Ddd(ominis) nnn(ostris) Valentiniano Valenti et Grat[iano ---] / Sextio Rustico v(iro) c(larissimo) procons{s}(ule) p(rovinciae) A[fr]i[cae et ---] / [--- v] (iro) c(larissimo) legato Numidiae balneae quae i[---] / redintegrate sunt devotione totius ordini[s ---] / [---] cur(ator) r(ei) p(ublicae) opus et sollicitudine et sumtibus adi[uvit EA-28 Ref. Loc. Cont. Date Text

EA-29 Ref. Loc. Cont. Date Text

CIL 8.25845 Abitina (Chouhoud el-Batin, Tunisia) Found out of context in several fragments. AD 376–377 (imperial titles) --- Valente Gr]atiano et Valentin[iano ---] / [--Aug]ustis thermis aestivis [---] / [--] orna[t]us constitu[erunt? ---] / [--- Hila]rius Hesperius v(ir) c(larissimus) proco[(n)s(ul) ---] / [--- cu]ratore r(ei) p(ublicae) et Minucio M[—

IRT 103a Sabratha (Sabratha, Libya) Altar later reused in the paving of a 6th c. basilica. AD 378 (prosopography) [Fl(aui) Vivi Benedicti v(iri)] p̣(erfectissimi) /totius integritatis modera/tionis iustitiae provisionis / fidei benignitatis fortitudinis / ac beneficentiae viro Fl(auio) Vivio / Benedicto v(iro) p(erfectissimo) praesidi prov(inciae) Tripol(itanae) / inter cetera beneficia sua quibus / omnem provinciam conpendiis re/mediis

EA-30 Ref. Loc. Cont. Date Text

AE (2004) 1681 Limisa (Ksar Lemsa, Tunisia) Lintel in limestone (findspot unknown). AD 387–388 (imperial titles) [Beatissimo saeculo(?) dominoru]m Augustorumqu[e n]ostrorum / [Valentiniani Theodosi Arcadi et [[[Maximi]]] Constantianas thermas vetustate / [conlapsas restitutas(?) et e]xcultas etiam sumptu publico / [--- vir cla]rissimus consularis p(rovinciae) V(aleriae) Byz(acenae) dedicavit / [curante ---]o flamine perpetuo curatore rei publicae

EA-31 Ref. AE (1949) 28 Loc. Carthage (Tunis, Tunisia) Cont. Large entablature block found in the Antonine Baths. Date AD 388–390 (imperial titles) Text [DDD(ominis)] nnn(ostris Valentiniano [The]odosio et Arcadio / [Iun]iorinus Polemius v(ir) c(larissimus) proconsule provinc/[iae //]E dignissimo decora[---] / [---] almae Karthagi[nis EA-32 Ref. Loc. Cont. Date Text

CIL 8.15204 = CIL 8.1412 Thignica (Aïn Tougga, Tunisia) An architrave block found in 15 fragments. AD 393 (prosopography) aquae]ductos taetra ac deformi caligine mersos et nullo felici aspect[u gaudentes(?) ---] valet in sple[ndidissimo municipio(?) ---] met[---] gemino provisionis [---] beneficio quae usui [privato(?) ero] gabatur lavacris praestitit quae hac viduata on[eribus illis iussit usui(?) f]ieri civibus / [--- proconsulatu ae] mili flori paterni v(iri) c(larissimi) et inlustris et eri fani geminiani v(iri) c(larissimi) leg(ati) c(ai) vib[---]n[--c]andido ffll(aminibus) p(er)p(etuis) dd(uumviris) [---] sum(p)tu public[o restituit(?) ---] et dedicavit. Note In this fragmentary inscription, local authorities rendered water that was used for private use to the lavacra. EA-33 Ref. AE (2010) 1789 Loc. Capsa (Gafsa, Tunisia) Cont. Large limestone block found during construction works near the casbah.

239

Late Antique Inscriptions Mentioning Baths Date AD 393–395 (imperial titles) Text Salvis ddd(ominis) nnn(ostris) / Theodosio Arcadio et Honorio / semper Augustis et trium/fatoribus ob splendorem / [t]hermarum Alexandrian/arum [ EA-34 Ref. Loc. Cont. Date Text

AE (1914) 57 = ILAfr 276 Thuburbo Maius (El Fahs, Tunisia) Base found in the curia on the forum. AD 395–408 (imperial titles) Salviani p(atroni or principalis) a(lmae) K(artaginis) / hortante felicitate temporum dd(ominorum) Au/ gg(ustorum)que nn(ostrorum) Arcadi et Honori inclytorum / principum ubique victorum proconsu/[[[latu ---]]] / [[--- ---]] / [---]inio Salviano Edilicio p(atrono) a(lmae) K(arthaginis) statuam / [vo]to [pa]triae officiorum etiam eius er/[ga e]am meritis indultam r(ei) p(ublicae) felix / T[h]uburbo Maius amantissimo civi / ac sui amanti quod etiam thermarum / hiemalium ex ima fundamentorum ori/gine usque [ad] fastigia culmen erexit idem / quoque [---] IOC [--- te]mporis vel usus vel / [---]N[---]O[---]TV omnium locorum / [---] IVA [ther]marum finis / INCI[---]S[--- su]mptu proprio oper ipen/sisque SV[---]IS EI[--- im]pensam cum amore / [---]MC[---]OSIA TVA / [

EA-35 Ref. CIL 8.25425 Loc. Hippo Diarrhytus (Sidi Abdallah near Bizerte, Tunisia) Cont. Mosaic inscription found in or near a probable bathhouse. Date End of the 4th c. (mosaics and paleography) Text Splendent tecta Bassiani Fundi, cognomine Baiae / invente lucisq[ue] magis candore relucen[t] / disposuit facere [---]DICATIN[---] / Oppositos m[---] / Nomine Sidon[ius ---] / iure sub a[---]. Note The Fundus Bassianus refers to a private estate, meaning these were privately owned baths; however, they could have been accessible to a larger clientele. EA-36 Ref. Loc. Cont. Date

ILAfr 285 Thuburbo Maius (El Fahs, Tunisia) Lintel found in the Winter Baths. End of the 4th or the beginning of the 5th c. (paleography) Text cellam s]oliarem cum solis omni etiam refuso instrumento aeris et plumbi firma refec[it] / [--- lapid]e(?) solidavit piscinam novam nomine cochleam redditis veteribus exceptoriis adiecit / [---] dedicavit G(aius) Optatianus fl(amen) p(er)p(etuus) cur(ator) r(ei) p(ublicae).

EA-37 Ref. Loc. Cont. Date Text

EA-38 Ref. Loc. Cont. Date Text

Foucher (1958) 134. Sidi Bou Ali (Tunisia) Mosaic inscription of the baths of a villa. 4th or 5th c. (mosaic) [S]abinianus / Senurianus / pingit et paimentavt bene lava/re.

CIL 8.948 Tubernuc (Aïn Tebornok, Tunisia) Reused in a modern doorway. 4th or 5th c. (imperial titles) [--- dd(ominis) nn(ostris) Valente et Valen]tiniano Augg(ustis) solium (a)estialium therm[arum] / [--- ---]is ut puro fonte pulc(h)rior redderetur aspe[ctus

EA-39 Ref. CIL 8.20990 Loc. Iol Caesarea (Cherchel, Algeria) Cont. Unknown. Date 4th or 5th c. (imperial titles) Text --- Theod]osio perpe[tuis Augustis ---] / [---]fe? thermar[um --EA-40 Ref. CIL 8.962 = ILAfr 321 Loc. Vina (Henchir Maden, Tunisia) Cont. Unknown. Date Second half of the 4th c. (title of the vicarius) Text Admini[stran]/tibus D[---] / v(iro) c(larissimo) amp(lissimo) pr[oco(n)s(ule) ---] / et Alexand[ro ---] / p(rimi) o(rdinis) c(omite) ag(ente) v(ices) p(raefectorum) p(raetorio) I[--- ]/nus f(lamen) p(er)p(etuus) ex [cur(atore)] / r(ei) p(ublicae) ad [---] / thermarum / posu[it d(ecreto) d(ecurionum) p(ecunia) p(ublica)]. EA-41 Ref. AE (2006) 1767 Loc. Thignica (Aïn Thougga, Tunisia) Cont. Large block broken into four fragments, reused in the wall of the citadel. Date AD 405 (prosopography) Text [---ddd(ominorum nnn(ostrorum Arcadi Honori et Theodosi] principum inclytorum semper Augustorum / [.. Flavio Pio)nio Diotimo v(iro) c(clarissimo) amplissimo proconsule provinciae A[fricae / --- ob] soliditatem et usum saluberrinum fundamentis parie[t ---] / Q(uintus) Vibul[en]ius Fabius Arianus fl(amen) p(er) p(etuus) cur(ator) reip(ublicae) [---] / --Note For Flavius Pionius Diotimus, see PLRE II, 368.

240

appendix 1

EA-42 Ref. ILAlg 2108 Loc. Madaurus (M’Daourouch, Algeria) Cont. Found in several fragments in the forum area and in the Large Baths. Date AD 407–408 (imperial titles and proconsul) Text Excellens glo[ria ddd(ominorum) nnn(ostrorum) Invic]/tissimorum prin[cipum Arcadi H]onor[i et Theodosi] / semper Auggg(ustorum) adm[inistrante P] omp(eio) Proc[ulo] / v(iro) c(larissimo) procons(ule) p(rovinciae) A(fricae) leg[ato Num(idiae)] Q(uinto) Thersio Cri[spino] / Megetio v(ir) c(larissimo) Cl(audius) S[isen]na Germanianu[s] / curator rei p(ublicae) cel[la]m balnearum lon[ga] / serie temporum ruina desolatam usib[usque] / lavacrorum den[e] gatam sumptu prop[rio et] / cam{o}eram cum suspensuris constructam novi[s] / ab splendido ordin[e] decretis titulis ded[icavit]. Transl. ‘The excelling glory of our lords, the invincible and foremost Arcadius, Honorius and Theodosius, everlasting Augusti. To the administrator Pompeius Proclus, vir clarissimus, proconsul of the province of Africa, legate of Numidia and to Quintus Thersius Crispinus Megethius, vir clarissimus, Claudius S[ise] na Germanianus, curator of public affairs (ordered) the room of the baths, which lay in ruins for a long period of time and were inaccessible in their lavacra, and the rooms constructed with new suspensurae (to be built) at his own expenses. He dedicated (these) following issued decrees by the magnificent Ordo.’ EA-43 Ref. Loc. Cont. Date Text

ILAlg 2109 Madaurus (M’Daourouch, Algeria) Found in the Large Baths. AD 407–408 (imperial titles) --- Ar]cad[i? ---] / [--- P]ompeio Pr[oculo ---] / [--Crisp]ino Meg[etio ---] / [--- ther]marum [- --] / [---] perpe[tu? ---

EA-44 Ref. ILAlg 2110 Loc. Madaurus (M’Daourouch, Algeria) Cont. Found near the Large Baths. Date AD 407–408 (prosopography) Text ---e]tiam ? luce [---] / [---] Cl(audii) Sisenae Ger[maniani ---] / [---]na operis suffecti [---] / [---]s pernicie [--EA-45 Ref. Loc. Cont. Date

ILS 5731 Membressa (Mejez el Bab, Tunisia) Square base as part of an altar. AD 412–414 (imperial titles and proconsul)

Text Salvis / dd(ominis) nn(ostris) Honorio et / Theodosio p(er)p(etuis) Augg(ustis) / administrante Q(uinto) Sen/tio Fabricio Iuliano / v(iro) c(larissimo) iterum procons(ule) v(ice) s(acra) i(udicante) / statuas et ornatum / piscinales conlocavit / M(arcus) Aurelius Resti/tutus ex toga(to) cur(atore) / r(ei) p(ublicae) cum splendi/do ordine suo. EA-46 Ref. CIL 8.25864 = ILAfr 492 Loc. Tichilla (Testour, Tunisia) Cont. Large slab of unknown provenance. Date AD 412–414 (imperial titles and proconsul) Text Dom[inis nostris --- --- Honorio et] / Theo[dosio --- ---] / thermas i[--- ]IALRIAETAIAII/TIORESTIRAN[--splen]doremque ex pe[c]unia / un[---]quartae FI[-- ]um pro ingenita sibi libe/ralitate IIISSIMIA[---] quae moenium pulico(rum) / civiat[is ---] VSCONDC(?) [--- sine ulla ---]onii ordinis civium/que conatione ad[ministrante ---] Q(uinto) Sentio Fabricio / Iuliano [---]is / III[---]sq(ue) dicavit. EA-47 Ref. CIL 8.1358; Thébert (2003) 497. Loc. Tichilla (Testour, Tunisia) Cont. Unknown. Date AD 408–423 (imperial titles and prosopography) Text Pollentes sin[e] fine imperio / dd(ominorum) nn(ostrorum) Honori et Theodosi p(er)p(etuorum) smp(er) Aug(ustorum), / administrante Felice nnodio v(iro) c(larissimo) amp(lissimo) / proc(onsule) p(rovinciae) A(fricae) v(ice) s(acra) i(udicante) cum irmo v(iro) c(larissimo) leg(ato) suo t(h)ermas (?) uas ita vetustas cum sra[ge --EA-48 Ref. CIL 8.13473 Loc. Carthage (Tunis, Tunisia) Cont. Marble slab found in 13 fragments on the Damous el-Karita site. Date 5th c. (Vandal rulers) Text Attalus ha[ec f]eci[t rebus sple]ndore paratis / quas fudi ga[udens igni]s vi tecta dicare / Vulcano qui [---] ius haec linqu[ere ---] / natis cum[---]s olim canente s[enecta] / mortali[--- ] vi cogar[ EA-49 Ref. Thébért (2003) 492, nr. 33. Loc. Moknine (Tunisia) Cont. Mosaic inscription in the apodyterium of a small bathhouse, possibly of a villa. Date 5th c. (mosaic) Text Quid pa/bes pal/les fru/er Bai/as quas tu / negabas fieri.

Late Antique Inscriptions Mentioning Baths Note The mosaic depicts a victorious driver of a chariot, waving a tablet bearing the text (Wiedler (1999) 327; Thébert (2003) 492, nr. 33). EA-50 Ref. Loc. Cont. Date Text

Ennabli (1986) 56–57; Thébert (2003) 493. Sidi Ghrib (Tunisia) Mosaic inscription on the raised border of the piscina. 5th c. (mosaic) Plus feci quam potui, minus quam volui. Si placet, commune est. / Si displicet nostrum est. Hic sunt tria vera ca(u)tu(s) sedes ebria. Transl. ‘I did more than I could, but less than I would have liked to. If it pleases, it is communal. If it displeases, its ours. Here are three words: Beware! Drunkseat!’ Note The mosaic inscription was found on the outer rim of the western piscina, turned towards the bathers in the frigidarium (Enabli (1986) 7). EA-51 Ref. Loc. Cont. Date Text

CIL 8.8509 Aïn Temouchent (Algeria) Mosaic inscription of a possible bathhouse. 5th c. (mosaic) Invida sidereo rumpantur pectora visu cedat et in nostris / lingua proterva locis hoc studio superamus avos gratumque / renidet aedibus in nostris summus apex operis feliciter. Note The mosaic depicts the head of Oceanus, surrounded by fish and four Nereids on seahorses with the text placed underneath (Bertherand (1856) 122). For a translation, see Berbrugger (1856) 123.

EA-52 Ref. AL 201 Loc. Carthage (Tunis, Tunisia) Cont. Unknown (included as an epigram in the Anthologia Latina) Date AD 496–523 (reign of Vandal King) Text De thermis Alianarum Hic ubi conspicuis radiant nunc signa metallis Et nitido clarum marmore fulget opus, Arida pulvereo squalebat cespite tellus Litoreique soli uilis harena fuit. Pulcra sed inmenso qui duxit cuhnina caelo, Ostendens pronis currere saxa iugis, Publica rex populis Thrasamundus gaudia vovit, Prospera continuans numine saecla suo. Paruit imperiis mutato lympha sapore Et dulcis fontes proluit unda nouos. Expavit subitas Vulcanus surgere thermas. Et trepida flammas subdidit ipse manu. EA-53 Ref. AL 202 Loc. Carthage (Tunis, Tunisia)

241 Cont. Unknown (included as an epigram in the Anthologia Latina). Date AD 496–523 (reign of Vandal King) Text Nobilis exsultat Baiarum fabrica thermis Et duplicat radios montibus aucta dies. Hoc uno rex fecit opus Thrasamundus in anno, Inclita dans multis munera temporibus. Hic senibus florens virtus renovatur anhelis, Hic fessos artus viva lavacra fovent. Miscentur pariter sociis incendia lymphis Et gelidos imbres proximus ignis habet. Utilis hic flamma est et nullos pascitur artus Optaturque magis per nova vota calor. lo Longior hic aegros morborum cura relinquit Nec lavat in vitreis hic moriturus aquis EA-54 Ref. AL 203 Loc. Carthage (Tunis, Tunisia) Cont. Unknown (included as an epigram in the Anthologia Latina). Date AD 496–523 (reign of Vandal King) Text Regia praeclaras erexit iussio moles, Sensit et imperium calx lapis unda focus. Inclusus Vulcanus aquis algentibus hic est Et pacem liquidis fontibus ignis habet. Cum lymphis gelidis gestat concordia flammam Ac stupet ardentes frigida nympha lacus. Uritur hic semper gaudens neque laeditur hospes Et vegetat medicus pectora fota vapor. Maxima sed quisquis patitur fastidia solis Aut gravibus madido corpore torpet aquis, His Thrasamundiacis properet se tinguere thermis: Protinus effugiet tristis uterque labor. EA-55 Ref. AL 204 Loc. Carthage (Tunis, Tunisia) Cont. Unknown (included as an epigram in the Anthologia Latina). Date AD 496–523 (reign of Vandal King) Text Publica qui celsis educit moenia tectis, Hic pia rex populis Thrasamundus vota dicavit, Per quem cuncta suis consurgunt pulchra ruinis Et nova transcendunt priscas fastigia sedes. Hic quoque post sacram meritis altaribus aedem Egregiasque aulas, quas grato erexit amore, Condidit ingentes proprio sub nomine thermas. Hic bonus inriguis decertat fontibus ignis, Hic etiam ardentis [nullus] timet ora camini, Plurimus hic imber gelidas adcommodat undas, Hic aestus levis est, hic nullum frigora torrent, Hic geminata dies per marmora fulget. EA-56 Ref. AL 205 Loc. Carthage (Tunis, Tunisia) Cont. Unknown (included as an epigram in the Anthologia Latina).

242 Date AD 496–523 (reign of Vandal King) Text Tranquillo nymfae de- Currite fluminis ortV. Hic nova flagranti s- Vccedite lumina FoebO Rupibus excelsis, ubi Nunc fastigia surgunT Aequanturque polo to- Tis praecelsa lauacrA Sedibus. hic magnis ex- Ardent marmora signiS, Ardua sublimes praev- Incunt culmina termaE, Muneraque eximius ta- Nti dat liminis auctoR Vnica continuae prae- Noscens praemia famaE. Non hic flamma nocet. v- Otum dinoscite carmeN, Discite vel quanto vi- Vat sub gurgite lymphA. Vandalicum hic renov- At claro de semine nomeN, Sub cuius titulo meri- Tis stat gratia factiS. Note The first, middle and last letters form the phrase ‘Thrasamundus cun(c)ta innovat vota serenans.’ (‘Thrasamundus renews his vow while he clears everything up’). EA-57 Ref. CIL 8.25362 Loc. Carthage (Tunis, Tunisia) Cont. Marble lintel found in two fragments reused in a synagogue and in a house. Date AD 530–533? (prosopography) Text Cerne salutiferas sp[lendent]i marmore Baias / qui calidos aest[us fran]gere quaeris aquis / hic ubi Vulcano Nep[tunus] certat amore / nec necat unda f[ocum n] ec nocet ignis aquas / gaude operi Gebam[unde tu]o regalis origo / deliciis sospes ute[re cum] populo. Note Gebamundus was probably a nephew of king Gelimer and may be identified as the Vandal commander in Procopius, Bell. 3.18 (see Gauckler (1907) 795). EA-58 Ref. Loc. Cont. Date Text

CIL 8.5352 = ILAlg 276 = EChrAfr-03, 193 = CLE 297 Calama (Guelma, Algeria) Lintel above a door of the baths. AD 539 (prosopography) Una et bis senas turres crescebant in ordine totas / mirabilem operam cito constructa videtur posticius / sub termas balteo concluditur ferro nu[ll]us malorum / poterit erigere man(um) Patrici(i) Solomon(is) insti[tu]tion(em) nemo / expugnare valevit defensio martir(um) tuet[u]r posticius ipse / Clemens et Vincentius martir(es) custod(iunt) introitum ipsu(m). Transl. ‘Twelve and one towers altogether rose up in a row; it seems a work of wonder, constructed so swiftly. The postern gate behind the baths is fastened with iron. No enemy could raise a hand against it. No one could take by storm the work of Patricius Solomon. The protection of martyrs secures this postern gate. The martyrs Clemens and Vincentius guard this entrance.’ (Fowden (1999) 47)

appendix 1 EA-59 Ref. AE (1933) 234–235 Loc. Carthage (Tunis, Tunisia) Cont. Lead lamella found in the ‘fountain of thousand amphorae’ (defixio or curse tablet). Date Unknown (4th c.?; late antique paleography) Text Ἀρθυλαιλαμ Σεμ/εσιλαμ Αεηιουω / Βαχυχ Βακαξιχυχ / Μενεβαιχυχ Ἀ/βρασαξ Βαζαβαχ/υχ Μενεβαιχ/υχ Ἀβρασαξ // Dom/ini Dei tenete / detinete Faler/nas ne quis illo/c accedere pos/s{s}it obligate / perobligate / Falernaru(m) ba/lineu(m) ab hac / die ne quis ho/mo illoc accedat. EA-60 Ref. Beschaouch(1977) 78 (fig.74). Loc. Bulla Regia (Hammam Daradji, Tunisia) Cont. Mosaic inscription on the edge of the piscina of a private villa. Date Unknown (late antique paleography, 5th c.?) Text Venantiorum Baiae EA-61 Ref. CIL 8.20579 Loc. Thamallula Antoniniani (Ras el Oud, Algeria) Cont. Found during construction works in a large building of Roman date, 400 m north-west of the ‘Byzantine fort’. Date Unknown (4th c.? Possible reference to the three emperors Valentinian / Valens / Gratian or Constantine II / Constantius II / Constans?). Text In his praed[iis ---] / fl(aminis?) Auggg(ustorum) p(er) p(etui?) [---] / balneum cy[---] / mre perp[--EA-62 Ref. Loc. Cont. Date Text

CIL 8.24090 Siagu (Ksar es Zit, Tunisia) Found in a possible bathhouse. Unknown (late antique paleography) ---]ici max[--- / --- thermas? d]ilpasas et defo[ormesS --- / ---s]oliumq(ue) fl(?) po[---

EA-63 Ref. Jeddi (1994) 275 Loc. Ouled Hafouz (Tunisia) Cont. Mosaic text found in a heated room of a bathhouse. Date Unknown (late antique mosaic). Text O(N?)EDAS NUM LABAS Note The second letter has been interpreted as a Greek sigma or a tilted N. Subsequently, the text has been reconstructed as ‘O sedas num labas’ or ‘Oned as(i)num labas’. EA-64 Ref. AL 108 Loc. Unknown.

243

Late Antique Inscriptions Mentioning Baths Cont. Unknown. Date Unknown (4th or 5th c.? comparison with similar North African inscriptions). Text De balneis. / Aspice fulgentis tectis et gurgite baias, / dant quibus aut parvum pictor et unda decus. / Namque gerunt pulchras splendentia culmina formas, blandaque perspicuo fonte fluenta cadunt. / Gaudia qui gemino gestit decerpere fructu et vita novit praetereunte frui, / hic lavet; hic corpus reparans menemque relaxans, / lumina picturis, membra fobit aquis. Note This epigram was included in the Anthologia Latina, but the type of text refers to a building inscription (Busch (1999) 234). The attribution to North Africa cannot be verified. EA-65 Ref. AL 109 Loc. Unknown. Cont. Unknown. Date Unknown (5th c.? comparison with similar North African inscriptions) Text De balneis / Fausta novum domini condens Fortuna lavacrum/ Invitat fessos huc properare viae./ Laude operis fundi capiet sua gaudia praesul, / ospes dulciflua dum recreatur aqua./ Condentis monstrat versus primordia nomen / auctoremque facit littera prima legi. / Lustrent pontivagi Cumani litoris antra / indigenae placeant plus mihi deliciae. Note The epigram was included in the Anthologia Latina, but the form and the proper names formed by the first and last letters of each line (‘Filocali’ and ‘Melaniae’, in bold, author’s addition) identify this text as a building inscription (Busch (1999) 231). The North African connection is proposed on the basis of the names, which may be linked to characters mentioned by Augustine (Busch (1999) 232).

Inscriptions from Palestina EP-1 Ref. Avi-Yonah (1963) 325 Loc. Scythopolis (Beit She’an/Baysan , Israel) Cont. Found in a suburban context some 70 m outside the Byzantine wall. Date AD 558–559 (local calendar) Text Θεόδωρος ὁ ποιμὴν / λουτρὰ καιυουργῶν νέμε[ι] / τοῖς τὴν ἄκραν νομοῦσι τῆς / λώβης νόσον / ἐν χρ(όνοις) ἰνδ(ικτιῶνος) ζ’ / ἔτους χκβ’ Transl. ‘Theodore the shepherd / allots, renewing them, the baths, / to those sick with very grievous / disease of leprosy / in the time of the seventh indiction / in the year 622’ (i.e. AD 558–559; transl. Avi-Yonah (1963) 325).

EP-2 Ref. CIIP II.1345 Loc. Caesarea Maritima (Har Qesari, Israel) Cont. Mosaic text found at the entrance of a bathhouse. Date Unknown (late antique mosaic) Text ΕΙΡΗΝΗΗΕΙΣΟ/ΔΟΣΣΟΥΚΑΙΗ/ΕΘΟΔΟΣΣΟΥ (ἐιρήνη ἡ ἐίσο/δος σου και ἡ / ἔξοδός σου) Note Mosaic of an unpublished bathhouse, located in square R 16–17. EP-3 Ref. CIIP II.1346 Loc. Caesarea Maritima (Har Qesari, Israel) Cont. Mosaic text found in the Area 1 Baths. Date Unknown (late antique mosaic) Text Ε Υ Τ Υ Χ Ω Σ Ε Υ Τ Ο Κ Ι Ω Κ Α Ι Ε Υ / Φ Ι Μ [ . ] ΑΤ Ο Ι Σ Φ Ι ΛΟΚΤΙΣΤΑΙΣ (ἐυτυχῶς Ἒυτυκιῳ καί Ἒυ/φιμ[í]ᾳ τοῖς φιλοκτίσταις) EP-4 Ref. Loc. Cont. Date Text

CIIP II.1347 Caesarea Maritima (Har Qesari, Israel) Mosaic text found in the Area 1 Baths. 6th c. (mosaic) ΚΑΛΟΚΕΡΙΑ ΕΠΙΚΗΟΣ ΕΥΤΥΧΙΣ … ΙΣ ΒΙΙΕΑ (Καλοκερία ἐπίκηος Εὐτύχις [---] ις ΒΙΙΕΑ) Note The central emblem of the mosaic depicts a woman holding a fruit basket. She is identified by the inscription as ‘the lady of the good season’, i.e. Summer. The rest of the central panel is filled with geometric shapes (see CIIP II, fig. 1347). Around the central panel, animals, palm trees, a building and a riding figure are depicted.

EP-5 Ref. Loc. Cont. Date Text

CIIP II.2090 (= SEG 56, 1891.1) Crocodilopolis (Tell Tanninim, Israel) Mosaic text found at the entrance of a bathhouse. 4th to 6th c. (late antique mosaic) ΕΙΣΕ[--] ΕΠΑ [--] (ἐίσε[λθε] ἐπ’ ἀ[γαθῷ]) Note According to the excavator, there were also shells with inscriptions (ΟΜΟΦ and ΤΡΙΠΤΙ[ ) used as tokens to enter the baths found on site (Stieglitz (2006) nr. 201).

EP-6 Ref. Loc. Cont. Date Text

CIIP III.2396 Ascalon (Ashkelon, Israel) Painted text found on the outer face of a wall of a pool. 5th or 6th c. (stratigraphy) ΕΙΣΕΛΘΕΑΠΟΛΑΑΥΣΟΝ ΚΑΙ [--] E[--] (ἐίσελθε ἀπόλαυσον καί [--]ε[--])

appendix 2

Late Antique Papyri Mentioning Baths The late antique papyri concerning baths and bathing habits are arranged chronologically, each papyrus having a code starting with P. The ostraca are grouped separately, each entry starting with an O-code. For each papyrus, the following information is presented: Ref. Reference to the papyrological corpus, journal or monograph in which the papyrus has been published. The abbreviations of the corpora can be found at the beginning of this book (p. xxi–xxii). Loc. The location where the papyrus has been found. The ancient findspot is mentioned in italics, followed in brackets by the name of the (closest) city / village in the vicinity.

Date Proposed date of the papyrus with the information on which this date has been based in brackets. The papyri that are included all date between AD 300 and AD 700. Cont. The type of document (legal note, private letter, etc.) and the content of the papyrus, focusing on the passage about baths and bathing habits. Where the papyrus was written in columns, the passage pertaining to baths will be referred to with the abbreviation ‘col.’. For papyri written as continuous texts, the passages will be referred to with ‘l.’ (line). Note Additional information about the papyrus.

Late Antique Papyri Mentioning Baths Code

Ref.

Loc.

Date

P-1

P. Oxy. 56, 3856

Oxyrhynchus (El Bahnasa)

Turn of 3rd–4th c. A business note in which the writer asks the addressees to round up the work on the baths and pass by three villages for fuel (to heat the baths?). It possibly concerns the note of a liturgical official who had delegated tasks concerning public baths.

P-2

P. Oxy. 1, 43 V (= W.Chr. 474 = C.Pap.Jud.III 475)

Oxyrhynchus (El Bahnasa)

Turn of 3rd–4th c. The allocation of troops within the city of Oxyrhynchus. Several streets and public buildings are named, including the ‘hot baths’ in the south-west near the theatre (col. 3, l. 10), the ‘divine baths’ in the south (col. 3, l. 24) and the ‘baths of Caesar’ in the north-east (col. 4, l. 24).

P-3

SB XVI, 12550

Near Alexandria

Turn of 3rd–4th c. Directions to a house, explaining that it is close to the former Baths of Claudianus (where the House of the Tulii now stands).

P-4

P. Oxy. 31, 2599

Oxyrhynchus (El Bahnasa)

Turn of 3rd–4th c. Private letter in which a woman asks her brother to buy and send her several items, including three pairs of bath-sandals (l.31: τρία σόλγεια τὰ εἰς βαλανεῖα).

P-5

P. Ryl. 4, 627

Hermopolis Magna (El Ashmunein)

Early 4th c.

List of daily expenses from the archive of Theophanes on his journey to Antioch. The expenses for the baths are mentioned as balanaria (col. 1, l. 15).

BL VIII 297; IX 230; XI 190; XII 170. For further daily accounts of the Theophanes archive, see also P-13 to P-15.

P-6

P. Oxy. 8, 1146

Oxyrhynchus (El Bahnasa)

Early 4th c.

Payment to Herakleidos, operator of the hypocaust (l.21: Ἡρακλειδ[ίω]νι ὑποκα[υ]στῇ)

BL XI 147; XII 138

P-7

P. Oxy. 36, 2798

Oxyrhynchus (El Bahnasa)

AD 304–305

A receipt issued by two cloakroom attendants (καψάριοι) written to their colleague Ammonios.

P-8

P. Oxy. 8, 1104

Oxyrhynchus (El Bahnasa)

AD 306

A letter from a prytanis to a logistès asking BL XI 147 to pay for the costs of the public baths (50 talents and 450 denarii of silver).

© koninklijke brill nv, leiden, ��20 | doi:��.��63/9789004419421_010

Cont.

Notes

245

Late Antique Papyri Mentioning Baths Late Antique Papyri Mentioning Baths (cont.) Code

Ref.

Loc.

Date

Cont.

Notes

P-9

P. Oxy. 12, 1499

Oxyrhynchus (El Bahnasa)

AD 309

A prytanis ordering a banker to pay three BL XI 150 attendants of the public bath (l. 2–3: παρα-χύταις δημ(οσίου) βαλανείου) their monthly wage (1 talent).

P-10

P. Oxy. 6, 896, col. I (= W. Chr. 48 = Sel. Pap. II 360)

Oxyrhynchus (El Bahnasa)

AD 316

BL IX 181; XI 145 A letter from a painter to the logistès asking for the necessary funds to restore the paintings in the Baths of Trajan Hadrian (l. 7–8: Ἁ̣ δ̣ρι ̣α̣ν̣ῶν θερμῶν), especially for the cold pools, in the sweat room (l. 12: [θ]όλου), in the entrances and exits of the entire colonnade and the four corridors around the sweat-bath. The letter was written by a public writer, as the painter himself was illiterate.

P-11

P. Oxy. 1, 53 (= P. Lond. III 751 descr.)

Oxyrhynchus (El Bahnasa)

AD 316

A report by a certain Aurelius Irenaeus on behalf of the guild of carpenters, to the logistès, mentioning repair works on the public warm baths.

BL XI 142. It probably concerned the works mentioned in P-10.

P-12

P. Oxy. 64, 4441, col. 6 & 9 (= SB III 6003)

Oxyrhynchus (El Bahnasa)

AD 316

Report to the logistès on the topography of Oxyrhynchus. We learn that the warm public baths lay in the same street as the vetch-seller’s shop and the beer-seller’s shop (col. 6, l. 14). The repairs to the baths pertained to the basins and pipes and was carried out by lead-workers (col. 9, l. 25–29).

BL XI 174; belongs to the same scroll as P-10 and P-11.

P-13

P. Ryl. 4, 629 R

Hermopolis Magna (El Ashmunein)

AD 317–323

The archive of Theophanes: monthly expenses for the baths while staying in Antioch (l. 202: 100 drachmai; l. 310–311: 200 drachmai with his friend Antoninus)

BL VIII 297; IX 230; XI 190f.; XII 170; belongs to the same archive as P-14 and P-15.

P-14

P. Ryl. 4, 630/637 (=C. Pap. Jud. III, 457c)

Hermopolis Magna (El Ashmunein)

AD 317–323

BL XII 170; belongs The archive of Theophanes: monthly expenses for the baths while staying in an to the same archive as unknown town (l. 40: 100 drachmai), in P-13 & P-15. Kata Hydata (l. 242: 100 drachmai) and Sidon (l. 357: 300 drachmai)

P-15

P. Ryl. 4, 639

Hermopolis Magna (El Ashmunein)

AD 317–323

The archive of Theophanes: monthly expenses for the baths while staying in unknown towns (l. 53: 200 drachmai; l. 112: 300 drachmai)

BL XI 191; XII 170f; belongs to the same archive as P-13 & P-14.

P-16

P. Oxy. 12, 1430

Heraclides (unknown location near Oxyrhynchus)

AD 324

A letter from the tesserarius to the stratègos of the Oxyrhynchite nome informing him that he has been partially reimbursed by the banker of public money for his delivery of fuel to the public baths. It concerns charcoal and flax.

BL VIII 245; IX 185; XI 150

P-17

P. Oxy. 14, 1741, col. 2

Oxyrhynchus (El Bahnasa)

AD 300–325

List of clothes, including bath-related objects (l. 28: βαλανάρια ἀπο̣φό�̣[ρητα)

BL X 144; XI 152

P-18

SB 24, 16000 (= SB 8 9902)

Panopolis (Akhmim)

AD 300–325

BL VIII 276; IX 205; List of occupations including two bath-attendants in the Baths of Claudius XI 176 and the Baths of Hermeios (l. 109 and 148: βαλανεύς)

246

appendix 2

Late Antique Papyri Mentioning Baths (cont.) Code

Ref.

Loc.

Date

Cont.

Notes

P-19

P. Coll. Youtie 2, 81 (= P. Oxy. 45, 3265)

Oxyrhynchus (El Bahnasa)

AD 326

A declaration of the glass workers’ guild BL XII 52 of Oxyrhynchus to the logistès, listing the price for the work of the warm rooms of the public baths (l. 9–10) and of the gymnasium (l. 14).

P-20

PSI 7, 804

Unknown

AD 337

An appeal for an immediate substitution of the prytanis, mentioning the fuel supply for the hypocaust of the baths (l. 4–5: εἰ]ς ὑπόκαυσιν τοῦ δημο[σίο]υ̣ β̣α̣λ̣[ανείο]υ̣)

P-21

P.Oxy. 6 892 R (= W.Chr. 49)

Oxyrhynchus (El Bahnasa)

AD 338

A letter from the logistès to a member of the council informing the latter that he is responsible for supplying wood for the construction of a public bathhouse and a gate.

P-22

P. Würzb. 15 R

Unknown

AD 341

A statement under oath by the ‘epimeletes BL VIII 513 of gold’ about receiving fines, including two ounces of gold of the baths (l. 10: βαλανίου χρυσίου οὐγ[κίας δύο)

P-23

P. Genova 1, 22

Oxyrhynchus (El Bahnasa)

AD 345

A contract for the rent of a topos in the neighbourhood of the hot baths (l. 10)

P-24

Stud. Pal. 20, 230

Unknown

AD 351

Payment of the building material for a bathhouse (l. 7: τ]οῦ βαλανί(ε)ου τοῦ γυμνασίου (δηναρίων μυριάδες) Βυ)

P-25

PSI 3, 217

Oxyrhynchus (El Bahnasa)

AD 357–362

Payment of a cloak attendant (l. 2: καψαρίῳ)

P-26

P. Oxy. 17, 2110

Oxyrhynchus (El Bahnasa)

AD 370

Reports of the debates held in the local senate mentioning an ex-gymnasiarch and senator as prytanis.

P-27

P. Lond. 3, 755 V (= P. Bagnall 43)

Unknown (Fayoum) AD 326–376

Inventory of columns existing in different buildings, with measurements such as height and thickness. The information on the columns in the baths (l.16) has been lost.

P-28

P. Osl. 1, 12

Unkown

AD 301–400

Magic bulla which had to be thrown in the furnace of a bathhouse in order to make the spell come true (l. 334–400).

P-29

P. Osl. 1,

Unknown

AD 301–400

Magic bulla which had to be thrown into the furnace of a bathhouse in order to make the spell come true (l. 75).

P-30

P. Oxy 6, 903

Oxyrhynchus (El Bahnasa)

AD 301–400

An accusation of a woman against her husband, including an episode in which the man took the bathing objects ‘out of security’ reasons (l. 29).

P-31

P. Oxy. 59, 3998

Oxyrhynchus (El Bahnasa)

AD 301–400

Private letter from a father to his daughter urging her to get hold of certain goods that have already been paid for. A list, added at the end of the scroll, mentions a washbasin and the bucket for the baths.

P-32

PSI 9, 1082

Oxyrhynchus (El Bahnasa)

AD 301–400

Private letter from a woman to her husband about sending him some clothes wrapped in his bath towel (l. 17: εἰς τὸ βαλανάριόν σου).

BL X 272; XII 277f.

BL IX 193; X 146; BL XI 159

BL XI 145

247

Late Antique Papyri Mentioning Baths Late Antique Papyri Mentioning Baths (cont.) Code

Ref.

Loc.

Date

Cont.

Notes

P-33

SB 18, 13307

Unknown

AD 301–400

Document listing prices for the protection P. Mich. inv. 4650 of buildings, including a bathhouse (l. 9: [εἰ]ς φυλακίαν βαλαν(ε)ίου).

P-34

SB 22, 15624 (=P. Lond. II 425)

Unknown

AD 301–400

Possibly a (semi-) literary text. The fragmentary condition only permits reading that a certain person was at the baths (ἐν το βαλα|νείῳ)

P-35

Stud. Pal.20, 109 R

Unknown

AD 301–400

A list of names, places and numbers mentioning βαλαν(ε)ῖον (l. 14).

BL IX 345/346

P-36

SB 12, 11075

Oxyrhynchus (El Bahnasa)

AD 401–450

A contract of marriage mentioning the dowry received by the husband. It included seven light bath towels without any ornaments (l.10: βαλανάρια ἰσχνᾶ λιτὰ ἑπτὰ).

BL VIII 366

P-37

P. Oxy. 34, 2718

Oxyrhynchus (El Bahnasa)

AD 458

A receipt for the acquisition of nails and BL VIII 261; IX 197 other materials for the construction of a public bath (l. 18: τοῦ δημοσίου λ(o)υτροῦ)

P-38

P. Oxy. 72, 4917

Oxyrhynchus (El Bahnasa)

AD 473

A financial transaction between a rich landowner and a bath attendant (l. 4: περιχύτης)

P-39

P. Flor. 3, 384

Hermopolis Magna (El Ashmunein)

AD 489

Lease contract of a bathhouse with an annex matronikon (L. 7: σὺν τῷ ματρωνικίῳ) for the period of 10 years and three months.

P-40

Stud. Pal. 20, 132

Unknown

AD 401–500

Document about the provision of chaff for the furnace of the baths (l. 4–5: εἰς ὑπόκαυσιν τοῦ δημ(οσίου) βαλαν(ε)ίου)

P-41

P. Oxy. 7, 1026

Oxyrhynchus (El Bahnasa)

AD 401–500

Document about the sale of household goods to pay off debts. Includes bath-towels (l. 14: βαλανάριον)

P-42

SB 26, 16762

Unknown

4th or 5th c.

Excerpt of a private letter mentioning a pair of bath slippers (l. 13).

P-43

P. Cair. Masp 2, 67143 R

Aphroditopolis (Aftih)

AD 538–547

An account of theft in the archive of Dioscoros, mentioning a bath-operator (l. 16: βαλανεὺς)

P-44

SB 26, 16817 (= PSI 9, 1061)

Unknown

AD 500–550

For baths and salt, see Fragment of a petition by the halourgoi (salt sellers). It mentions salt for the baths also Cod. Th XIV, 5, 1. (l. 17–18: εἰ�ς̣ τὴ�̣ ν χρείαν τοῦ βαλαν(ε)ίου).

P-45

P. Oxy. 36, 2780

Oxyrhynchus (El Bahnasa)

AD 553

Declaration of a certain Timothy, water-supplier of the public bath (l. 14–15: ὑδροπάροχο̣ς)̣ for his salary (two solidi).

P-46

P. Oxy. 16, 2015

Oxyrhynchus (El Bahnasa)

AD 555–556

BL VIII 253; XI 157 A receipt to the monks of St. Andrew’s monastery for the acquisition of a rope for the baths, probably in connection with the water supply by a well.

P-47

P. Oxy. 16, 2040

Oxyrhynchus (El Bahnasa)

AD 566–567

A list of contributions for the fuel of a public bath.

P-48

P. Cairo Masp. 1, 67009 V

Antinopolis (es-Sheikh Ebada)

AD 567–570

BL IX 41; XI 52 A draft of a second petition by the inhabitants of Antinoupolis for the dux of the Thebaid. They complain about the abuse of power by a strategòs, setting the entrance price of the public baths at 20 nummia (l. 22–23).

BL VIII 132; X 75; XI 81f.; XII 73

BL XII 137

BL VIII 73; IX 43

BL VIII 262; IX 197

BL IX 192; XII 144

248

appendix 2

Late Antique Papyri Mentioning Baths (cont.) Code

Ref.

Loc.

Date

Cont.

Notes

P-49

P. Oxy. 6, 915

Oxyrhynchus (El Bahnasa)

AD 572

A receipt for a lead-worker covering the expenses of lead and tin used in the restoration of the water pipes of a public bathhouse. The baths are said to be located in the suburbs.

P-50

SB 6, 9368

Oxyrhynchus (El Bahnasa)

AD 577–578 or AD The receipt of a bath-operator for lead 592–593 used in the repairs of the public baths.

P-51

P. Wisc. 2, 66

Oxyrhynchus (El Bahnasa)

AD 584

A receipt of an expensive amount of glass panes for the baths of a mansion outside the gate (l. 2: εἰς χρείαν τοῦ λουτροῦ τοῦ προαστίου).

P-52

SB 16, 12868 (= P. Lond. 3 p. LXXII no. 1304 a descr.)

Hermopolis Magna (El Ashmunein)

AD 592

Employment contract of a bath-operator BL IX 294 (l. 4).

P-53

BGU XIX, 2827 R

Hermopolis Magna (El Ashmunein)

AD 595

Employment contract of a bath attendant who started to work for an independent bath operator (l. 10), for the period of one year. The wage is unfortunately illegible.

P-54

P. Oxy. 16, 2006

Oxyrhynchus (El Bahnasa)

AD 501–600

A receipt for the wage of a bath attendant BL XI 157 (l. 2: βαλ[α]νεὺς τοῦ λουτροῦ). He was employed by an illustris.

P-55

P. Jena 2, 39

Unknown

AD 501–600

A document about the fuel supply, perhaps chaff, for the baths in the praetorium (l. 1: τοῦ λουτροῦ τοῦ πραιτω̣ ρί̣ ο̣�̣ υ̣)

P-56

SB 20, 14581

Near Hermopolis Magna (El Ashmunein)

AD 501–600

List of occupations, including the architect of a bathhouse (l. 4: οἰκοδ(ομῇ) βαλαν(ε)ίο).

P-57

Stud. Pal. 8, 1257

Near Hermopolis Magna (El Ashmunein)

AD 501–600

Receipt (?) for material for the baths.

P-58

CPR 19, 54

Unknown

4th–6th c.

Draft for the purchase of building material for a bathhouse, including wood, lead and tin. The translation for line 2 (εἰς τὰ πατήματα τοῦ δημο-σίου βαλαν(ε) ίο[υ]) is contested. It probably concerned wooden decorative doorsteps or stairs (Mayer (2004) 103).

P-59

P. Köln 4, 197

Unknown

5th or 6th c.

A list of assignments to repair a public bath (l. 1: τοῦ δημοσ(ίου) λουτροῦ).

P-60

P. Münch. 3, 143

Unknown

5th or 6th c.

List with expenses, including those for the baths (l. 3).

P-61

SB 14, 11358

Unknown

5th or 6th c.

Receipt for a service (?) provided in the public baths.

P-62

Stud. Pal. 8, 947

Arsinoe (Medinet el-Fayum)

5th or 6th c.

An order for the delivery of nails to be used for the doors of the stables of the baths (εἰς χρεί(αν) θύρ(ας) στάβλ\ου/ τ\οῦ/ δημο(σίου) λο\υ/ τρ\οῦ/).

P-63

SB 1, 4742

Arsinoe (Medinet el-Fayum)

Early 7th c.

A person named Balaneos (Βαλανέος) is mentioned in a letter.

P-64

P. Oxy. 6, 943

Oxyrhynchus (El Bahnasa)

AD 612–618

A decision about the rent of a bathhouse BL XI 146 (l. 4: τὸν φόρον τοῦ λο(υ)τροῦ).

BL X 197; XI 206

BL VIII 448; IX 340

249

Late Antique Papyri Mentioning Baths Late Antique Papyri Mentioning Baths (cont.) Code

Ref.

Loc.

Date

Cont.

Notes

P-65

P. Oxy. 16, 1921

Oxyrhynchus (El Bahnasa)

AD 621 (?)

A list of expenses, including for 50 glass panes for villa baths in Takona and another 80 for villa baths in Ophis (both referred to as λουτρόν). The texts mentions Persian officials.

BL VIII 251; IX 191; X 145

P-66

P. Ross. Georg. 3, 47

Arsinoe AD 628 (or 643 or (Medinet el-Fayum) 658)

BL VIII 291; belongs to A receipt for the remuneration paid to the σύμμαχος (clerk?) of the public baths the Justus archive (l. 1: τοῦ δημοσίου λουτροῦ) for his services. The text was written by a public writer.

P-67

P. Lond. 1, 113, 6 b

Arsinoe AD 633 (Medinet el-Fayum)

A lease contract of two rooms to an attendant of the public baths (l. 7–8: περιχύτης δημο(σίου) βαλανί(ε)ου).

P-68

Stud. Pal. 8, 980

Around Memphis (Mit Rahina)

Around AD 650

Payment of a bath-operator (δημ(οσίου) λουτρο(ῦ)).

BL VIII 449; IX 340

P-69

P. Apoll. 85

Unknown

AD 650–700

List of daily expenses, including those for the baths (l. 4: (ὑπὲρ) βαλανη̣ ).

BL VIII 11; XI 8

P-70

P. Iand. 4, 64

Unknown

AD 601–700

The payment of a bath-operator (βαλ[ανεὺς]).

P-71

P. Oxy. 16, 1925 R

Oxyrhynchus (El Bahnasa)

AD 601–700

A list of furniture and objects handed over to a summachos. A lengthy list of objects comes from the baths (ἐν τῷ λουτρῷ), including waterspouts in the shape of a lion’s head.

P-72

P. Ross. Georg. 5, 40

Unknown

6th or 7th c.

Note mentioning the expense of the water for the baths (ὑδροπαρόχ[ῳ] πλαγίων βαλανείων).

P-73

SB 1, 4905 (= SB 1, 5296 = Stud. Pal. 8, 1110)

Near Arsinoe 4th–7th c. (Medinet el-Fayum)

Payment order in wine for a certain Gerontios, attendant of the baths (l. 2–3: περιχύτῃ] τοῦ βαλαν(είου)).

BL VIII 317

P-74

SB 16, 12254

Unknown

7th or 8th c.

List of expenses, including those for the baths (l. 6: (ὑπὲρ) βαλαν(ευτ)ικοῦ).

BL X 212

O-1

O. Mich. 1, 102

Karanis (Kafr Al Massalat)

Early 4th c.

An account of expenses mentioning the wage of a bath-guard (?) (l. 5: βα[λανειοφ(?)]ύλαξιν)

O-2

O. Mich. 1, 219

Karanis (Kafr Al Massalat)

Early 4th c.

Taxation: one basket of chaff for the heating of the gymnasium.

O-3

O. Mich. 1, 221

Karanis (Kafr Al Massalat)

Early 4th c.

Taxation: four baskets for the baths.

O-4

O. Douch 3, 251 (= SB 16, 12358)

Kysis (Qasr Dush)

AD 350–400

Receipt for four matia of wheat for the rent of the baths (ὑ(πὲρ) μισθοῦ βαλανίου).

O-5

O. Douch 4, 367

Kysis (Qasr Dush)

AD 375–425

Payment order by the responsible of the baths (l. 3: ἐπιμ(ελητὴς) βαλ(ανείου)) to hand in two artabè of wheat to two soldiers (for unknown reasons).

O-6

O. Douch 5, 569

Kysis (Qasr Dush)

AD 375–425

Payment order of one artabè of wheat for the license of the baths (l. 5–6: ὑπὲρ φόρου βαλαν(ε)ίου).

O-7

O. Douch 5, 638

Kysis (Qasr Dush)

AD 375–425

Payment order of more than two artabè for the expenses of the public baths (l. 3: εἰς λόγ(ον) βαλανίου), to be paid to the epimeletès.

O-8

SB 12, 10990, 103

Abu Mena

5th or 6th c.

Payment for the bathhouse.

O-9

SB 1, 1970

Oxyrhynchus (El Bahnasa)

AD 501–600

Receipt for the delivery of brush wood for the heating of the baths (τοῦ δημοσ(ίου) λουτρ(οῦ)).

BL VIII 252; IX 192; X 145

BL XII 297

The responsible for the baths mentioned in this text is the same person as in O-6.

appendix 3

Lists of Late Antique Baths, Possible Baths and Continued Use of Baths A The Italian Peninsula, Sicily and Sardinia New Construction Site, Name of bathhouse

Construction date

Abandonment

Aquileia, Large Baths of Constantine II (C1)

First half of the 4th c.

5th c.

Brixia, Castello Baths (C2)

5th c.

6th c.

Brundisium, Baths of S. Pietro degli Schiavoni (C3)

4th c.

Unknown

Galeata, baths of the ‘Palace of Theodoric’ (VB)

Second half of the 5th c.

Middle of the 6th c.

Luna, Late Baths (C4)

4th c.

5th c.

Mediolanum, Baths of Herculeus (C5)

Late 3rd–early 4th c.

5th c.?

Mediolanum, Baths in via S. Maria Vale (C6)

4th c. to 6th c.

Unknown

Metapontum, baths

4th to 7th c.

Unknown

Ostia, Small Baths west of the Palazzo Imperiale (C7)

First half of the 4th c.

Unknown

Ostia, Baths in the Horreum (C8)

4th c.

Unknown

Ostia, Baths of Musiciolus (C9)

Late 3rd–early 4th c.

5th c.?

Ostia, Baths along the Via Marciana (C10)

Early 4th c.

Unknown

Ostia, Baths along the Via Severiana (C11)

4th c.

Unknown

Ostia, Small Baths along the Via Severiana (edifico b)

4th or 5th c.

Unknown

Ostia, Baths of Perseus (C12)

4th c.?

Unknown

Ostia, Baths of the Domus dei Dioscuri

Early 5th c.

Unknown

Ostia, Byzantine Baths (C13)

Late 4th–early 5th c.

Late 5th c.?

Ostia, Baths on the Via della Foce (C14)

Middle of the 5th c.

6th c.

Palazzolo, Villa Baths (VB)

Late 5th c.

9th c.?

Ostia, Baths on the Via dei Sepolcri

4th to 7th c.

Unknown

Piazza Armerina, Baths of the Villa del Casale (VB)

Early 4th c.

6th c.

Piazza Armerina, Small Baths south of the Villa del Casale (C15)

4th c.

Unknown

Portus, Baths of the Lantern (C16)

Late 3rd–early 4th c.

Late 5th c.

Portus, Small Baths northeast of Area F of the warehouses (C17)

4th to 7th c.

Unknown

Ravenna, Baths in the Via d’Azeglio (C18)

Late 3rd–early 4th c.

5th c.

Ravenna, Bagni del Clero (C19)

5th c.

9th c.?

Ravenna, Baths in the Via S. Alberto (C20)

6th c.

Unknown

Ravenna, Baths of Palazzolo (VB)

Late 5th c.

6th c.?

Rome, Baths of Diocletian (C21)

Early 4th c.

6th c.?

Rome, Baths of Constantine (C22)

Early 4th c.

6th c.?

Rome, Baths near the Scala Caci on the Palatine (C23)

Late 3rd–early 4th c.

5th c.

Rome, Baths of Elagabalus (C24)

Early 5th c.

Late 6th / early 7th c.

Rome, Baths in Via Ariosto (C25)

Early 4th c.

Unknown

Rome, Small Baths on the Palatine

Late 5th c.

9th c.?

Rome, Small Baths on the Forum Romanum (C26)

4th c. or later

Unknown

Rome, Baths near Porta San Lorenzo

4th to 7th c.

Unknown

San Giusto, Church Baths (C27)

5th c.

Late 7th c.

Turris Libisonis, Central Baths (C28)

Late 3rd–early 4th c.

Unknown

© koninklijke brill nv, leiden, ��20 | doi:��.��63/9789004419421_011

251

Lists of Late Antique Baths, Possible Baths and Continued Use of Baths New Construction (cont.) Site, Name of bathhouse

Construction date

Abandonment

Turris Libisonis, ‘Terme Pallottino’ (C29)

Late 3rd–early 4th c.

5th c.

Valentia, Mansio Baths (C30)

Late 3rd–early 4th c.

Early 5th c.

Volaterrae, Vallebuona Baths (C31)

Late 3rd–early 4th c.

Late 4th c.?

Volaterrae, Guarnacciane Baths of San Felice (C32)

Late 3rd–early 4th c.

Unknown

Site, Name of bathhouse

Construction date

Abandonment

Acireale (thermal?)

2nd or 3rd c.?

6th c. or later?

Agrigentum, baths in insula IV

Unknown

5th c.

Albintimilium

Unknown

5th c.?

Augusta Praetoria

Late 1st c.

4th c.?

Caesena

3rd c.

5th c.?

Canusium, Ferrara Baths

Late 2nd / early 3rd c.

5th or 6th c.

Canusium, Lomuscio Baths

2nd c.

4th c.

Carsulae

Late 1st c. BC

Middle of the 4th c.

Catania, ‘Terme della Rotonda’

3rd c.?

6th c.?

Comum, Baths in the Viale Lecco

Late 3rd c.

4th c.?

Cures Sabini

2nd c.?

4th c.?

Curinga

1st c.?

Late 4th c.

Forum Traiani, thermal baths

1st or 2nd c.

6th or 7th c.

Grumentum

Late 2nd / early 3rd c.

Late 4th c.?

Herdonia

2nd c.

5th c.

Iulium Carnicum

1st or 2nd c.

5th c.

Mesange, Malvindi-Campofreddo Baths

1st c.?

4th c.

Misterbianco

3rd c.?

4th c.?

Muralto

3rd c.

4th c.

Neapolis, Baths of the Stazione Toledo

Late 1st / early 2nd c.

5th c.

Nora, Central Baths

2nd c.

7th c.?

Opitergium

1st c.?

5th c.?

Ostia, Baths of the Six Columns

2nd c.

4th c. or later

Ostia, Baths of the Christian Basilica

2nd c.

Late 4th / early 5th c.

Ostia, Baths of Neptune

2nd c.

4th c. or later

Ostia, Baths of Porta Marina

2nd c.

Late 5th / early 6th c.

Ostia, Baths of Mithras

2nd c.

4th c.

Ostia, Maritime Baths

2nd c.

4th c.

Ostia, Forum Baths

2nd c.

Late 5th / early 6th c.

Ostia, Baths of the Lighthouse

2nd c.

5th c.

Ostia (Isola Sacra), Baths of Matidia

2nd c.

6th c.?

Ostia, Baths of the Philosopher

3rd c.

4th c.?

Pollena Trocchia

1st c.?

5th c.

Rome, Baths of Agrippa

1st c. BC

Late 5th c. or later

Rome, Baths of Nero

1st c.

Late 5th c. or later

Rome, Baths of Titus

1st c.

5th or 6th c.

Continued Use of Existing Baths during Late Antiquity

252

appendix 3

Continued Use of Existing Baths during Late Antiquity (cont.) Site, Name of bathhouse

Construction date

Abandonment

Rome, Baths of Trajan

Early 2nd c.

Late 5th c.

Rome, Lateran Baths

2nd c.

4th c.?

Rome, Baths under the Piazza dei Cinquecento

2nd c.

Late 5th / early 6th c.

Rome, Baths of Helena

3rd c.

4th c. or later

Rome, Baths of Caracalla

Early 3rd c.

6th c.?

Rome, Baths of Decius

Mid 3rd c.

5th c.?

San Gaetano di Vada

Late 1st / early 2nd c.

4th or 5th c.

Sentinum, Terme di Santa Lucia

1st c.

4th c.

Suasa, Baths of the Domus dei Coiedii

2nd c.

4th c.?

Sybaris, Baths of the Parco del Cavalo

1st c.

5th c.

Syracuse, Baths of Dafne

6th c.

Teanum Sidicinum

1st c.?

5th c.

Tridentum, baths under Palazzo Crivelli

Unknown

5th c.

Tyndaris

Unknown

Late 3rd / early 4th c.

Venusia

2nd c.

Late 4th c.

Vicus Augustanus, Public baths

Early 3rd c.

4th c. or later

Site, Name of bathhouse

Construction date

Abandonment

Ariminum, Palazzo Gioia

4th c.

Unknown

Brixia, Baths in the Via Gasparro da Salò

5th c.

Unknown

Mediolanum, Via Brisa

4th c.

Unknown

Ravenna, basins in the Via Rasponi

6th c.

Unknown

Ravenna, so-called Palace Baths of Theoderic

2nd c.?

4th c.?

Rome, Temple of Minerva Medica

Early 4th c.

6th c.?

Rome, Baths near Santa Croce in Gerusalemme

4th c.

Unknown

Spoletium, Palazzo Mauri

Late 6th c.

Unknown

Ticinum, San Tommaso Baths

Late 3rd / early 4th c.

Unknown

Verona, Piazza Vescovado

Late 4th c.

6th c. or later?

Construction date

Abandonment

Althiburos, Baths of the House of Asclepeia (C34)

Late 3rd / early 4th c.

End of 4th c.

Baten Zammour

Early 4th c.

Unknown

Belalis Maior, Baths of Theseus and the Minotaur (C35)

Late 3rd / early 4th c.

Unknown

Bulla Regia, Baths Northwest of the Theatre (C36)

4th c.

5th c.

Bulla Regia, Small Northeast Baths (C37)

4th c.?

Unknown

Cincari

Early 4th c.

Unknown

Possible Late Antique Baths

B Roman North Africa New Construction Site, Name of bathhouse Africa Proconsularis

253

Lists of Late Antique Baths, Possible Baths and Continued Use of Baths New Construction (cont.) Site, Name of bathhouse

Construction date

Abandonment

Carthage, Balnea Privata (C43)

Late 4th / early 5th c.

Unknown

Carthage, Odeon Hill Baths (C44)

Mid 4th c.

Early 6th c.

Carthage, Baths of the Okba Ibn Nefaa Street (C45)

4th or 5th c.

Unknown

Carthage, Byzantine Baths (C46)

Early 6th c.

7th c.?

Carthage, Baths in the Baal Hammon Street

6th c.

7th c.

Carthage, Douar Chott Baths (C47)

4th to 7th c.

Unknown

Kerkouane (C48)

Late 3rd / early 4th c.

Unknown

Kef Glia

4th to 7th c.

Unknown

Ouled Hafouz

4th to 7th c.

Unknown

*Sidi Ghrib, Baths of the Marine Thiasos (VB)

4th c.

6th c.

Sufetula, Baths of Basilica 2 (C55)

Late 5th / early 6th c.

Unknown

Sufetula, Baths nr. 26 (C56)

Late 5th / early 6th c.

Unknown

Sufetula, Baths near Basilica 5

Late 5th / early 6th c.

Unknown

Thenae, Baths of the Months (C60)

Late 3rd / early 4th c.

Unknown

Thenae, Small Southeast Baths (C61)

Late 3rd / early 4th c.

6th c.

Theveste, Baths (C62)

Early 4th c.

Unknown

*Theveste, Church Baths (EB)

4th / 5th c.

Unknown

*Tipasa, Baths of the Basilica (EB)

Late 4th c.

Unknown

Sila, Baths of Skylla

4th or 5th c.

Unknown

Sitifis, Baths of the Triumph of Venus (C53)

Late 4th / early 5th c.

Unknown

Sitifis, Baths of the 5th Century (C52)

Early 5th c.

Late 6th / early 7th c.

Cuicul, East Baths (C38)

4th c.

Unknown

Cuicul, Baths of the House of Europe (C39)

4th c.

Unknown

Cuicul, Baths of the House of the Donkey (C40)

4th c.

Unknown

Cuicul, Baths of the House of Castorius (external)(C41)

4th c. to 7th c.

Unknown

*Cuicul, Baths of the Baptistery (EB)

Late 4th / early 5th c.

Unknown

Lambaesis, Baths of the Hunters (C49)

Late 3rd / early 4th c.

Unknown

Lambaesis, Small Baths South of the Asclepius Temple (C50)

4th c.?

Unknown

Oued Athmenia, Baths of Pompeianus (C52)

Late 4th / early 5th c.

Early 6th c.?

Thamugadi, Small North Baths (C57)

4th or 5th c.

Unknown

Thamugadi, Small Northeast Baths (C58)

5th or 6th c.

6th c.?

Thamugadi, Baths North of the Capitolium (C59)

Late 3rd / early 4th c.

Unknown

*Thamugadi, Baths of the Donatist Cathedral (EB)

Late 4th / early 5th c.

6th c.?

*Thamugadi, Baths of the Byzantine Fortress (FB)

6th c.

Unknown

Ain En-Ngila (C33)

4th c.

Unknown

Tubactis Municipium (C63)

4th c.

Unknown

Mellaha (C51)

Second half 4th c.

Unknown

Byzacena

Mauretania Caesariensis

Mauretania Sitifensis

Numidia

Tripolitania

254

appendix 3

Continued Use of Existing Baths during Late Antiquity Site, Name of bathhouse

Construction date

Abandonment

Belalis Maior, Baths of the Forum

2nd c.

5th c.

Bulla Regia, Baths of Iulia Memmia

2nd c.?

5th c. or later

Bulla Regia, Northeast Baths

3rd or 4th c.?

4th c. or later

Carthage, Antonine Baths

2nd c.

4th c.

Mactaris, Large East Baths

2nd c.

5th c.

Madauros, Large Baths

3rd c.?

5th or 6th c.

Numluli

3rd or 4th c.?

5th or 6th c.

Pupput, Baths of the Crater

2nd or 3rd c.

4th c. or later?

Thuburbo Maius, Summer Baths

2nd or 3rd c.

Mid 5th c.

Thuburbo Maius, Winter Baths

2nd or 3rd c.

5th or 6th c.

Thuburbo Maius, Baths of the Labyrinth

2nd or 3rd c.

5th c.

Thuburbo Maius, Baths of the Stars

2nd or 3rd c.

5th c.

Thuburbo Maius, Baths of the Capitolium

2nd or 3rd c.

4th c. or later

Thugga, Aïn Doura Baths

2nd or 3rd c.

4th c.

2nd or 3rd c.

4th c.?

Iol Caesarea, Large West Baths

2nd or 3rd c.

4th or 5th c.?

Iol Caesarea, Baths of the Muses

Unknown

4th c. or later

Rusguniae, Southeast Baths

2nd c.?

4th c. or later

Cuicul, Large South Baths

2nd c.

4th c.

Lambaesis, Asclepieium Baths

2nd c.?

4th c.

Thamugadi, Large South Baths

2nd c.?

5th or 6th c.

Thamugadi, Small South Baths

Unknown

6th c.?

Sabratha, Seaward Baths

1st c.

6th c. or later

Sabratha, Baths of the Theatre

2nd c.

Late 6th or early 7th c.

Sabratha, Baths of Oceanus

2nd c.

Late 4th c.

Sabratha, Baths of Region VII

2nd c.

Middle of the 4th c.

Leptis Magna, Schola Baths

3rd c. or later?

4th c. or later?

Leptis Magna, Hunting Baths

2nd c.

4th c.

En-Naggàza

late 3rd c.

4th c.?

Construction date

Abandonment

4th c.

Unknown

Mid 4th c.

6th cent.

Late 3rd or early 4th c.

Unknown

Africa Proconsularis

Byzacena Sufetula, Large Baths Mauretania Caesariensis

Numidia

Tripolitania

Possible Late Antique Baths Site, Name of bathhouse Africa Proconsularis Neapolis, building at Sidi Mahrsi Byzacena Ammaedra, building under Basilica 1 Mauretania Caesariensis Icosium, building under the Church of St. Philip

255

Lists of Late Antique Baths, Possible Baths and Continued Use of Baths

C Cyrenaica New Construction Site, Name of bathhouse

Construction date

Abandonment

Apollonia, Late Roman Baths (C65)

4th c.

5th or 6th c.

Cyrene, Byzantine Baths (C67)

4th c.

7th c.?

Gasr Bandis

4th to 7th c.

Unknown

Gabu Iunes

5th to 7th c.

Unknown

Gasr Khuraybah (C68)

5th to 7th c.

Unknown

Gasr Mismar (C69)

5th to 7th c.

Unknown

Gasr Sherbin

4th to 7th c.

Unknown

Limnias

4th to 7th c.

Unknown

Mechili

4th to 7th c.

Unknown

Mghernes (C70)

4th to 7th c.

Unknown

Ptolemais, City Baths (C71)

4th c.

6th c.

Ptolemais, Baths of the House of Paulus (C72)

Late 5th / early 6th c.

7th c.?

Siret Ain Relles

6th c. or later

Unknown

*Taucheira, Fortress Baths (FB)

7th c.

7th c.

Taucheira, Byzantine Baths (C73)

6th c.

7th c.

Wadi Senab (C74)

4th to 7th c.

Unknown

Zawiet el Argub

4th to 7th c.

Unknown

Site, Name of bathhouse

Construction date

Abandonment

Erythron

3rd c.

Late 4th c.

Site, Name of bathhouse

Construction date

Abandonment

Abu Mena, South Baths (C75)

Early 5th c.

8th c.

Abu Mena, North Baths (C76)

Early 6th c.

Late 7th c.

*Abu Sha’ar, Fortress Baths (FB)

4th c.

Late 4th / early 5th c.

Alexandria, Kom al-Dikka Baths (C77)

Late 4th / early 5th c.

7th c.

Alexandria, Baths under the Governmental Hospital

4th to 7th c.

Unknown

Alexandria, Baths in Anubis Street (C78)

4th or 5th c.

Unknown

Alexandria, Baths of the Cinema Majestic (C79)

Late 4th / early 5th c.

7th c.

Alexandria, Baths behind the Water Company (C80)

5th c.

Middle of the 6th c.

Alexandria, Baths of the Mehattet Square (C81)

5th c.

Late 6th / early 7th c.

Clysma (C82)

4th to 7th c.

7th c. or later

Ezbeth Fath Allah (C83)

6th c.

7th c.

Gherra / Mohameddia (C84)

5th or 6th c.

Unknown

Continued Use of Existing Baths during Late Antiquity

D Egypt New Construction

256

appendix 3

New Construction (cont.) Site, Name of bathhouse

Construction date

Abandonment

Karanis, Large Baths (C85)

4th to 7th c.

Unknown

Karm Kandara (C86)

4th to 7th c.

Unknown

Kom el Ahmar (C87)

4th to 7th c.

8th c.

Kom el-Dosheh (C88)

5th or 6th c.

6th c.

Marea, Byzantine Baths (C89)

Late 5th c.

8th c.

Marea, Central Baths (C90)

5th to 7th c.

Unknown

Mergham (C91)

5th to 7th c.

Unknown

Mit Abul Kum (C92)

4th to 7th c.

Unknown

Merghib

4th to 7th c.

Unknown

*Nag el-Hagar, Fortress Baths (FB)

4th c.

Unknown

*Pelusium, Fortress Baths

Late 3rd/ early 4th c.

Unknown

Psenemphaia (Kom Trougah) (C93)

4th to 6th c.

6th c.

Sersena (C94)

4th to 6th c.

Unknown

Taposiris Magna, Byzantine Baths (C95)

Mid 5th c.

7th c.

Teiba (C96)

6th c.

7th c.

Tell Kanaïs, site IV

4th c.

6th c.

*Tell el-Makhzan (EB)

4th c.

8th c.?

Tell el-Ruhban (C97)

4th to 7th c.

Unknown

Xois, Roman Baths

5th to 7th c.

Unknown

Site, Name of bathhouse

Construction date

Abandonment

Bitylion?

2nd–4th c.

4th c.

Pelusium, second bathhouse

2nd c.

8th c.

Thebes, Imperial Baths

2nd h. 2nd c.

Middle of the 4th c.

Trimithis

1st or 2nd c.?

End of the 4th c.

Site, Name of bathhouse

Construction date

Abandonment

Alexandria, Cricket Ground

4th to 7th c.

Unknown

al-Barnugi

4th to 7th c.

Unknown

Bubastis, Bath 1

1st–4th c.

Unknown

Bubastis, Bath 2

1st–4th c.

Unknown

Kellis

4th-5th c.?

Unknown

Kom al-Baroud

4th to 7th c.

8th c.

Kom Kobbeiz

5th or 6th c.?

Unknown

Nicopolis, North-west Baths

1st–4th c.

Unknown

Nicopolis, South-west Baths

1st–4th c.

Unknown

Tell Ishnik

1st–4th c.

Unknown

Continued Use of Existing Baths during Late Antiquity

Possible Late Antique Baths

257

Lists of Late Antique Baths, Possible Baths and Continued Use of Baths

E Palestine New Construction Site, Name of bathhouse

Construction date

Abandonment

*Ain Gharandal, Fortress Baths (FB)

4th c.?

5th c.?

Ashkelon, City Tell Baths

4th c.

5th or 6th c.

Beth Yerah, Baths (C98)

4th c.

Late 6th / early 7th c.

*Bir Madhkur, Fortress Baths (FB)

4th c.

Unknown

Caesarea Maritima, Area 1 Baths (C99)

4th c.

Late 6th / early 7th c.

Caesarea Maritima, Villa Suburbana Baths (C100)

Mid 6th c.

Middle of the 7th c.

Diocaesarea / Sepphoris, Western Baths

Late 3rd / early 4th c.

Unknown

El-Burj

Late 3rd / early 4th c.

Late 6th / early 7th c.

Elusa

4th to 7th c.

Unknown

Gadara, City Centre Baths (C102)

Early 4th c.

First half of the 7th c.

Gadara, Herakleides Baths (C103)

4th c.

Unknown

Ha’on (C104)

Late 6th / early 7th c.

Late 7th c.

Horbath Zikhrin, Area C Baths

4th to 7th c.

Unknown

Horbath Zikhrin, Area F Baths (C105)

4th to 7th c.

Unknown

*Jerusalem, Mount Scopus Baths (EB)

4th to 7th c.

Unknown

Jerusalem, Notre Dame Monastery Baths (C106)

4th to 7th c.

Unknown

Jerusalem, Old City Baths (C107)

4th to 7th c.

7th c.

Jerusalem, Ophel Baths

Late 6th / early 7th c.

Early 7th c.

*Jerusalem, Third Wall Baths (EB)

6th c.

9th c.

Kabul (C108)

4th c.?

Late 7th / early 8th c.

Khirbet Lasan

Late 4th c.

6th c.

*Legio, Fortress Baths (FB)

Early 4th c.

Middle of the 4th c.

Ma’ale Adumin, Monastery Baths (EB)

5th c.?

Unknown

Mampsis (C109)

4th to 7th c.

Unknown

Nicopolis (C110)

4th c.

7th c.

Oboda (C111)

4th to 7th c.

Unknown

Rama (C112)

Late 3rd / early 4th c.

5th or 6th c.

Rehovot (C113)

4th to 7th c.

Unknown

Scythopolis, Southern Baths (C114)

Late 3rd / early 4th c.

Early 6th c.

Scythopolis, Western Baths (C115)

4th c.

Late 6th / early 7th c.

Tel Sera

5th c.?

5th or 6th c.

Tiberias

4th c.

11th c.

Umm el-Amr, Baths near Monastery (C116)

5th c.

Unknown

*Umm el-Hajar (FB)

5th c.?

5th or 6th c.

*Yotvata, Fortress Baths (FB)

Late 3rd / early 4th c.

Late 4th c.

Zikhron Ya’aqov (C117)

4th c.

5th c.

Dharih, Late Baths (C101)

258

appendix 3

Continued Use of Existing Baths during Late Antiquity Site, Name of bathhouse

Construction date

Abandonment

Aila

2nd or 3rd c.

4th c. or later

Eleutheropolis, Baths near the Theatre

3rd c.

4th c.?

Hippos, South Baths

Early 3rd c.

Late 4th / early 5th c.

Humeima

1st–3rd c.?

4th c.?

Horbat Mesar

2nd c.

7th or 8th c.

Jerusalem, Mar Elias Baths

3rd c.?

Late 4th / early 5th c.

Pella

1st c.

Late 6th / early 7th c.?

Ramat Rachel

3rd c.

Early 7th c.

Sahir al-Baqar

1st c.?

4th c.?

Scythopolis, Eastern Baths

2nd c.

6th c.?

*Tamara, Fortress baths

3rd c.

4th c.

Thana / Thornia

1st c.?

4th c.?

Site, Name of bathhouse

Construction date

Abandonment

Ashkelon, El Qabu Baths

5th c.

7th c.

Possible Late Antique Baths

Baqa el-Gharbiya

5th c.

6th c.

Barqa

4th to 7th c.

Unknown

*Be’er Shema, Fortress Baths (FB)

4th to 7th c.

Unknown

Be’er Sheva, Bate Hamalka Street

4th to 7th c.

Unknown

*Be’er Sheva, Fortress Baths (FB)

4th to 7th c.

Unknown

Beth Nehemya

4th c.?

5th or 6th c.

Deir el-Balah

4th to 7th c.

Unknown

En-Qobi/Ein el-Qabu

Late 6th / early 7th c.

Unknown

(Tel) Hefer/Tell el-Ifshar, northwest of tell

4th to 7th c.

Unknown

Horbat Hanot / Khirbet el-Khan, roadstation baths

4th to 7th c.

Unknown

Horbat Minim / Khirbet el-Minya, baths south-east of Umayyad palace

4th to 7th c.

Unknown

Horbat Tarbenet

4th to 7th c.

7th c.

Jaffa/Yafo, Ganor Compound

4th to 7th c.

7th c. or later

Jerusalem, Beit Hanina

Late 6th / early 7th c.

Unknown

Kefar Uriyya

4th to 7th c.

Unknown

Khirbat Burin, Qalansuwa junction

4th to 7th c.

Unknown

Khirbat Ta’ena

4th to 7th c.

13th c. or later

Lydda / Diospolis

Late 6th / early 7th c.

Unknown

Na’aran

4th to 7th c.

Unknown

Nahal Ashan

4th to 7th c.

Unknown

Nahf

Late 6th / early 7th c.

Early 8th c.

Neve Shalom

Late 6th / early 7th c.

Unknown

Ptolemais, Building in Trumpeldor Street

Late 6th / early 7th c.

Unknown Unknown

Qalandia

4th c.

Ramat Ha-Sharon

4th to 7th c.

Unknown

Ramla, Area I2

7th c.

8th c.

Ramla, Area K1

7th c.

8th c.

Tel Aviv, HaAretz Museum

4th to 7th c.

Unknown

Tel Qasile

6th c.

7th c.

Gazetteer

⸪

Introduction The gazetteer presents a sample of late antique baths in Roman Italy, Sicily and Sardinia, North Africa, Cyrenaica, Egypt and Palestine. The different baths are grouped by region and then arranged alphabetically following their ancient names (or as found in the discussion). Much as in Yvon Thébert’s catalogue,1 each newly built bath is briefly presented, giving the following information: – excavation history of the bathhouse – relevant bibliography – construction date – continued use of the bathhouse, date of abandonment and possible reuse of the building – a description of the bathhouse plan – information about the decoration of the different rooms – information about the heating system and water management – Construction technique – Surface area (only for C-numbers) – GPS coordinates (only for C-numbers) Each late antique bath for which a plan was available has been given a C-number (catalogued). The plans were re-examined and updated on the basis of earlier publications and, when possible, by the author’s own observations. Each plan is reproduced following a key, specially created by the author to present a uniform and clear corpus of plans for easy comparison. The size of the baths is always given in relation to a scale-bar of 5 m (chosen with the smaller baths in mind), except for the large imperial-style thermae of several thousand square metres. Besides information on the securely dated late antique baths, a short overview of possible late antique baths is also given for each region. These buildings have sometimes been interpreted as bathhouses; although the nature of the archaeological evidence does not meet with the basic criterion for identifying a bath as proposed at the beginning of this study (see Introduction). In most cases, either a hypocaust or a pool has been recognized, but there is no conclusive evidence for the other element. Some regional private baths belonging to a villa show clear signs of public use. These have been grouped under a separate heading. Baths belonging to an ecclesiastical complex or baths of an army fortress have been included in the gazetteer. Although not ‘public’ in the strict sense of the word (see Introduction), these baths were intended to be used for several people at the same time, i.e. a communal use, even

1  Thébert (2003) 125–284.

if this was restricted to a specific target group. Both categories are described following the same scheme as presented above. Finally, the gazetteer presents an overview of the High Imperial baths that were still in use during Late Antiquity. As some of these baths continued to be the most important bathing facility of a city or neighbourhood, it is important to understand if and how these baths were altered to ensure their survival. Missing from this overview are the High Imperial baths of the urban case-studies presented in the discussion. These baths are discussed as part of the urban infrastructure of these cities. It is important to note that the evidence for dating the baths is not always secure. For this reason, the evidence is rated with an A to D system, loosely based on the dating classes in Luke Lavan’s work on late antique public space.2 The following system is used in this book: – (A): Secure date provided by scientific analyses (radiocarbon dating, dendrochronology, etc.), inscriptions and / or contextual finds (ceramics, coins, small finds). – (B): Secure terminus ante quem / post quem on the basis of stratigraphy or inscriptions. – (C): Approximate date provided by general construction technique, architectural styles, use of spolia or decorative schemes. – (D): Approximate date on the basis of regional or site developments. As Lavan rightly points out, such an ‘evaluation’ of the evidence should not be seen as a harsh judgement of former archaeological research. It merely points out the availability of the evidence and engenders a cautious approach when dealing with this data. The general maps showing the location of the different archaeological sites and of the inscriptions have a uniform layout, making use of the same key. There is no separate map for the location of the papyri, as the find spots (Oxyrhynchus, Hermopolis, etc.) are also shown on the map of the archaeological sites. The maps of the cities that have been discussed as case-studies are shown in a uniform layout, for which the key is also given below. The plans of the other cities are adapted from earlier publications, sometimes with small changes according to the key of the case-studies. The figures are borrowed from secondary literature. In some specific examples of baths within a private dwelling, but with a (semi-)public use,

2  I express my sincere gratitude to Dr. Luke Lavan for kindly providing me with a draft of his chapter on dating in his upcoming book (Lavan, forthcoming).

262 the different entrances are marked with arrows of a different colour. The green arrows show the public (street) entrance, the red arrows the entrance from the house and the yellow arrows the service entrance. In some plans of baths, the

Gazetteer temperature of the pools is represented by different colours: blue for cold, orange for tepid and red for warm. The plans of a selected number of baths have been adapted following the key of the newly built baths.

Baths in the Italian Peninsula, Sicily and Sardinia New Construction C1. Aquileia (Reg. X), Large Baths of Constantine II

Excavation history: The earliest excavations on the site known as the ‘Braida Murada’ were in 1922–1923 under Giovanni Brusin. The substantial mosaics were believed to be part of a large public building, possibly a bathhouse. New excavations in the 1980s confirmed the function of a bathhouse, placing the mosaics in the cold section of the building. The campaigns of the Università di Udine—starting in 2002 and still on-going—have revealed the extent of the baths (ca. 2 hectares), as well as a construction date. Bibliography: Brusin (1922); Brusin (1923) Bertacchi (1981); Bertacchi (1990); Lopreato (1994); Fales et al. (2003) Rubinich (2006); Rubinich (2007); Rubinich (2008); Rubinich (2013) Construction: (A) On the basis of an inscription found in the vicinity of the baths and mentioning Constantine II (EI-8) and of a coin of the same emperor found in the preparation layers of a floor, the baths have been dated to AD 337–340.3 Continued use and abandonment: (A) Restorations were still being carried out in the first half of the 5th c. and the walls were still standing in the 6th. In this last phase, the baths were reused to accommodate modest houses, until the vaults fell down in the 7th c. The site remained occupied until the 12th c., although the type of activities that went on here cannot be identified.4 Plan: As the excavations were still on-going at the time of writing, the plan is not yet entirely clear. The cold rooms were situated to the east. Test pits underneath the floors revealed that no anterior construction took place, meaning that the baths had been constructed ex novo on a virgin piece of land on the fringes of town.5 It consisted of a large rectangular hall (40 × 22 m), surrounded by 6 rectangular piscinae. To the north and to the south, rectangular aulae (31.5 × 22 m) may have served as apodyteria. To the east of the frigidarium, the remains of a large open-air swimming pool were partially uncovered (10 m in width, length unknown). The heated section extended towards the west, but the exact layout is still unclear. Several rooms with hypocausts have been identified. The caldarium undoubtedly lay on the same axis as the frigidarium and probably had an apsidal alveus on its western side.6 3  Rubinich (2013) 89. 4  Rubinich (2007) 135–37; Rubinich (2008) 161; Rubinich (2013) 89–90. 5  Rubinich (2006) 154–55 contra Bertacchi (1981) 63. 6  Rubinich (2013) 88.

C1

Aquileia (Reg. X), Large Baths of Constantine II (After Rubinich (2008) 162, fig. 1)

Decoration: The cold rooms were paved with polychrome mosaics (in the aulae) and opus sectile (in the frigidarium). The mosaics depicted the busts of athletes, similar to those found in the Baths of Caracalla in Rome.7 The inscription refers to athletic games. The cold pools were adorned with polychrome mosaics of different types of stone, including porphyry and serpentine. In the northern aula, a marine theme of Neptune on a chariot with Nereids and Tritons was also found. In the southern aula, a hunting scene points to eastern influences.8 A small room to the south was adorned with a black and white geometric mosaic, possibly dating from a restoration phase in the late 4th or early 5th c. Several of the mosaics were also restored at a late date with marble slabs (5th or even 6th c.?).9 The walls of the cold rooms were finished with slabs of marble and other types of stone (including pink granite). The upper walls and the intrados of the vaults were probably embellished with wall paintings and glass mosaics, as many painted plaster fragments and tesserae were found in the destruction layers of the building. The capitals that were 7  Lopreato (1994) 99 and pl. XLVI.2, XLVII.1–2. 8  Rubinich (2013) 86. 9  Fales et al. (2003) 221–22; Rubinich (2013) 87.

264

Gazetteer

found belong to the Asian-Corinthian type. Several statue fragments came to light during the excavations, most of them dating to the 1st and 2nd c.10 It is unclear whether these statues had been relocated to adorn the late antique baths or were brought there at a later date to be burnt in medieval lime kilns. Technology: As most parts of the heated section are still buried, little can be said about the heating system and water management. The hypocaust pillars were made out of round bricks. At least 4 furnaces heated the supposed caldarium. Construction technique: Opus testaceum Bathhouse surface: Unknown (incomplete plan; exceeding 1,000 m2) GPS coordinates: 45°46’11.13”N, 13°21’53.74”E

C2. Brixia (Brescia, Reg. X), Castello Baths

Excavation history: In the courtyard of the fortress of Brescia, a small bathhouse was discovered during systematic excavations in the 1980s. Bibliography: Breda (1984); Brogiolo (1993). Construction: (B) The stratigraphy of the excavations point to a phase of use in the 5th and 6th c. Continued use and abandonment: (B) The building was later transformed into a latrine. Plan: The building (5.7 × 2.1 m) consisted of three successive rooms (apodyterium / frigidarium -tepidariumcaldarium?). It was built against a pre-existing water reservoir. Decoration: The walls seem to have been coated with a simple stucco. Technology: The caldarium was heated by two parallel channels and tubuli. There was only one small basin. Construction technique: Opus testaceum Bathhouse surface: ca. 12 m2 GPS coordinates: 45°32’33.19”N, 10°13’30.58”E

C2

C3. Brundisium (Reg. II, Brindisi), San Pietro degli Schiavoni Baths

Excavation history: In Brundisium (Brindisi), an excavation in the 1960s uncovered part of a Roman insula that was thoroughly remodelled in Late Antiquity. A bathhouse was inserted in the Imperial period structures, its praefurnium partially encroaching upon the street. Bibliography: Cocchiaro et al. (1990); Lippolis and Baldini Lippolis (1997). Construction: (C) The type of opus listatum used to construct the walls saw the archaeologists date the building to the early 4th c. Continued use and abandonment: (C) It is not known when the baths fell out of use, but the entire insula shows signs of continuous occupation until the Early Middle Ages. Plan: The bad state of preservation and concise publication of the excavations makes it difficult to reconstruct the internal organisation of the rooms. A room with a small apsidal pool has been interpreted as the frigidarium. Decoration: In some of the rooms, marble slabs decorated the walls. Technology: At least 4 rooms seem to have been heated by a hypocaust. The small pool was drained by a sewage channel leading to the street. Construction technique: Opus caementicium? Bathhouse surface: ca. 300 m2 GPS coordinates: 40°38’20.17”N, 17°56’40.53”E

Brixia (Brescia, Reg. X), Castello Baths (After Brogiolo (1993) 72, fig. 50)

C3 10  Fales et al. (2003) 242–59.

Brundisium (Brindisi, Reg. II), Baths of S. Pietro degli Schavioni (After Lipolis and Lipolis (1997) tav. LXXV)

265

Gazetteer

C4. Luna (Reg. VII, Luni), Late Baths

Excavation history: The bathhouse was found during systematic excavations of the Soprintendenza during the 1970s. Bibliography: Ward-Perkins (1977); Ward-Perkins (1978). Construction: (C) An earthquake at the beginning of the 4th c. seems to have dealt a severe blow to a city already in decline. The capitolium and the basilica were probably abandoned around this time. A bathhouse was built upon the levelled remains of the former, perhaps at the end of the 4th or the beginning of the 5th c.11 Continued use and abandonment: Unknown.

Plan: The baths are only partially known. In the foundations of the walls, architectural elements, probably of the capitolium, were found. The remains of what appears to be a cold room with a semicircular pool were identified.12 To the north-west of this, a room with a hypocaust reveals the beginning of the heated section. The second heated room possibly had a semicircular pool. Decoration: The cold room was paved with white marble slabs. Technology: A sewage channel ran in an east-west direction to the south of the baths, coming from the nearby late antique fish market.13 A second sewer ran from the west side of the baths, possibly to drain the heated pools. Construction technique: Opus caementicium Bathhouse surface: Unknown (incomplete plan) GPS coordinates: 44° 3’52.12”N, 10° 1’3.68”E

C5. Mediolanum (Reg. XI, Milan), Herculean Baths

Excavation history: Parts of the large complex were first discovered in 1841 during construction works near modern Corso Europa. Smaller works revealed new rooms in the 1960s and 1970s, before archaeological excavations uncovered most of the cold section between 1985 and 1988. Bibliography: Ceresa Mori (1987); Ceresa Mori (1990); David (1996).

C4

Luna (Luni, Reg. VII), Late Baths (After Ward-Perkins (1978) 35, fig. 10)

C5 11   Ward-Perkins (1978) 38. The link between the late 3rd c. domus dei Mosaici and the baths seems implausible due to the great distance between both buildings, see Haug (2003) 207, n. 921 contra Durante and Gervasini (2000) 13.

Mediolanum (Milan, Reg. XI), Herculean Baths (After Ceresa Mori (1990) 460, fig. 2a)

12   Ward-Perkins (1978) 33, fig. 10. 13  Pia Rossignani (1985) 62–63.

266 Construction: (B) The so-called ‘Terme Erculee’ were mentioned by Ausonius (see chapter 2). Brick stamps and amphora types used in the construction of the baths give a terminus post quem of the late 3rd or early 4th c.14 Fragments of ceramic found in the preparation layers of the mosaics date from the early 4th c.15 Those mosaics, which are contemporaneous with the construction, date from the late 3rd or early 4th c., with only some evidence for later restorations, also during the 4th c.16 An inscription found in the frigidarium is often interpreted as a dedication by Maximian (AD 285), who allegedly gave his name (Maximianus Herculeus) to the baths.17 Continued use and abandonment: It is unknown when the complex fell out of use (5th c.?), but the building probably collapsed after a fire. Plan: The plan of the baths is partially known through excavations, but has been severely damaged by building projects in the course of the 20th c. They covered a surface of around 4,500 m2 in the ancient city centre. It seems that the bathhouse was preceded to the north by a large colonnaded palaestra. The perimeter wall had several semicircular apses and had monumental access with portico to the north. The frigidarium was rectangular, but had a large semicircular apse to the north, protruding into the palaestra. It probably contained a pool. To the east and west of the frigidarium, there was a rectangular room (apodyterium or basilica thermarum). From the frigidarium, one passed into a succession of heated rooms flanking a central tepidarium. Both the flanking rooms and the tepidarium were connected to the caldarium by a long narrow corridor. The caldarium itself had two flanking semicircular apses for alvei. A third pool may have been lodged in an apse in the south wall.18 The layout has obvious similarities with the imperial baths at Trier (Germany, Constantinian in date) and the Baths of Constantine in Rome (see below, C22). Decoration: The information about the decorative scheme is restricted to the decoration of the floors, as the walls were dismantled in later phases. Polychrome mosaics floors with geometric design were found in the eastern apodyterium and the eastern heated rooms. The piscina of the frigidarium was paved with marble slabs. A large statue of Hercules, dated to the 2nd c., was also found on site.19

14  Ceresa Mori (1990) 101. 15  Ceresa Mori (1987) 149. 16  David (1996) 96–104. 17  Some doubts have been raised about this interpretation. The inscription may be dated to the reign of Commodus, see Ceresa Mori and Sartori (1997) 31–32. Yet, this does not exclude the reuse of the inscription in the Tetrarchic period. 18  Ceresa Mori (1990) 100. 19  M  useo Civico Archeologico di Milano, inv. 1143.

Gazetteer Technology: Some sewage channels were found underneath the floor of the frigidarium, but the other aspects of water management and heating techniques are poorly understood. The water for these thermae was possibly supplied by the Aqualunga aqueduct, entering the city from the north-east.20 Construction technique: Opus testaceum Bathhouse surface: ca. 4,500 m2 GPS coordinates: 45°27’52.20”N, 9°11’45.95”E

C6. Mediolanum (Reg. XI, Milan), Baths in Via S. Maria Valle

Excavation history: During construction works in 1963, a building with hypocausts and basins was discovered between the Via Torino and the Via San Maria Valle. Other walls and mosaics, presumably from the same building, were found in the vicinity, in Via Soncino and underneath Piazza S. Maria Valle. Besides the hypocaust and the basins, the toponym of the nearby Via Bagnera seems to confirm the interpretation as a bathhouse.21 Bibliography: Lusuardi Siena (1986); Haug (2003) 425 (with anterior bibliography). Construction: (C) The mosaics were dated between the 4th and 6th c., although the building itself underwent many restorations.22

C6

Mediolanum (Milan, Reg. XI), Baths in Via S. Maria Valle (After Anonymous (1990) tav. CLXXXIV, fig. 246)

20  Caporusso (1990) 94. 21  Caporusso (1990) 95. 22  David (1996) 131–32.

267

Gazetteer Continued use and abandonment: Unknown. Plan: The building seems to have been oriented along an east-west axis. In the south, a room with a hypocaust was identified, connecting to a round pool (diam.: 5.1 m). To the north-east of the latter, part of a semicircular pool can be identified. Decoration: Only some fragments of a mosaic floor were recovered during the excavations. Technology: The hypocaust pillars consisted of round tiles. No other information is available. Construction technique: Opus testaceum Bathhouse surface: Unknown GPS coordinates: 45°27’38.97”N, 9°11’57.54”E Metapontum (Reg. III), Baths Excavation history: Unknown. Bibliography: Gianotta (1980). Construction: (D) The bathhouse in Metapontum (Metaponto) has sometimes been dated to Late Antiquity due to the use of spolia and the small apsidal pools.23 Continued use and abandonment: Unknown. Plan: Unknown.

Construction: (B) A stamp on a lead fistula, mentioning the praefectus urbi and his wife, dates one of the phases of use to AD 330–350.24 However, it is uncertain whether this fistula belonged to the initial construction phase or to a later restoration. Continued use and abandonment: Unknown. Plan: The excavated part of the baths consisted of a rectangular room upon a hypocaust with two horseshoe-shaped alvei. Decoration: The interior of the small pools were decorated with marble slabs. Technology: The hypocaust pillars were made of square tiles; the tubuli were rectangular.25 Construction technique: Opus listatum Bathhouse surface: Unknown (incomplete plan) GPS coordinates: 41°45’12.17”N, 12°16’50.75”E

C8. Ostia (Reg. I), Baths in the Horreum (III, xvII, 1)

Excavation history: This region of Ostia was excavated in the 1950s without publishing the results.

C7. Ostia (Reg. I), Baths West of the Palazzo Imperiale

Excavation history: During geophysical prospection in 2000 by the German Archaeological Institute in Rome, the remains of what appeared to be a small bathhouse were discovered in the unexcavated fields west of the Palazzo Imperiale. Test trenches confirmed that the structure was a bathhouse. Bibliography: Heinzelmann (2001); Martin et al. (2002).

C7

Ostia (Reg. I), Small Baths West of the Palazzo Imperiale (After Martin et al. (2002) 273, fig. 16) c8

23  Gianotta (1980) 78 and pl. XXIII.

Ostia (Reg. I), Baths in the Horreum (After Poccardi (2006) 178, fig. 6)

24  Heinzelmann (2001) 325–27; Martin et al. (2002) 273–74. 25  As can be seen in Heinzelmann (2001) 324, fig. 10.

268 Bibliography: Poccardi (2006). Construction: (B) The construction technique of the walls, a late type of opus listatum, points to a late antique construction date (4th c. or later). Furthermore, the baths were built upon the Diocletian-Constantinian raising of the walking level.26 Continued use and abandonment: Unknown. Plan: The small bathhouse was built in a horreum along the Via dei Aurighi. The bathers entered through the main gate of the horreum.27 The portico that preceded this gate was delimited by thin walls to form two rooms on either side of the entrance. The westernmost room served as a praefurnium for the baths; the easternmost is of unknown use. The first room of the baths was unheated. It had a water supply pipe in its east wall, probably supplying a small basin or a labrum. In the north-western corner, there may have been a second basin (unknown dimensions). The bather continued south into a suite of three heated rooms, of which only the southernmost had pools. Both the rectangular (1 × 1.4m) and the apsidal alveus (1.6 × 1.1m) were heated by a furnace. These small and highly functional baths could not have accommodated more than 10 people at a time. Its use was perhaps limited to a specific group of people or perhaps to the people that lived or worked in the other rooms of the horreum. Decoration: Due to the ruinous state of the remains and the absence of an excavation report, the decoration of the baths is difficult to reconstruct. There is no evidence for iron clamps in the walls, making it likely that the walls were coated with plaster and then painted, rather than adorned with marble slabs. Technology: The hypocaust consisted of square pillars, while the walls were heated by tubuli. A water reservoir may have been lodged in the south-east room of the horreum.28 It is unknown how the water was transported to the baths and the sewers in the Via Degli Aurighi were already out of use when the baths were built. Construction technique: Opus listatum Bathhouse surface: ca. 60 m2 GPS coordinates: 41°45’9.22”N, 12°17’2.50”E

C9. Ostia (Reg. I), Baths of Musiciolus

Excavation history: Along the Via Severiana, some 150 m east of the insula of the thermae of Porta Marina, a small bathhouse was uncovered during the excavations of the nearby synagogue. The excavations of the baths were carried out

26  Gering (2010) 102, n. 34. 27  The following description of the bathhouse is based on the author’s observations on site, in combination with the plan presented by Poccardi (2006) 178, fig. 6. Measurements were taken on site. 28  Poccardi (2006) 178, fig. 6.

Gazetteer

C9

Ostia (Reg. I), Baths of Musiciolus (After Turci (2014) fig. 4)

in 1979 by the Soprintendenza, but were never finalised. Small restoration works were carried out in the 1980s. Bibliography: Floriani Squarciapino (1961); Turchi (2014). Construction: (C) The mosaics, stolen during restoration works in the 1980s, were dated to the end of the 3rd or beginning of the 4th c.29 The type of opus listatum and the type of mortar seems to imply the same time frame.30 Continued use and abandonment: (C) Small restorations may have taken place as late as the 5th c.31 Plan: The bathhouse was built within a Hadrianic structure that was severely altered in Severan times. The bathers entered the building from the Via Severiana into a small vestibule, separated from a large hall by two columns. A narrow door in the north-east corner led into the frigidarium, which consisted of a rectangular space with a northsouth orientation and a floor in cocciopesto. A well in the northern wall was a relic of the first phase of the building, but was reused for the baths. The southern part of the frigidarium, delineated off by two columns, had a mosaic floor with the depiction of marine fauna. The eastern part, with a bench along its eastern and southern wall, had a simple black and white geometric mosaic. The north-eastern corner of the frigidarium was occupied by an apsidal piscina (ca. 3.2 × 3.4 m; 1.2 m deep). A second rectangular piscina was lodged in the south-western corner (ca. 1.4 × 2 m; 1.2 m deep). The heated rooms lay to the south and only the last 29  Floriani Squarciapino (1987) 113. 30  Turci (2014) §33. 31  Turci (2014) §32 on the basis of wall construction.

Gazetteer one had two single-person alvei (semicircular: diam. 1.1 m; rectangular: 1.5 × 1 m; at least 0.8 m deep). The apse of the middle heated room did not lodge a pool. It may have been used as a small sweat alcove.32 Decoration: Great care was taken to embellish the interior of the baths with marble and mosaics. The marble columns in the cold rooms and the marble slabs used to decorate the pools were spolia. The mosaic of the frigidarium depicted marine animals, such as dolphins, fish and jellyfish. The polychrome mosaic in the central heated room depicted the busts of 4 athletes—named Faustus, Ursus, Luxsurius and Pascentius—and a gymnasiarch named Musiciolus. The mosaics in the other heated rooms consisted of geometric and floral designs. Technology: Only in the last two heated rooms were the walls were heated by tubuli (of different forms and dimensions).33 The hypocaust pillars were made of standard square tiles. A boiler may have been placed on top of the caldarium furnace. The alvei may have drained directly onto the furnace floor, but the pool floors are too damaged to identify outlets. It is unknown how the baths were supplied with water. The well in the frigidarium could not have supplied all the pools and may only have been used for cleaning the baths or for short ablutions. On satellite imagery, a possible large cistern consisting of three rectangular rooms can be identified some 40 m north of the baths.34 Construction technique: Opus listatum Bathhouse surface: ca. 300 m2 GPS coordinates: 41°44’57.18”N, 12°17’17.64”E

C10. Ostia (Reg. I), Baths along the Via Marciana (Iv, Ix, 1)

Excavation history: A small bathhouse was discovered on the junction of the Via Marciana and the Via Severiana. The excavations by the Soprintendenza in the 1970s were never published. Recently, the entire insula was re-examined by the University of Bologna. Bibliography: David et al. (2014). Construction: (C) The opus listatum used to construct the bath-related walls point to a late antique construction date (4th c.).35 The absence of any published data hampers a clear understanding of this building. Recent excavations in the structure north of the baths, revealed continued activity in the 5th c.36 No such evidence is available for the baths. Continued use and abandonment: Unknown. 32  Turci (2014) § 30. 33  Although the tubuli have an average height of 34 cm, their cross section varies from rectangular to square and round. A type measuring 7 × 7 × 34 cm was also found in the Baths of Elagabalus in Rome, dated to the 5th c. (see above C25, observations by the author). 34  Martin et al. (2002) 268, fig. 10. 35  David et al. (2014) §25. 36  David et al. (2014) §31–39.

269

C10 Ostia (Reg. I), Baths along the Via Marciana (After Poccardi (2006) 178, fig. 7)

Plan: The baths were inserted in a pre-existing building, probably a shop, and can best be compared to the baths of the horreum (see above, C8). What remains of these baths are two heated rooms and a service corridor.37 The cold section probably lay south of the heated rooms, bordering the Via Severiana. The building was probably accessed from this road or from the Via Marciana. The heated section consisted of an L-shaped room. It had an apsidal (1.5 × 1.25 m; 1.10 m deep) and a rectangular alveus (1.2 × 1.4 m; 1.1 m deep), both of which had very thin internal steps. The access to the rectangular pool must have been rather difficult, as the borders of the pool were quite high and without any steps facilitating entry. Therefore, the pool might have been a simple basin that provided hot water for ablutions.38 Decoration: The floors and the walls of the heated rooms, as well as the inside of the pools, were adorned with white marble slabs, some of which are still visible.

37  As there is no published excavation report, the description of the baths is based on observations and measurements taken by the author on site. 38  However, the pool / basin seems to have been equipped with a testudo, which would be quite strange (and costly) for a simple ‘warm water reservoir.’ One would also expect a simple basin to be quite shallow.

270

Gazetteer

figure 61 Baths along the Via Marciana, seen from the west photo: author, April 2015

Technology: The two pools were heated by their own furnace. A third stokehole heated the room itself. The type of hypocaust cannot be identified, as observations on site are hampered by vegetation and soil accumulation. The walls were heated by the short type of tubuli, although these did not cover the entire surface of the wall. Rows of three adjacent tubuli were lodged in recesses in the walls of the rooms and on the northern wall of the apsidal alveus (fig. 61). The water for the pools was probably supplied by a well, found in the northern corner of the praefurnium.39 The inlets for the water supply of the pools are still visible in the walls. The lead or terracotta pipes were probably torn out after the abandonment of the baths. There does not seem to have been a boiler above one of the furnaces, as any type of supporting structure is absent. Alternatively, the bath water was only heated by means of the hypocaust. For the rectangular alveus, a (semi?) testudo may have been used.40 The water of the apsidal alveus was drained through an outlet in its western corner and was probably used to clean the underlying furnace. The water of the rectangular pool could have been drained through the presumed testudo. Traces of a sewage channel can still be distinguished in the praefurnium floor, heading towards the sewer of the Via Marciana. 39  Shown on fig. 7 in Poccardi (2006) 178. 40  Maréchal (2017) 182–84.

Construction technique: Opus listatum Bathhouse surface: Unknown (incomplete plan; not exceeding 500 m2) GPS coordinates: 41°44’58.46”N, 12°17’9.45”E

C11. Ostia (Reg. I), Small Baths on the Via Severiana

Excavation history: Across the Via Marciana, in the insula belonging to the Baths of Porta Marina, a small bathhouse was discovered in the 1970s. The excavations were only briefly described.41 Bibliography: Floriani Squarciapino (1974) Construction: (C) On the basis of the opus listatum technique of the walls, the baths have been dated to the 4th c. or later.42 Continued use and abandonment: Unknown. Plan: Just as the Baths in the horreum (see above, C8) and the baths in the Via Marciana (see above, C10), this complex was built within pre-existing structures. The entrance lay on the Via Severiana and led into a vestibule. From here, the bathers entered the frigidarium with its two large piscinae, one apsidal (ca. 3 × 3.6 m; 1.2 m deep) and one rectangular (ca. 3 × 2.1 m; 1.2 m deep). The heated rooms lay south and were almost square in plan. The easternmost room, the 41  The description of the baths is based on the observations and measurements of the author made on site. 42  Pavolini (2006) 181.

Gazetteer

271

figure 62 Satellite imagery of Ostia (Italy), showing a street (arrows) heading towards the Baths of Perseus (A). Also depicted are the Baths of Musiciolus (B), the Baths of Edificio b (C), the Baths of Porta Marina (D), the Small Baths along the Via Severiana (E) and the Baths along the Via Marciana (F). Google Earth, June 2007

C11

Ostia (Reg. I), Small Baths on the Via Severiana (After Poccardi (2006) 180, fig. 10)

caldarium, had an apsidal (1.5 × 1.8 m; 1.2 m deep) and a square alveus (s.: 1.9 m). Decoration: The floors of both cold and warm rooms, as well as the inside of the pools, were paved with white marble slabs, some of which have survived to this day. The walls are too damaged to reconstruct any form of decoration. The three niches in the walls of the apsidal piscina must have contained statues. In the western praefurnium,

marble column shafts can be seen, but it cannot be confirmed if these came from the baths. Technology: The type of hypocaust could not be determined. The tubuli are of a rather large type (7 × 13 × 38 cm) in all three rooms. The caldarium was heated by two furnaces, located under the alvei. A square room east of the northern praefurnium was probably used to store the firewood. No traces of a boiler installation were found. The rounded form of the furnace openings could point to the use of testudines. The water supply was probably ensured by a branch of the aqueduct that also serviced the thermae of Porta Marina. A reservoir can still be seen some 15 m east of the baths, although there is no evidence that it was connected to the baths.43 The waste water from the piscinae was collected in a sewer than ran under the frigidarium and probably south towards the sewer of the Via Severiana. A water collector can still be seen in the centre of the frigidarium floor. The alvei may have been drained by outlets above the furnaces. Construction technique: Opus listatum, opus testaceum Bathhouse surface: ca. 130 m2 GPS coordinates: 41°44’58.09”N, 12°17’10.52”E

43  The reservoir is not contemporaneous with the thermae of Porta Marina, as it is constructed in an opus listatum similar to the one used in the small baths.

272

Gazetteer a larger insula and seem to have had a separate entrance from the street, we can assume that the baths had a public or semi-public use. This articulation of baths and living units is reminiscent of the Baths in the Palazzo Imperiale, the Baths of the Seven Sages and the Baths of Buticosus (see above, chapter 3, p. 115 and p. 122–123). Decoration: Unknown. Technology: Unknown. Construction technique: Opus testaceum Bathhouse surface: Unknown GPS coordinates: 41°45’3.82”N, 12°17’30.88”E

C13. Ostia (Reg. I), Byzantine Baths (IV, IV, 8)

C12 Ostia (Reg. I), Baths of Perseus (After Ricciardi and Scrinari (1996a) 183, fig. 299; plan without scale)

Excavation history: The building was first excavated during large-scale clearing works in the 1930s, before being reexamined by the Soprintendenza in the 1950s and the 1970s.45 Bibliography: Poccardi (2006). Construction: (C) The Byzantine Baths have sometimes been attributed to the 2nd c. because part of the complex was built on top of earlier imperial structures.46 However, the majority of the walls relating to the bath elements— those of the pools, heated rooms and service corridors— are constructed in opus listatum dated to the turn of the 4th and 5th c.47 Some earlier walls in opus testaceum or opus mixtum were reused in the cold rooms, while traces of an even older wall in opus reticulatum were incorporated in the southern part of the heated rooms. Continued use and abandonment: Unknown.

C12. Ostia (Reg. I), Baths of Perseus

Excavation history: Outside the Porta Laurentina, a bathhouse was found that supplanted a Hadrianic structure. The complex was excavated, but the results were never published. The remains were reburied afterwards. Bibliography: Not available. Construction: (D) On the basis of the plan, the baths have been dated in the 4th c.44 The resemblance to the Baths of the House of the Dioscuri (see above, chapter 3, p. 128–129) and the last phase of the ‘Byzantine’ Baths (see below) supports a late antique phase of use. Continued use and abandonment: Unknown. Plan: Satellite imagery reveals that the baths were located along an important traffic artery, connecting the Via Severiana with the suburbs outside the Porta Laurentina (fig. 62). In the absence of an excavation report, the plan is difficult to interpret. It seems that a frigidarium had a small circular and a larger rectangular pool, while the heated room (caldarium) had two small semicircular alvei. Information on the technical aspects and the decoration is lacking. As the baths are implanted within the structure of 44  Ricciardi and Scrinari (1996a) 183.

C13 Ostia (Reg. I), Byzantine Baths (After Calza et al. (1953) pl. 7 and 12)

45  The baths have also been called the Terme di Via del Tempio Rotundo or the Terme della domus di Giove fulminatore. 46  Poccardi (2006) 177. 47  Heres (1982) 511–14.

Gazetteer

273

Plan: The complex was the largest newly built bathhouse of Late Antiquity.48 It was entered by a long corridor giving onto the Via del tempio rotundo.49 The cold room was conceived as a large hall with 6 columns dividing it into three naves. It was equipped with an exceptionally large apsidal natatio (ca. 12 × 5 m; 1.4 m deep) and two apsidal piscinae (ca. 3.2 × 6 m; 1.2 m deep). To the south, the bathing itinerary split up into two separate suites of rooms. The eastern one started in a narrow corridor, at the end of which a circular pool (diam.: 1.7 m; 0.8 m deep) was added at a later date. To the south, three almost square heated rooms ended at a caldarium with two apsidal alvei (ca. 2.6 × 2.7 m; 1.2 m deep). In the central heated room, the western wall was pierced, possibly at a later date, to connect with the western suite of heated rooms. The latter had essentially the same layout, except that the northernmost heated room had a semicircular apse. The two alvei were slightly smaller than those of the eastern caldarium, but equally deep (1.2 m). The western suite seems to have had a separate entrance to the west: a narrow door led into a corridor with a large semicircular apse to its south and an adjacent apsidal piscina (1.7 × 1.9 m; 1.2 m deep). This part of the baths possibly dated to the first phase of the building, perhaps as the private bath suite of a house dating from the High Empire. The double design of these baths has often been interpreted as a separation of the sexes.50 The eastern wing of the baths, including also the large cold hall with the natatio, has been labelled the men’s section, while the smaller western wing has been called the women’s baths. If a strict separation of sexes was the case, then one would expect that the sections were fenced off. It would be interesting to verify if the communal hall connecting the cold hall with the heated rooms could be closed by a door. The ruinous state of the remains and the dense vegetation impede any observation on site. Decoration: The cold hall was paved with a mosaic of large tesserae (fig. 63), while the natatio and the piscinae were embellished with marble slabs. The types of marbles and their disposition confirm the late 4th or early 5th c. date.51 The walls and the floors of the heated rooms and the heated pools were also covered with marble. The upper parts of the wall, however, may have been decorated with wall paintings. Technology: Both the eastern and the western suite of heated rooms had a hypocaust with the traditional square pillars. The tubuli were mainly of a large type in the caldaria (internal measurements: 7 × 10 × 38 cm) and of the small

type (6 × 9 × 34 cm) in the other rooms. Each alveus was heated by its own furnace. The western semicircular alveus of the western suite of rooms was the only pool not to be heated by tubuli. The water for the baths was stored in a large reservoir east of the complex.52 The connection with the baths, as well as the distribution within the bathhouse, is unclear. The waste water of the piscinae must have been collected by a sewer running under the frigidarium. The large apisdal piscina was drained by a sewer running east.53 The ruinous state of the alvei does not permit identification of the outlets. They were probably drained by a sewage channel running south of the baths and along the western side. Construction technique: Opus testaceum, opus listatum Bathhouse surface: ca. 1,000 m2 GPS coordinates: 41°45’10.80”N, 12°17’18.71”E

48  c a. 1000 m2 according to Nielsen (1993b) 6. 49  The description of the baths is based on on site observations by the author. Measurements were taken on site. 50  Nielsen (1993a) 57; Pavolini (2006) 197. 51  Pensabene (2007) 537.

52  Ricciardi and Scrinari (1996a) 173. 53  This sewer can be seen on the photographs of the 1970s excavations (Archivio Fotografico Inv. 6298, B3809).

figure 63 Byzantine Baths in Ostia (Italy), detail of the mosaic of the courtyard, scale is 50 cm photo: author, March 2015

274

Gazetteer

figure 64 Baths on the via della Foce in Ostia (Italy), oval alveus of the caldarium seen from the southeast, scale is 50 cm photo: author, March 2015

Ostia (Reg. I), Baths of ‘Edificio b’ Excavation history: In the late 1970s, some test trenches were opened along the Via Severiana. Some 70 m east of the insula of the thermae of Porta Marina, a structure (edificio b) from the Hadrianic period was transformed into a bathhouse. Bibliography: Pavolini (1980). Construction: (B) Stratigraphic excavations of the previously unknown structure revealed that this transformation took place in the 4th or 5th c.54 Continued use and abandonment: Unknown. Plan: Due to the limited extent of the trenches, only a small part of the building came to light. The Hadrianic walls were pierced to construct furnaces and two round basins in opus listatum were added to the existing room. They partially encroached on the Via Severiana. The use of tubi fittili was also attested. Decoration: Unknown. Technology: Unknown.

C14. Ostia (Reg. I), Baths on the Via della Foce (I, XIX, 5)

Excavation history: The small, but well-preserved baths in the Via della Foce were unearthed in 1931 and restored

54  Pavolini (1980) 116.

C14 Ostia (Reg. I), Baths on the Via Della Foce (After Calza et al. (1953) pl. 1)

in 1951. Unfortunately, these interventions have not been published. Bibliography: Heres (1982) 434–39. Construction: (C) The construction date of the baths has been placed in the 4th or 5th c. on the basis of the type

275

Gazetteer of walls, the plan and the high level on which the complex was built.55 Furthermore, the row of shops preceding the baths has been dated to the reign of Diocletian by tile stamps, giving a terminus post quem for the construction of the baths cutting through these.56 Continued use and abandonment: Unknown. Plan: The plan is characterized by a preference for the curvilinear forms and its small pools.57 The floor of the baths lies around 1.4 m higher than the surrounding Severan walking level. This is a result of raising the banks of the Tiber, undertaken in the early 4th c.58 The hypocaust level and the praefurnia were built directly on top of this layer, which forced the actual bathing rooms to be built at an even higher level. One could enter the baths by a passage cut through pre-existing shops or directly from the Via della Foce, through a room south-east of the actual bath rooms. The frigidarium was square in plan, but had two lateral pools and a semicircular apse in its southern wall. The circular pool (diam.: 2.6 m) and the apsidal pool (3 × 3.6 m) were both 1.45 m deep. Two narrow doorways in the northern wall led into an unheated, square, transitional room (east) and a round heated room (west) with an apsidal pool (1.3 × 1 m; 1.2 m deep). From the latter, one passed into an irregularly shaped heated room and then on into the square caldarium with its three alvei. These three pools had different shapes: rectangular with rounded ends (2.25 × 0.94 m), oval (1.5 × 1.2 m; fig. 64) and apsidal (1.7 × 1.2 m). All three alvei were 1.2 m deep and the steps to access them had a limited width (between 15 and 20 cm), making them unsuited to use as a bench (fig. 64). The western tub had an elongated shape, resembling a bath tub. Some fragments of the fallen vaulting now lie west of the baths. While it is impossible to reconstruct the type of vault on the basis of these fragments, we can note that they contain pumice to lighten the weight. The location of these baths—along an important road—and the direct street access point to the public use of this building. It could easily accommodate between 20 and 30 persons. The design was adapted to the available building space, but that did not stop the architects from choosing a visually pleasing composition. The frigidarium was the most important room of the building. In the heated rooms, the focus lay on the single-person pools. Decoration: The decoration of the walls and floors has not been preserved. However, the imprint of the marble slabs in the piscinae is still visible (fig. 64). Furthermore, some small fragments of the white marble slabs are still visible in 55  Becatti (1953) 155; Heres (1982) 434–39; Pavolini (2006) 128. 56  Meiggs (1960) 552. 57  The description of the baths is based on on site observations by the author. Measurements were taken on site. 58  Gering (2010) 102.

the alvei. The walls of the first heated room are still coated with a water-resistant mortar. No imprints of marble slabs are visible, making it likely that the walls were simply coated with plaster and then painted. A statue or perhaps a labrum may have stood in the large apse in the frigidarium. Technology: The hypocausts of the three heated rooms rested upon square pillars. The walls were heated by tubuli of different dimensions (fig. 65).59 The use of different types of tubuli was perhaps due to reuse.60 The water for the baths was stored in a reservoir to the west of the baths.61 It was built within an abandoned horreum and was probably fed by a branch of an aqueduct. The distribution of the water within the baths is unclear, as the walls are not sufficiently preserved to identify the supply pipes. The water of the alvei must have been disposed of by an outlet above the furnaces that heated these pools, as no other outlets can be seen in the bottom or the walls of the pools. In this way, the waste water probably cleaned the furnaces as well. Sewage channels can still be seen along the western side of the building, probably running off to the sewer in the Via della Foce. The piscinae had outlets in their outer walls. Construction technique: Opus testaceum, opus listatum Bathhouse surface: ca. 200 m2 GPS coordinates: 41°45’13.51”N, 12°17’3.58”E

C15. Piazza Armerina (Sicily), Small Baths South of the Villa del Casale

Excavation history: In recent excavation campaigns (2004– 2012), a small bathhouse was discovered some 100 m south of the well-known Villa del Casale, on the southern end of a large piazza with a colonnaded portico that has come to light during the same campaigns (fig. 66). The excavators argue that the small baths formed part of the pars fructuaria of the villa complex and that the large Villa del Casale was thus not only a luxury retreat, but indeed ‘un’ unità residenziale, amministrativa e produttiva’.62 Bibliography: Pensabene (2010); Carloni et al. (2014); Pensabene (2014); Pensabene and Baresi (Forthcoming). Construction: (A) The stratigraphic excavations of the bathhouse made clear that it was built during the 4th c. and was

59  The internal dimensions of the separate box tiles in the caldarium and central heated room averaged 8 cm in length, 11–12 cm in width and 38 cm in height. The box tiles in the southernmost heated room averaged 5.5–6 cm in length, 9.5–10 cm in width and 34 cm in height. The tubuli that can be seen today were found in situ during the excavations, as is confirmed by the photos in the Archivio Fotografico Inv. 5668, B3179 and Inv. 5669, B3180. 60  The reuse of ceramic building material in these baths is attested by the use of bricks with stamps from the time of Hadrian, see Bloch (1953) 219. 61  Ricciardi and Scrinari (1996a) 156. 62  Pensabene (2010) 5.

276

Gazetteer

figure 65 Baths on the via della Foce, Ostia (Italy), detail of large tubulus found in the caldarium, scale is 10 cm photo: author, March 2015

contemporaneous with the construction of the luxurious villa. Continued use and abandonment: (A) Several restoration phases took place until the 6th c., when the building was reused for agricultural purposes. During the 9th c., the praefurnium and part of the caldarium was converted into a pottery kiln.63 Plan: The exact location of the entrance of the baths is unknown. However, the bather first entered a trapezoidal hall with peristylium, which may have combined different functions, including that of dressing room. To the west, one entered the frigidarium, which had a single large apsidal piscina on its western side. It was covered by a cross vault. In a later phase, a small semicircular pool was added along the northern wall of the frigidarium, creating also two square rooms to its east and west.64 In the last phase of the baths (6th c.?), a wall was erected in the middle of the frigidarium, running north to south, possibly to sustain the vault. The semicircular piscina must have fallen out of use at this time. By a doorway in the south-east corner of this cold room, the bathers entered the heated section. One room on hypocaust had a small semicircular pool, while a second served as an antechamber for the caldarium, C15 Piazza Armerina (Sicily) Small Baths south of the Villa del Casale (After Pensabene (2014) 10, fig. 2)

63  Carloni et al. (2014) 577; Pensabene (2014) 13. 64  Pensabene (2014) 17, fig. 12.

277

Gazetteer

figure 66 Archaeological site of the Villa del Casale at Piazza Armerina (Sicily), with the Villa Baths (A) and the Small South Baths (B) in Pensabene (2014) 10, fig. 1; with permission

which in its turn had a semicircular alveus on its western side. The plan of these small baths shows similarities with the late antique baths of the Middle East, especially with those of the Northern Syrian plateau (see above, chapter 4, p. 187–188). Decoration: The peristylium hall was equipped with 12 marble columns of the Corinthian order. The frigidarium had a polychrome mosaic floor with at least three phases. The second phase showed a zigzag pattern (‘rainbow-style’) and an emblema with a pair of bath sandals. The mosaic of the first heated room has mainly disappeared. Only a fragment with a polychrome design (‘a treccia’) was found. In the last phase of the baths, the floor was paved with irregular slabs of marble and terracotta tiles. The large piscina of the frigidarium was embellished with marble crustae, while the later semicircular cold pool had a mosaic with wavy patterns against its sides. The semicircular pool of the first heated room had a mosaic with wavy pattern on its bottom. Technology: Due to the later reuse of the building, the heated section was heavily destroyed. Parts of the hypocaust system, with pillars made of square bricks topped by bessales and cocciopesto, and of the wall heating (tubuli) could nevertheless be identified. The praefurnium was located south of the caldarium. The small semicircular pool of the first heated room had an outlet in its north-east corner. Construction technique: Opus caementicium Bathhouse surface: 500 m2 GPS coordinates: 37°21’49.66”N, 14°20’2.79”E

C16. Portus (Reg. I), Baths of the Lantern (‘Terme della Lanterna’)

Excavation history: The baths were discovered in the 1820s by topographers, who drew some reconstruction drawings. Recent re-examinations in the 2000s by the Soprintendenza identified 7 different phases in the building’s history. Bibliography: Not available at the time of writing. Construction: (A) The stratigraphic excavations by the Soprintendenza made clear that the bathhouse was built in an earlier Julio-Claudian complex, which was subsequently incorporated into a Trajanic building. The baths were constructed at the end of the 3rd or in the early 4th c. Continued use and abandonment: (A) Several restorations were identified until the baths fell out of use at the end of the 5th c., when the building was partially demolished to make way for a new defensive wall.

C16 Portus (Reg. I), Baths of the Lantern (‘Terme della Lanterna’), (Sketch plan after observations on site)

278

Gazetteer

figure 67 Small pool in a heated room of the Baths of the Lantern in Portus (Italy) photo: author, March 2016

Plan: The different restoration phases make it difficult to reconstruct the original layout of the baths.65 The bathers entered an oblong hall from the south, which connected to a rectangular room to the east (apodyterium?) and an apsidal room to the north. The actual bathing rooms lay west of the entrance. The cold room had two piscinae, one rectangular in the south-west corner and one apsidal on the south side. A small doorway, probably pierced through the west wall in a later phase, gave access to a round heated room, perhaps a sudatorium, which still awaits excavation. To the north of the cold room, three heated rooms can be identified. One possibly entered these rooms through a rectangular heat-lock. The easternmost heated room had a semicircular pool against its northern wall, which stood directly above a furnace (fig. 67). The middle heated room was devoid of any pools or special features, but was also heated by its own furnace. The westernmost room should be interpreted as the caldarium. It had a rectangular alveus on its northern side, just above a furnace, and an apsidal alveus on its southern side. The rectangular one may have been reduced in size at some point. Decoration: The apsidal room north of the entrance had a black and white mosaic with geometric and floral patterns, 65   The description is based on the author’s own observations on site, as well as the general excavation plan by the Soprintendenza presented during the Roman Archaeology Conference in Rome in 2016.

including 4-petal flowers and swastikas. It is unclear to which phase this mosaic belonged. The inner sides of the piscinae were possibly embellished with marble slabs, as imprints in the mortar seem to suggest. The walls of the round room, at least the lower part, also seem to have been clad with marble slabs. Technology: The service corridor ran along the northern side of the baths. Each heated room seems to have had its own furnace. There is no information on the type of hypocaust. The walls were heated by tubuli with a rectangular cross section. Construction technique: Opus testaceum Bathhouse surface: ca. 500 m2 GPS coordinates: 41°46’41.79”N, 12°15’19.90”E

C17. Portus (Reg. I), Small Baths to the north-east of area F of the Magazzini Traianei

Excavation history: Area F of the so-called ‘Magazzini Traianei’ in Portus has been investigated by the École française de Rome and the Soprintendenza in 2011. A small test pit to the north-east of area F revealed the remains of a small bathhouse, possibly intended for the people working at the harbour.66 Bibliography: Bukowiecki et al. (2012). Construction date: (D) The small-scale test pit did not reveal any datable artefacts. However, the use of the special 66  Bukowiecki et al. (2012) §8.

279

Gazetteer

C17

Portus (Reg. I), Small Baths to the north-east of area F of the Magazzini Traianei (After Bukowiecki et al. (2012) fig. 1; plan without scale)

type of late antique tubuli, observed by the author on site and also found in Ostia, in combination with the presence of a semicircular pool, point to a late antique date. Continued use and abandonment: Unknown. Plan: To this day, only a small semicircular pool and some small rectangular rooms were identified. The pool and adjacent room were heated by tubuli and thus presumably had a hypocaust.67 Technology: The heating system made use of the same type of large broad tubuli found in the baths with a late antique phase in Ostia.68 Construction technique: Opus testaceum Bathhouse surface: Unknown GPS coordinates: 41°46’35.86”N, 12°15’13.12”E

C18. Ravenna (Reg. VIII), Baths in Via d’Azeglio

Excavation history: During the construction of a garage in the old city centre of Ravenna (between Via D’Azeglio and Via Barbiani), important archaeological remains were uncovered (map 35). The excavation campaigns in 1993–1994 by the Soprintendenza yielded finds from the 2nd c. BC up until the modern period, including the remains of a small bathhouse. Bibliography: Montevecchi (2004); Montevecchi and Leoni (2004). Construction: (B) The stratigraphic excavations in the Via d’Azeglio offer a rare insight into the urban developments in Ravenna. The archaeologists uncovered a street and two houses dating from the High Empire in the lowest 67  Bukowiecki et al. (2012) fig. 5. 68  Maréchal (Forthcoming).

C18 Ravenna (Reg. VIII), Baths in Via d’Azeglio (After Montevecchi (2004) 39, fig. 37)

reachable strata of the excavations. At the end of the 3rd c., one of the houses was destroyed by fire and a small bathhouse was built upon the remains.69 Continued use and abandonment: (B) At some point during the 5th c., the largest pool of the bathhouse was filled with debris and a mosaic was laid on top of it. The Christianinspired mosaic emblema of ‘the Good Shepherd’ seduced the excavators to cautiously suggest that the remaining pool might have been used in a Christian context, possibly as a baptismal font.70 However, the theme of the shepherd does not necessarily suggest a Christian connotation, as the pigeons could also point to ‘una funzione onorifica, il che escluderebbe riferimenti ad una scena pastorale di genere’ (‘an honorific function, which would exclude references to such a [Christian] pastoral scene’).71 Plan: Only a part of the baths was uncovered by the excavations, making the plan unclear. The entrance to the building was not found, which makes it difficult to determine whether these baths were public or private. The room with the apse may have been an apodyterium.72 It was accessed from the south-east by a rectangular room with an earthen floor (service room?). To the west, the old impluvium of the house was transformed into a small pool or reservoir. It is unclear if this part of the old domus still belonged to the baths. To the north-east, two rooms, each with one pool, could be identified. The full dimensions of the largest pool

69  Montevecchi (2004) 16 and 17, fig. 2; Montevecchi and Leoni (2004) 48. 70  Montevecchi and Leoni (2004) 52. 71  Maioli (2003b) 18. 72  Montevecchi and Leoni (2004) 48.

280 are unknown, but it was least 1 m deep. The smaller pool measured 2.6 × 1.9 m. Decoration: The presumed apodyterium had an opus sectile floor. Both pools had a plain white mosaic floor and marble slabs against the sides. Technology: Tubuli were found in the access steps of the pool. However, these may only have been used as building material rather than as a conduit for hot air.73 There is no reason to assume that the pool was filled with hot water either. In a later phase, the largest pool was reduced in size and the smaller one was filled in. Construction technique: Opus testaceum, opus caementicium Bathhouse surface: Unknown (incomplete plan, not exceeding 500 m2) GPS coordinates: 44°25’7.21”N, 12°11’40.32”E

C19. Ravenna (Reg. VIII), Baths of the Banca Popolare (‘bagni del clero’)

Excavation history: The baths were discovered during the construction of the Banca Popolare in 1980 (map 35). The high ground water table made the excavations very difficult, especially to reach the lower lying levels, impeding a good reconstruction of the older phases of this part of town. Bibliography: Bermond Montanari (1984–1985); Maioli (1988); Pellacchia (2014).

Gazetteer Construction: (B) The baths were built upon a domus dating from the Hadrianic period that perished in a fire in the early 4th c.74 A terminus post quem based on the stratigraphy of the site points to an early 5th c. date. According to the 9th-c. writer Agnellus, they were built by the bishop of Ravenna in the middle of the 6th c. to replace an older bathhouse.75 Continued use and abandonment: (B) The stratigraphy related to the baths revealed that the building was used from the 5th to the 9th c., while the restorations of the walls suggest at least three different building phases perhaps dating to the 5th, early 6th and late 6th c.76 The baths were still in use at the time of Agnellus in the 9th c. Plan: The complex was built on wooden foundation posts and against a pre-existing wall, perhaps the defensive wall of the city. Two small heated rooms on a hypocaust each had a small semicircular alveus. One of these was lined with tubuli. The service corridors and praefurnia probably lay behind the alvei. To the west, a room with three apses was equipped with water supply channels and drains. In the smallest of the apses, a small semicircular pool was found. It was covered with a vault made out of tubi fittili. To the south of the two semicircular alvei, a rectangular pool coated with cocciopesto and with an internal step may have belonged to a cold room. Decoration: Two of the small semicircular alvei were still embellished with marble slabs at the time of discovery. The other pools may have had a simple mortar coating. Technology: The walls were heated by tubuli, while several water channels were also found. There is no information on the furnaces. Construction technique: Opus testaceum Bathhouse surface: Unknown (incomplete plan) GPS coordinates: 44°24’56.48”N, 12°11’54.07”E

C20. Ravenna (Reg. VIII), Baths in Via S. Alberto

Excavation history: During works on a sewage channel in 1983, the remains of two superimposed bathhouses were discovered on the Via S. Alberto. Bibliography: Maioli (1988). Construction: (C) The baths in the northern part of the trench were dated to the 2nd or 3rd c. on the basis of the type of walls. However, a lead fistula connected these baths to the late antique aqueduct, pointing to a continued use of this complex until at least the 4th c. To the south, the remains of a second bathhouse were preliminarily dated to the 6th c., as the walls were constructed with reused bricks, a typical phenomenon for Ravenna in this century.77

C19 Ravenna (Reg. VIII), Baths of the Banca Popolare ‘bagni del clero’ (After Maioli (1988c) 76, unnumbered fig.) 73  c ontra Montevecchi and Leoni (2004) 50.

74  Maioli (1988) 71. 75  Agnellus, Lib. Pont. Eccl. Rav., Victor, 28. 76   Bermond Montanari (1984–1985) 22; Maioli (1988) 77; Pellacchia (2014) 208 on the basis of building materials and literary references. 77  Maioli (1988) 90.

Gazetteer

map 35

Ravenna in the 6th c., with location of baths and the most important public areas and buildings after Manzelli (2000) fig. 152 1: Public building of unknown function 6: Church S. Croce 2: Circus? 7: Church S. Vitale 3: Church S. Apollinare Nuovo 8: Church S. Andrea Maggiore 4: Basilica Apostolorum 9: Basins in Via Rasponi 5: Church S. Agatha Maggiore

281

282

Gazetteer

Excavation history: The Baths of Diocletian have never been the subject of systematic archaeological investigations. However, detailed drawings by Renaissance artists showing substantial standing remains of the complex have contributed greatly to our knowledge of the building.79 Small test trenches at the end of the 19th c. and more recently in the 1980s and 1990s revealed parts of the heated sector, service corridors and peripheral structures.80 Bibliography: Paulin (1890); Candilio (1985); Candilio (1989); Candilio (1994).

Construction: (A) A large inscription at the entrance of the baths commemorated how Maximian, co-emperor with Diocletian, ordered the construction of the baths in AD 298 and how the works were completed in AD 306 (EI-1).81 Continued use and abandonment: (A) The baths are still mentioned in the second half of the 5th c. by Olympiodorus (see chapter 2), but probably fell out of use during the 6th c., when the aqueducts were cut during Belisarius’ siege (see chapter 2). It is not known whether the baths were reopened after the Byzantine conquest.82 The frigidarium was later transformed into the Church of Santa Maria degli Angeli, designed in the 16th c. by Michelangelo and altered in the 18th c. by Luigi Vanvitelli. Large parts of the baths were demolished in the 1860s, when the area was re-developed to create the first Termini train station. Plan: The Baths of Diocletian (map 5), built on the Viminal Hill, were the largest public baths ever to be built in the Roman empire. The surface of the complex covered no less than 140,600 m2 and served the residents of the Viminal, Quirinal and Esquiline hills. The plan fits well within the development of the large imperial type of baths in Rome. The similarities with the Baths of Trajan and those of Caracalla are undeniable.83 A wall with several exedrae and halls encompassed a spacious garden, which must have been adorned with statues and fountains (fig. 68). A semicircular cavea at the southwestern side of the enclosure wall formed a type of theatre, the shape of which can still be recognized in the modern piazza Esedra. By analogy with the Baths of the Caracalla and the Baths of Trajan, the halls and exedrae of the enclosure wall probably housed libraries, lecture halls and nymphaea.84 The bath block itself was formed by a main axis consisting of a natatio, frigidarium, tepidarium and caldarium in a north-south orientation, flanked on either side by apodyteria, a basilical hall and a suite of 4 heated rooms with an east-west orientation.85 The bathers entered the bath block by 4 doors in the northern side of the building. The two outer doors opened into a tripartite hall, followed to the north by an oval heated room. The northeastern parts are still standing to this day. By their position and their commodities—such as heating by hypocaust— these can be interpreted as dressing rooms. The two inner entrances opened into rectangular halls flanking the large natatio. The latter is thought to have contained over 5 million litres of water and was the largest open-air pool in

78  The total length of the building is estimated to be at least 73 m, according to Maioli (1988) 89. 79  The drawings of E. Dupérac (from 1575) and in the Codex Destailleur (HDz 4151 fol. 41; 16th c.), the sketch of A. Dossio (from 1569) and the etching of Piranesi (18th c.) are especially informative (see Krencker et al. (1929) 279–82; Yegül (1992) 167–68, figs. 184–187). 80  Paulin (1890) 15–16; Candilio (1985); Candilio (1989); Candilio (1994).

81  According to the passio of S. Marcellus, I, III (Act. Sanct. Ian. II, 369, 374), dating from the 6th or early 7th c., the baths were built by Maximian, in the name of Diocletian, using the forced labour of his Christian subjects, see De Spirito (1999a) 58. 82   Coates-Stephens (2003a) 173. 83  Yegül (1992) 168–69; Nielsen (1993a) 55. 84  Yegül (1992) 169. 85  Krencker et al. (1929) 279 on the basis of the renaissance drawings.

C20 Ravenna (Reg. VIII), Baths in Via S. Alberto (After Manzelli (2000) 158, fig. 104)

Continued use and abandonment: Unknown. Plan: The overall layout of the baths in unknown due to the limited size of the excavation trenches. Two rooms with apsidal pools were identified. Some 35 m to the south, a small rectangular room of 3 m long, equipped with a hypocaust, was found.78 Decoration: The walls were coated with cocciopesto. In the destruction debris of the building, marble slabs and some fragments of a mosaic with glass tesserae were found. The base of a column made out of bricks was also identified. Technology: The irregularly placed pillars of the hypocaust were made out of rectangular tiles. The southern room had a drain ejecting waste water onto the adjacent street. Construction technique: Opus testaceum Bathhouse surface: Unknown (incomplete plan) GPS coordinates: Exact coordinates unknown (remains have disappeared)

C21. Rome (Reg. I), The Baths of Diocletian

283

Gazetteer

C21 Rome (Reg. I), Baths of Diocletian (After Nielsen (1993b) 89, fig. 59)

the Roman empire. From the flanking halls or the dressing rooms, one could enter the colonnaded basilical halls (basilicae thermarum) that probably functioned as indoor exercise courts.86 The frigidarium was the centrepiece of the building. It was cross-shaped and had 4 cross-shaped piscinae in the spaces between the arms of the cross (fig. 69). On the east and west sides, a set of two tripartite halls connected the frigidarium to the basilicae thermarum. From these halls a doorway gave access to an octagonal room, followed by three heated rooms with an east-west orientation and ending in the central caldarium. The south-west octagonal room has withstood the ravages of time and was transformed into a planetarium in the 1920s (now known as the aula ottagona). It is possible that these octagonal rooms were equipped with pools.87 The first heated room had the form of an oval intersecting with a rectangle, while the second and third room were simply rectangular. The suit of rooms starting with the octagonal room and ending with the caldarium rooms were all large sun-lit lounges in which the bathers could relax, have a massage or sweat out the bad humours of the body.88 The caldarium, according to the Anonymous Destailleur, was rectangular, but divided into three parts by columns protruding out of the walls.89 The southern wall lodged three 86  Brödner (1983) 232. 87  Candilio (1994) 166. 88  The rooms have been attributed specific functions like unctoria, tepidaria, sudatoria and districtaria (e.g. Nielsen (1993b) 89, fig. 59). However, there is no evidence for these specific attributions. It is more likely that these rooms were used for similar purposes, albeit in a different architectural setting, much like the description of the heated rooms given by Lucian, Hippias 6. 89  Anonymous Destailleur, Hdz. 4151, fol. 41.

alvei, the central one having a rounded end. The lateral walls each had a single semicircular alveus. The northern wall had a central semicircular apse flanked on either side by a rectangular alveus. From the apse, two narrow doorways led into the tepidarium. The latter had a plan of a circle intersecting with a rectangle and had two lateral alvei. By two narrow doorways, the bathers could reach the frigidarium. A remarkable feature was the corridor separating the frigidarium from the heated rooms, serving probably as a service corridor and as an insulating space. Decoration: The decoration of the rooms inside the bathing block must have been spectacular. The walls have long been stripped of their marble revetments, but the imprints of the iron clamps holding the slabs are still visible. Glass tesserae found during excavations point to the use of wall mosaics on the inside of the vaults. Architrave blocks, friezes and other architectural elements were carved out of white marble. Some of these elements probably belonged to a large temple complex that once stood on the location of the baths.90 Some of the columns were monolith blocks of red granite or porphyry. The numerous niches that can be seen in the standing walls reveal the presence of statues. In the western basilica, a statue of Venus and a male torso were found. Especially the northern and southern walls enclosing the natatio must have resembled the elaborate façades as could be found in the scaenae frons of theatres or in nymphaea. In the basilicae thermarum, polychrome geometric mosaics embellished the floors, while floors in opus sectile were also found. Technology: The remains of the heating system and the water distribution are scanty. The hypocaust probably consisted of rectangular pillars made out of square tiles, 90  Candilio (1994) 175–83.

284

Gazetteer

figure 68 Reconstruction drawing in perspective view from the northeast of the Baths of Diocletian in Rome (Italy) in Paulin (1890)

while the wall heating was provided by tubuli.91 The service corridors were an underground network of broad galleries running along the southern side of the building.92 The water for the thermae was supplied by a branch of the Aqua Marcia aqueduct. A large trapezoidal cistern of 91 m in length was found on the north-eastern side of the baths in 1876.93 The impact of a building project of this size can hardly be underestimated. Small excavations underneath the aula ottagona revealed that the complex was built on private houses. Large parts of existing town fabric were destroyed to make way for this enormous ‘palace for the people’ in a region where there had not been large thermae before. Just as for the Baths of Caracalla, the efforts on a financial level, but also on the level of labour and building materials, must have been an important stimulus for the economy of Rome. It is even possible that brick production, which had dwindled in the course of the 3rd c., was revitalised to build the complex.94 Construction technique: Opus testaceum Bathhouse surface: ca. 38,750 m2 (140,600 m2 with gardens) GPS coordinates: 41°54’13.08”N, 12°29’51.67”E

C22. Rome (Reg. I), Baths of Constantine

Excavation history: The remains of the baths have disappeared almost entirely. Our knowledge of the plan is derived from 16th c. drawings, such as the one by Palladio. 91  Judging from the sketches by Anonymous Destailleur, Hdz. 4151, fol. 50. 92  A  nonymous Destailleur, Hdz. 4151, fol. 52. 93  Coarelli (2008) 323. 94  Coarelli (2008) 324.

C22 Rome (Reg. I), Baths of Constantine (After Nielsen (1993b) 91, fig. 62)

Gazetteer

figure 69 Frigidarium of the Baths of Diocletian in Rome (Italy), now turned into the Church of Santa Maria degli Angelli photo: author, March 2015

285

286 Small excavations in the 1870s and the 1980s confirmed the basic layout. Bibliography: Fiorelli (1877) 204; Fiorelli (1878) 233; Vilucchi (1985); Vilucchi (1989). Construction: (A) Construction probably began under Constantine’s rival Maxentius, as brick stamps found during small test trenches confirm.95 Continued use and abandonment: (A) An inscription recording the restoration of the baths by the praefectus urbi Petronius Perpenna Magnus Quadratianus in AD 443 (EI-46) could still be seen on site in the 15th and 16th c.96 The baths may still have been in use during the reign of Theoderic, as brick stamps from this period have been identified.97 During the Middle Ages, several churches and private houses were built inside the ruins of the baths. Most of the remains of the baths were destroyed during the Renaissance, when several palazzi were built on this location. Remains of the foundations can still be seen in the cellars of these buildings. Plan: The Baths of Constantine were built on the Quirinal hill, only 800 m south-west of the Baths of Diocletian (map 5). Investigations revealed that the baths were constructed by levelling existing insulae of domus and warehouses. The size of the building plot strongly influenced the layout of the baths. The plan stands out from the other imperial thermae by the absence of an enclosure wall. The architects were probably forced to fit the plan within the available building plot, leaving little space for spacious gardens. The design of these baths was conceived in terms of depth, rather than in breadth.98 Only a large semicircular palaestra to the north and the cavea to the south were added to the bathing block (on Palladio’s plan). The north part of the complex consisted of an open-air natatio flanked on each side by large halls. The cross-shaped frigidarium with 4 pools in the corners strongly resembled the cold section of the Baths of Diocletian.99 The basilical halls seem to have been rotated 90° and were directly linked to the frigidarium, just as in earlier thermae such as those of Nero.100 The articulation between the cold and the heated section is not well understood. It seems that an oblong hall separated the two parts. To the east and west of this hall, a round room served as a passage to connect the hall to the suit of east-west oriented heated rooms at the south end of the baths. The octagonal room at both corners is reminiscent of the aula ottagona of the Baths of Diocletian. The following two rectangular heated rooms led to the round caldarium with its three alvei. By two narrow doorways in

Gazetteer the south wall, the bathers could pass the quatrefoil tepidarium back to the cold rooms. Decoration: The excavations revealed fragments of the interior decoration. The rectangular heated rooms had mosaic floors and slabs of porphyry against the walls. When the Via Nazionale was constructed in 1870s and 1880s, several statues were found, such as the Dioscuri (now on Piazza Quirinale), a larger-than-life size statue of Constantine and Constans and two river gods (now in Campidoglio Square). In contrast to popular belief, the bronze statues of the ‘Boxer’ (‘il pugilatore in riposo’) and the ‘Hellenistic prince’ (‘il principe ellenistico’), (both in the Museo Nazionale al Palazzo Massimo) were not found in the baths, but in structures dating from the imperial period.101 Technology: The technical aspects of the thermae, such as the heating system and the water management, are unknown. The hypocausts of the heated rooms were presumably made out of the standard rectangular pillars, but there is no information on the wall heating.102 Construction technique: Opus testaceum Bathhouse surface: ca. 9,000 m2 (18 ,000 m2 including gardens) GPS coordinates: 41°53’56.50”N, 12°29’13.74”E

C23. Rome (Reg. I), Baths near the Scala Caci

Excavation history: In the 1870s, a small bathhouse was discovered in a cryptoporticus on the south-west side of the Palatine (map 5). The building was re-examined by the Soprintendenza in the 1980s. Bibliography: Cassatella and Iacopi (1990). Construction: (C) The cryptoporticus was constructed in the 1st c. AD to shelter tabernae. The baths were constructed at the turn of the 3rd and 4th c., using a different type of construction technique with smaller bricks.103 Continued use and abandonment: (B) Small test trenches made in the 1980s placed the abandonment of the building in the 5th c.

C23 Rome (Reg. I), Baths near the Scala Caci on the Palatine (After Cassatella and Iacopi (1990) 136, fig. 11) 95  Coarelli (2008) 308. 96  Krencker et al. (1929) 282. 97   C IL 15.1655; Vilucchi (1985) 357. 98  Yegül (1992) 172; Nielsen (1993a) 56. 99  Yegül (1992) 171. 100  Krencker et al. (1929) 283.

101  Heymans (2013) 238. 102  Fiorelli (1877) 204. 103  Cassatella and Iacopi (1990) 137. Several bricks of a 2nd c. module were reused in the construction of the baths.

287

Gazetteer Plan: In a first phase, the baths consisted of a simple succession of a cold room to the west and three heated rooms to the east. Only the easternmost room, the caldarium, had a rectangular alveus. In a second phase, this caldarium was abandoned and transformed into a praefurnium, while the frigidarium was turned into a tepidarium with a small basin. The new caldarium was equipped with two alvei, one semicircular in the old doorway to the west and one rectangular. Decoration: Some of the mosaic pavements were still visible when the baths were discovered at the end of the 19th c. A mosaic of Neptune was found in the original caldarium, while simple geometric patterns adorned the floors of other rooms. There is no information on the decoration of the walls. Technology: The suspended floors were built on the standard rectangular pillars. There is no information on the type of wall heating. The water supply is also unknown. From the boilers, lead fistulae in the walls supplied the alvei with water. In the second-phase caldarium, the rectangular alveus was equipped with a testudo. The water was drained through terracotta pipes, joining a pre-existing sewer to the north of the baths. Construction technique: Opus testaceum Bathhouse surface: ca. 250 m2 GPS coordinates: 41°53’21.61”N, 12°29’4.02”E

C24 Rome (Reg. I), Baths of Elagabalus (After Giorgi (2013) 81, fig. 10)

Excavation history: At the foot of the northern side of the Palatine, just south of the Temple of Venus and Rome and along the Via Sacra, a late antique bathhouse was discovered during excavations in the 1870s (map 5). Since these early digs, several small-scale excavations and architectural studies have been carried out, for example by the Escuela Española en Roma de Arqueología e Historia (1982–1992) and the Soprintendenza in collaboration with La Sapienza University of Rome (2000s). The entire complex was called the terme di Elagabalo, although the baths constituted only a small part of the structure.104 Bibliography: Mar (2005); Saguì (2009); Saguì (2013); Giorgi (2013); Saguì and Cante (2015). Construction: (A) The street onto which part of the baths protruded was dated by stratigraphic excavations to the early 5th c.105 Continued use and activity: (A) The baths went out of use in the late 6th or early 7th c., when the praefurnia were filled in with debris.106 Plan: The baths were part of a late antique building that was built within the shell of a horreum from the Severan period. The building had a representational function, with

fountains and a stibadium (a semicircular platform used as a couch for dinner parties). Rather than a domus, the building was probably an assembly hall for a corporation or association.107 The baths occupied the north-west corner of the building, and probably had a separate street entrance from the Via Sacra.108 The rooms succeeded each other in a simple linear north-south configuration. To the north, a frigidarium had two small pools, one square and one semicircular. By a doorway to the south, it gave access to a first heated room, which was divided by a partition wall in two parts, both without pools. To the south, a small caldarium had an apsidal and rectangular alveus (fig. 70). The service corridor surrounded the heated parts and was partially built over the earlier stibadium. Decoration: The floors of the bath rooms and inside of the pools, were embellished with marble slabs. Technology: The suspended floor rested on square pillars made of bessales. Three furnaces heated the caldarium, with two of these directly heating an alveus. The bipartite tepidarium was heated by two furnaces, one for each part. The walls were heated by rectangular tubuli. The bath made use of the Hadrianic sewer running underneath the building to discard the waste waters. In the rectangular pool of the caldarium, a lead fistula at the bottom gave directly into this sewer. No traces of supply channels have been found. It is likely that the baths were supplied by the Aqua Claudia that fed the Palatine.109 A large amount of tubi fittili, used

104  Saguì (2009) 235. 105  Mar (2005) 290. 106  Giorgi (2013) 56.

107  Mar (2005) 292; Saguì (2013) 149–50. 108  Giorgi (2013) 61. 109  Saguì and Cante (2015) 53–54.

C24. Rome (Reg. I), Baths of Elagabalus

288

Gazetteer

figure 70 Baths of Elagabalus in Rome (Italy), small east alveus of the caldarium seen from the west photo: author, April 2014

for constructing the vaults, was found in the destruction layers of the baths. Construction technique: Opus testaceum Bathhouse surface: ca. 250 m2 GPS coordinates: 41°53’25.52”N, 12°29’20.86”E

C25. Rome (Reg. I), Baths in the Via Ariosto

Excavation history: During the construction of the Via Ariosto in the first months of 1875, the remains of a small bathhouse were found (map 5).110 Bibliography: Lanciani (1875) 77–82; Barrano and Martines (2006); Guspini (2007). Construction: The stamps found on the tegulae of the hypocaust could be dated to the period of Diocletian and Constantine.111 Continued use and abandonment: Unknown. Plan: The remains are only known by an undated sketch of Lanciani, making it difficult to reconstruct the plan.112 The 110  There has been some confusion about the name of these baths in earlier research, as the remains were known as the ‘bagni Via Merulana’, ‘bagni presso Piazza Dante’ or ‘bagni presso la Via Galileo’ before the modern Via Ariosto was laid out, see Häuber (2014) 655–60. 111  Lanciani (1875) 81. 112  Plan reproduced in Häuber (1990) 88, fig. 71; See now Guspini (2007) 2, fig. 2.

cold room seems to have consisted of an oval room with an apsidal piscina on its west side and a rectangular piscina on its east side. To the north of this frigidarium, there is a round room of unknown function. By a narrow doorway to the south, the bather continued into an oblong room with a semicircular pool on its west side (and possibly also on its east side). This could be interpreted as the tepidarium. A doorway to the south gave access to a rectangular room, which was possibly heated. A second heated room might have been located to its east. From the latter, an oblique doorway led into the third heated room, presumably the caldarium. We can reconstruct a semicircular alveus on its south side and perhaps a smaller rectangular alveus in a recess on the east side. Along the south of this room lay probably the service corridor. Recent excavations in the Via Ariosto have revealed a domus, to which the baths may have belonged.113 Decoration: The walls had been stripped of their revetment. The geometric black and white mosaics and several marble statues known as the ‘Esquiline Group’ that were found during excavation, possibly belonged to an earlier domus.114 The statues, including fragments of three different 113  Barrano and Martines (2006) 3; Häuber (2014) 664. 114  For the ‘Esquiline group’, see Lanciani (1875) 79–80. For remarks, see Haüber (2014) 663–64.

289

Gazetteer

C26. Rome (Reg. I), Small Baths on the Forum Romanum

Excavation history: During the excavations on the Forum Romanum in 1882, Rodolfo Lanciani found a small bathhouse, along the Via Sacra, between the Temple of Romulus and the Temple of Faustina and Antonius (map 5). The remains were severely damaged in a subsequent search for burials. The results of these excavations were never published. Bibliography: Palombi (1988). Construction: (C) The construction technique points to a late antique construction date.117 Continued use and abandonment: Unknown. Plan: The plan is difficult to reconstruct from the 19th-c. plan of the remains. It seems that there was a frigidarium to the north. Its walls were constructed directly on the marble paving of an earlier building. A rectangular cold pool (2.4 × 0.8 m) was located in the south-west corner. To the south, two rooms with a suspended floor were heated by a furnace to the east. The eastern room had a small oval pool (max. 2 × 1.5 m), perhaps heated by a separate furnace. Decoration: The supposed frigidarium was paved with terracotta tiles, while the piscina had a revetment of marble slabs. Technology: Several wells were found in the vicinity of the bathhouse, but a chronological link between baths and wells could not be proven.118

C25 Rome (Reg. I), Baths in Via Ariosto (After Guspini (2007) 2, fig. 2)

Venus statues, an Asclepius, an ephebe, a head of Jupiter and a young boy wrestling with a lion do not post-date the Severan period.115 It is worth noticing that other architectural elements in precious stone were found, including column fragments in giallo antico and porta santa. Among these elements, Lanciani describes a ‘tazza circolare (diam. 2.05 m) forse ad uso di fontata’ (‘a circular bowl possibly used as a fountain’).116 The context of the find is not specified, but if it was used in the baths, it could have served as a labrum. Technology: No information is available on the water management or heating systems. Construction technique: Opus testaceum Bathhouse surface: ca. 550 m2 GPS coordinates: 41°53’25.37”N, 12°30’14.92”E 115   Coates-Stephens (2001b) 219. 116  Lanciani (1875) 80; also Haüber (2014) 447, fig. 74 A-B.

C26 Rome (Reg. I), Small Baths on the Forum Romanum (After Palombi (1988) 82, fig. 2)

117  Palombi (1988) 85. 118  Palombi (1988) 83, n. 33.

290

Gazetteer

Construction technique: Unknown Bathhouse surface: Unknown (min. 100 m2; not exceeding 500 m2) GPS coordinates: 41°53’31.09”N, 12°29’12.51”E Rome (Reg. I), Baths near Porta San Lorenzo Excavation history: During the construction of a new sewer in 1874, the remains of a bathhouse were discovered near the Porta San Lorenzo (map 5, nr. 16). Bibliography: Albertoni (1992). Construction: The construction technique of the walls in fragments of reused brick and unequally cut tuff points to a late antique construction date.119 Continued use and abandonment: Unknown. Plan: A room of 7.6 × 7.8 m had a mosaic floor supported by a hypocaust system. A small pool with three steps supports the interpretation as a bathhouse. However, on the basis of the available information, it cannot be confirmed if these baths had a private or public character. Decoration: The large room had a black and white mosaic featuring marine creatures and two Nereids. The pool was paved with white marble slabs. Technology: The hypocaust system is not described in the excavation report.

C27. San Giusto (Reg. II), Church Baths

Excavation history: Rescue excavations in the late 1990s near Borgo San Giusto (near Foggia, Apulia) revealed a small rural settlement and a late antique ecclesiastical complex, including a three-aisled church with adjacent baptistery, a smaller funerary church and a small bathhouse. Two other churches, dating from the same period, were found in a 5 km radius, which could point to an increase in population or an intense ‘Christianisation’ of the countryside.120 Bibliography: Volpe et al. (2007); Volpe et al. (2013). Construction: (A) The ceramics that were found during the excavations, point to a 5th c. date for the entire complex, with restoration phases in the middle of the 6th c. Continued use and abandonment: At the end of the 7th c., the site was reused for poorly built houses. The churches and baptistery were used as a burial ground. Plan: The articulation of the bathhouse to the churches and the baptistery reveal that they were added at a later date. The orientation of the baths differs slightly from the other structures. One entered the baths by a door in the southeast, giving access to a rectangular room with a bench against its south wall and a round basin in its western wall

119  Albertoni (1992) 342. 120  Volpe et al. (2007) 220.

C27 San Giusto (Reg. II), Church Baths (After Volpe et al. (2007) 225, fig. 3)

(diam.: 1.4 m; 0.5 m deep). The latter has been interpreted as a pediluvium.121 Additionally, it might have enabled the bathers to perform simple ablutions, such as pouring water on the head or limbs. By a doorway in the north wall, one entered the heated rooms. It is unclear whether the division of the hypocaust in three different parts matched a division into three different rooms or whether there was only one heated room. In the most eastern part of the heated section, a rectangular pool (1.88 × 1.3 m; 1 m deep) might be reconstructed. To the west of the heated room, an apsidal alveus might have existed.122 During the 6th c., several changes were made to the original design. The western part of the heated room was transformed into a praefurnium and the round basin in the cold room was abandoned to make way for a reservoir. This reservoir could be heated, generating warm water for a labrum and steam that was led directly into the heated room. Decoration: The decoration of the baths was rather modest. The floors were paved with terracotta tiles, while the walls were covered with slabs of limestone. Technology: The suspended floor of the heated room rested on square pillars made of terracotta tiles and the northern wall was heated by tubuli. The furnace for this first phase was not discovered, but it probably lay at the western end of the bathhouse. There is no information on the water supply. The water for the pools was probably transported 121  Volpe et al. (2007) 226. 122  Volpe et al. (2007) 227.

291

Gazetteer in buckets. No wells or reservoirs were found in the immediate vicinity of the baths. However, the baths undoubtedly made use of the same hydraulic system as the adjacent baptistery.123 The waste water of the round basin was discarded by a sewage channel running east underneath the cold room. The water from the west pool of the heated room was drained by an outlet in the west wall, piercing the western outer wall of the baths. Construction technique: More or less horizontal layers of field stones and reused bricks bonded by a sandy mortar. Bathhouse surface: ca. 70 m2 GPS coordinates: 41°26’3.98”N, 15°25’23.13”E

C28. Turris Libisonis (Porto Torres, Sardinia), ‘Terme Pallottino’

Excavation history: During an excavation campaign near modern Porto Torres in 1941–1942, archaeologist Massimo Pallottino found the remains of a modest bathhouse. In 2009, a new excavation campaign by the Sardinian universities of Sassari and Cagliari enlarged the site and focussed on the surroundings of the bathhouse. Bibliography: Carboni et al. (2012). Construction: (C) On the basis of construction technique (opus vittatum) of the walls and the style of the mosaics, the bath was dated to the late 3rd / early 4th c. Continued use and abandonment: (B) The baths fell out of use during the 5th c., when a fire destroyed most of the building. In a later phase, a small pottery kiln may have been built in the vicinity.

Plan: The entire extent of the bathhouse is unknown, as part of a modern road destroyed the north-west side of the building. The northernmost room was partially cut by this road, but its hypocaust system and rectangular piscina (5 × 4.9 m) could still be seen. The threshold to the next room was made with spoliated slabs of marble. This second room (9.5 × 5.2 m) also had a hypocaust. It ended in a semicircular apse on its eastern side. The third room, also heated by hypocaust, was similar in size and shape to the second room, but was also preserved on its western end, revealing a semicircular apse. To the south of this room, there may have been a fourth heated room or a praefurnium, but the modern delimitation of the archaeological site impedes further investigation. A small basin south-east of the third room may have been a pool of an unidentified room to the south.124 To the east of this basin, a water reservoir (5.7 × 4.2 m) was found. In a newly opened trench in which the excavators hoped to find structures surrounding the baths, a room on hypocaust, fragments of tegulae mammatae and the possible remains of a water channel were found. The extent of (the heated section of) the baths may have been quite impressive. Decoration: The rectangular piscina and the second heated room had a mosaic floor with geometric patterns. Technology: The service corridors of these baths have not been located and neither the furnaces that heated the warm section. The hypocaust pillars were made with square tiles and the walls were heated—or rather insulated—by tegulae mammatae. The cavities inside the walls of the third room ‘per l’allogiamento dei tubuli’125 may rather have been recesses that lodged the chimney flues. The water management is poorly understood, with the exception of the possible reservoir mentioned above. No sewers or water supply channels were discovered. The tegulae mammatae found in the newly opened trench may have belonged to an older phase or older bathhouse in the vicinity. Construction technique: Opus listatum Bathhouse surface: Unknown (at least 450 m2) GPS coordinates: 40°50’15.65”N, 8°23’47.56”E

C29. Turris Libisonis (Porto Torres, Sardinia), Central Baths

C28 Turris Libisonis (Porto Torres, Sardinia), ‘Terme Pallottino’ (After Carboni et al. (2012) 2632, fig. 3)

Excavation history: The bath was probably excavated between 1925 and 1939, although the first articles discussing its architecture and mosaics only appeared in the 1960s. The building was also described in a general study of Turris by the University of Sassari. Bibliography: Maetzke (1962); Maetzke (1964); Boninu et al. (1984) 13–18.

123  Volpe et al. (2007) 223, 227.

124  Carboni et al. (2012) 2633. 125  Carboni et al. (2012) 2632.

292

C29 Turris Libisonis (Porto Torres, Sardinia), Central Baths (After Boninu et al. (1984) tav. XIV)

Construction: (B) A small stratigraphic excavation underneath the piscinae of the frigidarium yielded a terminus post quem of the late 3rd or early 4th c. The mosaics point to a similar period.126 Continued use and abandonment: Several building phases were identified, although none could be dated. The shops to the west of the bath, following the same orientation, date from the 5th c.127 Plan: The entrance to the baths, located along an eastwest oriented street, was preceded by a portico of brickbuilt columns, also found in some baths in Cyrenaica (see below, esp. C73). From this portico, one entered the frigidarium which had a cruciform plan. In the east and west arms of the cross there was a rectangular piscinae, accessible by three steps. Reservoirs on the north-west and north-east corner of the frigidarium gave the room the outward appearance of a rectangle. The heated rooms lay to the south of the frigidarium, oriented on an east-west axis. The middle room may have been a tepidarium, while the westernmost apsidal room may have been a solarium. The easternmost room was the caldarium, which also had a cruciform plan. The northern and southern arms lodged rectangular alvei, while the eastern arm ended in an apsidal 126  Boninu et al. (1984) 13–17. 127  Boninu et al. (1984) 18.

Gazetteer alveus. The praefurnium was located east and south of the caldarium. The cryptoporticus south of the bathing block may have pertained to an older (2nd c.) bathhouse that was identified in the small-scale excavations underneath the piscinae.128 Decoration: All of the rooms of this bathhouse, excluding the service rooms, had mosaic floors. The patterns included simple geometric or floral patterns. On the crumbled fragments of the vaults once covering the piscinae and the caldarium, evidence of wall mosaics made out of glass paste was found. The walls of the piscinae and alvei may have been clad with marble slabs. Technology: The only remains of a hypocaust were recognized in the caldarium. The pillars were made of limestone monoliths, while the walls were heated by tegulae hamatae. Two furnaces can be identified, one underneath the southern rectangular alveus and one underneath the apsidal alveus. The piscinae and alvei seem to have been supplied by reservoirs adjacent to these pools, but it is unknown how these were fed. The drainage channels coming from the alvei seem to head east, where they probably connected to the street’s sewer network. Construction technique: Opus testaceum Bathhouse surface: ca. 1,600 m2 GPS coordinates: 40°50’15.65”N, 8°23’47.56”E

C30. Valentia (Reg. II, Valesio), Mansio Baths

Excavation history: The site of Valentia (Valesio, Apulia) is located some 16 km south of Brindisi, in the Salento region of Apulia. In Late Antiquity, the site is called mutatio Valentia on the Itinerarium Burdigalense, revealing that the settlement had become a relay station on the Via Traiana.129 In the 1960s, a suspensura was discovered during smallscale excavations of the Soprintendenza archeologica di Taranto. The site was further excavated by a Dutch team in the late 1980s. Bibliography: Boersma and Yntema (1987); Boersma (1995). Construction: (A) Test pits underneath the floors of the bathhouse discovered late 3rd c. ceramics.130 Continued use and abandonment: (A) The building was abandoned at the beginning of the 5th c., when layers of debris with 5th c. ceramics filled in the rooms. The site was only reused in the 13th c., when a fortified farm was built over the baths.131 The baths of Valentia must have been a welcome facility for travellers between Brundisium (Brindisi) and Lupiae (Lecce). It might have offered rooms to spend the night, stables and an eatery. These facilities

128  Boninu et al. (1984) 17. 129  Itinerarium Burdigalense, PL 8, col. 794. 130  Boersma (1995) 26. 131  Boersma (1995) 40–41.

293

Gazetteer

C30 Valentia (Valesio, Reg. II), mansio baths (After Boersma (1995) 95, fig. 125)

could have been located in the northern part of the building and on an upper storey.132 Plan: The bathhouse covered an area of about 620 m2 and was probably located along a road running east of the building. The walls were built in opus caementicium, using local calcarenite stones for the facing of the walls. The parts of the baths that came into contact with the fire and hot gasses from the furnace were made of brick. One entered the complex by a portico in the north-eastern corner. The two large rooms on the north side were not excavated, but could have been stables or rooms belonging to an inn. They were not directly connected with the rooms of the baths. From the portico, the bathers could enter a square room (4.05 × 3.65 m) with an opus signinum floor that probably served as an apodyterium, or could reach the frigidarium by passing along an oblong corridor. The frigidarium (5.85 × 8.05 m) probably served multiple functions, but did not have pools or fountains. A piscina (ca. 2 × 2.5 m; 0.8 m deep) was situated in a room east of frigidarium. By a doorway to the south, one could continue the bathing itinerary in the three east-west oriented heated rooms. The first of these (3.9 × 4.5 m) was indirectly heated, meaning it was most likely a tepidarium.133 The following room (3.8 × 4.5 m) to the west had its own furnace in the southern side. The elevated temperature in this room 132  There is evidence for the upper storeys of baths being used for houses, cf. the central baths in Herculaneum or the Baths of the Seven Sages in Ostia. See also Seneca’s complaints in Ep. 56, 1–2. 133  Boersma (1995) 80–82.

would have made it a sudatorium.134 The last room (6.5 × 5.7 m) had a furnace to the west, heating both the hypocaust and the rectangular alveus (2.1 × 2.8 m) in the west side. A second rectangular alveus (2.2 × 1.8 m) was lodged in a recess in the north wall. Next to it, a small niche, reaching down to the floor level, contained a water basin that could have served for quick ablutions.135 Directly to the north and west of the caldarium, several service rooms were floors of beaten earth were found. A latrine (3.1 × 2.1 m) with 4 seats was built in the angle formed by the piscina and the tepidarium. It was not connected to one of the bath rooms, but had a separate entrance. The frigidarium, the piscina room and the heated rooms were covered by barrel vaults made out of rubble concrete. The collapsed parts were found inside the rooms, on top of the layers of debris. The other rooms were probably covered by flat or pitched roofs, as many roof tiles were found during the excavation. The praefurnia might even have been covered with tubi fittili. Decoration: The most remarkable element in the decorative scheme of the baths was the mosaic floor of the frigidarium, made out of pebbles and opus signinum. In the centre of a white field, there was an emblema of a kantharos directed to the north. As in the apodyterium, there were traces of mortar and plaster against the walls of the frigidarium. The room with the piscina had a plain white mosaic floor and marble veneer against the walls. The piscina itself was embellished with white marble slabs. The floors of the heated rooms were covered with mosaics, but the design has not survived. The lower parts of the walls were possibly finished with a marble veneer; the upper parts were coated in mortar and finished with plaster. No traces of paintings survive, although the base layer of plaster on the walls certainly made it possible.136 Several types of decorative stone, such as breccia, granite and marble, were found during the excavations, as well as fragments of window pane. The exterior of the bathhouse was also coated with plaster. Traces of red paint have been found on the lower parts. Technology: The suspensurae of the heated rooms rested on square pillars made of terracotta tiles. As the tepidarium did not have its own furnace, its hypocaust was connected to that of the sudatorium by two bays. A chimney in the east wall of the tepidarium ensured the draught of the hot air. In the sudatorium and the caldarium, one of the pillars was replaced by a single stone monolith and in the latter a repair was made by using a tubulus filled with mortar. The walls of both rooms were heated by tubuli, which were rectangular in section. The water was obtained from 134  Boersma (1995) 89. 135  Boersma and Yntema (1987) 119, fig. 47. 136  Boersma (1995) 158.

294

Gazetteer

a well located in the service court north of the caldarium. It fed the boiler mounted on the caldarium’s furnace and probably also the piscina. The combined capacity of all the pools reached 7,470 litres (7.47 m3). No traces of reservoirs or cisterns were found. The distribution to the pools is unclear, as no pipes were observed. The water disposal of the alvei is unknown, as no sewage channels have been found. An outlet at the bottom of the pools possibly gave directly onto the bath rooms’ floor, as the floors of the caldarium and sudatorium seem to have sloped gently towards the tepidarium. The piscina was emptied by a drain giving directly into the adjacent latrine. It is interesting to note how the hypocaust and the pools were built after the shell of the building. This may have been the work of specialized teams that were hired to install the bath-specific elements.137 Construction technique: Opus caementicium Bathhouse surface: ca. 620 m2 GPS coordinates: 40°30’30.64”N, 18° 1’58.37”E

C31. Volaterrae (Reg. VII, Volterra), Vallebuona Baths

Excavation history: In the porticus behind the theatre of Volaterrae (Volterra), a bathhouse was excavated in the 1960s and 1970s (fig. 71). The results and finds of this research were never published. Bibliography: Corvo (2000). Construction: (C) The structure has been dated to the late 3rd / early 4th c. on the basis of the plan and mosaics. Continued use and abandonment: (C) One of the furnaces fell out of use at the end of the 4th c., but it is not known whether the entire bathhouse was abandoned at the same time. Plan: The building seems to have been erected in a single phase, covering an area of around 1,000 m2. The linear plan was the obvious choice in the limited building plot within the porticus. However, the orientation of the heated rooms to the north seems odd. This is partially because of the local topography of Volterra, with the afternoon sun only hitting this part of the terrain.138 Furthermore, aspects of water management (water supply and waste water disposal) that cannot be reconstructed may have necessitated this abnormal orientation. The location of the main entrance, probably to the south, directly opposite the theatre, probably played an important role.139 The bathing itinerary followed the standard pattern, starting in a large multipurpose hall with a large apse of unknown function (districtarium or unctorium?) and an adjacent latrine. The frigidarium was almost square in plan (9 × 9.2 m) and had lateral piscinae. Both had an apsidal 137  Boersma (1995) 153 without giving further evidence. 138  Corvo (2000) 106. 139  Corvo (2000) 102.

C31 Volaterrae (Volterra, Reg. VII), Vallebuona Baths (After Corvo (2000) 64, fig. 3.2)

shape, the western pool (4.1 × 3.9 m) being slightly larger than the eastern one (2.8 × 3.5 m). By two doorways to the north, the bathers passed an oblong room (2.5 × 10.2 m) with apsidal endings that only could have functioned as a heat lock between the warm and the cold rooms. The first heated room was almost square in plan (6.1 × 6.3 m) and had no pools. It was heated by a furnace to the west. The second heated room must be identified as the caldarium, as it had three heated pools, the western one being added at a later date. Each alveus was heated by its own furnace. To the south-east of the caldarium, a third heated room stood at the end of the bathing route. It was circular in shape and did not seem to have had any pools. This was probably the sudatorium. Decoration: The entrance hall had a polychrome mosaic floor with geometric patterns. The walls of the frigidarium were covered with marble slabs, as was the interior of the piscinae. The bottom of the western pool was covered with a simple mosaic (a black raster on a plain white background). In the pools, large quantities of glass tesserae in different colours were found, pointing to wall mosaics adorning the intrados of the vaults. The oblong transitional

295

Gazetteer

figure 71 Theatre of Volaterrae (Italy) with the late antique Vallebuona Baths in the peristyle square in Munzi and Terrenato (2000) 21, fig. 21; with permission

room had a mosaic with geometric patterns, including the ‘knot of Solomon’. Little remains of the decoration in the heated rooms. The two apsidal alvei still retained their interior decoration with marble slabs. Several fragments of marble crustae were found in the round room. Technology: The rooms were heated by a standard hypocaust system with pillars made out of square bricks. Only in the round room were tubuli found in situ. In the first heated room, there is no evidence of wall heating. Each heated room had its own furnace, with the caldarium

having three, one underneath each alveus. These pools may have been heated by testudines.140 The water supply of the baths was not discovered and neither were there any traces of water supply channels within the walls. Several drains were found in different rooms, but the sewage channels could not be explored. The only certain channel connects

140  Corvo (2000) 95.

296 the drain of the western piscina with the sewer running along the porticus. The alvei were probably emptied by an outlet above the furnace.141 The sewer beneath the latrine ran northward due to the slope of the terrain, and was not flushed by waste water from the pools. Construction technique: Opus listatum Bathhouse surface: ca. 1,000 m2 GPS coordinates: 43°24’13.58”N, 10°51’35.39”E

C32. Volaterrae (Reg. VII, Volterra), Guarnacciane Baths (Baths of San Felice)

Excavation history: Besides the baths at the Vallebuona site, a second bathhouse was found in Volterra. The site was first excavated in 1758 and later in the 1870s and 1880s, but the results were not published. Bibliography: Not available.

Gazetteer Construction: (C) On the basis of the plan and the mosaics, it has been dated to the end of the 3rd or the beginning of the 4th c.142 Continued use and abandonment: Unknown. Plan: The plan is known through a drawing of 1884, showing a large hall with a possible pool in its centre, a frigidarium with two apsidal pools and what seems to be a circular bathing route with several round and apsidal rooms.143 A small rectangular alveus can also be discerned. Decoration: Unknown. Technology: At least three of the rooms were heated by a hypocaust and by tubuli. Construction technique: Opus testaceum? Bathhouse surface: Unknown GPS coordinates: 43°24’7.98”N, 10°51’19.40”E

Possible Late Antique Baths Ariminum (Reg. VIII, Rimini), Palazzo Gioia The harbour town of Ariminum (Rimini, Emilia-Romagna) was an important trade centre since the Republican period. It was directly connected with Rome from 220 BC onwards by the Via Flaminia. In Late Antiquity, the city remained an important strategic centre, controlling the bottleneck between the Apennines and Adriatic coast close to the capital Ravenna.144 During excavations in the old palazzo Gioia near the modern piazza Cavour, the remains of sumptuous mosaics with marine themes and an apsidal pool were found. On the basis of the mosaics, the structures were dated to the Constantinian era.145 The large mosaic of the cold room had a scene of a marine thiasos and a scene of two winged Victories flanking an emblema of a gorgon’s head. The mosaic that surrounded the apsidal piscina consisted of intricate polychrome geometric figures. No traces of a hypocaust system have been found. Further to the south, a structure with several small apses may have been part of the caldarium.146 The interpretation as a bathhouse is based on the theme of the mosaics and the presence of a pool.147 Brixia (Reg. X), Baths in Via Gasparro da Salò In the southern suburbs of Brixia (Brescia), remains of a large structure were found. A mosaic inscription and a lead fistula date the construction to the 5th c.148 The structure has been interpreted as a bathhouse, but the exact dimensions and the

C32 Volaterrae (Volterra, Reg. VII), Guarnacciane Baths (Baths of San Felice) (After Corvo (2000) 64, fig. 3.2; plan without scale)

141  Corvo (2000) 101.

142  Corvo (2000) 107. 143  Published in Munzi (2000) 194, fig. 7.2. 144  Negrelli (2008) 7, 46–47. 145  Zuffa (1962) 123. 146  Zuffa (1962) 125. 147  The structures have also been interpreted as the remains of a rich villa with fountains, see Maioli (1987) 216–21. 148  Mirabella Roberti (1963) 271; Rossi (1990) 153.

Gazetteer layout of the rooms are unknown. A polychrome geometric mosaic with the text was found in one of the cold rooms. In the vicinity, a branch of the city’s aqueduct was found.149 No other information is available.150 Mediolanum (Reg. IX, Milan), Building in Via Brisa A building with a central round room and several apsidal halls was found in Milan along the Via Brisa. The construction type of the walls with reused bricks and thick layers of mortar points to a 4th c. date.151 The round room (diam.: 20.7 m) had a circular colonnade encompassing a basin or possibly a cold pool.152 Several of the rooms surrounding the central round room had a hypocaust system. Several marble crustae were found during the excavation. The building has been interpreted as a bathhouse or as a part of the Palatium.153 Ravenna (Reg. VIII), the basins in Via Rasponi During works on the sewage channel, at a depth of 4 m, two semicircular basins in cocciopesto were found (map 35). The type of wall construction points to a 6thc. date.154 The limited size of the trenches made it impossible to identify the structure, but the pools suggest it was part of a bathhouse or possibly a nymphaeum.155 Ravenna (Reg. VIII), the so-called ‘Baths of the Palace of Theoderic’ When a large residential structure was found on the Via di Roma in 1908, it was immediately labelled as the ‘Palace of Theoderic’. The residence of the Ostrogoth King was known from literary sources, so when an opulent palace-like building was found in the old city centre next to the 6th-c. basilica of Sant’Apollinare Nuovo, the local citizens of Ravenna did not hesitate to draw their conclusions. The archaeologists, however, connected the building to 4th-c. architecture known from Rome.156 In fact, recent research has shown that a 4th-c. phase incorporated an Imperial period residence.157 The new palace was subsequently extended in the 5th and 6th c.158 A large central peristylium with an octagonal fountain was surrounded by several rooms, including an aula regia with an opus sectile floor and an aula triabsidata with a mosaic of Bellerophon and the Chimaera. To the south of the peristylium, a late antique bathhouse was allegedly identified.159 149  Panazza (1959) 137–38. 150  Rossi (1990) 153–54 also reports late antique baths under the duomo vecchio (3rd or 4th c., without further details) and possibly in the Via Trieste, south-west of the forum. 151  Mirabella Roberti (1959) 79. 152  Mirabella Roberti (1959) 78. 153  Haug (2003) 420–21. 154  Maioli (1988) 85. 155  Haug (2003) 208. 156  Ghirardini (1917); Ghirardini (1918). 157  Manzelli (2003) 69. 158  Maioli (1988) 81 based on the mosaics. 159  Maioli (1988) 84.

297 A recent re-examination, however, attributed the rooms with a hypocaust to the earliest phases of the building (1st and 2nd c. AD).160 Instead of a private bathhouse, the rooms are now identified as heated living rooms of the Imperial period villa. Rome (Reg. I), The Temple of Minerva Medica There has been much scholarly debate about the function of the so-called ‘Temple of Minerva Medica’ (fig. 72; map 5, nr. 15), located on the Esquiline hill between the modern Via Giolliti and the railway tracks leading to Roma Termini station.161 Several excavation campaigns over several decades were unable to answer questions about the exact use.162 The hypocaust system—found in the unpublished excavations in the 1940s and 1950s—has been interpreted as part of a bath room, but is more likely to be the heating system of a triclinium.163 The structure was dated to the first 30 years of the 4th c. on the basis of tile stamps from the reigns of Maxentius and Constantine.164 The only identifiable remains consist of a large decagonal hall, crowned by a dome with its apex presumably at 32 m above the floor level. On each side of the decagon there was an apsidal niche. The walls of the rooms showed traces of tubuli. The decoration was quite luxurious: the floors were embellished with opus sectile, the walls were covered with marble slabs and the dome showed traces of glass wall mosaics. In the foundation of the walls, fragments of marble statues were reused as building material. In a later phase, some apsidal rooms were added around the decagonal hall. The absence of any pools or other water-related elements makes the interpretation as a bathhouse problematic. Rome (Reg. I), Santa Croce in Gerusalemme Small-scale rescue excavations underneath the piazza Santa Croce in Gerusalemme in 2000 revealed part of an apsidal room with a hypocaust and tubuli (map 4, nr. 17). The floor of the room was covered with a layer of cocciopesto. An adjacent service corridor with a furnace was also identified. Ceramics and a coin that were found in a sewer, as well as a stamp on one of the tiles, date the complex to the 4th c.165 The building may have been part of the imperial palace (the Sessorium) of Constantine and his mother Helena, which was located in the Santa Croce area.166

160  Manzelli (2003) 72–73; Novara (2003) 59. 161  The building was called the Temple of Minerva, because a statue of Minerva was erroneously attributed to this site in the 16th c. 162  Biasci (2003). 163  Barbera et al. (2014) 264 correcting Barbera et al. (2007) 17. 164  Barbera et al. (2007) 2. 165  Di Meo and Zaccagnini (2000) 247. 166  Di Meo and Zaccagnini (2000) 246.

298

Gazetteer

figure 72 Plan of the so-called ‘Temple of Minerva Medica’ in Rome (Italy), perhaps a bathhouse or, more likely, a heated reception hall after Barbera et al. (2007) 7, fig. 6b

Spoletium (Reg. VI, Spoleto), Palazzo Mauri The consolidation works of the Palazzo Mauri in Spoleto revealed a polychrome mosaic depicting vines and animals and partially made with glass tesserae.167 Stylistically, the mosaic belongs to the late 6th c.168 In the centre of the pavement, a marble drain points to a water related function of the room. Another trench in the vicinity of the mosaic yielded a water basin coated with cocciopesto. While no evidence for a heating system was found due to the limited extent of the excavations, it is possible that the mosaic and the basin were part of a bath complex. However, the theme of the mosaic pushed some scholars to interpret the room as a baptistery.169 Ticinum (Reg. XI), Building under San Tomasso Ticinum (Pavia, Lombardy) is one of the Roman cities that lived its heyday in Late Antiquity. Augustus patronized building activities in this small municipium, but it was its large neighbour Milan that became the provincial centre. The city was sacked by Attila the Hun in AD 452, but revived under the Ostrogoth King Theoderic. According to the Anonymous Valesianus, Theoderic was responsible for the erection of thermae (see chapter 2, p. 61). In AD 540, Ticinum took over the role of Gothic capital from Ravenna. Under Lombard rule, the capital’s name was changed to Papia. In 1895, the remains of 167  Salvatore (2004) 60. 168  Costamagna (2009) 21. 169  Costamagna (2009) 20.

a late antique ‘bathhouse’ were discovered near the Church of San Tommaso (fig. 73), in the centre of the ancient city.170 The type of plan, construction technique and lead water conduits point to a late 3rd or early 4th c. date.171 Only the cold section of the supposed bath building survived. A large hall, at least 5 m in length, ended in a triapsidal exedra. A semicircular apse (diam.: 0.8 m) was present in the preserved wall. Running along the outside was a type of vaulted service corridor.172 The interpretation as a bathhouse rests mainly on the fact that a lead water conduit ending in a cistern was discovered near the remains a year prior to the excavations. The alleged bathhouse was also served by such lead water conduits. In medieval times, the city’s main fountain was, not coincidentally, located beside the San Tommaso church. A sewer also ran under the room.173 Verona (Reg. X), Piazza Vescovado Verona was an important city in Late Antiquity. During a small-scale excavation of the piazza Vescovado in 1992, just north-east of the duomo, the remains of what appeared to be a late antique bathhouse were found. The stratigraphic excavations revealed that the building was built at the end of the 4th c., on top of an ancient domus. The building was in use 170  Bullough (1966) 83. 171  Arslan and Bossi (1968) 306; Hudson (1990) 176. 172  Arslan and Bossi (1968) 307, fig. 2. 173  Bullough (1966) 83, 85.

Gazetteer

figure 73 Plan of the alleged Baths of San Tommaso in Ticinum (Italy) after Bullough (1966) 86, fig. 3

throughout the Gothic period and perhaps on into Lombard rule.174 The limited excavation only enabled the archaeologists to uncover part of a large room (ca. 90 m2), traversed by a water channel. The walls were coated with hydraulic mortar. In a late phase of use, a hearth was dug into the floor, perhaps to heat the water for pools.175 There is no information available on the rest of the building or on the surrounding town fabric.

Villa Baths with a Possible Public Use Galeata (Reg. VI), Baths of the Palace of Theoderic Excavation history: In 1998, the University of Bologna extended the German excavations of 1942 of the so-called ‘Palace of Theoderic’ near modern Galeata (fig. 74). The new excavations (1998–present) brought to light a bathhouse, presumably private, near the large villa uncovered by the Deutsches Archäologisches Institut (DAI). Bibliography: Villicich (2002); Villicich (2012); Villicich (2014).

174  Cavalieri Manasse and Bruno (2003) 61; Anonymous Valesianus (6th c.) recorded the restoration of baths in Verona by the Gothic king Theoderic (see chapter 2, p. 61). 175  Cavalieri Manasse and Bruno (2003) 61.

299 Construction: (A) The small finds and the ceramics found inside the baths and in the foundation layers, dated this bathhouse to the early 6th c.176 Continued use and abandonment: (B) The baths probably fell out of use in the course of the 7th c., when layers of debris filled the basin of the courtyard.177 Plan: (fig. 75) The bathhouse was located 50 m south-east of the pavilion and was connected to it by an L-shaped covered ambulacrum. The fact that the bathhouse was not attached to the main residence could favour some accessibility to the larger public.178 The ambulacrum might have been the privilege of the dominus and his entourage, and a side entrance could have existed.179 The ambulacrum opened onto an enclosed courtyard with a rectangular, but shallow pool in its centre. This pool was probably a decorative element, rather than an actual piscina or natatio.180 Two diaetae (summer dwellings) were found to the north and to the south of the courtyard, presumably to be used in the hot summer months. The actual bath rooms, which extended to the north-east, could hence be interpreted as a ‘winter bath’.181 The bather entered a square room with a pisicina on its north and south side, labelled apodyterium by the excavators, but having the characteristics of a frigidarium. The room, or at least the piscinae, was covered by a vault made of tubi fittili. From the apodyterium, one climbed some steps to arrive in an axis of three rooms, interpreted as frigidarium, tepidarium and laconicum. The cold room had a rectangular shape with an apsidal ending, in which a semicircular pool must have been located.182 By a broad flight of steps, one continued north into the tepidarium. The heated section of the baths did not withstand the ravages of time. A small heat lock connected the tepidarium to the octagonal caldarium, which had three rectangular alvei on its sides. Decoration: The destruction layers yielded several marble crustae and tesserae, which must have adorned floors and walls. The basin of the courtyard was paved with slabs of 176  Villicich (2014) 242. 177  Villicich (2014) 249. 178  Similarly, the villa ‘ad duas lauros’ at Centocelle near Rome, on the Via Labicana, had a separate large bathhouse, linked by an ambulacrum, in addition to the private baths attached to the complex, see Volpe (2007b). This large bath may also have a public use. 179  See the reconstruction in Villicich (2014) 248, fig. 7. 180  Villicich (2014) 248. 181  Villicich (2014) 248; however, the terms ‘summer’ and ‘winter’ baths do not seem to be appropriate in this case, as ‘summer’ baths still had a heated section. The courtyard is reminiscent of the late antique type of baths in the Middle East (see chapter 4, p. 187–188). 182  Villicich (2014) 249. This room may in fact have been a tepidarium, indirectly heated by the following heated rooms. The ‘apodyterium’ with its lateral piscinae would then assume the function of frigidarium.

300

Gazetteer

figure 74 Archaeological site of the Villa of Theodoric in Galeata (Italy) after Villicich (2012) 10, fig. 14; with permission

figure 75 Plan of the Baths of Theodoric in Galeata (Italy) after Villicich (2012) 10, fig. 15

sandstone. Several fragments of window pane were also found. Technology: The pillars of the hypocausts of the heated rooms were made with square tiles. There is no mention of tubuli or any other form of wall heating. The lead fistulae that drained the cold pools of the ‘apodyterium’ were still in place, giving into a sewer underneath the floor towards the south-west. Palazzolo (Reg. VIII), ‘Palace Baths’ Excavation history: Some 8 km north of Ravenna, the remains of a private estate were found under the Church of Santa Maria infra balneum. Bibliography: Bermond Montanari (1972); Maioli (1988). Construction: (D) According to Agnellus, a small hunting palace was built in Palazzolo in AD 493.183 The construction technique, as well as the plan, point to a 6th c. date. 183  Agnellus, Lib. Pont. Eccl. Rav. as quoted in Holder-Egger (1878) 303.

Continued use and abandonment: After the 6th c., the building may have had a different use.184 Agnellus again claims that the bath was destroyed in the 9th c. to salvage building materials. Plan: The palace was rectangular in plan with the bathhouse in its south-west corner. An octagonal room with a hypocaust may be interpreted as a caldarium, while a smaller room with hypocaust and wall heating could have been the sudatorium.185 Decoration: The walls were covered with marble slabs. Technology: The hypocaust made use of standard square pillars and the walls were heated by tubuli. Water for the presumed pools was probably supplied by the nearby river rather than by aqueduct.186

184  Bermond Montanari (1972) 215. 185  Maioli (1988) 93; Maioli and Montevecchi (2003) 84 favour an interpretation as tepidarium and caldarium respectively. 186  Maioli (1988) 92–93.

Gazetteer Piazza Armerina (Sicily), Large Baths of the Villa del Casale Excavation history: The first excavations of the villa started in the 1930s under the supervision of Paolo Orsi. However, it was only in the 1950s that the first results were published, when Gino Gentilli restarted the works. The splendid mosaics got most of the attention, while small finds and stratigraphy were neglected. Several campaigns have been carried out, most recently in the late 2000s focussing on the immediate surroundings of the villa. Bibliography: Gentili (1956); Ampolo et al. (1971); Wilson (1983); Pensabene (2010). Construction: (A) Test pits dug during excavations by Andrea Carandini’s team in the 1970s confirmed the early 4th c. construction date that had been proposed on the basis of the mosaics.187 Continued use and abandonment: (B) There are indications that the baths were still (partially) in use during the 6th c., with some minor restoration carried out. The villa may have become the property of the Church at this point.188 The frigidarium might even have been reused for Christian cult, judging from the high quantity of oil lamps with Christograms found here.189 Plan: (fig. 76) The bathhouse was located in the north-west part of the villa as an integral part of the original design. Even if the baths were a part of this luxurious private estate (fig. 66), they were probably open to ‘outsiders’ as well, seeing that they had a separate access connecting to the exterior of the villa.190 Whether the baths were open to all the inhabitants of the surrounding countryside, or access was restricted to the clientes of the dominus cannot be verified. It connected to the central peristylium of the house and to the monumental entrance porch of the villa (fig. 66). The entire structure was erected in opus caementicium, while the vaults were constructed with tubi fittili. The large oblong room with apsidal endings (length: 22 m) probably functioned as an entrance vestibule with a variety of possible functions (dressing room, exercise court, lounge area, etc.).191 From this room, the bathers continued north into the octagonal frigidarium. Each of the sides of the octagon had an apsidal niche, except for the northern and southern side, which opened onto piscinae. The northern pool had an elongated shape with apsidal ending, while the southern one had a trefoil shape. Through a square room to the west, acting as a heat lock, one entered the heated rooms. The first room also had an oblong shape with apsidal endings (length: 18 m). Two doorways gave access to a suite of three rectangular rooms lying on the same north-south 187  Ampolo et al. (1971) 169–175, 179. 188  Pensabene (2010) 13. 189  Pensabene (2014) 12. 190  Wilson (1983) 20. 191  Wilson (1983) 20.

301 axis. The northernmost and southernmost room each had an alveus, while the middle room may have had a heated labrum. Decoration: The villa is famous for its well-preserved mosaic floors and the bathhouse is no exception. The long multifunctional entrance hall was adorned with granite columns and a mosaic depicting a scene in the Circus Maximus in Rome.192 The rendering of the Circus supported the idea that the villa was owned by a wealthy Roman senator or even an emperor.193 In the mosaic of the frigidarium, a marine scene involved Cupids fishing on the back of dolphins and other sea-creatures such as Nereids. In one of the niches, a mosaic depicts a boy exiting the baths. The heat lock was adorned with a mosaic of a man being anointed with oil by a servant, while a third person carried an oil flask and strigiles. The room has therefore been interpreted as an unctorium.194 The floors of the oblong room and the three rectangular rooms have not been preserved. The walls of the oblong room were plastered and painted to look like marble. Wall decoration in the heated rooms has not survived. Technology: There were 5 furnaces for the heated section: one for each of the rectangular rooms and two for the oblong room. The three furnaces of the rectangular rooms may have supported boilers. The pillars supporting the suspensura were made of square tiles. The praefurnia were also covered by tubi fittili vaulting. The water for the pools was stored in a large reservoir located some 75 m north of the villa, which was fed by an aqueduct. A channel led the water to the large apsidal piscina. The pipes for the distribution to the other pools and boilers have been found inside the walls, but have not been included in the published plans.

Continued Use and Abandonment of Imperial Period Baths during Late Antiquity In several other towns of Italy, bathhouses built during the High Empire were still in use during Late Antiquity. Unfortunately, the early 20th-c. excavations paid little attention to these late antique phases. As can be deduced from the epigraphic evidence (see chapter 2), several bathhouses were restored during the 4th c. The list of baths mentioned below, although not exhaustive, demonstrates how different types of baths in different types of settlements were kept in good use.195 192  Gentili (1956) 19; Pensabene (2009) 87–88. 193  For an overview of the emperor theories and the objections, see Wilson (1983) 86–92. 194  Gentili (1956) 21. 195  The corpus presented here is not the result of systematic research into every Roman bathhouse ever found in the

302

Gazetteer

figure 76 Plan of the Villa Baths in Piazza Armerina (Sicily) after Nielsen (1993b) 104, fig. 87

Acireale (Sicily) The baths of Acireale were probably thermal baths fed by the sulphurous spring of San Severa. Only two large rectangular rooms have survived, both heated by a hypocaust and with flues in the walls.196 The entire complex may have been vast, with a large pool found south of the heated rooms. The baths probably had a long lifespan, starting in the Imperial Period and possibly continuing into Late Antiquity or even Early Medieval times. Acireale is still today popular for its thermal baths. Agrigentum (Sicily, Agrigento) The remains of a Roman bathhouse, possibly private, were discovered in 2015 in insula IV.197 Only the caldarium and a praefurnium were identified so far, but the excavations are ongoing. In a city that is famous for its Hellenistic buildings, these are the first baths to be found. A preliminary study of the ceramics dates the abandonment of the building in the 5th c.

Italian Peninsula. The list was compiled on the basis of earlier studies—esp. Nielsen (1993b); Anonymous (1990); Haug (2003) 204–15; Volpe et al. (2007), Cerca (2008)—and information gathered during the research into late antique baths. In the following overview, only the primary excavation reports or descriptions have been consulted. For exhaustive bibliographical information, see the authors mentioned above. 196  Cosentini (1966) 9–12. 197  The first results of this excavation were unavailable at the time of writing. The find was reported in several local newspapers.

Albintimilium (Reg. IX, Ventimiglia) The city of Albintilium fell under Roman control in the period between 180 and 90 BC and survives today as the modern Ventimiglia. During the Lombard-Byzantine period, occupation shifted to the upper parts of the city. A High Imperial bathhouse remained in use until possibly the 5th c., when burials took place within its walls.198 Around the same time, the theatre was reused for housing. Augusta Praetoria (Reg. XI, Aosta) The baths under the Scuola Normale in Aosta were probably built at the end of the 1st c. AD. During the late 3rd or early 4th c., a heated apsidal room was added to the original plan.199 Restorations to the heated rooms were carried out, with a final phase with reused materials possibly commissioned after the baths had fallen out of use.200 A possible bathhouse southwest of the forum (insula 34) may still have been in use in the 4th c.201 Caesena (Reg. VIII, Cesena) The bathhouse in the Via Tiberti in modern Cesena was built during the 3rd c. on top of a smaller bathhouse.202 An inscription dating from the reign of Probus or Carus recalls the restoration of the balneum Aurelianum.203 The baths may still have been in use in the 4th and even during the 5th c., when the 198  Pallarès (1998) 29. 199  Mollo Mezzena (1992) 274 and 293, fig. 6. 200  Haug (2003) 372. 201  Mollo Mezena (1992) 274. 202  Maioli (1998) 76. 203  Maioli (2000) 498.

Gazetteer building was embellished with new mosaic floors. The definite abandonment of the building occurred before the 6th c., when an entirely new structure was erected on its location.204 The mosaics of the 5th-c. phase included geometric shapes and ‘knots of Solomon.’ Canusium (Reg. II, Canosa) In Canusium (Canosa di Puglia), two High Imperial bathhouses have been found. The so-called ‘terme Ferrara’ probably date to the Severan period, although restorations in opus listatum point to a late phase of use, perhaps stretching to the 5th and 6th c. A small semicircular pool, constructed against the northern wall of the supposed frigidarium, may have belonged to this last phase.205 A second bathhouse, called the ‘terme Lomuscio’, also dates to the 2nd c. AD. The baths were altered several times and seem to have been in use in the 4th c., when a new heated room, constructed in opus listatum and with its own praefurnium, was added.206 It is not surprising that public facilities such as baths were kept in good use until at least the 4th c., as Canusium was still an important trade centre on the Via Traiana during Late Antiquity. The epigraphic output during the 4th and even 5th c. is evidence of a continued urban dynamism.207 Carsulae (Reg. VI, San Gemini) The baths of Carsulae (near San Gemini) have come under renewed investigation between 2006 and 2013. The building was constructed in the Late Republican or Early Imperial period, but only fell out of use in the middle of the 4th c. The abandonment of the baths, for unknown reasons, was dated by ceramics and coins found in the destruction layers. An apsidal alveus in one of the heated rooms belongs to a later phase of the baths.208 Several furnaces were also blocked during the lifespan of the baths, reversing the bathing itinerary and transforming heated rooms into cold rooms.209 Catania (Sicily) Two public baths have been found in Catania, both dating from the High Empire. The ‘Terme dell’Indirizzo’ probably date from the late 2nd or 3rd c., but the lack of stratigraphic excavation data impedes further analysis of its lifespan.210 North of the theatre, the so-called ‘Terme della Rotonda’ probably date from the 1st or 2nd c. Test pits and restorations after the 204  Maioli (1998) 76–79. 205  Cassano and Bianchini (1992) 730–33. 206  Bertocchi and Bianchini (1992) 736. 207  Grelle (1992) 821–22. 208  Whitehead (2010) 7. 209  The latest results of the excavations (2011–2013) had not been published at the time of writing, although a summary was available on the Fasti website (AIAC-1027). 210  Branciforti (2013).

303 damage caused by Second World War bombing revealed that the bath was transformed into a church during the 6th c.211 Comum (Reg. XI, Como) In Comum (Como), the bathhouse in viale Lecco was probably built in the second half of the 3rd c.212 It is likely that the baths were still in use in the (early) 4th c. An octagonal room with rectangular niches on its sides was connected by an oblong room with opposing semicircular niches to a hexagonal room with at least one semicircular pool. Cures Sabini (Reg. IV, Passo Corese) The bathhouse at Cures Sabini (Passo Corese) was constructed in the Imperial period.213 An inscription (EI-29) mentioning the construction or repair of the baptisterium and another room found reused in a modern mansion in the vicinity may refer to this building.214 A recent re-examination of the site, however, has not revealed any datable late antique phases yet, but a possible lime kiln was built after the baths fell out of use.215 Curinga (Reg. III) During the 1960s, a bathhouse near the modern town of Curinga was partially destroyed by the construction of a new road (fig. 77). Fortunately, an architectural study of the remains was undertaken before the partial demolition. The building was again excavated in the 2000s by the Soprintendenza, but no excavation report was published.216 The construction date of the baths is uncertain. On the basis of the technique of the walls (opus testaceum and opus listatum) and on the basis of the use of tubi fittili for the vaults, the baths have been cautiously dated to the turn of the 3rd and 4th c.217 A coin from the reign of Diocletian might provide a terminus post quem, but the archaeological context in which it was found is not clear. A re-examination of the walls in the 2005–2006 campaign placed the original construction in the 1st c. AD.218 The stratigraphy of the excavations in 2005–2006 makes clear that the baths had two separate restoration phases in the 4th c. These did not alter the layout of the baths, but only reinforced certain walls and possibly introduced tubi in the vault construction. The complex fell out of use at the end of the 4th c. (debris and ceramics in the supply channels), and was 211  Branciforti and Guastella (2008). 212  De Angelis d’Ossat (1990) 163. 213  Cavalieri and Lenzi (2015) 243. 214  Fiorelli (1878) 29. 215  Cavalieri and Lenzi (2015) 242–43. 216   The unpublished excavation report (Anonymous (2013)), plans and photographs are available on the website of the Beni Culturali di Calabria (http://www.beniculturalicalabria. it/bandi.php?id=38&action=VIEW). 217  Arslan (1966) 47. 218  Anonymous (2013) 5.

304

Gazetteer

figure 77 Plan of the Baths of Curinga (Italy) after Arslan (1966) fig. 3

subsequently spoliated during the 5th c.219 All marble slabs seem to have been removed. During the 5th to 7th c., however, some renewed activity in the frigidarium, including the construction of two small basins, point to a renewed use of the building.220 Forum Traiani (Sardinia, Fordongianus) The bathhouse in Forum Traiani was actually a thermal bath near the Aquae Hypsitanae, probably constructed in the late 1st or early 2nd c. AD 221 In the late 3rd or early 4th c., the thermal baths were restored, adding a small semicircular pool, and enlarged with a ‘normal’ bathing wing making use of ‘normal’ water, at least for the cold pools.222 The most characteristic elements of these new rooms, in contrast with the older complex, is the use of opus listatum, a small semicircular pool in addition to a larger rectangular pool in the frigidarium and the use of opus sectile. The baths may have been abandoned at a late date, in the 6th or 7th c.223

219  Anonymous (2013) 18–20. 220  Anonymous (2013) 21. 221   I L Sard. 194; Serra and Bacco (1998) 1216–28. 222  Serra and Bacco (1998) 1226, 1229–38 (dating on the basis of wall construction and stratigraphy). 223  Serra and Bacco (1998) 1233.

Grumentum (Reg. III, Grumento Nova) The baths of Grumentum (Grumento Nova) were located along a broad road. A room with cylindrical hypocaust pillars and tubuli should probably be interpreted as the caldarium.224 The analysis of the materials found during the excavation and of the wall paintings point to a Republican or Augustan construction date. The building was still in use in the late 4th c.225 Herdonia (Reg. II, Ordona) The ancient site of Herdonia (Ordona) was located along the Via Traiana. By the 1st c. AD, Herdonia was equipped with a forum, amphitheatre and balneum.226 The construction of the Via Traiana in the first half of the 2nd c. meant an important boost for the city’s development. Its centre was embellished with a new forum, macellum and temple.227 In the second half of the 2nd c., a modest bathhouse (ca. 1,500 to 2,000 m2) with palaestra was built along the main traffic artery (fig. 78). Around the middle of the 4th c., the baths were restored along with the Via Traiana, possibly after an earthquake.228 The interventions included the blocking off of doors, repairs to the mosaic floors with larger tesserae, and restoration of some 224  Nava (2001) 977. 225  Nava (2001) 977; Nava (2002) 763, without further evidence. 226   C IL 12.3188. 227  Volpe (2000b) 512. 228  Flavia et al. (2000) 131–43.

305

Gazetteer

figure 78 Plan of the baths of Herdonia (Italy) in Rocco and Turchiano (2000) 203, fig. 227; with permission

hypocaust pillars with square and circular tiles. The reduction of the size of the tepidarium probably aimed to reduce the heat loss from the heated to the cold section.229 A new praefurnium was also created to heat both tepidarium and caldarium. The rectangular apse on the east side of the presumed caldarium, opposite the semicircular alveus, may have been an additional alveus. During the first half of the 5th c., some small rooms of unknown function were added west of the heated section. The baths fell out of use in the second half of the 5th c., when layers of alluvium covered the floors and the wall paintings crumbled on top of this. The building was then reused for other purposes, possibly housing.230 Iulium Carnicum (Reg. X, Zuglio) The Alpine town of Iulium Carnicum (Zuglio) had a bathhouse near the forum, dating from the Imperial period, which was possibly still in use in the late 4th c.231 The Imperial period bath was thoroughly reorganised in the 2nd c., but no traces of this late antique reorganisation can be discerned. The baths were probably abandoned after a fire.232 Mesange (Reg. II) In Mesange, a small town some 13 km south-west of Brindisi, a bathhouse was excavated in 1986–1987 (Malvindi 229  Flavia et al. (2000) 161. 230  Flavia et al. (2000) 150–59. 231  Moro (1956) 74; Oriolo (2001) 284, without further evidence. 232  Moro (1956) 72–74.

Campofreddo site). It was probably constructed in the 1st c. AD, but possibly remained in use until the 4th.233 The plain white mosaics were restored with slabs of marble, while the hypocaust pillars were renovated by using stone monoliths.234 Photographs taken during the excavation show a small semicircular pool, but there is no mention about it being a later addition.235 Misterbianco (Sicily) In the small town of Misterbianco, some 5 km west of Catania, a bathhouse stands in apparent isolation from other Roman structures (fig. 79). It has been dated to the Severan era, only because this was a period of great expansion in the region.236 However, the type of wall construction—opus caementicium with very little brick—could point to a later date (turn of the 3rd and 4th c.?).237 According to Tomasello, no complete bricks could be identified on site.238 This probably points to the reuse of brick fragments, a widespread practice in Late Antiquity. In the second phase of the building, the use of bricks disappeared entirely. The layout of these baths followed a double row type, sometimes called a ‘block type’. Two independent entrances, which were connected, gave access to a heated room. Only the 233  Lipolis and Lipolis (1997) 333; Cocchiaro (1998). 234  Cocchiaro (1988) 167. 235  Cocchiaro (1988) plates. 236  Tomasello (1984) 202–203. 237  Wilson (1983) 70. 238  Tomasello (1984) 200.

306

Gazetteer

figure 79 Plan of the Baths of Misterbianco (Sicily) after Tomasello (1984) 199, fig. 11

eastern one gave access to the caldarium with its two alvei. In a second phase, dated without convincing evidence to the 4th or 5th c., a large frigidarium with two adjacent apsidal piscinae was constructed.239 The entrance door to the north-eastern room was blocked up, transforming the itinerary into an axial row type with a reversing itinerary. The date of this building, as well as its interpretation as a private bath of a yet undiscovered villa, is unconvincing.240 However, parallels with the plans of some small baths in Cyrenaica (see below, C73, C74) could support a late antique date.241 Nevertheless, the compact plan with two parallel rows of rooms is not unique to Late Antiquity. Muralto (Reg. XI) In the Alpine vicus of Muralto, rescue excavations in the 1980s discovered several workshops and a small bathhouse. The settlement, located on an important traffic artery and possibly specialized in glass production, had a continued occupation throughout Late Antiquity and into the Middle Ages.242 The baths were dated to the second half of the 3rd c., but were probably still functioning at the beginning of the 4th.243

239  Tomasello (1984) 202–203. 240  Tomasello (1984) 197. 241  The similarities have not gone unnoticed by Tomasello (1984) 198, n. 47. 242  Biaggio Simona and Janke (2008). 243  Donati (1990) 243.

Neapolis (Reg. I, Naples) The construction of the new metro line in Naples uncovered the remains of a Roman bathhouse (near Stazione Toledo), dated by stratigraphic excavations to the late 1st or early 2nd c. The baths were restored in the 4th c., before falling out of use during the 5th, when the ruins were reused as a burial ground. The reuse of several Roman walls in 16th and 17th c. buildings makes a reconstruction of the plan and of the restoration phases difficult.244 Nora (Sardinia, Pula) Recent excavations in the large baths of Nora (Pula) have refined the chronology proposed by the excavations of the 1950s. The bathhouse was built on top of 2nd c. houses.245 The insula of the baths was occupied until the 7th c., although it is not specified whether the baths were still in use or had a different function.246 A second small bathhouse in Nora was probably built in the Imperial period, but may have had a 4th c. phase judging from the mosaics. The absence of stratigraphic excavations impedes further refinement of the chronology. Opitergium (Reg. X, Oderzo) The baths in modern Oderzo, 50 m east of the forum, should be attributed to the High Empire, although archaeological evidence to pinpoint the century is lacking.247 Ceramics found 244  De Caro (2001) 884–85. 245  Frontori (2014) 6–7. 246  Frontori (2014) 4. 247  Busana (1995) 60 on the basis of the plan.

Gazetteer during the excavations date between the 1st and 5th c. AD, giving the hypothetical lifespan of the building.248 The limited size of the excavation trench has only revealed part of the heated section—at least 4 rooms on a hypocaust, making these baths quite large—thus making it difficult to discern possible late antique restorations or alterations. Pollena Trocchia (Reg. I) The bathhouse in Pollena Trocchia, on the north slopes of Mount Vesuvius, was built upon layers of ash from the AD 79 eruption. The bathhouse fell out of use shortly before another eruption (AD 472) covered the baths with another layer of ash.249 It was used for burials by the mid-5th c.250 The fact that no traces of a settlement were found in the immediate vicinity of the baths encouraged excavators to believe that the baths were part of a private estate. Any such villa has yet to be discovered. San Gaetano di Vada (Reg. VII) The baths in San Gaetano di Vada were located behind a row of shops that probably belonged to the harbour (fig. 80). They were likely built in the late 1st or early 2nd c. AD. Excavations by the University of Pisa in the 1980s revealed that the building was used until the 4th or even 5th c.251 It is difficult to link any reorganisations of the plan, such as the addition of a new caldarium with two small alvei, to this late phase of use. Sentinum (Reg. VI, Sassoferrato) The ‘Terme di Santa Lucia’ were located just outside the southern city gate of Sentinum and probably built in the Neronian or Flavian era. Important restoration works and a reorganisation of the bathhouse took place during the 2nd c. The building was gradually abandoned during the 4th c. and heavily spoliated at the end of the 4th and beginning of the 5th c.252 Suasa (Reg. VI, Castelleone di Suasa) The ‘domus dei Coiedii’ was located opposite the forum of the municipium, just west of the theatre and amphitheatre. The house itself was built in the Late Republic and enlarged in the 2nd c.253 During this phase, a bathhouse (fig. 81) was added to the hortus, which had become a peristyle courtyard at this point.254 Even if the baths were located within the walls of the garden, the remote location of the baths in position to the house and the presence of street entrance

248  Busana (1995) 62. 249  De Simone et al. (2011) 62–63. 250  De Simone et al. (2012) 198–99. 251  Menchelli and Vaggioli (1987) 127. 252  De Marinis et al. (2008) 210. 253  Dall’Aglio et al. (2007) 191–93. 254  Giorgi (2012) 347.

307 close to the baths may point to a semi-public use. The baths were restored in the 3rd and 4th c.255 Sybaris (also called Copia Thurii, Reg. III, Sibari) In the south-west part of ancient Sybaris (Sibari), a bathhouse dating to the 1st c. AD was found on a southern side road of the plateia maritima (Parco del Cavallo site).256 The baths may have been in use until the 5th c., before being dismantled for building material in the 6th.257 Syracusae (Siracuse, Sicily) The ‘Baths of Dafne’ have been the subject of debate. Traditionally identified as the bathhouse where Constans II (AD 641–668) was murdered, there is no stratigraphic evidence confirming a construction date.258 The plan shows some small semicircular pools added to the original layout. Teanum Sidicinum (Reg. I, Teano) During the construction works for a high-speed train, the remains of a large bathhouse were found in the ancient site of Teanum Sidicinum. The baths could perhaps be identified with the balneum Clodianum, known from a 1st c. inscription.259 The complex was built on terraces and must have been quite large, possibly around 8,000 m2. It was only abandoned in the 5th c.260 Tridentum (Reg. XI, Trento) The archaeological remains of the Roman period in modern Trento are scarce, especially those pertaining to public buildings. The only remains of a possible bathhouse seem to have been in use in the 4th and 5th c. The building, located on the southern edge of the city, close to the city wall, underneath the modern Palazzo Crivelli, was reduced in size at this time.261 Tyndaris (Sicily, Tindari) The baths in the coastal settlement of Tyndaris lie close the 4th c. Christian basilica. The mosaics of the last phase of use of the baths date to the late 3rd c.262 Venusia (Reg. II, Venosa) The occupation of Venusia (Venosa) shifted south during the Medieval period, meaning large parts of the Roman town were not built upon. In 1956, a bathhouse was found along an

255  Giorgi (2012) 347. 256  Guzzo (1976) 620. 257  Guzzo (1986) 532. 258  On Constans II, see Theophanes, Chron. 1.535. For the excavations, see Cultrera (1954); Messsina (1993). 259   C IL 10.4792. 260  De Caro (2001) 898–99, without further evidence. 261  Ciurletti (2003) 40, n. 16. 262  Fasolo (2014) 58–60.

308

Gazetteer

figure 80 Plan of the Baths of San Gaetano di Vada (Italy), scale is 4 m in Menchelli and Vaggioli (1988) 122, fig. 94; with permission

figure 81 Plan of the House of the Coiedii in Suasa (Italy) after Giorgi (2012) 358, fig. 14a

important traffic artery leading to the theatre.263 On the basis of the type of wall construction and mosaics, the baths have been dated to the 2nd c., but were renovated until the late 4th c. During the first half of the 5th c., the presence of burials within the building confirm abandonment. A small semicircular pool added to one of the heated rooms is the only confirmed late antique element.264

Vicus Augustanus (Reg. I, Castel Porziano) The vicus Augustanus is located on the premises of the modern Castel Porziano estates, 16 km south-west of Rome and 9 km east of Ostia. Since Roman times, it has been a favourite resort of wealthy elites, emperors, popes and Italian presidents. While most of the bathhouses discovered were part of luxurious private villas, a larger structure with possible palaestra has been interpreted as a public bath (fig. 82).265 It

263  Salvatore (1984) re-examining older excavations. 264  Salvatore (1984) 82–84.

265  On the villa sites, see Marzano (2007) 315, 327. For the public bath, see Claridge (1985a); Claridge (1985b).

309

Gazetteer

figure 82 Plan of the Baths at Castel Porziano (Italy) in Claridge (1985a) 221, fig. 3; with permission

was located in a small vicus along the Via Severiana, connecting Portus (and Ostia) to Terracina. Brick stamps found in the sewage channel date to the Severan period.266 The baths were probably renovated during the Constantinian era, with buttresses reinforcing the vaults of the heated rooms and the large northern alveus of the caldarium replaced by two smaller single-person alvei. Probably in the late 4th c., the baths were reduced to about one fifth of their original size.267 The 266  Claridge (1985b) 72. 267  Claridge (1985b) 75.

sudatorium was transformed into a frigidarium with two single-person pools, while the caldarium was divided into a small tepidarium and caldarium. The furnaces of the other heated rooms were closed off, while stairs were constructed in the old frigidarium. It is possible that the building was turned into a private house at this time, with the baths reduced to a small private facility.268

268  Claridge (1985a) 221.

Baths in Roman North Africa New Construction C33. Ain En-Ngila (Libya)

Excavation history: The bathhouse was excavated in 1926– 1927 by the Italian Colonial Department of Antiquities. Bibliography: Bartoccini (1929a). Construction: (C) On the basis of the palaeography of a mosaic text and wall paintings, the baths have been dated to turn of the 3rd and 4th c. Continued use and abandonment: Unknown. Plan: One entered the baths through a long corridor, flanked by a latrine to the west. At the end of this corridor, the bather passed through a doorway into two rooms with low benches along the walls (apodyteria?). Continuing south, an almost square frigidarium had two rectangular piscinae lying on opposite ends of the room. A doorway supposedly connected this frigidarium to the aforementioned latrine.269 The connection to the adjacent southern room, which was heated by a hypocaust, is unclear. To the north-west of the latter, an L-shaped heated room should be identified as the caldarium. The apse in its northern side probably lodged an alveus, which would then lie above the furnace heating

this room. To the north of these rooms, a separate wing, also accessible from the corridor, may have been used for massages or epilation. In the largest room, two squarish basins might have been pools. Decoration: Fragments of painted wall plaster and partial slabs of marble were recovered during the excavation. The large central room with benches had a black and white mosaic with rosette design. The frigidarium floor was embellished with a polychrome mosaic with a rosette in its centre and gallinaceans in its angles. A mosaic with the text ‘bene laba’ was identified between the apodyterium and frigidarium. Technology: The hypocaust pillars were built with square tiles, with no mention of tubuli. The sewer that passed underneath the latrine may well have collected waste water from the piscinae as well, although these aspects of water disposal were not investigated at the time of excavation. Construction technique: Opus caementicium Bathhouse surface: ca. 350 m2 GPS coordinates: Exact coordinates unknown

C34. Althiburos (Henchir Medeina,Tunisia), Baths of the House of Asclepeia

Excavation history: The house with adjacent bathhouse was excavated during the 1970s by the Direction des Antiquités de Tunisie. Bibliography: Ennaïfer (1976). Construction: (C) The bathhouse in Althiburos was built adjacent to a large domus on the outskirts of the town, some 200 m north of the forum. The manner in which the walls abutted the existing walls of the villa makes clear that the baths were added in a later phase. The house is dated to the beginning of the 3rd c. on the basis of its mosaics, while the baths may have been added at the end of this century or the beginning of the 4th.270 Continued use and abandonment: (B) The baths probably fell out of use at the end of the 4th c., when debris with late 4th c. ceramics clogged up the sewers. Plan: In the first phase, the baths seemed only accessible by the open court of the house. Later, a paved porch was built to the east, directly onto the street.271 The original doorway to the house was blocked up and a new entrance was

C33 Ain En-Ngila (Libya), (After Bartoccini (1929a) 102, fig. 33)

269  Bartoccini (1929a) 103.

270  Ennaïfer (1976) 143–49; Thébert (2003) 129. 271  The excavators saw this paved entrance as a service door for carts bringing in the firewood, due to the tracks of wheels visible in the pavement (Ennaïfer (1976) 89). However, the effort in constructing this impressive paved porch, as well as the fact that this paving continued in the vestibulum, where a door to the domus was located, identify this entrance as a main doorway. From the vestibule, a narrow doorway led to the unpaved service area.

311

Gazetteer

C34 Althiburos (Henchir Medeina, Tunisia), Baths of the House of Asclepeia (After Thébert (2003) 582, Pl. XXXII.4)

created that gave access to the same room as the ‘public’ entrance. The square frigidarium (s: 4.4 m) had a large piscina (3 × 4.7 m; 1.1 m deep). At the beginning of the heated section there was a small room with a single-person pool (1.7 × 1.2 m; 0.7 m deep). The caldarium had a semicircular (r: 2 m) and a rectangular alveus.272 A thin wall made out of tubi fittili in the caldarium demarcates the south-western corner as a separate area. Decoration: The piscina was clad in white marble slabs. Mosaics, mainly black and white geometric motifs, covered the floors of the bathing rooms. Slabs of travertine were used as wall decoration. Technology: The hypocaust pillars were made out of limestone monoliths. A sewer coming from the house crossed the service area and probably collected waste water from the alvei.273 Construction technique: Opus africanum Bathhouse surface: ca. 170 m2 GPS coordinates: 35°52’32.06”N, 8°47’13.00”E Baten Zammour (Tunisia) Excavation history: A small bathhouse was discovered near a spring in a residential neighbourhood of the modern village (near Gafsa, Tunisia). Bibliography: Wiedler (1999) 295; Khanoussi (2001). Construction: (C) The mosaics have been dated to the first quarter of the 4th c.274 It cannot be verified whether the mosaics belonged to the original construction phase or pertain to a later restoration. Continued use and abandonment: Unknown 272  Following Thébert (2003) 130 contra Ennaïfer (1976) 91–91 who reconstructs a sudatorium (1.8 × 1.4 m) instead of a rectangular alveus. 273  Visible on plan CLXXIV in Ennaïfer (1976). 274  Wiedler (1999) 295; Khanoussi (2001).

Plan: Unknown. Decoration: Several mosaics including geometric and floral designs, a Venus with fishing erotes and an athletics game were discovered. Technology: Unknown.

C35. Belalis Maior (Henchir el-Faour, Tunisia), Baths of Theseus and the Minotaur

Excavation history: The bathhouse was investigated by the Direction des Antiquités de Tunisie in the 1970s. Bibliography: Mahjoubi (1978); Thébert (2003) 132–33. Construction: (A) The cold section mosaics can be attributed to the 4th c. This date is supported by ceramics found in the test pits reaching the foundation levels of the building (middle of the 4th c.).275 Continued use and abandonment: Several restoration phases can be identified, although it is impossible to date the different phases.276 Plan: The Baths of Theseus and the Minotaur, named after a mosaic in the frigidarium, were located just south of another bathhouse (the Forum Baths, see below, p. 358), separated only by a 3 m wide decumanus. The modest size (ca. 150 m2) saw the excavators label these baths as private, although no other elements corroborate such an interpretation.277 On the contrary, no domus has been discovered in the vicinity and the baths did have a street entrance. When entering, one had to turn left to reach the actual baths. A small corridor with a cold pool (ca. 2 × 2 m) preceded the spacious frigidarium. Due to its large size, the room probably served other functions as well (dressing room, anointing). East of the cold room, a rectangular space with a square piscina (ca. 2 × 2m) completed 275  Mahjoubi (1978) 225–26. 276  Thébert (2003) 132. 277  Mahjoubi (1978) 211.

312

Gazetteer The floors of the other cold rooms were embellished with polychrome geometric mosaics—in particular stars, octagons and lozenges—while the floors of the piscinae were covered with plain white mosaics of large tesserae.281 Technology: The heating system is difficult to reconstruct due to the bad state of preservation. The water supply was at least partially ensured by a well situated east of the baths. A channel led from this well to the two square cold pools. There is no information about waste water disposal. Construction technique: Opus africanum Bathhouse surface: ca. 150 m2 GPS coordinates: 36°45’49.38”N, 9°15’30.05”E

C36. Bulla Regia (Hamam Daradji, Tunisia), Baths North-west of the Theatre

Excavation history: The baths were discovered during the large-scale excavation in Bulla Regia between 1959 and 1962. An architectural study of the building was published in the 1980s. Bibliography: Hanoune et al. (1983); Thébert (2003) 134–35.

C35 Belalis Maior (Henchir el-Faouar, Tunisia), Baths of Theseus and the Minotaur (After Mahjoubi (1978) pl. XXVII)

the cold section. This strange layout meant that the cold pools lay outside the frigidarium. The heated section was found severely damaged. All the hypocausts had disappeared and no type of wall heating could be recognized. The two rectangular rooms that originally had suspended floors were each heated by a furnace in their southern side. The easternmost room had an apsidal ending in the south, probably lodging an alveus.278 On the level of the hypocaust, an east-west oriented wall separated the northern parts of both heated rooms. The excavator suggests these walls might have demarcated heated pools, although no evidence of boilers or reinforced hypocaust pillars to support these pools was identified.279 Fragments of tubi fittili were found in the mortar foundations of the mosaics, suggesting that terracotta tubes were used to construct the vaults. Decoration: The decorative scheme fits well within the North African contexts of the 4th c.280 The central panel of the frigidarium depicted Theseus killing the Minotaur. 278  Thébert (2003) 133. 279  Mahjoubi (1978) 211. 280  Mahjoubi (1978) 225; Dunbabin (1978) 249.

C36 Bulla Regia (Hamam Daradji, Tunisia), Baths North-west of the Theatre (After Hanoune et al. (1983) 64, fig. 3)

281  For a full description of the mosaics, see Mahjoubi (1978) 211–26.

313

Gazetteer

map 36

Bulla Regia (Tunisia), location of public baths and the most important buildings after Beschaouch et al. (1977) 13, fig. 3 1: Forum 2: Theatre 3: Christian basilica 4: House of the Hunt 5: Byzantine Fortress 6: Baths of Iulia Memmia 7: Large South Baths 8: Small Northeast Baths 9: Baths of the Venantii 10: Baths East of the Theatre

Construction: (C) The excavations of 1959–1962 did not yield solid archaeological information, so the date of the building has only been deduced from the style of the mosaics. Most of these should be placed in the 4th c.282 Continued use and abandonment: (C) The many restorations of the mosaic floors point to a long use of the building, perhaps well into the 5th c.283 After the disaffection of the baths, they were put to a different use. The blocking 282  Hanoune et al. (1983) 87. 283  Hanoune et al. (1983) 87.

up of the doorway between the northern and the southeastern heated room should probably be attributed to this phase.284 Plan: The Baths North-West of the Theatre were one of the 5 public bathhouses discovered in Bulla Regia (map 36). The baths were accessible by a small side street of a cardo and lay in the theatre quarter. By passing through a vestibule, one entered an oblong room that may have served as the dressing room. The large edge around the central 284  Thébert (2003) 135.

314 geometric panel of the mosaic may indicate that benches were placed against the wall.285 The octagonal frigidarium had 4 semicircular niches on opposite sides and the lateral piscinae. The north-eastern one was rectangular with an apsidal ending (0.9 m deep), while the south-western one had a trefoil plan (1.2 m deep). A doorway (1 m wide) in the eastern side of the frigidarium could be closed with a door and led further south to the heated section. The first room (4.6 × 2.9 m) had a hypocaust system, but no separate furnace. A door in the south-east corner gave access to a second heated room (2.6 × 2.8 m). It ended in a large apse to the south, probably containing an alveus.286 The third heated room (3.8 × 5 m) was accessible from the two other heated rooms. An alveus was lodged in the apse. Decoration: Few fragments of painted wall plaster were found. All of the floors had mosaics. The frigidarium mosaic imitated a marble cipollino floor, while a central octagonal emblem might have reflected the highest point of the vault.287 In the heated rooms, the polychrome mosaics had both floral and figurative designs (kantharoi), although few large fragments survived the demolition of the suspensura. The floors of the piscinae were covered with plain white mosaics. In the rooms outside the actual bathing rooms, mosaics with geometric designs or marble imitation were found. Marble sewer grates were found in the oblong room preceding the frigidarium. Technology: The baths were heated by a hypocaust system, but it is not known whether a type of wall heating was present. Underneath the alvei, a base of bricks supported the pools, rather than the usual perpendicular walls in continuation of the furnace mouth. There is no information on the water supply or waste water disposal. There was a small well in the service corridor east of the apsidal piscina. Construction technique: Opus africanum Bathhouse surface: ca. 800 m2 GPS coordinates: 36°33’32.28”N, 8°45’23.24”E

C37. Bulla Regia (Hamam Daradji, Tunisia), Small North-east Baths

Excavation history: Located in the north-eastern part of the excavated remains of the city (map 36), the baths were partially excavated in the 1920s. Bibliography: Carton (1922); Beschaouch et al. (1977) 79. Construction: (D) The baths may have been built at the same time as the nearby church, which dates to the 4th c. or later.288 The plan also has some characteristic late antique features, including several small semicircular and rectangular pools and niches.

285  Hanoune et al. (1983) 74. 286  Hanoune et al. (1983) 74. 287  Hanoune et al. (1983) 79. 288  Carton (1922) 332.

Gazetteer

C37 Bulla Regia (Hamam Daradji, Tunisia), Small North-east Baths (After Beschaouch et al. (1977) 79, fig. 75)

Continued use and abandonment: Several small hearths dating from the Islamic period were found on the bath room floors.289 Plan: The itinerary is unclear. The cold room was located in the north-east and had a semicircular pool. A doorway in the south-west corner gave access to two rectangular rooms (tepidaria). The northernmost had a small rectangular pool in its northern side. This northernmost room connected to two heated rooms, also on a north-south axis, that both had a small apsidal alveus on their western side, located above a furnace. From the southernmost room, the bathers could rejoin the frigidarium by passing through two (heated?) rooms, the latter with two semicircular niches in its eastern and western side. Decoration: The frigidarium had a mosaic floor which depicted a figurative scene, according to Carton. The heated rooms had mosaic floors with geometric patterns. Technology: There is little information about the hypocaust. The alvei had tubuli. To the north-west of the baths, a section of an aqueduct was found. Construction technique: Opus africanum Bathhouse surface: Unknown (at least 340 m2) GPS coordinates: 36°33’32.28”N, 8°45’23.24”E

289  Carton (1922) 333.

315

Gazetteer

Cincari (Henchir Toungar, Tunisia)

Excavation history: The baths were accidentally discovered during construction works in Henchir Toungar near Thuburbo Minus (Tebourba). Bibliography: No publication available. Construction: (C) The building was dated to the early 4th c. on the basis of its mosaics.290 Continued use and abandonment: Unknown. Plan: Unknown. Decoration: A mosaic, possibly from the frigidarium, depicted hunting scenes. Technology: Unknown.

C38. Columnata (Toukria / Sidi Hosni, Algeria)

Excavation history: The Baths of Columnata were discovered by a farmer in 1882. After he excavated the building, the site was visited by members of the Societé historique algérienneé, who described the bath and surrounding archaeological remains. The remains of some houses, a necropolis and a wall were also discovered in the area. Bibliography: Gavault (1883a); Thébert (2003) 194. Construction: (D) The simple linear plan is reminiscent of Byzantine fortress baths (see chapter 4, p. 184) or of the semi-public baths in Volubilis (see below, p. 377-381), while the small semicircular apse in the second heated room

could be the single-person pool that is found in many late antique baths. The use of tubi fittili to construct the vaults could also point to a late antique date. A coin dating from the reign of Constantius II proves that there was still some 4th c. activity in this small settlement.291 Continued use and abandonment: Unknown. Plan: Four rectangular rooms lay on a single axis. The bathers probably entered the baths through the frigidarium, which must also have served as an apodyterium. It had an apsidal piscina. The following three rooms had a hypocaust, but possibly only the last one was directly heated by a furnace. A small semicircular apse in the second heated room probably contained a water basin or small pool.292 The third heated room probably had an alveus above the furnace, as can be deduced from the hydraulic mortar and calcareous deposits described by Gavault.293 Decoration: The brief description by Gavault does not mention any decorative elements. Technology: The small basin in the second heated room had an outlet at its bottom. The bathhouse was located in a natural depression in the landscape, perfect for collecting rainwater. According to the local farmers, a water channel and a fountain were also found in the vicinity. The hypocaust pillars were made with square tiles and reached a height of 1 m. There is no mention of any wall heating. Construction technique: Unknown Bathhouse surface: Unknown (no dimensions are given) GPS coordinates: Exact coordinates unknown

C39. Cuicul (Djemila, Algeria), East Baths

Excavation history: The East Baths are part of an insula located north-east of the new forum. This quarter was excavated between 1943 and 1949. Bibliography: Allais (1953); Thébert (2003), 198–99. Construction: (C) On the basis of the mosaics, the baths were initially dated to the 3rd c.294 However, better knowledge of late antique North African mosaics pushed the construction date into the 4th c.295 Continued use and abandonment: Unknown. Plan: The baths lay along a street oriented north-east / south-west, starting at the new forum and leading to an unexcavated part of town (map 15). Six small steps led down from this street into the bathhouse (ca. 620 m2). A long (10 m) and narrow entrance passage in the north leads to a spacious cold room (14 × 9 m) with an apsidal cold pool (6.4 × 5.3 m; 1.3 m deep). South of this frigidarium, lay an oblong room (14 × 3.4 m) that may have served as a dressing or massage room. It may have had a separate door to the C38 Columnata (Toukria / Sidi Hosni, Algeria), (After Gavault (1883) pl. II, plan without scale) 290  Wiedler (1999) 301–302.

291  Gavault (1883a) 237. 292  Thébert (2003) 194. 293  Gavault (1883a) 234. 294  Allais (1953) 65. 295   Blanchard-Lemée (1975) 223–24.

316

C39 Cuicul (Djemila, Algeria), East Baths (After Thébert (2003) 627, pl. LXXVII.3)

outside in its south wall, but the ruinous state of the building does not allow confirmation.296 From the frigidarium a small door led to the service area, while a larger doorway gave into the first of a set of heated rooms, each with its own heated pool. The first room had a rectangular pool (1.7 × 1 m) against its western wall, located above the only furnace. The semicircular pool (2.4 × 1.5 m) of the second room was similarly arranged, although the furnace did not have a water boiler. The pools of the last room (9 × 3.3 m) are only conjectural as no traces of them have survived.297 In a later (undated) phase, this last room was made into a cellar, while the other heated rooms were subdivided into 4 rooms by a new wall cutting north-west / south-east through them. The heating system of the first room was removed. The oblong room was made into a stable (4 troughs were discovered), while the frigidarium possibly went out of use. These interventions are interpreted as turning these public baths private.298 Decoration: The floor of the oblong room was covered with a simple, floral-geometric, black and white mosaic. The access to the piscina was embellished by columns and Corinthian pilasters, while the floor was laid out in mosaics and a red marble slab crowned the margin. The walls

296   See plan in Allais (1953) 51 contra Thébert (2003) 627, pl. LXXVII.2. 297  Thébert (2003) 199. 298  Allais (1953) 62.

Gazetteer above were decorated with polychrome wall paintings, but too damaged to reconstruct. Technology: As the bathhouse was restored and modified several times, it is difficult to reconstruct the heating systems and aspects of water management. The heated rooms had a hypocaust, while the walls were, at least in the last phase of the baths, heated by creating a hollow space using tiles with a protruding rim. The pools were heated by direct contact with the furnace and at least the pool of the first room received heated water from a boiler. The water for the boiler was supplied by a small cistern under the street to the north. This was alimented by a pipe that ran under the street. For water disposal, a sewer was found that started from the pool of the second heated room and ran south to the piscina to collect its waste water. There may have been a latrine east of the piscina. The channel eventually ended in the large sewer starting at the new forum and passing under the oblong room. This main sewer made use of the natural slope of the terrain to dump waste water into the gorge north of the city. Construction technique: Opus africanum Bathhouse surface: ca. 620 m2 GPS coordinates: 36°19’18.37”N, 5°44’11.20”E

C40. Cuicul (Djemila, Algeria), Baths of the House of Europe

Excavation history: The house was discovered during the large-scale excavations of the early 20th c. (1916), but no publication followed.

C40 Cuicul (Djemila, Algeria), Baths of the House of Europe (After Blanchard-Lemée (1975) fig. 49)

317

Gazetteer

figure 83 Plan of the House of Europe in Cuicul (Algeria) with semi-public bathhouse in box after Blanchard-Lemée (1975) fig. 49

Bibliography: Ballu (1917); Allais (1939); Blanchard-Lemée (1975), 140–51; Thébert (2003) 201–202; Maréchal (2016). Construction: (C) The mosaic of the abduction of Europe in the triclinium, which gave the house its name, was dated in the 4th c.299 The bathhouse was probably built later than the house, but in the same century.300 Continued use and abandonment: Unknown. Plan: The bath was incorporated into the large domus, but was accessible by two different entrances, both onto a street (fig. 83). The north-east entrance was located in a secondary cardo, diagonally opposite the Baths of the Capitolium (map 15). It led into a small chamber that could have served as a porter’s lodge. By going up three steps, one reached a vestibule that probably served as a dressing room (9 × 5.3 m) with a simple concrete floor. The south-western entrance was accessible from the cardo maximus and opened onto a large room (8.8 × 5.4 m) with a small latrine against its north wall. Through a doorway in the south-east corner it gave access to the same vestibulum. The vestibule was separated from the cold room by two rectangular pillars. The square frigidarium (s: 6 m) had a column in each corner, possibly supporting a cross vault. A square pool (s: 2 m) with an apse (r: 0.8 m) was coated with lime mortar. In the eastern corner of the room, a doorway gave access to the heated rooms, all with a hypocaust. The first room (5.7 × 2.8 m) was indirectly heated, while the second (3.2 × 5 m) and third room (3.3 × 1.9 m) had their own furnace. As the second room was larger, the temperature was probably lower than in the last room. Although not recognized by 299  Allais (1939) 44. 300   Blanchard-Lemée (1975) 151.

the excavators at the time of excavations, the second room probably had a pool.301 There is evidence of a water boiler above its furnace.302 From the frigidarium, a small doorway in the northern corner led to the praefurnium, directly accessible from the street. This service room connected with a small set of bath rooms to the west. Interpreted as a bath for the staff, it comprised a small unheated chamber (1 × 1.6 m) with a cold tub, an indirectly heated room (1 × 1.5 m) and a heated space (2.4 × 1.2 m; including alveus) with heated pool.303 Decoration: The frigidarium floor was covered with a mosaic and the walls with marble slabs. The decoration of the heated section was lost when the rooms changed function. Technology: There is little information about the heating system. Only the hypocausts of the three heated rooms are mentioned. We know nothing about possible wall heating. For the small ‘service baths’, a single furnace and simple suspended floor without wall heating seems to have sufficed. The water supply was guaranteed by several reservoirs. The piscina was filled by a reservoir north-west of the frigidarium, while the boilers for the heated parts were probably fed by a small reservoir north of the service room. There is no evidence for waste water disposal in the heated part, while the cold pool was probably drained from the west, passing under the latrines to the sewer in the cardo. Construction technique: Opus africanum Bathhouse surface: ca. 280 m2 GPS coordinates: 36°19’22.48”N, 5°44’3.93”E 301  Ballu (1917). 302  Thébert (2003) 202. 303  Ballu (1917) 282.

318

Gazetteer

C41. Cuicul (Djemila, Algeria), Baths of the House of the Donkey

Excavation history: Like most of Cuicul, the House of the Donkey was excavated in the early 20th c. Bibliography: Ballu (1910); Ballu (1921); Blanchard-Lemée (1975) 29–32, 58–59, 85–100; Thébert (2003) 200–201; Maréchal (2016). Construction: (C) The baths were inserted in the house at a later phase. On the basis of the mosaics, it seems this occurred at the end of 4th or beginning of the 5th c.304 Continued use and abandonment: The baths had several building phases, but the lack of archaeological data makes these difficult to date. Plan: The house is situated just south of the old forum, along the cardo maximus and attached to the rear of the temple of Venus Genetrix (map 15). The baths occupy the north-western corner of the mansion (ca. 110 m2, fig. 84) and were accessible from the cardo maximus. Two passages between the house and the baths were located in the large entrance room at the beginning of the bathing route. In contrast to the House of Europe, the southern door gave directly into the core of the house. Nevertheless, it is difficult understand which parts of the house were private and which were (semi-) public when the baths were in use. A sort of vestibule was created in the portico of the cardo. From here, one could enter the service room or an oblong room that may have served as a dressing room to the baths. The eastern end in particular, separated by two columns, may have been used as such. Through a doorway to the north-east, one entered the rectangular cold room (10.5 × 4 m), which had two cold pools at opposite ends. The semicircular one (1.8 × 2.15 m; 1 m deep) at the western end protruded into the vestibulum and was coated in a waterproof mortar. The eastern pool had a rectangular shape with a small apse (2.75 × 1.6 m; 0.7 m deep). Two rectangular niches in the northern wall may have been used to store bath utensils.305 By a narrow doorway in the north-east corner of the cold room, the bathers entered the heated part of the baths, continuing in an east-west direction. A first room (3.5 × 1.9 m), its mosaic floor still intact at the moment of excavation, may have been equipped with a hypocaust, although no traces seem to have survived after the mosaic was removed.306 If there was a hypocaust, this room must have been indirectly heated. In the second room (3.5 × 2.2 m), the hypocaust is unmistakably present and was heated by a furnace in its southern corner. Yvon Thébert argues that a small heated pool may have been present, as the furnace was surmounted by a boiler.307 As the space is 304  Dunbabin (1978) 256. 305  Ballu (1910) 105. 306   Blanchard-Lemée (1975) 30 contra Ballu (1921) 235. 307  Thébert (2003) 201.

C41 Cuicul (Djemila, Algeria), Baths of the House of the Donkey(After Blanchard-Lemée (1975) 24, fig. 4)

rather restricted, the boiler may only have served a tap or may have served the pool of the next heated room (also considered by Thébert). This third heated room (3.3 × 1.8 m) was interpreted as one large pool, although Thébert reconstructs a small caldarium with tub.308 In either case, a separate furnace heated the hypocaust of this room. Decoration: The floor was covered with a polychrome mosaic consisting of 73 medallions sprouting from two kantharoi. In each medallion, an animal or a human figure was represented, including several kinds of bird, deer, elephants, dogs, etc.309 The medallions were organised in 5 bands, the middle one consisting of 9 larger medallions. In these large medallions, we find the representation of a donkey with the text ‘ASINUS NICA’ (‘The victorious donkey’). Furthermore, we find a naked man, probably a slave, with a bucket in each hand, a man with a small box (bath utensils?), a man dressed in a cloak and a man with a piece of cloth in his hand (a towel?). The whole composition has strong connotations with the cult of Bacchus (grapes, peacocks, kanthorai) and hunting (wild animals, dogs, horses, boars) and may have been an allusion to the otium of the upper class.310 On the mosaic floor of the first heated room, another donkey with the same text was represented. The donkey was seen as a symbol of good luck in both classical and Christian circles.311 The bottom of the eastern piscina was covered with a white mosaic, while the sides were covered with lime mortar. 308   Resp. Ballu (1910) 105; Blanchard-Lemée (1975) 31 contra Thébert (2003) 201. 309  For a full description, see Blanchard-Lemée (1975) 85–96. 310   Blanchard-Lemée (1975) 93. 311   Blanchard-Lemée (1975) 96–100.

319

Gazetteer

figure 84 Plan of the House of the Donkey in Cuicul (Algeria) with semi-public bathhouse in box, scale is 10 m after Blanchard-Lemée (1975) 24, fig. 4

Technology: The heating system is difficult to reconstruct, due to the interventions during the first excavations. The two small furnaces were sufficient to heat the restricted surface. The water for the pools was stored in a reservoir behind the eastern pool of the frigidarium. It is unknown if additional means of water supply existed. The waste water disposal of both cold and heated pools was secured by a sewer that came from the fountain of the house and ran under the frigidarium. It probably ended in the sewer of the cardo maximus. Small test pits in the frigidarium suggest that these drainage channels pre-existed the baths.312 The architect would have designed the baths with the drains in mind. Construction technique: Opus africanum Bathhouse surface: ca. 110 m2 GPS coordinates: 36°19’18.15”N, 5°44’7.10”E

C42. Cuicul (Djemila, Algeria), Baths of the House of Castorius

Excavation history: The house was excavated in the early 20th c. Bibliography: Ballu (1912); Blanchard-Lemée (1975) 161–66; Thébert (2003) 199–200; Maréchal (2016). Construction: (D) The House of Castorius was built after the southern part of the early city wall was dismantled at the beginning of the 3rd c.313 There were two baths discovered in this house. The ‘internal baths’ belonged to the construction phase of the house, while the ‘external baths’ 312   Blanchard-Lemée (1975) 57–58. 313   Blanchard-Lemée (1975) 199.

C42 Cuicul (Djemila, Algeria), Baths of the House of Castorius (After Blanchard-Lemée (1975) fig. 62)

were built later. On the basis of the layout and the fact that the baths protruded onto the street, a date in the 4th c. or later could be proposed. Continued use and abandonment: Unknown. Plan: The house was constructed between the northern wall of the new forum and a temple (fig. 85). It was one of the largest houses discovered in Cuicul and was named after the apparent owner. The ‘external baths’ (or ‘petits thermes’) were implanted on a part of the public road, which connected the old with the new forum. The small scale of these baths, with a total surface not exceeding 60 m2, have led scholars to believe that they could only have been used by the staff or slaves.314 However, the street access may point to use by more than just the members of 314   Blanchard-Lemée (1975) 165.

320

Gazetteer

figure 85 Plan of the House of Castorius in Cuicul (Algeria), with adjacent semipublic baths (in box) and private baths (in shaded box) after Blanchard-Lemée (1975) fig. 62

the household.315 Furthermore, the number and size of the pools may also have surpassed the requirements for simple ‘household baths’ (compare to the ‘household baths’ in the House of Europe). The only link to the house was through a long and narrow corridor. The bath only consisted of one unheated and one heated room. The frigidarium had an apsidal (1.2 × 2.1 m) and rectangular pool (1.5 × 1.9 m). The heated room (4.8 × 2.7 m) was accessed from the cold room by going up three steps, equalling the height of the hypocaust pillars. At the south-eastern end, a pool (0.9 × 2.7 m) was located just above the only furnace. From the frigidarium, one could follow a narrow corridor leading to a small latrine and on to the older ‘internal’ bathhouse. It is unclear whether these were still in use at that time. Decoration: Both cold and heated pools were embellished with a simple black mosaic. Technology: There is little information on the heating system and water management. The waste water from the alveus may have been channelled to a sewer passing through the praefurnium, coming from the internal baths and running north.316 Construction technique: Opus caementicium Bathhouse surface: ca. 35 m2 GPS coordinates: 36°19’17.13”N, 5°44’8.95”E

C43. Carthage (Tunis, Tunisia), balnea privata (‘Villa aux bains’)

315  Maréchal (2016) 128. 316  Visible on fig. 62 in Blanchard-Lemée (1975).

317  Hansen (2002) 116–17. 318  Hansen (2002) 31.

Excavation history: The bathhouse was discovered in the 1990s during the Swedish excavations of the UNESCO ‘Pour sauver Carthage’ project, on a stretch of land between the Byrsa and Junon hill. Bibliography: Hansen (2002). Construction: (C) The baths were part of an insula, which could only be dated by the mosaics, pointing to a late 2nd or early 3rd c. date. Important reorganisations of the entire insula, including a complete restructuring of the bathhouse to open it to the public, was dated at the late 4th / early 5th c.317 Continued use and abandonment: (B) The building was dismantled in the Islamic Period to salvage building materials.318 Plan: In the first phase, the bathhouse (fig. 86) consisted of a courtyard, a combined apodyterium-frigidarium, a tepidarium and a caldarium with a single alveus. Some small changes were made to the heated section—possibly with the construction of a smaller alveus—and to the service corridors, before a thorough reorganisation changed the entire layout. In the second phase, the bathhouse got a new entrance from the larger decumanus running north of the insula. It was S-shaped, so that no direct line of sight revealed the interior of the baths. The large pool at

321

Gazetteer

figure 86 Plan of phase 1 of the balnea privata in Carthage (Tunisia) in Hansen 2002, 116, fig. 96; with permission

C43 Carthage (Tunis, Tunisia), balnea privata (After Hansen (2002) 117, fig. 97)

the beginning of the heated section was replaced by two smaller single-person pools, one of which was now accessible from the frigidarium. The caldarium was reduced in size and the large alveus was abandoned in favour of two small single-person alvei. To the west of this bath, separated by a communal service corridor, a second suite of bath rooms was built. These were only accessible from the central corridor of the courtyard, although an access from the unexcavated west side of the insula—along the cardo maximus—cannot be ruled out. The rooms lay on a north-south axis and comprised a small unheated room (1.12 × 3.16 m)—possibly an apodyterium—and two small heated rooms.319 The first

room (1.12 × 3.18 m) had a hypocaust but no wall heating. The second room (1.77 × 3.22 m) was heated by a single furnace and possibly had a small alveus (1.25 × 1.7 m) in its north-west corner.320 The excavators doubted that these baths ever functioned, as no traces of soot were found in the hypocaust or in the furnace. The fact that a private bath, only accessible from the inside of a building, was ‘opened up’ to the public by creating a street entrance, is reminiscent of late antique houses in Cuicul (C40, C41, C42) and possibly Volubilis (see below, p. 377–381), which were equipped with large ‘semi-public’ baths in the 4th c. The creation of a separate bathhouse (west wing), perhaps for exclusive use by the owners, and

319  Hansen (2002) 79–80.

320  Hansen (2002) 76–77.

322 the reorganisation of the pools in the original bathhouse probably fits within this new ‘public use’ of the east wing. Decoration: The decoration of the eastern baths has been attributed to the second (public) phase, consisting of a polychrome mosaic—grapes, a pomegranate and pigeons surrounded by acanthus leaves—in the southern entrance vestibule. The small pools were paved with white marble slabs. The western baths were badly damaged and no traces of their decoration were found. Only the unheated room retained its polychrome mosaic with geometric patterns. Technology: The hypocaust pillars were made of square bricks or stone slabs, mainly local sandstone and limestone. No traces of tubuli have been found in either bath. Only the alvei of the eastern baths had lateral flues. The hypocaust floor underneath the presumed alveus of the western baths lay slightly lower than the hypocaust floor of the caldarium. A small ramp connected both floors, presumably to facilitate the circulation of the hot air from the lower to the higher part.321 The water for the pools was stored in a reservoir located above the service corridor and perhaps above the small latrine. Six other cisterns were also in use in the insula, although not all were probably used for supplying the baths. These cisterns were likely filled by rainwater collected on the roof.322 The combined volume of cisterns and reservoirs that provided water for the baths is estimated at 153 m3. During the second phase, the large well (diam.: 1.73 m) south of the east baths probably fed the two reservoirs.323 The conduits leading water into the pools or to the presumed boilers above the furnaces have not been found. The latrine was flushed with the water from the alvei, while water from the small cold pools and the pool in the frigidarium were drained by a sewer heading north towards the street. Construction technique: Opus africanum Bathhouse surface: ca. 165 m2 (east wing), ca. 38 m2 (west wing) GPS coordinates: Exact coordinates unknown

C44. Carthage (Tunis, Tunisia), Odeon Hill Baths

Excavation history: On the crossroad of the Kardo IV E and the Decumanus VI N, an excavation of Trinity University (San Antonio, USA) in the early 1990s discovered the remains of a small bathhouse and a cryptoporticus. Bibliography: Garrison et al. (1993). Construction date: (B) The bath was built on the premises of an earlier peristyle-courtyard house. The construction level seems to postdate a destruction layer in the house dated by ceramics to the middle of the 4th c. Continued use and abandonment: (B) The entire site fell out of use in the first quarter of the 6th c., when layers of 321  Hansen (2002) 78–79. 322  Hansen (2002) 107. 323  Hansen (2002) 67–68.

Gazetteer

C44 Carthage (Tunis, Tunisia), Odeon Hill Baths (After Garrison et al. (1993) 253, fig. 2)

debris—including datable ceramics—filled the rooms. The site was reused as a burial ground in the later 6th and 7th c. Plan: The poor preservation of the walls and doorways and the limited size of the excavation make it difficult to label these baths as private or public. The existing outer walls of the house were reused to create at least three heated rooms. An apse containing an alveus probably identifies one of these as a caldarium. Against the eastern wall of the same room, part of a ‘sitz-bath’ was also discovered. The pool can only be reconstructed partially. The excavators suggest that 6 to 7 people at a time could bathe in this pool, although this assumption seems to rest on generalities about communal pools rather than on the reconstructed dimensions of the pool in question.324 Decoration: The pool had a plain white mosaic floor. Technology: Little is known about the water management of these baths. No traces of wall heating have been identified either. The cryptoporticus of the adjacent house made use of tubi fittili, but no evidence of these building materials has been found inside the baths. Construction technique: Opus caementicium Bathhouse surface: Unknown (at least 250 m2) GPS coordinates: Exact coordinates unknown

C45. Carthage (Tunisia), Baths in Okba Ibn Nefaa Street

Excavation history: The Baths in Okba Ibn Nefaa Street were described briefly by the excavator in the 1920s. Bibliography: Saumagne (1925); Rossiter (2009).

324  Garrison et al. (1993) 255.

323

Gazetteer

GPS coordinates: Exact coordinates unknown (remains now disappeared)

C46. Carthage (Tunisia), Byzantine Baths

C45 Carthage (Tunis, Tunisia), Baths in Okba Ibn Nefaa Street (Hypothetical sketch of the frigidarium, no scale)

Construction: (D) The remains are described as a baroque architecture, typical of the 4th or 5th c.325 Continued use and abandonment: Unknown. Plan: No plan has ever been published, but a sketch of the cold room can be proposed on the basis of the detailed description. The baths were located along the Cardo IV East (map 19). A small tetrastyle porch, partially encroaching onto the street, gave access to a rectangular vestibule. A doorway in its western side led to a rectangular room ending in a large niche flanked on either side by a smaller one. From this room, a doorway gave access to the frigidarium (north) and the caldarium (south). The caldarium was a rectangular room (4.6 × 2.9 m) on hypocausts. To the south, a doorway might have led to a sweat room.326 The frigidarium was square in plan (s: 4.4 m), but had two semicircular niches in its south side and a trefoil room to its north side. The trefoil room housed a quatrefoil pool (1.27 m deep), and was covered entirely in plain white mosaics. In each of the apses of the trefoil room there was a large niche (1.2 m high; 0.6 m wide; 0.4 m deep). Decoration: Unknown. Technology: At the base of the eastern niche of the frigidarium, a water conduit supplied the pool. An outlet for waste water was found in the floor of the western lobe. It was connected to a sewer passing west of the frigidarium. Construction technique: Opus caementicium Bathhouse surface: Unknown

325  Rossiter (2009) 185. 326  Saumagne (1925) xcvii.

Excavation history: The site of the large Antonine baths, which was later reused for the construction of the Byzantine baths, was excavated in the 1940s and 1950s by the ‘Commission des fouilles’ of the French Ministry of Foreign Affairs. Bibliography: Picard (1946–1949); Picard (1950); Picard (1955–1956); Lézine et al. (1956); Lézine (1968); Thébert (2003) 141–43. Construction: (B) If we identify the bathhouse that arose on the site of the Antonine baths as Procopius’ ‘Thermae Theodorianae’ a construction date shortly after AD 533 could be proposed.327 Alexandre Lézine claimed that the Antonine baths (fig. 42) fell out of use when the vault of the frigidarium collapsed, probably during the 4th c.328 The heated section of the baths may have been used until AD 425, when a fire destroyed most of the building.329 In the subsequent period, a layer of alluvial clay covered the destruction debris, pointing to a long period of abandonment. When Byzantine forces took control of the city, the construction of a large public bathhouse seems to have been one of the top priorities. Continued use and abandonment: (C) The bathhouse probably fell out of use in the 7th c., when a pottery kiln was constructed in the central pillar of the underground room supporting the caldarium.330 Decorative elements such as marble and granite column shafts and entablatures were salvaged as building material or burnt in lime kilns during the Islamic period.331 Plan: The new bathhouse was constructed on the remains of the old Antonine Baths (fig. 42), more specifically above the old heated section. Even if the Byzantine Baths covered only around one seventh of the total surface of the Antonine Baths, it was still a large complex by the standards of the period. Indeed, if one compares the size of these baths to other contemporary public baths in North Africa, we can deduce the importance of Carthage, in population size and wealth. The strange disposition of the rooms, a double linear row, was conditioned by the standing remains of the previous building and a will to exploit as much of the existing structures as possible.332 The baths consist of two perfectly symmetrical parts, joining in a communal caldarium. One entered the complex 327  Procop., Aed. 6.5.10 (named after Justinian’s wife). 328  Lézine (1968) 71–72. 329  The layer of ash, long attributed to a fire caused by Vandal raids, has been dated on the basis of coins and oil lamps (Lézine et al. (1956) 426). 330  Lézine et al. (1956) 430. 331  Lézine (1968) 72. 332  Thébert (2003) 143.

324

Gazetteer

C46 Carthage (Tunis, Tunisia), Byzantine Baths (After Lézine (1969) 36, fig. 9)

by taking a staircase at the western or eastern end of the building. From each side, a succession of three cold rooms was installed in the former hexagonal ‘dressing room’, the rectangular heated pool and the hexagonal districtarium.333 In the middle room, the old niches in the walls were converted into shallow pools, interpreted as pediluvia by Lézine.334 Continuing towards the centre of the complex, two heated rooms preceded the caldarium, communal to both the eastern and western wings. The first heated room was installed in the old tepidarium, although its form had been altered to include lateral apses. The second room, small and rectangular, was constructed in the former service courts. The caldarium, with lateral apsidal endings, occupied the space of the former ‘exit tepidarium’. The poor conservation of the remains does not allow for identification of the heated pools. Fragments of large stone basins were found, possibly labra, which may have been spolia of the preceding Antonine Baths. Decoration: The types of decoration that have been found include mosaic floors depicting acanthus leaves, fruits and birds. In the large latrines, a mosaic with a marine scene has been found. There is little information on other forms of decoration.335 Technology: Little is known about the technical aspects of these baths. Each of the 5 heated rooms had its own furnace. The hypocaust pillars were made out of brick, but the type of wall heating could not be reconstructed. The water supply was ensured by large reservoirs at the western and

333  The hexagonal dressing room, interpreted as such by Lézine (1968), should perhaps be seen as a solarium, judging from its protruding position and its three large windows. 334  Lézine (1968) 67. 335  The excavation reports of the 1940s leave little doubt about the excavations’ strategies and interests: ‘Le déblaiement des Thermes d’Antonin a été poursuivi (...) avec des moyens mécaniques importants, prêtés par la Direction des Travaux publics. Malheureusement, il n’y a à signaler aucune découverte sculpturale ou épigraphique notable.’ (‘The clearing of the Antoine Baths was continued (...) with important mechanical aid, lent by the Office of Public Works. Unfortunately, there is not one discovery of a sculpture or inscription worthy to be mentioned.’) (Picard (1949) 619; translation by the author).

eastern ends of the bathing rooms, which had originally been small lounges. Construction technique: Opus caementicium Bathhouse surface: ca. 2,025 m2 GPS coordinates: 36°51’16.00”N, 10°20’5.68”E Carthage (Tunisia), Baths in the Baal Hammon Street Excavation history: In 1988, a part of a Byzantine bathhouse was discovered during rescue excavations in the Baal Hammon street (map 21, nr. 1). Bibliography: Ellis (1989). Construction: (D) The architectural remains as well as the construction level of the baths, at the same level as the Byzantine street, point to a 6th c. date. Continued use and abandonment: In the 7th c., mud brick walls divided the building into several units. The baths were probably transformed into houses. Plan: The remains included a basin (0.5 m in width), the base of a column, a mortar floor and a sewer. Decoration: Unknown. Technology: Besides the pool with its waterproof mortar coating, few bath elements were identified.

C47. Carthage (Tunisia), Douar Chott Baths

Excavation history: At the south-west corner of the socalled turris aquaria, a complex of 4 large cisterns, a large bathhouse was partially excavated in the 1920s. Bibliography: Saumagne (1928) 638–45. Construction: (D) The bathhouse has been dated to Late Antiquity due to the reuse of building material and the recovery of artefacts with Christian decoration. The brick stamps found in the building mention Christian names.336 Continued use and abandonment: Unknown. Plan: The unusually large size of these baths prompted the excavators to identify the complex with the baths of Thrasamund (see EA-52 to EA-56) or Justinian’s Thermae Theodorianae (see above, p. 323n327).337 There is no material evidence to support such claims, while the peripheral location makes it unlikely that we should identify the 336  Saumagne (1928) 643–44. 337  Saumagne (1928) 644.

325

Gazetteer

C48 Kerkouane (Tunisia), (After Courtois (1954) fig. 3)

C47 Carthage (Tunis, Tunisia), Douar Chott Baths (After Saumagne (1928) 639, fig. 1)

complex with Augustine’s baths of Gargilius (see, chapter 2, p. 50). Stone robbing destroyed much of the walls, but the robber trenches give an idea of the original layout of these baths. An oval pool with semicircular niches was constructed on top of an earlier building. The latter was used as a foundation by filling up the rooms with mortar. South-west of this pool, a large rectangular room was identified as the frigidarium. South of this, two heated rooms on a hypocaust were discovered, delimited by a service corridor to the south. West of the heated rooms, a floor on a lower level drained by a lead pipe reveals the existence of a heated pool. Decoration: The frigidarium was paved with greenish marble slabs. The wall and floor decoration of the heated rooms was stripped in antiquity. Only the geometric mosaic floor of the oval piscina and the plain white mosaic of the alveus have been preserved. Technology: The hypocaust pillars were built out of square or round tiles and reached a height of 1 m, while the walls were heated using tubuli. One of the alvei was drained by a lead fistula. Construction technique: Opus caementicium Bathhouse surface: Unknown (possibly exceeding 500 m2) GPS coordinates: Exact coordinates unknown (remains have disappeared)

C48. Kerkouane (Tunisia)

Excavation history: The baths were found and excavated in 1952 by the ‘Commission des fouilles’ of the French Ministry of Foreign Affairs. Bibliography: Courtois (1954); Thébert (2003), 143–44.

Construction: (C) On the basis of the mosaics, a late 3rd or early 4th c. date has been proposed.338 Continued use and abandonment: Unknown. Plan: The small bathhouse was found in a rural setting, with few traces of other structures in the vicinity. As was the case in Sidi Ghrib and Oued Athmenia (see below, p. 345– 347 and C52), these baths may have belonged to a villa or could have served a small rural population. The plan was a simple linear succession of a cold room ( frigidarium) and two heated rooms (tepidarium and caldarium). The frigidarium had an apsidal piscina (3.1 × 2.4 m; 1 m deep), while the caldarium had two apsidal alvei (north: 2.7 × 2 m; 0.8 m deep; west: 2 × 2.5 m). Decoration: The mosaics consisted mainly of geometric patterns, but a depiction of a bird in front of a tree was also found. A pair of sandals was reproduced at the entrances to the frigidarium and the tepidarium.339 Technology: The tepidarium and caldarium were heated by hypocausts, but there was no trace of wall heating. The vaults were constructed with tubi fittili. Courtois does not mention evidence of water supply or waste water disposal. Construction technique: Opus caementicium Bathhouse surface: ca. 150 m2 GPS coordinates: Exact location unknown Kef Glia (Tunisia) Excavation history: Some 10 km north-east of Thugga, on a promontory off the road connecting El-Kef to Tunis (N5), the remains of several Roman and Byzantine buildings were identified during the survey of the University of Trento (Italy) in collaboration with the Institut National du 338  Courtois (1954) 202. 339  For a description of the mosaics, see Courtois (1954) 199–202.

326 Patrimoine de la Tunisie (1994 to 2000, 2008 and 2012). The promontory, which dominated the access to the Khalled valley, was the site of a Byzantine fortress and several elements linked to agricultural production (oil presses), as well as 7 cisterns and a bathhouse. Bibliography: de Vos Raaijmakers and Attaoui (2013) 127–30 and 356–57. Construction: (C) The opus vittatum walls of the baths may point to a late antique or Byzantine construction date.340 The surface finds of the site, covering the entire Roman and the early Islamic periods, permit such a dating.341 Continued use and abandonment: Unknown. Plan: As the building has not been excavated, only a brief description of the building is provided. At least two rooms had a hypocaust, located on the south-western side of the building. A semicircular niche in the north-west corner of the building showed traces of water pipes removed after the abandonment of the building. This may have been a schola labri.342 At least one room was covered by a cross vault, parts of which were found inside the building. Decoration: Unknown. Technology: The hypocaust pillars were made with square tiles. There is no information on the water management of the site, except for the aforementioned water pipes.

C49. Lambaesis (Tazoult, Algeria), Baths of the Hunters

Gazetteer

C49 Lambaesis (Tazoult, Algeria), Baths of the Hunters (After Thébert (2003) 635, pl. LXXXV.2)

Excavation history: The Baths of the Hunters in Lambaesis (Lambèse/Tazoult, Algeria) were named after the ‘officers of the Hunters of Africa’ that unearthed the complex in 1880–1881. The Service des monuments historiques de l’Algérie investigated the building in 1938. Bibliography: Krencker et al. (1929) 215–17; Leschi (1938– 1940); Thébert (2003) 212–13. Construction: (D) Yvon Thébert argued that the layout of the plan points to a late date, and the three-apse plan of the caldarium was popular in late antique architecture (see also below C53).343 The only datable element that has emerged from the early excavations is the tile stamp of Legio III Augusta, stationed in the city. While there is no attestation of the legion in the city after AD 293, it could be that a stock of tiles was still used (or reused) at a later date. An inscription dated to the 3rd c. and linked to the Asclepius cult was probably reused in the baths.344 Continued use and abandonment: Unknown. Plan: The baths lie 100 m north-east of the new forum and some 150 m east of the Asclepeium. The plan is dominated

by the oblong frigidarium with its large rectangular pool and a rectangular space to its south (ca. 13 × 10 m). The function of the latter is unclear. It may have been a room for physical exercise or perhaps a lounge area.345 In this case, we can imagine that the small side chambers east and west of the hall were rooms for massages and anointing. From the central frigidarium, the bathers started a clockwise circular itinerary. A first rectangular room to the south-west of the frigidarium may have been a tepidarium. The second heated room was octagonal in shape and had two semicircular niches on its eastern side, but seemingly lacked pools.346 The caldarium had an oblong plan, with semicircular alvei on its short ends and semicircular apses in the middle of the long sides. One entered and exited the room by two separate doorways in the small eastern apse. Before returning to the frigidarium, one passed a small rectangular exit-tepidarium with a semicircular apse which may have lodged a small pool or labrum.347 A separately accessible latrine east of the bathhouse was probably flushed by waste water from the piscina. Decoration: The report by Leschi does not discuss the decorative scheme.

340  Several sites of Vandal and Byzantine date in the immediate vicinity were erected in the same opus vittatum technique, see de Vos Raaijmakers and Attaoui (2013). 341  de Vos Raaijmakers and Attaoui (2013) 130. 342  de Vos Raaijmakers and Attaoui (2013) 129. 343  Thébert (1983) 107; Thébert (2003) 212. 344  Leschi (1938–1940) 267–69.

345  The room was accessible from the outside by a separate entrance, which supports the idea of a room for physical exercises, following the theory of Thébert (1991). 346  Thébert (2003) 213 contra Krencker et al. (1929) 217. 347  Krencker et al. (1929) 217.

327

Gazetteer

C50 Lambaesis (Tazoult, Algeria), Small Baths south of the Asclepeium (After Thébert (2003) 635, pl. LXXXV.5)

Technology: There is little information on the hypocaust system or the type of wall heating. The use of tubi fittili for the vaults is attested in the service area. Construction technique: Opus africanum? Bathhouse surface: ca. 1,400 m2 GPS coordinates: 35°28’47.66”N, 6°16’12.29”E

undated bathhouse (fig. 87, see below, p. 360).349 Both baths have been linked to the healing cult of Asclepius: the sick had to bathe before entering the sanctuary.350 One entered the baths by a large, almost square hall to the north (7.4 × 6.75 m), possibly used as multifunction space (apodyterium, unctorium). A large doorway to the south led into the frigidarium, which had two lateral semicircular piscinae. In the centre of the southern wall of the frigidarium, a narrow door led into the first heated room, which in its turn gave access to a heated room to the east and to the west. Yvon Thébert reconstructed a semicircular pool in the outer walls of both of these rooms. The doorway between the eastern room and the frigidarium may have facilitated the itinerary for people entering the heated section.351 From the western room, one accessed a fourth heated room, also with semicircular pool on its eastern side. To the south of this room lay the caldarium with a semicircular exedra in its south wall and a semicircular pool on both its eastern and western end. Decoration: No information about the decorative scheme is available. Technology: The caldarium was heated by three furnaces: one directly underneath each alveus and one just east of the central exedra. The room to the north of the caldarium was also heated by a furnace, located underneath its pool. The three small rooms to the north had a hypocaust, but may have been indirectly heated. A sewer ran along the southern wall of the caldarium’s praefurnium.352 Construction technique: Opus africanum? Bathhouse surface: ca. 400 m2 GPS coordinates: 35°28’41.88”N, 6°16’2.35”E

C50. Lambaesis (Tazoult, Algeria), Small Baths south of the Asclepeium

C51. Mellaha (Libya)

348  Thébert (2003) 214.

349  Ballu (1915) 112–13. 350  Cagnat (1923) 86. 351  Thébert (2003) 214. 352  Visible in Cagnat (1923) tav. V.

Excavation history: The baths were excavated during the large-scale investigations of Lambaesis in the 1920s by the Service des monuments historiques de l’Algérie. Bibliography: Cagnat (1923). Construction: (D) No archaeological evidence permits a dating of construction, but Yvon Thébert considers the numerous semicircular apses and pools as indications for a late date, possibly 4th c.348 The three-apsidal caldarium resembles the caldarium of the Baths of the Hunters in the same city (C49). The overall plan is reminiscent of the 4th c. Valebuona Baths in Volaterrae (C31). The small single-person pool early in the heated section is also a popular late antique element, although the medical context of the sanctuary could account for the high number of singleperson pools. Continued use and abandonment: Unknown. Plan: The bathhouse of around 400 m2 was discovered south of the Temple of Asclepius, in the vicinity of another

Excavation history: In 1924, a farmer found the remains of a Roman building is his field just outside the oasis of Tajoura (Libya), east of the Libyan capital Tripoli. The excavation by the Italian Colonial Department of Antiquities identified it as a small bathhouse, perhaps belonging to a villa or a hamlet. Bibliography: Bartoccini (1927). Construction: (B) On the basis of the ceramic finds and a coin of Constantius II, Renato Bartoccini dated the building to the second half of the 4th c. Continued use and abandonment: Unknown. Plan: Only part of the bathhouse was excavated, so it is difficult to identify the overall plan and locate the entrance. According to Bartoccini’s description, 4 rooms were discovered. The two rooms to the north, as well as the room just to the south-west of these, were unheated. A small

328

figure 87 Sanctuary of Asclepius in Lambaesis (Algeria). The Small South Baths (IV) and the Sanctuary Baths (in shaded box) can be identified after Cagnat (1923) tav. V

Gazetteer

329

Gazetteer

C51 Mellaha (Libya), (After Bartoccini (1927) 240, unnumbered fig.)

single-person semicircular pool was located on this room’s southern end (1 m deep). A doorway in the west wall led into a heated room, but the walls were too damaged to reconstruct the heating system. Decoration: The easternmost cold room and the room just to its south had polychrome mosaic floors with geometric pattern. On the threshold between these two rooms, there was also a mosaic of geometric design. Technology: The small cold pool was drained by a sewage channel that joined the sewage channel from the room just in front of it. No other information can be found in the excavation report. Construction technique: Opus caementicum Bathhouse surface: Unknown (at least 30 m2) GPS coordinates: Exact coordinates unknown

C52. Oued Athmenia (Algeria), Baths of Pompeianus

Excavation history: In 1877, a large bathhouse was found by peasants in a field around the modern village of Oued Athmenia (Algeria). The complex was excavated by the Société archéologiue de la province de Constantine. The same society also re-investigated the remains in 1928–1929. Bibliography: Poulle (1878); Krencker et al. (1929) 224; Alquier and Alquier (1930); Thébert (2003) 218–20. Construction: (C) On the basis of the mosaics, the building could be dated to the end of the 4th or the beginning of the 5th c.353 Continued use and abandonment: (C) The horse-mosaic in the caldarium is sometimes dated to the early 6th c. on the basis of mosaic style and of the typology of the depicted building.354

353  Dunbabin (1978) 267. 354  Sarnowski (1978) 16.

C52 Oued Athmenia (Algeria), Baths of Pompeianus (After Poulle (1878) pl. XIX)

Plan: As no considerable Roman ruins were found in the vicinity, the excavators concluded that the baths must have belonged to a large villa that had yet to be discovered or that had possibly been destroyed during building works in 1856–1857.355 Other scholars believed that the southern part of the building constituted the actual living quarters;356 however, the small surface of this living space in comparison to the baths, makes such an interpretation unlikely.357 The impressive size of the complex convinced most scholars that this bathhouse also had a public use. It is interesting to notice that the original excavation report does mention ‘quelques bourgades composées d’un petit nombre de maisons’ (‘some hamlets consisting of a small number of houses’) and that several latifundia were found further along the road to Sétif.358 These baths may have served a rich landowner and a small rural community, as may have been the case in Sidi Ghrib (see below, p. 345–347). The most noticeable feature of the plan is the large apsidal pool of the frigidarium. This remarkable construction actually consisted of a smaller apsidal piscina (4 × 3.4 m) surrounded by a larger natatio (3.5 m in width). Both pools are enclosed by an ambulatory (2 m in width) that may have looked onto a garden. Fikret Yegül sees the pavilionlike layout of the piscina-in-natatio as a reference to certain 355  Poulle (1878) 435, 447. 356  Alquier and Alquier (1930) 290–91. 357  Thébert (2003) 218. 358  Poulle (1878) 435.

330

Gazetteer

figure 88 Plan of the Baths of Pompeianus in Oued Athmenia (Algeria) depicting the mosaic floors in Yegül (1992) 248, fig. 297 after an unpublished drawing

water sanctuaries or island pavilions.359 The frigidarium was the heart of the building, while the square room south of it and the ambulatory may have fulfilled the function of the oblong rooms encountered in other late baths (see C49; C61). The heated section is more elaborate than those of the baths discussed so far, especially with addition of a round sudatorium at the beginning of the bathing itinerary. The caldarium probably had two alvei, one of which was semicircular, while the other could have been rectangular.360 It is difficult to interpret the protruding room with convex sides on the northern side as an apodyterium361 if we accept that the main entrance was not situated on the northern side, but rather on the southeastern side. The room with lateral apses, called the tablinum by Alquier and Alquier, may have been the actual entrance with dressing rooms. Thébert even wonders if the large open space, known by excavations to be unpaved, should be identified as a palaestra.362 In fact, the entire southern section was probably a part of the baths, rather than living quarters. An alternative interpretation could be that this part was restricted to a specific clientele or the masters of a domus. The open court may have been a garden in this case.363 359  Yegül (1992) 249. 360  Krencker et al. (1929) 224; Thébert (2003) 219. 361  As interpreted by Poulle (1878) 444; Alquier and Alquier (1930) 300. 362  Thébert (2003) 219. 363  Krencker et al. (1929) 224.

Decoration: The decorative schemes merit some attention (fig. 88). The floor mosaic in the centre of the frigidarium depicted a large estate with a tower and walls enclosing a garden. The lower registers are presenting a hunting scene. The panel is oriented towards the heated section, which could support the hypothesis that bathers entered through the western entrance. On the lateral sides of this panel, located in the recesses created by the Corinthian columns, the scenes of marine creatures with Nereids on their backs recall similar representations in the Sidi Ghrib baths. In the room with the rectangular alveus, a scene of a domina seated between trees and pastures is accompanied by the text ‘FILOSOFO FILOLOCUS’ (‘the most beloved location for the philosopher’). On the lower register, a ‘SEPTUM VENATIONIS’ (‘park for hunting’) announces the scene in the frigidarium, while a ‘PECVARI LOCUS’ (‘the corral’) fits within the overall pastoral theme of the mosaics. In the mosaic of the caldarium, the race horses of Pompeianus, and possibly the stables are shown. All 6 horses are named: ALTUS’ (‘the powerful’), ‘PULLENTIANUS’ (‘the stallion’), ‘DELICATUS’ (‘the delicate’), ‘SCHOLASTICUS’ (‘the wise’), ‘TITAS’ (‘the giant’) and ‘POLYDOXUS’ (‘the victorious’). There are special words of appreciation for the latter: ‘vincas non vincas, te amamus, Polydoxe’ (‘Whether you win or not, we still love you, Polydoxus’). The name of the presumed owner of the baths, Pompeianus, is spelled out above the scene. In the other rooms, intricate polychrome geometric mosaics embellish the floors. The concentric circles of the mosaic in the round floor must have reflected

Gazetteer the optical effect of the dome covering the room. The excavators also stress the large quantities of marble that were recovered at the site. Door jambs, plinths and the walls were covered in mainly white marble slabs. Technology: The baths functioned according to the normal Roman standards. The suspensurae (wrongly used to denote wall heating by Poulle) are built with brick pilae, while the wall heating was ensured by tubuli.364 In an interesting passage, the excavators mention that they found a stock of 200 to 300 of these suspensurae (sic)—resembling bottomless beer pots according to Poulle—in an unspecified room of the baths.365 As there are no pictures of the find, it is hard to confirm whether these terracotta elements were actual tubuli or were tubi fitilli. The presence of such a ‘stock’ might be due to parts of the wall heating being restored or as ‘back up’ to make ad hoc repairs of malfunctioning parts. It could also have been the ‘leftovers’ of a building phase. The water management of the site is also well known. The water came from two large reservoirs located west of the baths, which were presumably fed by a small source. By underground supply channels, the water was sent directly to the piscina and natatio. A separate branch of this supply channel led to the open courtyard, probably to water plants or feed a fountain.366 The alvei may have been fed by a separate water channel. The lead pipes that led the water in the heated pools were still visible at the moment of excavation. Water disposal was accomplished through a network of sewers, passing by a small latrine, and eventually discharging into a nearby stream. Construction technique: Opus mixtum Bathhouse surface: ca. 1,600 m2 (including courtyard) GPS coordinates: Exact coordinates unknown Ouled Hafouz (Tunisia) Excavation history: During construction works of a private residence in the town of Ouled Hafouz (Tunisia), a large mosaic of Leda (?) resting on hypocaust pillars was found. A small salvage excavation was carried out, but remains unpublished, except for a description of the mosaic. Bibliography: Jeddi (1994). Construction: (C) The baths seem to have had several building phases, as traces of an earlier mosaic were found beneath the Leda mosaic. The style of the Leda mosaic points 364  Poulle also mentions ‘suspensurae’ (i.e. tubuli) in the corners of the frigidarium, presumably to mildly heat this cold room (Poulle (1878) 445). The presence of these tubuli is difficult to explain. As it is hard to imagine that these tubuli may have affected the temperature in such a large room, they should perhaps be interpreted as simple constructional elements. The use of tubuli as building material in walls and steps has been attested in several sites, see for example Ostia (Baths of the Swimmer, Small Baths on the Via Severiana). 365  Poulle (1878) 443–44. 366  Poulle (1878) 453.

331 to a date between the 4th and 6th c., but with a preference for the latter. A comparison with the style of the Early Umayyad paintings in Qusair Amra has been proposed, hinting at a 7th c. date.367 Continued use and abandonment: Unknown. Plan: The limited size of the excavations provides little knowledge about the actual building. Decoration: The mosaic represented a nude goddess (Venus?) with a Cupid and Leda holding a swan. Text was inserted at the bottom of the scene (EA-63), asking ‘why do you not take a bath?’ Technology: The hypocaust pillars were made of standard square tiles, but no other information is available. Sila (Bordj al-Ksar, Algeria), Baths of Skylla Excavation history: Little is known about the small town of Sila, 32 km south-east of Constantine. The town may have been a castrum of Cirta. It seems to have been the seat of a bishop during the 5th c. and may also have hosted a Byzantine fortress.368 In the early 20th c., a small bathhouse was found in the lower part of the site. Bibliography: Gsell (1905). Construction: (C) A mosaic was dated to the 4th or 5th c., although it is difficult to determine if it belonged to the original construction phase.369 Continued use and abandonment: Unknown. Plan: There was at least one rectangular piscina (4 × 3.8 m) and two small rooms on a hypocaust, heated by a single furnace. Decoration: The most important feature of the building was a large mosaic, of which only a third has been preserved (4.68 × 6.45 m). It depicted Skylla in a marine setting, including Nereids, dolphins and seahorses. The piscina had a plain white mosaic floor. 367  Jeddi (1994) 278. 368  Pringle (1981) 300. 369  Dunbabin (1978) 249; Wiedler (1999) 250. Already Stephane Gsell had proposed a late antique date, with a scorn that was typical for the scientific climate: ‘La mosaïque que nous venons de décrire est d’un bien mauvais art; le dessin en est incorrect, la technique grossière et négligée, la couleur assez terne; les visages sont sans expression, les détails des corps et des draperies indiqués par des lignes raides et disgracieuses. Ce morceau doit être du troisième ou du quatrième siècle. C’est l’ouvrage d’artisans maladroits, à peine digne des thermes du chef-lieu de canton qu’était Sila. Cependant l’image de Scylla figurée au milieu de ce pavement, lui donne un certain intérêt.’ (‘The mosaic we just described is artistically quite bad; the drawing is incorrect, the technique is crude and sloppy, the color is quite dull; the faces are expressionless, the details of the bodies and draperies are indicated by stiff and disgraceful lines. This piece must be of 3rd or 4th-c. date. It is the work of clumsy craftsmen, barely worthy of the baths of the canton capital that was Sila. However, the image of Scylla depicted in the middle of this floor gives it a certain interest.’) (Gsell (1905), 7; translation by the author).

332

Gazetteer

Technology: The excavation report mentions no special features of the hypocaust or the water related elements of the baths.

C53. Sitifis (Sétif, Algeria), Baths of the 5th c.

Excavation history: A small bathhouse was discovered during the construction of a park in Sétif, with excavations carried out between 1977 and 1984. Bibliography: Mohamedi and Fentress (1985); Fentress (1989); Mohamedi et al. (1991); Thébert (2003) 223–24. Construction: (A) Originally, a Roman bathhouse occupied an entire insula until it fell out of use in the 5th c. An occupation layer of 40–50 cm accumulated in the abandoned building in the period between AD 410 and AD 425.370 The old walls were torn down and a new bathhouse in opus africanum was built on the southern part of the insula. Continued use and abandonment: (B) It seems to have been in use until the end of the 6th c. or the first quarter of the 7th, when rubble covered the mosaic of the frigidarium.371 Pushing the use of the baths into the Islamic period is difficult, as evidence for continued bathing practise is lacking.372 Plan: One entered the baths by an oblong vestibule in the east. No trace of paving was found. The sturdy eastern wall of the vestibule suggests that it was covered by a shed roof sloping down to the east. To the north, a doorway led into a rectangular room, which was severely damaged on its northern side during excavations. However, a trench could be identified here, dug into the foundation layers, possibly to construct some steps leading down into an open area. The zone to the east of the baths, which had a thin mortar floor covered by a layer of sand, was identified as a palaestra.373 From the vestibule, a large doorway to the west, flanked by a pair of columns, gave access to the frigidarium. The absence of sockets in the threshold makes it unlikely that there was a door here. The room (8.9 × 8.8 m) had two apsidal pools, one to the north (3.6 × 3.2 m) and one to the south (3.6 × 2.8 m). They were both 0.7 m deep. In the 4 corners of the room, cubiform column bases were found. These must have been additional supports for the cross vault covering the room.374 From the cold room, one passed into to a heated section through a doorway in the west side. In contrast to the doorway east, the latter was not marked by a meander mosaic before the threshold. The heated section consisted of only a single room on a hypocaust. The pillars alone withstood the ravages of time, making it 370  Fentress (1989) 322. 371  Mohamedi and Fentress (1985) 471. 372  E. Fentress, personal communication. 373  Mohamedi et al. (1991) 63. The presence of a palaestra would be rather strange, as the combination of baths and a sporting ground is uncommon in late antique contexts (see chapter 4). 374  Mohamedi et al. (1991) 85.

C53 Sitifis (Sétif, Algeria), Baths of the 5th c. (After Mohamedi et al. (1991) 60, fig. 13)

difficult to identify a possible pool. The walls were probably heated, although the excavation report does not mention the presence of tubuli. The possible absence of a hot water pool could mean that the heated section only had the function of a sweat chamber. A portable brazier may have created additional heat. In the middle of the lower part of the south wall, the characteristic open arch reveals the presence of a furnace. It was heated from an L-shaped praefurnium. Its floor, made out of battered earth, was lying 1.2 m lower than the floor of the frigidarium. Decoration: The northern room giving access to the presumed palaestra had a geometric mosaic. The floor of the cold room was embellished with a polychrome mosaic. The central emblema (3.68 × 3.77 m) depicted Venus emerging from the water, carried by two marine creatures and surrounded by fish. Two flying amores spread a cloth above the head of the goddess, while in the bottom corners two fishermen can be discerned.375 The border of the mosaic consisted of a geometric pattern imitating opus sectile. The diagonal pattern of the geometric motives, as well as the diagonal composition of the emblema reflected the diagonal structure of the cross vault. The wavy patterns of the mosaics on the bottom of the pools were meant to highlight 375  For a full description, see Mohamedi et al. (1991) 73–77.

333

Gazetteer the water surface.376 The mosaics against the sides of the pools imitate marble slabs, especially pavonazetto. In the destruction layers of the frigidarium, several fragments of mural paintings were found. Again, marble slabs were imitated. The popularity of marble imitation is attested in other late antique buildings in Sitifis.377 The vault was probably painted sky blue. It is remarkable that floral patterns, a recurrent theme in 5th c. mosaics, were absent. Technology: The heating system was very basic. A single furnace heated the hypocaust of the small caldarium. A type of wall heating was also present. This can be inferred from the walls of the room having a continuous mortar coating. If there had not been some kind of wall heating, the level of the suspended floor would have been visible against the walls.378 The ruinous state of the caldarium did not permit the locating of an alveus. Furthermore, there is no sign of masonry that supported a boiler in the praefurnium. If there was a small pool, it was heated only by direct contact with the furnace. The water supply is unclear. To the west of the northern piscina, a reservoir was supported by a vaulted room (internal dimensions: 4.1 × 3 m). Its location next to the pool points to a direct connection between the two, although all traces of piping have disappeared. It is also unclear how this reservoir was filled. A well from the preceding period to the east of the baths may still have been in use.379 Waste water from the piscinae were collected by a channel starting at the bottom south of the northern piscina and running south under the frigidarium. It connected to the drain of the southern pool and left the baths under the south-east corner, to join the sewer in the street south of the baths. Construction technique: Opus africanum? Bathhouse surface: ca. 108 m2 GPS coordinates: Exact coordinates unknown

C54 Sitifis (Setif, Algeria), Baths of the Triumph of Venus (After Février et al. (1970) fig. 28)

Excavation history: The baths were discovered and excavated in 1958–1966 by the Service des monuments historiques de l’Algérie. Bibliography: Lassus (1962); Thébert (2003) 223. Construction: (C) The Bath of the Triumph of Venus was dated on the basis of its mosaics, pointing to a late 4th or early 5th c. date. Two churches found in its vicinity were dated by funerary inscriptions to similar dates. The geometric motif on the panel border in the frigidarium is exactly the same as the mosaic pattern embellishing

a tomb in one of the churches, dated by an inscription in AD 389–412.380 Continued use and abandonment: Unknown. Plan: This small bathhouse fitted within a rectangle of 12 × 9 m. It is unclear whether the bath was connected to the structures to the east, which may have been a domus, or if it was accessible directly from the street. It consisted of only three rooms, arranged on a single north-south axis. The northernmost room (5.5 × 5.3 m) was obviously the most important one. We can recognize a frigidarium with a single cold pool in its south-eastern corner. By a doorway in the south-western corner, one entered the heated section. The first room was rectangular and heated by a single furnace in the east side. By a doorway in the centre of its southern side, one could reach the second heated room with its trefoil plan. There were probably small pools in one or more of the three apses. The only furnace was also located to the east. Decoration: The frigidarium floor was embellished with a well-preserved mosaic of Venus and amores.381 As the mosaic was meant to be seen from the east, Yvon Thébert

376  Mohamedi et al. (1991) 77. 377  Mohamedi et al. (1991) 84. 378  Mohamedi et al. (1991) 61. 379  Fentress (1989) 329.

380  Lassus (1962) 102–104. 381  For a description, see Lassus (1962) 104.

C54. Sitifis (Sétif, Algeria), Baths of the Triumph of Venus

334

Gazetteer

suggested that the main entrance to the baths must have been located in the east side of the building.382 Technology: There is no information on the water supply or disposal. The heating system seems to have been restricted to two furnaces, each heating the hypocausts of a warm room. Construction technique: Opus africanum Bathhouse surface: ca. 350 m2 GPS coordinates: Exact coordinates unknown

C55. Sufetula (Sbeitla, Tunisia), The Baths of Basilica 2

Excavation history: The baths were unearthed in 1913, when they were interpreted as the private baths of a house.383 Bibliography: Merlin (1915); Duval (1971a) 304–309; Thébert (2003) 157–59. Construction: (D) The bath has the same orientation and is built on the same late antique street level of Basilica 2, making it likely that it was built in the same period.384 The basilica is usually dated at the end of the 5th or beginning of the 6th c., but hard archaeological evidence is lacking.385 Continued use and abandonment: Successive restorations can be discerned, but are difficult to date. Plan: The Baths of Basilica 2 are located just north of the Basilica of Vitalis, along one of the cardines leading to the north of the city (map 23). The building was erected in opus caementicium using local limestone or sandstone, with some restorations made in brick. The plan fits within a rectangle of 27 × 20 m for a total surface of around 520 m2. There were two entrances to the vestibulum, with the west one cordoned off by a wall and small fountain to impede indiscrete looks of the passers-by. From here, a narrow passage led to the latrines, which were also accessible by a door giving directly onto the street. The large paved vestibulum (4.3 × 16.5 m) gave access to the frigidarium. This large room (15.1 × 6 m) had one large rectangular pool (6 × 4.2 m) and a smaller one with apsidal ending (2.7 × 2 m). A small room (3.3 × 2.8 m) in the north-east corner of the room may have been used as dressing room or massage-room.386 Continuing south, a succession of 4 heated rooms awaited the bathers. The first room (3.8 × 3.1 m), accessed through a narrow door (0.96 m wide) to reduce heat loss, was indirectly heated by the hypocaust of the neighbouring room. The hypocaust of the apse to the south was removed at a later stage. The room immediately east (4.8 × 3.1 m) had a suspended floor and tubuli, indirectly heated by the furnace of the room to the south. The apse of the eastern end was deprived of a hypocaust, but was equipped 382  Thébert (2003) 223. 383  Merlin (1915) 37; due to this early interpretation, the baths were included as private baths in de Haan (2010) 351 (A. 254). 384  For remarks, see Thébert (2003) 157–58. 385  Duval (1971a) 292. 386  Duval (1971a) 307.

C55 Sufetula (Sbeitla, Tunisia), Baths of Basilica 2 (After Duval (1971) dépliant VII)

with a small pool (1.1 × 0.96 m; 0.7 m deep), accessed by a single step. The third heated room had two heated pools, the eastern apsidal one (3.2 × 2.7 m) being later walled up to function as a reservoir. The western rectangular pool (2 × 3 m) was located directly above the furnace heating the entire room. The last heated room also had two heated pools, both rectangular but of different dimensions. The small one in the south (1 × 1.2 m) was only large enough to accommodate one person at a time, while the eastern one (2.1 × 2 m) could welcome more people.387 Both were located directly above a furnace. Decoration: The decoration, poorly preserved, seemed to be quite inconspicuous. The large vestibulum was paved with slabs of local limestone. The second heated room originally had marble paving. The pools were covered with plain white mosaic. Technology: The water supply of the baths is unknown; however, an aqueduct probably reached the city in the north, and the Wadi Sbeitla was situated nearby. A large reservoir was located just south of the latrines, and must have stored the water of the pools and the fountain in the

387  Duval (1971a) 309.

335

Gazetteer latrine.388 The transformed pool in the third heated room probably supplemented storage capacity when the small plunge was built in the second heated room. Waste water disposal is badly understood. A drain pipe was observed in the south-east corner of the apsidal piscina of the frigidarium, continuing west under the cold room to join the sewer of the latrines.389 These probably emptied into the street sewer. The drain of the small plunge in the second heated room turned south, where it probably joined the drains of the pools of the fourth heated room. The heating system seems to have been badly damaged even before the clearing works of 1913. The pilae of the hypocaust were made out of square tiles. The dimensions or form of the tubuli have not been published. The fourth heated room was arguably the hottest, as it had two furnaces and faced directly south. Furthermore, its surface was smaller than the third heated room, which was only heated by a single furnace. The pool of the latter must have been of a more temperate heat, as it was larger than the two pools of the fourth room. The second heated room was indirectly heated and must have been slightly warmer than the first, which was situated even further from the furnace. The first room was also set at a right angle to the other three rooms, while the passage to the second room was narrow. This ‘peripheral position’ of the first room made it a perfect ‘heat lock’ or buffer between the hot and the cold sections. Construction technique: Opus caementicium Bathhouse surface: ca. 590 m2 GPS coordinates: 35°14’32.28”N, 9° 7’9.75”E

C56. Sufetula (Sbeitla, Tunisia), Baths nr. 26

Excavation history: The baths were partially excavated in 1950 and 1954 and restored in 1969, but these interventions were never published. Bibliography: Merlin (1946–1949); Thébert (2003) 157. Construction: (C) The baths are located in the south-east quarter of the site, in a part that saw important building activity from the 4th c. onwards (map 23).390 The construction level and the mosaics point to a date in the 5th or possibly 6th c.391 Continued use and abandonment: Unknown. Plan: The building was small, fitting in a rectangle of 22 × 24 m. Due to their small size, these baths have sometimes been interpreted as private baths.392 However, no house has been discovered in the vicinity. The street access also favours a public facility. It faced north-west, probably accessible from a side-street of the main road. One entered through an 388  Duval (1971a) 309. 389  Indicated on plan VII in Duval (1971a). 390  The Arch of Diocletian was built at the eastern end of the main road that crossed this part of town. 391  Thébert (2003) 157. 392  Picard (1949) 633; Duval and Baratte (1973) 102.

C56 Sufetula (Sbeitla, Tunisia), Baths nr. 26 (After Thébert (2003) 602, pl. LII.2)

oblong hall and a small room that formed an entrance ‘en baïonnette’. A wall prevented the person entering from directly looking into the bathing part. From here, one entered a large hall, partially paved, which was divided into two parts by a pair of double columns. The southernmost part was equipped with an apsidal cold-water pool (ca. 2.6 × 3 m). From this frigidarium, two doorways led south-east. The easternmost opened onto a small furnace room, servicing a pool. The fornacator apparently had to pass through the cold room to fire up the furnace.393 The semicircular pool (ca. 1.5 × 3 m) was reached by taking the southernmost doorway and may have been constructed at a later date.394 Continuing south, a rectangular room with no further amenities was followed by an almost square room with apsidal heated pool, directly heated by a furnace. Decoration: On the inner wall of the piscina, a polychrome mosaic represented fish and other marine animals. The figurative band was bordered by a simple ribbon pattern below and a stylized wave pattern on top. The bottom was paved with limestone slabs (observation by the author on site). Technology: The two furnaces that were discovered heated alvei. There is no information about a possible hypocaust or wall heating. The water management of the baths also 393  We must assume that this praefurnium was only accessible through the frigidarium, as no other doors have been found. If the pool was a later addition, the praefurnium must have been installed in an existing room (an unctorium?) because this must have been the easiest solution (instead of piercing a furnace through the eastern wall). 394  Duval and Baratte (1973) 103.

336

Gazetteer

remains in question. The pool of the frigidarium may have been fed by a reservoir lying directly north-east of it, but the structure is too badly damaged to confirm this. Construction technique: Opus africanum Bathhouse surface: Unknown (at least 530 m2) GPS coordinates: 35°14’16.36”N, 9°7’13.79”E Sufetula (Sbeitla, Tunisia), Baths near Basilica 5 Excavation history: Unknown. Construction: (D) The baths can be attributed to Late Antiquity on the basis of their plan and their location, along the same road as the Baths nr. 26, just north of the Byzantine fort nr. 27 and some 20 m east of basilica 5 (map 23, nr. 4). Bibliography: No publication available. Continued use and abandonment: Unknown. Plan: The building was centred on a rectangular frigidarium with two pools, one rectangular and one apsidal.395 Decoration: The floor and walls of the rectangular pool were covered with white mosaics. Technology: Unknown.

C.57. Thamugadi (Timgad, Algeria), Small North Baths

Excavation history: These baths were excavated in 1902 by the Service des monuments historiques de l’Algérie. This small bathhouse was located along the cardo maximus near the northern gate of the ancient walls. Bibliography: Ballu (1903a); Germain (1969) 11–12; Thébert (2003) 229–30. Construction: (C) On the basis of mosaics, a 4th or even 5th c. date is proposed.396 It is difficult to determine if these mosaics belong to the original building phase. Continued use and abandonment: Unknown. Plan: The bathhouse was implanted in an existing insula of 20 × 20 m, resulting in the modest surface of 400 m2. While located along the cardo maximus, the principal entrance of the baths lay on the east side, from a small cardo parallel to the city’s main north-south axis. At the time of digging, the eastern part of the building had collapsed into the underlying cellar, making a reconstruction of the rooms difficult. The frigidarium was probably square (s: 8.5 m) and had a rectangular pool (ca. 4 × 3 m). The apodyterium may have been south of the cold room. Flanking the piscina to the east and the west may have been rooms for massages or anointing. By a doorway in the west side, one entered the heated section, first by passing through a small vestibule with a square room of unknown function to the south. The first warm room (6.25 × 4.35 m) was heated by a single furnace to the north-west. A doorway in the west side gave access to a similar heated room (6.25 × 4.7 m) with a single 395  Béjaoui (2004) 24. 396  Germain (1969) 12.

C57 Thamugadi (Timgad, Algeria), Small North Baths (After Thébert (2003) 649, pl. XCIX.3)

furnace. The third heated room (6.77 × 4.13 m) was connected to the other two rooms and had two heated pools. The semicircular one (r: 2.5 m) in the south was heated by two furnaces, while the rectangular one (ca. 2.1 × 5 m) in the west by only one. The elliptic form of one of the heat channels makes clear that a boiler was placed on top of it.397 A service corridor ran along the west side of the heated rooms. Another corridor running north-south harboured a small latrine and was accessible from the outside by a door to the north. At its southern end, the corridor turned east towards the praefurnia of the second and third heated room. A small door in the south side of the building joined this part of the corridor, passing a second larger latrine to the east. Decoration: The floor of the piscina was embellished with a simple black mosaic with coloured circles. The small vestibule had a plain white mosaic floor. On the floor of the semicircular alveus, a mosaic with a geometric pattern and a white background was found. Technology: Little is known about the heating system, except for the furnaces and connected hypocausts described above. Wall heating can be supposed for all three rooms, yet the excavation reports make no mention of specific techniques. A water conduit is known to pass along the cardo maximus, although its trajectory at the northern end of the street is unclear.398 The sewers underneath the baths are partially known. They drain the pools and pass under the latrines.399 Construction technique: Opus africanum 397  Thébert (2003) 230. 398  Lohman (1979) 168, fig. 1. 399  Indicated on pl. VIII in Ballu (1903b).

337

Gazetteer Bathhouse surface: ca. 500 m2 GPS coordinates: 35°29’11.55”N, 6°28’7.72”E

C58. Thamugadi (Timgad, Algeria), Small North-east Baths

Excavation history: The complex was discovered in 1907 and subsequently excavated by the Service des monuments historiques de l’Algérie. Bibliography: Ballu (1908); Thébert (2003) 234–35. Construction: (C) The only evidence for dating this complex is a set of Corinthian capitals in a Greek-Byzantine style. Even if these could have been a later addition, they inform us that this building was still occupied in the 6th c.400 Continued use and abandonment: Unknown. Plan: These small baths were located on the north-east angle of the old walls. The bathhouse was built after the walls lost their defensive function, as it was built against them. It is difficult to assess the articulation of the baths within the urban fabric. The area north of the baths, just outside the walls, has not been investigated. To the south, the bathhouse was built against an insula. These modest baths covered a surface of around 650 m2 and were structurally divided into eastern and western wings. The entrance lay at the northern end of the so-called ‘east boulevard’, the street running between the easternmost insulae of the town and the east wall. Before climbing the stairs to enter the baths, a door immediately to the east of these stairs gave access to a square praefurnium in which three furnaces of the east wing were identified. Taking the stairs led to a narrow corridor running east-west. At the western end probably lay some utility rooms and the L-shaped service corridor, narrowed by a semicircular pool protruding out of the heated rooms. By taking a passage immediately north of the stairs, one entered the actual bathing rooms. A rectangular room east of this passage may have served as an apodyterium or unctorium. The frigidarium was a large hall (13.2 × 6.7 m) paved with limestone, with a rectangular pool to the south and a smaller apsidal pool (ca. 2.3 × 2.2 m) to the east. A doorway in the west side of the room gave access to the heated area. The first two rooms lay on a north-south axis and were heated by a single furnace in the south-western corner of the southern room. From the south-east corner of the latter, a passage led back to a square room which was connected both to the piscina and the frigidarium. The third and fourth heated rooms were accessible by a narrow doorway in the north-west corner of the second room. The third room was heated by a single furnace to the south, while the fourth room had two furnaces, one of which was located under the semicircular pool in the south side. From the frigidarium a small passage just north of the apsidal 400  Thébert (2003) 234.

C58 Thamugadi (Timgad, Algeria), Small North-east Baths (After Thébert (2003) 652, pl. CII.2)

piscina led to the east wing. By passing a small latrine and descending some stairs, one arrived in an irregularly shaped room. The strange shape is due to the original town wall curving to the south at this point. On the south end, a doorway gave access to a long corridor (18 × 3.5 m) that had a separate street entrance through the city wall. The southern end of the corridor and a southern room have unclear delimitations and unclear functions. The bathing rooms consist of a set of three heated rooms to the west of the corridor. The connection between these rooms and with the corridor are uncertain. Yvon Thébert reconstructs doors between these three rooms and a door between the southernmost room and the corridor.401 We would have the north-south succession of a caldarium (4.15 × 5.6 m), a sudatorium (2 × 32 m) and a tepidarium (4.2 × 5.6 m). There is no indication of a pool in this eastern wing of the baths. One could wonder if these rooms were actual bathing rooms. Interpreting the eastern wing as the women’s baths does not resolve the issue of the pools.402 Perhaps this wing was used for the ‘peripheral’ activities of bathing, such as anointing, massage, depilation, etc. Decoration: The excavation report mentions few details concerning the decorative scheme of these baths. The Corinthian capitals in ‘Byzantine’ style merited mention.403 The frigidarium was paved with large limestone flagstones, while the only mosaics were found in the middle of the three heated rooms of the east wing. Technology: Except for the hypocausts found in the heated rooms, little is known about the heating systems in these baths. Only in the small ‘sudatorium’ in the east wing has wall heating been recognized, although it is not specified if these were tubuli or a different type of system. The limited width of the doorways between the heated and the cold 401  Thébert (2003) 235. 402  Thébert (2003) 234. 403  Ballu (1908) 245.

338

Gazetteer

figure 89 Plan of the House North of the Capitolium in Thamugadi (Algeria) with adjacent bathhouse (in box) after Germain (1969) 106, fig. 15

C59. Thamugadi (Timgad, Algeria), The Baths of the House North of the Capitolium

C59 Thamugadi (Timgad, Algeria), Baths the House North of the Capitolium (After Thébert (2003) 663, pl. CXIII.3)

sections of both the east and west wing reveals the typical concern of heat retention. There is no information about the water supply. Ballu mentions a reservoir near the large piscina, but it seems he is referring to the square room west of it.404 Nothing is mentioned of water disposal. There are no traces of water pipes in the north-east area of the city and most insula in the vicinity had their own well.405 Yet, it is doubtful that the piscina could have been filled without the supply of an aqueduct. Construction technique: Opus africanum? Bathhouse surface: ca. 650 m2 GPS coordinates: 35°29’12.66”N, 6°28’13.24”E

404  Ballu (1908) 244. 405  Lohman (1979) 168, fig. 1.

Excavation history: A large complex, perhaps a house, with a separate bathhouse, was discovered by the Service des monuments historiques de l’Algérie in 1915. Bibliography: Ballu (1916); Germain (1969) 106–114; Thébert (2003) 250–51. Construction: (C) The mosaics in these baths point to a late 3rd or an early 4th c. date.406 The plan of the baths strongly recalls the Baths North-West of the Theatre in Bulla Regia (see above, C36). The latter were dated in the 4th c., which would corroborate the date proposed for this bathhouse. Continued use and abandonment: Unknown. Plan: These baths were located outside the old city, in the region that developed west of the centre, 130 m north of the capitolium. As the bathhouse lay next to a large structure, perhaps a house, they were interpreted as private baths (fig. 89). However, the baths had a separate entrance from the street, which implies their use was not restricted to the users of the neighbouring edifice.407 One entered the baths through a portico along the street. A narrow passage led to a door that gave access to an oblong room (14.7 × 7.2 m), sometimes interpreted as an exercise area. A door in the south-east corner of this room may have given directly onto the street.408 A recess in the south-west corner and a small room in the north-west corner may have been used as storerooms for oil and towels. A large doorway in the western wall gave into the octagonal frigidarium (width of 7.2 m; s: 3 m). On 4 of the sides of the octagon, there were semicircular niches (r: 1 m), which 406  The 3rd-c. date is proposed by Dunbabin (1978) 275; Thébert (2003) 251 proposes the 4th c. 407  Contra Thébert (2003) 251 who proposes a ‘private’ function. 408  Thébert (2003) 663, pl. CXIII.3.

339

Gazetteer may have contained benches or statues. The south-western niche was a doorway that connected the cold room to a kind of courtyard south of the baths, in which a basin or possibly a pool (6 × 3 m) was located.409 The two pools of the frigidarium, one to the north and one to the south, were of a curvilinear design. A doorway to the west led to 4 heated rooms, all with a hypocaust. The almost square heated rooms (s: ca. 3.2 m) were arranged so that they formed a larger square, with each room connected to the two neighbouring rooms. Only the room connecting to the frigidarium was indirectly heated and had no wall heating. The southernmost rooms were equipped with heated pools placed directly above the furnaces. Decoration: The floor of the oblong room was covered with a terracotta mosaic. The frigidarium floor was covered in a geometric mosaic of black, white and yellow colours. On the threshold between the cold and the warm section, a mosaic depicted two pairs of bath clogs accompanied by the text ‘BENE LAVA’.410 The text was readable for the persons entering the warm rooms. For the people exiting these rooms back into the frigidarium, the words ‘salvum lavisse’ may have been spelled out, but the mosaic is too damaged to verify.411 A black and white geometric mosaic adorned the floor of one of the heated rooms, while the semicircular pool of the south-western room was embellished with marble slabs. The attention paid to the mosaics could reveal a public function of the building.412 Technology: The heated section should be interpreted according to the number and disposition of the furnaces. The warmest room must have been the north-western one, as the surface to be heated by the hypocaust and wall heating was rather small.413 The south-western room was heated by two furnaces, but had a larger surface due to its two pools. The south-eastern room must have been moderately heated, as the ratio of furnace to surface was the largest here. The north-eastern room was indirectly heated through the doorway with the south-eastern room and on the hypocaust level. The poor preservation of the walls at hypocaust level does not permit reconstruction of the connections between the hypocausts. There is no information about the water supply, although the combined volume of the pools, both heated and unheated, must have necessitated a considerable amount of water. Reservoirs could have been located on the roof. Water disposal is also unknown. A sewer that went around the south and west sides of the heated section, collecting the drainage from the pools, would seem the most logic choice. 409  Thébert (2003) 251. 410  Ballu (1916) 241. 411  Germain (1969) 116. 412  Germain (1969) 118. 413  Thébert (2003) 251.

Construction technique: Opus africanum Bathhouse surface: ca. 350 m2 GPS coordinates: 35°29’5.68”N, 6°27’55.55”E

C60. Thenae (Thyna, Tunisia), Baths of the Months

Excavation history: The baths were discovered in 1961 by construction workers and partially excavated in 1963 by the Direction des Antiquités de Tunisie. Bibliography: Fendri, M. (1964); Fendri, N. (1964); Thébert (2003) 161–62. Construction: (D) In the preliminary report, a lifespan between the 3rd and 4th c. was proposed. This conclusion was based on the discovery of a monetary treasure, comprising 320 coins of the period AD 297 to AD 311, which had been buried in a corner of the second heated room.414 Most of the coins showed little traces of use. As the treasure formed a homogenous unit, the excavators concluded that the treasure had been hidden at the beginning of the second decade of the 4th c., when the baths had fallen out of use or when they were restored.415 However, the room in which the coins were found was thoroughly reorganised. The hypocaust floor was taken out and a new system with wall heating was put in place. It is not specified where the treasure was found, making the archaeological context unclear. Yvon Thébert proposes a construction date in the late 3rd / early 4th c.416

C60 Thenae (Thyna, Tunisia), Baths of the Months (After Fendri (1964) plan 2) 414  Fendri, N. (1964) 59–67. 415  Resp. Fendri, M. (1964) 57; Fendri (1985) 161. 416  Thébert (2003) 161–62.

340 Continued use and abandonment: (C) The mosaics point to a long use of the building, judging from ‘le caractère nettement « décadent » des mosaïques de seconde main par rapport aux pavements initiaux’. (‘the slightly “decadent” character of the second-hand mosaics in comparison to the original pavement’).417 Plan: We can note that the plan has the same type of large hall adjacent to the frigidarium as found in the Baths of the Hunters in Lambaesis (C49). It could be accessed by a separate entrance to the west, passing a latrine, and perhaps also from a rectangular room to the east (apodyterium?). The frigidarium was reached by passing through several rectangular rooms of unknown function, the largest of which (north of the piscina) might have been the apodyterium. The actual frigidarium had a rectangular piscina which was flanked by two columns on its northern and southern side. The small single person pool was not located at the beginning of the heated section (e.g. below C62), but inside the frigidarium itself. According to the excavators, the water of this small pool was mildly heated by a boiler in a small room north-west of the frigidarium. The boiler was supposedly heated by a channel coming from the hypocaust of the caldarium. The heated rooms were accessed by a small doorway in the western side of the frigidarium. The first room was a tepidarium, as no wall heating was discovered, only chimney flues in the wall opposite the furnace. The second heated room had no specific features (sudatorium?), while the third room had a rectangular alveus along its western wall. A small sudatorium with its own furnace was constructed at the end of the bathing route (north-west of the caldarium). Decoration: The floors of the cold section were embellished with mosaics (geometric and floral designs), while the walls were painted with figurative scenes of women and the ceilings with geometric paintings and marble imitations. The long use of the baths can be deduced from the several layers of wall paintings in these rooms (up to 4). Around the piscina, mosaic medallions with marine scenes (Nereids on dolphins) were depicted. The heated rooms were embellished with marble slabs against the walls, sometimes running up until the start of the vault. In the caldarium, the mosaic was found that lend its name to the baths, depicting the personifications of the months January to April. Instead of window glass, lapis specularis was used. Technology: Each heated room had its own furnace. The hypocausts consisted of the standard square pillars. Only the tepidarium did not have tubuli wall heating. It seems that the heated section was thoroughly reorganised at a certain point, when the second heated room was equipped with wall heating and the caldarium was enlarged, possibly by

417  Fendri, M. (1964) 55; translation by the author.

Gazetteer creating the alveus.418 The boiler disposition in the room north-west of the frigidarium is a unique feature. The water in the boiler seems to have been heated by caldarium hypocaust and must have been intended for the small single person pool in the frigidarium. This small pool with tepid water is reminiscent of the similar pool in the second cold room of the baths at Theveste (see below, C65). The water for the pools was, at least partially, supplied by rooftop reservoirs collecting rainwater. The pools were emptied by outlets connecting to a sewer that passed the latrine before entering the city sewer. Construction technique: Opus caementicium Bathhouse surface: ca. 900 m2 GPS coordinates: 34°39’0.83”N, 10°40’59.70”E

C61. Thenae (Thyna, Tunisia), Small South-east Baths (‘thermes des pugilistes’)

Excavation history: A bathhouse was discovered when a small test pit was opened near the Roman city wall of Thenae in 1954. It was subsequently excavated by the Direction des antiquités de Tunsie and the Institut des Hautes-Études de Tunis. Bibliography: Thirion (1957). Construction: (C) The oldest mosaics found in the baths seem to date from late 3rd / early 4th c.419 Continued use and abandonment: (C) Some of the mosaics pertaining to the last phase of decoration (venatio scene) seem to date to the late 4th / early 5th c.420 Plan: The Small South-East baths lay in the south-east part of the city, 150 m from the city walls. The building can be subdivided into an eastern and a western part. The eastern part, lying on a lower level, may have been part of a domus of the second half of the 3rd c.421 The western part consisted of a bathhouse, fitting within a rectangle of 27 × 21 m. The bathers accessed the baths from the west, through a north-south oriented corridor. The plan was dominated by the cold room, consisting of three parts. The part in the centre of the building had two apsidal pools and a polychrome mosaic of floral motifs with exotic fauna in medallions. South of this room, the second part had a rectangular pool to the west and a large apse to the south. An almost square room to the south-west of this second frigidarium also had a small pool, which was adapted several times. The heated section was rather small, but nevertheless comprised 4 rooms, which probably had to be visited following a clockwise itinerary. The first two rectangular rooms had no pools and may be interpreted as an entrance tepidarium and a sudatorium. The third room, the caldarium, 418  Fendri, M. (1964) 51. 419  Thirion (1957) 231; Dunbabin (1978) 273. 420  Dunbabin (1978) 273. 421  Thirion (1957) 210.

341

Gazetteer

the proportions are inaccurate (…) and the attitude utterly clumsy.’), coming to the logical conclusion ‘un tel mépris de la vraisemblance et de l’unité de l’ensemble ne se rencontre pas avant le IV e siècle.’(‘such a contempt for realism and the unity of the whole is not encountered before the 4th century.’).423 The bottom of the semicircular pool was embellished with plain grey-green mosaics made of large tesserae. Among the recovered wall paintings in the cold section, some geometric patterns and marble imitation can be seen. In the heated rooms and above the semicircular piscinae, the walls were clad with marble slabs. Technology: The hypocaust of the corridor was made with small walls perpendicular to the north-south walls. There were no tubuli in this part, only chimney flues to ensure the draught. The tubuli in the caldarium were not side by side, but at regular intervals. Water supply pipes were found in the north and west walls of the caldarium. A large cistern, south of the baths, stored the water for the pools and may have been fed by rainwater.424 Construction technique: Opus caementicium Bathhouse surface: ca. 570 m2 GPS coordinates: 34°38’48.74”N, 10°40’53.54”E

C62. Theveste (Tébessa / Henchir Safia, Algeria)

C61 Thenae (Thyna, Tunisia), Small South-east Baths (After Thirion (1957) general plan)

had a semicircular and rectangular alveus, a feature that can also be discerned in the small neighbourhood baths in Thamugadi (see above chapter 3, p. 146–148). The construction of an indirectly heated north-south corridor to the east of the heated rooms enabled the architects to reduce direct heat loss through the doorways to the cold section. Decoration: The mosaic decoration on the floors of the cold section was mainly geometric and floral (craters with flowers, see also Sidi Ghrib, p. 347–348), with some figurative medallions. The latter included two amores on a sailing boat, wild animals (some type of feline and possibly a secretary bird killing a snake), a combat scene between two boxers and a possible venatio with two bears. In the large apse of the frigidarium, a building with two towers (a villa?) was depicted. The style of execution and the themes point to specialist mosaicists with links to the schools of Antioch in Syria.422 The appreciation of the excavators leaves little doubt about the contemporary views on late antique mosaics: ‘Le dessin, dans son ensemble, est raide et naïf; les proportions sont inexactes (...) et l’attitude pleine de gaucherie.’ (‘The drawing as a whole is stiff and naive; 422  Thirion (1957) 242.

Excavation history: The Baths at Theveste were found and subsequently damaged by army troops in the first decades of the 20th c. The site was investigated by Jean Lassus. Bibliography: Lassus (1959); Thébert (2003) 205–206. Construction: (C) The baths were dated to the 4th c. on the basis of their mosaics.425 Continued use and abandonment: Unknown. Plan: Due to their small size, Lassus interpreted these baths as a private complex of a large villa.426 However, no trace of this supposed villa has been found. Early descriptions of the region describe the presence of a small Roman hamlet covering some 2.5 ha.427 The entire building has a very compact plan and presents itself as a rectangle, with only a small semicircular pool projecting from the south side. The baths were accessible through a colonnaded space to the east. The bathers would pass by a latrine, before entering the presumed dressing room. The square frigidarium (s: 4 m) was the centre of the building, with all the other rooms constructed around it. The heated rooms were entered by passing through a second cold room with a single-person pool and then continuing west through a long heated corridor which gave access to two interconnected heated rooms to the north. 423  Thirion (1957) 231; translation by the author. Dunbabin (1978) 273 proposes the second half of the 3rd to the 5th c. 424  Thirion (1957) 211. 425  Lassus (1959) 335; Dunbabin (1978) 272; Wiedler (1999) 262–63. 426  Lassus (1959) 325. 427  Thébert (2003) 205.

342

Gazetteer

C63. Tubactis Municipium (Gasr Ahmed, Libya)

C62 Theveste (Tébessa / Henchir Safia, Algeria), (After Lassus (1959) general plan)

There was probably only one heated pool in the presumed caldarium in the north-west corner of the complex.428 The small semicircular pool, large enough for only one person and located at the beginning (or end) of the succession of heated rooms, is reminiscent of the small pool with the same location in the Baths of Sidi Ghrib (see below, p. 346). For the vaulting of the building, tubi fittili were used. Decoration: The mosaics, only found intact in the unheated parts, consisted of polychrome geometric motifs. The mosaic on the bottom of the large piscina was a simple yellow field. The walls and the ceilings where covered in a white gypsum plaster. No traces of paint have been identified. Technology: The water for the alveus was heated in a boiler located west of the caldarium. It also provided hot water for the small semicircular pool south of the frigidarium, which is reminiscent of the warm water labra found in the apodyterium of some early Roman baths. The single person pool was drained by an outlet at the bottom giving into the sewer of the nearby latrine. The water for the pools was stored in a large reservoir north of the baths. The distribution system from the reservoir to the pools was not described by Lassus. Construction technique: Opus caementicium Bathhouse surface: ca. 150 m2 GPS coordinates: Exact coordinates unknown

428  Thébert (2003) 206.

Excavation history: In 1922, a small Roman-style bathhouse was discovered near modern Gasr Ahmed, just outside Misrata (Libya), when army barracks were being constructed. The building was subsequently investigated by the Italian Colonial Department of Antiquities. Bibliography: Bartoccini (1927). Construction: (B) On the basis of the decoration, such as stucco mouldings and mosaics, and small finds, including a coin from the reign of Julian, the lifespan of the bathhouse was placed in the 4th c. Continued use and abandonment: Unknown. Plan: To the east of the baths, a luxuriously decorated room with a second floor was discovered, prompting Renato Bartoccini to interpret the bathhouse as part of a (urban) villa.429 The exact structural links between bath and presumed domus are not clear. There was possibly an entrance to the baths from the street and possibly one from the house. Such a small yet luxurious bathhouse near the sea is reminiscent of public baths such as the Hunting Baths in Leptis Magna (see below, p. 360) or private baths of maritime villas such as the Villa Silin.430 The description of the baths by Bartoccini is brief. The building was oriented towards the sea. The first room was

C63 Tubactis Municipium (Gasr Ahmed, Libya), (After Bartoccini (1927) 214, unnumbered fig.)

429  Bartoccini (1927) 217. 430  Salza Prina Ricotti (1971).

343

Gazetteer covered by a barrel vault and probably functioned as a combined frigidarium and apodyterium. In its north side, a doorway may have led to a domus. The southern part of the room may have contained a pool (vague indication on the aforementioned figure). The second room, presumably the tepidarium due to its small door connecting to the frigidarium and wider door to the heated rooms, had a flat roof. To the west, two rooms on a hypocaust can be interpreted as the sudatorium and the caldarium.431 The first heated room had a flat roof with a round oculus in its centre, presumably to regulate the heat.432 It had a rectangular pool along its entire south wall, which also rested on the hypocaust. By an oblique passage, one reached the caldarium. This second heated room had two pools, one semicircular and one rectangular (no dimensions are given by Bartocini). Decoration: The cold rooms had mosaic floors, depicting polychrome geometric designs. On the insides of the pools, traces of marble slabs could still be seen at the time of excavation. The vaults were decorated with colourful moulded stuccos, one of which had the form of a seashell. Fragments of wall painting were also found. Technology: Little information is given on the hypocausts. The walls of the heated rooms were apparently heated by box-like tubuli. Water was supplied at least partially through rainwater, as gullies on the roof seem to imply. There is no information on water disposal. Construction technique: Opus caementicium? Bathhouse surface: ca. 40 m2 GPS coordinates: Exact coordinates unknown

C64. Uthina (Oudhna, Tunisia), Baths of the Laberii

Excavation history: The Baths of the Laberii were excavated at the end of the 19th c. by the Service des antiquités et des arts de Tunsie. Bibliography: Gauckler (1896); Gsell (1898); Gsell (1899); Thébert (2003) 185–86. Construction: (C) The baths possibly date from the end of the 3rd or the beginning of the 4th c., rather than from the early 3rd.433 The style of the mosaic of Orpheus found in the frigidarium points to this period (fig. 90).434 The coins that were found in the sewers confirm that the baths were in use in the 4th c., the most recent one dating from the reign of Constantine.435 Continued use and abandonment: (B) The baths were out of use by the 5th or 6th c., when a pottery workshop was installed in the building. A large amount of ceramics with

431  These are labelled caldarium I & II in Bartoccini (1927) 215. 432  Bartoccini (1927) 215 relying on the description by Vitr., De arch. 5.10.5. 433  Thébert (2003) 185 contra Gauckler (1896) 215. 434  Dunbabin (1978) 241. 435  Gauckler (1896) 215.

C64 Uthina (Oudhna, Tunisia), Baths of the Laberii (After Gauckler (1896) 216, fig. 11)

Christian symbols was covered when a fire caused the tubi fittili vault to collapse.436 Plan: No remains of a domus or villa have been found in the immediate vicinity of the Laberii Baths. The plan is characterized by the large frigidarium. It had two pools, the smaller one probably being a later addition. An oval pool to the south of the frigidarium was originally divided into two parts by a small wall, creating an oval basin and a small ‘shower’.437 In a second phase, this room was transformed into a heated room, the ‘shower’ becoming a small heated pool. It was preceded by a small transition room which probably acted as ‘heat trap’ between this room and the frigidarium. The original warm rooms were laid out in a semicircular itinerary, beginning in a large rectangular room in the south-east of the building (tepidarium?), continuing north to two smaller rooms, an apsidal side and ending in the caldarium with its two apisdal pools. A shortcut enabled the bathers to return from the caldarium to the ‘shower’ room. Decoration: The decorative scheme of the building had ‘Byzantine characteristics’, depicting flowers, garlands and 436  Gauckler (1896) 215; Gsell (1899) 61–62. 437  Gauckler (1896) 225.

344

Gazetteer

figure 90 Baths of the Laberii in Uthina (Tunisia), drawing of the mosaic of the frigidarium depicting Orpheus enchanting wild animals in Gauckler (1896) 218, fig. 12

roses. The large figurative mosaic scene in the frigidarium (fig. 90), depicting Orpheus and animals, was accompanied by the text: ‘IN PRAEDIS LABERIORUM LABERIANI ET PAULINUS’.438 It identifies the building as belonging to members of the gens Laberia. But as Thébert remarked, this does not mean that these baths were only for private use.439 Indeed, the text recalls the inscription of the Baths of Iulia Felix in Pompeii, in which a balneum is leased for a period of 5 years.440 The plain white mosaics 438  Gauckler (1896) 221; CIL 8.24019. 439  Thébert (2003) 185. 440   C IL 4.1136 = ILS 5723.

in the piscinae are a common decoration for the bottom of pools. Marble was still an important element in the decoration: slabs embellished the walls and some floors, while statues of draped figures and an amor on a dolphin stood respectively in the niches of the cold room and the piscina. Technology: It seems that the water supply was—at least partially—ensured by reservoirs on the roof, as terracotta water pipes have been found in the remains of the collapsed vaults. It is unclear which heated rooms had tubuli as wall heating. Construction technique: Opus africanum? Bathhouse surface: ca. 700 m2 GPS coordinates: 36°36’27.68”N, 10°10’21.78”E

Gazetteer

Possible Late Antique Baths

345

Sidi Ghrib (Tunisia), Baths of the Marine Thiasos Excavation history: Sidi Ghrib is a small hamlet 30 km east of Tunis and close to the village Borj el Hamri. The bathhouse discovered here in 1975 has yielded some of the finest late antique mosaics in North Africa. Between 1985 and 1992, the adjacent villa complex, already partially visible in the 1970s, was further excavated. Bibliography: Ennabli (1986); Ennabli and Neuru (1994); Thébert (2003) 149–51. Construction: (A) Test pits underneath the mortar floors of the baths have yielded pottery fragments with a terminus post quem of AD 360.445 The mosaics, which form a coherent entity with the architecture and hence seem contemporaneous with the construction of the building, should be dated in the 5th c.446 Continued use and abandonment: (A) Part of the villa was destroyed by fire at the beginning of the 6th c., but was subsequently reused for industrial purposes until the early

7th c.447 It is possible that the baths ceased to function after the fire. During the last phase, the pools and doorways of the frigidarium were blocked, but the vaults were still intact. These later collapsed on layers postdating these blockages. In the Early Islamic period, the tomb of a marabou was installed in the corridor.448 Plan: (fig. 91) The size of the building and the fact that there were two entrances may point to semi-public use. It is problematic that little is known about the surrounding settlement pattern of Sidi Ghrib. Did the bathhouse serve a small community living dispersed in the countryside? The inscription of the piscina hints at a communal use (EA-50): si placet commune est, si displicet nostrum est (‘if it pleases it is communal, if it displeases it is ours’). If the baths were accessible to ‘outsiders’, who would actually have the privilege of bathing here? Did the dominus allow the farmers to visit his baths? Were the baths open only on special occasions? The theme of the mosaics may be informative. The central theme of the mosaic in the apse is the wedding of Neptune with Amphitrite. Other than the obvious link between baths and the aquatic world, this theme might imply the use of the complex for the ‘nuptial baths’.449 The west entrance to the baths connected indirectly to the villa, while the east entrance just opened onto the surrounding estate. Both entrances were S-shaped, which prevented one to look directly inside the bathhouse. Once inside, a long corridor (20 × 3 m) with an east-west orientation led to the centre of the building. Two small recesses (3 × 1.75 m), one in the western part and one in the eastern part, may have been used as dressing rooms.450 The mosaics found in these recesses depicted a dominus with a friend and a servant leaving for bird hunting (west) and a domina seated and dressing up with the help of two maids (east). The excavator interpreted the west entrance as the ‘male’ entrance and the east one as the ‘female’.451 In the middle of the corridor, a large apse opened up to the north, while a large passageway to the south led into the cold room. The apse had 6 columns placed against the walls, probably to support the elongated vault. The columns at both sides of the doorway were aligned with the columns of the frigidarium. The latter divided the cold room into three naves. Each corner of the room was fenced off by a type of choir screen, judging from the imprints in the floors and the walls. There were two lateral cold pools. The one in the west had a rectangular form with lateral sides curved outwards (2.2 × 2.6 m). It was 1.1 m deep. The east pool had

441  Wiedler (1999) 293. 442  Devoulx (1875) 420. 443  Wiedler (1999) 237. 444  Wiedler (1999) 328–29. 445  Thébert (2003) 149 refining the chronology by Ennabli (1986) 55–56. 446  Thébert (2003) 149.

447  Ennabli and Neuru (1994) 216–17. 448  Ennabli (1986) 50–51. 449  Ennabli (1986) 44. The nuptial baths were already existent in Ancient Greece, see Håland (2009) 133. The tradition still exists in Islamic hammams. 450  Ennabli (1986) 4–5. 451  Ennabli (1986) 31.

Ammaedra (Haidra, Tunisia), building under Basilica 1 Excavations underneath Basilica 1 in Ammaedra (Haidra, Tunisia) revealed a mosaic with marine scenes, dated in the middle of the 4th c., which may have belonged to a bathhouse.441 Icosium (Algier, Algeria), building under the Church of St. Philip During construction works in 1844 for the Church of St. Philip in modern Algiers, two large mosaics were discovered, depicting birds and theatre masks inside a geometric field of octagons and stylised crosses. The mosaics were constructed on top of cisterns, which seemed of an anterior date.442 The mosaics may have belonged to a bathhouse and were dated at the turn of the 3rd and 4th c.443 Neapolis (Nabeul, Tunisia), building at Sidi Mahrsi A mosaic depicting roosters and marine scenes with Nereids and sea monsters was found in a room with a basin in the outskirts of modern Nabeul (Tunisia) in 1967. The marine theme of the mosaic, as well as the presence of a pool, makes it likely that this room should be interpreted as part of a (private?) bathhouse. The mosaics point to a 4th c. date.444

Villa Baths with a Possible Public Use

346

Gazetteer

figure 91 Plan of the Baths of Sidi Ghrib (Tunisia) after Ennabli (1986) 3, plan

a hexagonal shape (with each side curved inwards, s: 1.3 m) and was 1.2 m deep. It was surrounded by a broad edge and 6 columns, one on each angle of the hexagon. From a small doorway lying on the south end of the north-south axis of the frigidarium, one could access the heated section. A rectangular room acted as a heat lock, preventing the hot air from the warm rooms to pass directly into the cold section. It had a small unheated pool (1.25 × 1.75 m) against its west side. The first room on hypocaust was circular (diam.: 3.5 m). Its doorway to the heat lock did not lie in the same axis as the passage to the cold room. The

furnace on the south side was blocked at a later stage. By a short corridor to the east, one reached a rectangular room (3.5 × 2.5 m) on a hypocaust with two heated pools. The semicircular pool (r: 1.25 m) was built above a furnace. The rectangular pool (3.5 × 1.5 m) on the north side is problematic, as a doorway connecting this room to the frigidarium was blocked up or pierced in a later phase, making this pool respectively a later addition or an original feature. By a doorway in the east side, one reached the hexagonal caldarium. On the north-east and south-east side of this hexagon, a rectangular pool with rounded lateral sides was

Gazetteer

347

installed. Each one was built just above a furnace. A third furnace was found on the eastern side, between the two pools. The three remaining sides that did not have a pool or a doorway had a niche. When the original caldarium fell out of use, the doorway with the rectangular room was blocked up and the latter became the new caldarium. It is unclear whether the door to the frigidarium was also blocked up at this stage, making it possible to construct a second rectangular alveus, or if, on the contrary, the rectangular alveus was removed to allow a new door to the frigidarium. As Thébert pointed out, it is unlikely that the bathing route was reversed, making the round room the new caldarium.452 This round room never had any pools. However, the same author argues that the doorway connecting the frigidarium and the rectangular heated room belonged to the original phase and was later blocked up. This seems rather unlikely, as the original architectural design seems strict. Constructing a passage in a room that was meticulously planned in a corner that was demarcated by a screen would devalue the symmetrical layout of the entire building. Thus it seems that an ‘intermediary’ phase might have existed, when the linear reversing route was changed into a circular one by opening a new doorway between the cold and heated sections, resulting in the destruction of the rectangular alveus (fig. 92). When the original caldarium fell out of use, the rectangular room took over this function and the doorway to the frigidarium was blocked up again to reduce heat loss. A small bench was constructed where the rectangular alveus once stood. The furnace of the round room was blocked up, so that it became an indirectly heated tepidarium. The main bathing rooms were covered by vaults. Remains of tubi fittili imbedded in lime mortar have been found on top of the abandonment layers of the building. The type of vaulting depended on the shape of the room. The caldarium and the round room were probably covered by a dome, while the frigidarium may have been covered by a cross or barrel vault.453 Decoration: The mosaic floors of the entire building fitted within an aquatic theme. On the thresholds of both entrances, a mosaic represented fountains in the form of a crater and surrounded by rose branches. In the rectangular niches of the corridor, mosaics representing otium-related scenes of the dominus and domina were identified. In the western niche, the domina is seen seated on a chair, holding a mirror and surrounded by servants, probably preparing her for the bath. In the eastern niche, the dominus is seen preparing for the hunt. In the long corridor, the apse, the central nave of the frigidarium and in front of the piscinae, a homogenous 452  Thébert (2003) 151. 453  Ennabli (1986) 25–26.

figure 92 Baths of Sidi Ghrib (Tunisia), different phases of use after Ennabli (1986) 3, general plan

348 theme was presented. The central mosaic of the frigidarium extended to the corridor and was crowned by the mosaic of the apse. It depicted the marriage of Neptune and the Nereid Amphitrite (top register), who were accompanied by Triton and other Nereids mounted on sea creatures (on the lower registers in the frigidarium and on the registers in the corridor). In the mosaic of the apse, Venus was riding a ‘sea lion’ while an amor presented her with a small box. On the panels before the piscinae, similar aquatic scenes were partially destroyed. The western panel probably depicted Venus carried by two creatures (half man, half animal). The identification of Venus would fit well with the inscription found just above, as the love goddess would be a personification of the state of ebria (‘intoxication’) mentioned in the text.454 The eastern panel was also badly damaged, but still showed a male figure in front of a mountain that poured a white liquid from an oenochoe. Above his head, the name ‘GAVRVS’ was spelled out. On the basis of iconographic parallels in Italy, Michèle BlanchardLemée identified the person as Bacchus, standing in front of the mons Gaurus, a hill in the bay of Naples reputed for its excellent white wine.455 The same scholar also refers to a letter by Symmachus, probably published shortly before the baths were built, in which a horned Bacchus inhabited the summit of Gaurus.456 The poem then continues as a praise of Baiae, with its divine waters, clear skies and hot vapours. As we have seen in the literature and epigraphy (see chapter 2), the link of a particular bathhouse with the famous Baiae is often made, especially in North Africa. The implicit reference to Baiae made by the mosaic would not only fit this panel within its bathing context, it would also connect it with the larger marine theme of the mosaic in the frigidarium and beyond. In the 4 fenced-off corners of the frigidarium, a woman carrying a vase or a basket was depicted. This may have been the 4 seasons. The pools, both unheated and heated, were decorated with plain white mosaics. On the lower outer step of the western piscina, a mosaic inscription conveyed a message to the bather (EA-50). On the walls surrounding the eastern piscina, a mosaic imitated opus sectile panels. The heated rooms had geometric mosaics. The floor mosaic of the first small rectangular room also imitated opus sectile. The hexagonal caldarium had a mosaic consisting of a polychrome field of floral and vegetal motifs sprouting from vases and a central emblema depicting the head of Neptune, his moustache ending in dolphins and streams of water pouring out of his mouth forming the sea below. The rectangular niches of the hexagon and the sides

454   Blanchard-Lemée (1988) 371. 455   Blanchard-Lemée (1988) 375–77. 456  Symmachus, Ep. I.8. See Blanchard-Lemée (1988) 380–83.

Gazetteer of the south-east alveus were embellished with geometric patterns. Technology: The heating system made use of the classic hypocaust system consisting of square pillars supporting large tiles. The wall heating did not use tubuli, but flat tiles and terracotta spacers to create the ‘double wall’. These were attached to the walls using iron studs. The number of furnaces of each room determined their function. The round room not only had one furnace, but also a restricted surface, making it a dry heat room. The room with the semicircular alveus must have been more moderately heated, as a single furnace heated a larger surface. The evaporation of the water in the pools made it a wet heat room. The caldarium was arguably the hottest and wettest room, having three furnaces and two pools. The furnaces that were located underneath the pools all had brick piers protruding into the praefurnia to support water boilers. The excavation report mentions evidence of terracotta piping inside the walls of the heated rooms to conduct the warm water from these boilers into the alvei. The boilers must have been filled manually by bath personnel. A reservoir (5.5 × 3 m) was discovered south of the western piscina. It provided water for that piscina and for the adjacent cold plunge of the small rectangular room. The eastern piscina was probably supplied by a small reservoir lying just east of it.457 It was drained by a lead pipe at the bottom of its east side. The semicircular alveus was drained by an outlet in its southern side, while the north-eastern alveus of the caldarium was drained by an opening giving directly onto the floor.

Other Private Baths with Possible Public Access Several large extra-urban villas or large domus seem to have had a private bathhouse in Late Antiquity. However, some of these baths may also have been accessible for a larger public, such as the Baths of Sidi Ghrib and Oued Athemenia in the countryside, or the baths of the large houses in Cuicul and other cities.458 Unfortunately, the archaeological remains often do not permit the reconstruction of the precise layout of the baths, their articulation to the house and the relation to the surrounding area.459 The presence of a separate entrance, not giving into the house, but to the outside, is important to 457  Ennabli (1986) 9. 458  Maréchal (2016). 459  The early or limited excavation of the site, followed by degradation in modern times, is often the main reason for the difficult interpretation: see the private baths discussed in Wiedler (1999) 217–375. Most of the private baths in urban contexts have not been properly studied, see the catalogue and index by de Haan (2010) 248–75 (K31-K44) and esp. 339–56 (A197-A276).

349

Gazetteer identify a possible ‘public’ access (see Introduction). The fact that most of these baths have been dated on the basis of the mosaics poses an additional problem, as it is often unclear from early excavation reports whether the late antique mosaics belonged to the original construction phase of the baths or rather to a late antique restoration. Future excavations of some of these ‘private’ baths may perhaps prove the public access. The villa baths in Auzia (Aïd Sidi Belkassem, Algeria) is known from a mosaic depicting Leda and a swan, dated to the late 3rd or 4th c.460 The baths of the House of the Molphonii in al-Haouria, 45 km south-west of Kairouan, were excavated in 1906 and never properly published.461 A mosaic of Minerva and Neptune was accompanied by an inscription (EA-12), while another mosaic depicted Oceanus with putti and seahorses. Both mosaics were dated to the first half of the 4th c.462 The baths of the Fundus Bassianus near Hippo Diarrhytus (Bizerte, Tunisia) were excavated in 1902. The bathhouse itself has never enjoyed much scholarly attention, as eyes are turned to the large polychrome mosaics of two racehorses from the heated room and a marine scene with a text (EA-35). The first has been dated to the early 4th c., while the second is attributed to the 4th or 5th c.463 A fragment depicting the mythological horse Arion, probably from the frigidarium, may even date from the 6th c.464 The villa baths in Kalaa Kabira, a small village 12 km west of Sousse (Tunisia), were excavated in 1957, but a plan was never published.465 The mosaic of the frigidarium, depicting vases (kantharoi) and vines, was dated to the 4th c.466 A bathhouse probably belonging to a rural villa was also excavated in the vicinity of modern Moknine (Tunisia). The mosaic from the presumed apodyterium depicted a charioteer and circus horses, surrounded by vines, and accompanied by an inscription (EA-49). The mosaic was dated to the 5th c.467 The baths of the House of the Peristylium in Pupput (Soul el Abiod, Tunisia) were not accessible from the street and date to the 4th c.468 In Sakiet as-Zit, a village 7 km north of Sfax (Tunisia), a possible villa with a private bath was found in 1953. The Orpheus mosaic was dated to the early 4th c.469 In Thubursicu Numidarum (Khamissa, Algeria), elevated to municipium under Trajan and the seat of a bishopric twice visited by Augustine, a possible private bath was dated in the 4th c. on the basis of the mosaics.470 460  de Haan (2010) 339–40. For the mosaic: Mercier (1870). 461  Dubiez (1906). 462  Dunbabin (1978) 261; Wiedler (1999) 312–13. 463  Dunbabin (1978) 268; Wiedler (1999) 314. 464  Dunbabin (1978) 268. 465  Foucher (1958b) 32. 466  Wiedler (1999) 310. 467  Thébert (2003) 492. 468  Ghiotto (2003) 225; Thébert (2003) 146–47; de Haan (2010) 260–62. 469  Wiedler (1999) 330–31. 470  Ballu (1906) 185–90; Souville (1976) 917; Wiedler (1999) 266.

Ecclesiastical Baths Cuicul (Djemila, Algeria), the Baths of the Baptistery Excavation history: Unearthed in the 1910s, this exceptional bathhouse has never been the object of a detailed study. Bibliography: Monceaux (1922); Allais (1938) 59; Thébert (2003) 202–203. Construction: (D) In the late 4th / early 5th c., a basilica was built near the southern edge of town, 150 m south of the theatre (map 16).471 Simultaneously, the basilica was complemented with a baptistery and adjacent bathhouse. Continued use and abandonment: Unknown. Plan: (fig. 93) As there was no separate entrance from the streets to the baths, scholars have suggested that this bathhouse was only accessible when catechumens received the sacrament of baptism. The person would first wash his body, before he would cleanse his soul.472 It is difficult to check if the baths already existed before the baptistery, as post-excavation restorations impede a clear reading of the remains.473 The articulation of the baths within the surrounding buildings does not permit conclusions. However, the baths seem to have been in use at the same time as the baptistery, because it would be difficult to explain the articulation of both buildings if the baths were already out of use.474 It is likely that such fully functional buildings were used as ‘normal’ baths by members of the congregation or the clergy. Simply closing the eastern door in the frigidarium would prevent bathers from entering the baptistery. The concentration of ecclesiastic buildings has led some scholars to believe that Cuicul may have been a site of pilgrimage.475 A bathhouse in such a context would not be unusual, as the xenodochiae and diaconiae in Rome are known to have had baths.476 The baths were only accessible by passing through the ecclesiastical complex. They covered a total surface of around 200 m2. From the basilica, one entered the baptistery by a room that was delimited to the north by the rounded outer wall that encircled the font and to the south by a recess secluded by three columns. This recess may have functioned as dressing room. A semicircular doorway to the west gave access to a corridor that led to the cold room (ca. 8 × 6 m). Two rows of two columns divided this room into three naves. On the south side was a cold apsidal pool with a small room with a basin to the west. To the north, a doorway gave access to 4 heated rooms. The two southernmost rooms were indirectly heated. The eastern one had a semicircular pool (r: 1 m). The two northernmost rooms were directly heated 471  Monceaux (1922) 381–83. 472  Monceaux (1922) 400–401. 473  Thébert (2003) 203. 474  Thébert (2003) 203. 475  Monceaux (1922) 407. 476  See also Stasolla (2002a) 29–31; Stasolla (2008) 894–95.

350

Gazetteer

figure 93 Plan of the Baths of Baptistery in Cuicul (Algeria) after Thébert (2003) 631, pl. LXXXI.1–4

by a furnace. Pools may have been located just above them.477 The doorways between these 4 rooms are difficult to reconstruct, but probably formed a reversing route starting in the south-east room and ending in the north-east one. From the frigidarium a doorway in the east led back into the baptistery. A latrine was attached to the south side of the bathhouse, but does not seem to have been connected to it. Decoration: The brick walls of the baths were coated with a plain white plaster. It is difficult to identify traces of paint. Technology: Little is known about the heating system and water distribution. The first description of the building does not discuss the hypocaust, nor does it mention wall heating. Only two furnaces have been found, one of which was equipped to carry a boiler. It is interesting to notice that an excavation report of the nearby basilica uncovered a large bronze ‘platter’.478 An inscription on the object mentions it weighs 310 Roman pounds (101 kg). However, its present weight is much lower, suggesting a part of the object is missing. It seems clear that this bronze platter with raised semicircular ending is a part of a testudo used

477  Thébert (2003) 203. 478  Monceaux (1923) 94 calling the object a ‘bénitier,’ a vessel for holy water.

in the nearby baths.479 The water supply is unknown, while only drains coming from the frigidarium and the latrines were found.480 Theveste (Tébessa Khalia, Algeria), ‘Church Baths’ Excavation history: The site was discovered in 1866 and partially excavated by the French army in the 1880s. Between 1902 and 1903, the bathhouse, two churches and an olive oil production unit were further excavated (fig. 94). Bibliography: Chedé (1882); Ballu (1903) clxx–clxxi; Boucher (1954); Christern (1976). Construction: (D) The bathhouse is thought to have belonged to the phase of the church building (4th or 5th c.).481 Continued use and abandonment: (D) Basilica II seems to have been fortified in the Byzantine period, but it is unknown if the other buildings of the site, such as the baths, were still in use.482 Some of the architectural ornaments of the round room display Byzantine workmanship.483 The entrance of the baths was heavily disturbed by squatter occupation dating from the Islamic period. 479  Monceaux (1923) 94, fig. 2; Nielsen (1993b) 66, fig. 26. See already Thébert (2003) 203. 480  Indicated on the plan in Leschi (1953). 481  Following Yegül (1992) 243. 482  Boucher (1954) 177. 483  Boucher (1954) pl. III, fig. 7 and 8.

Gazetteer

figure 94 Plan of the ‘Church site’ in Tebessa Khalia (Algeria) with small baths (in box) after Yegül (1992) 245, fig. 293

351

352

Gazetteer

figure 95 Plan of the Baths of Tebessa Khalia (Algeria) after Christern (1976) fig. 31

Plan: (fig. 95) The baths were entered from the east, possibly through a portico. A small corridor gave into a surprisingly large round room (diam.: 16.4m), with an apsidal piscina (2.51 × 8.56 m) on its south-east side. A low bench running along the wall may have supported a colonnade, as column bases were found on top of it. The floor was made of mortar. Chedé mentions that at the time of excavation thin boards (4 cm) of wood were found on top of the floor.484 This impressive round room may have been covered by a simple wooden roof or tubi fitilli dome.485 Such an impressive room, which could have acted as an apodyterium / frigidarium, may have had a more general function as assembly hall as well, perhaps even serving as a baptiserium on some occasions.486 To the west, a corridor with two lateral doors connected to the actual bathing rooms. A rectangular frigidarium had a long apsidal piscina on its south side, and a semicircular apse on its north side. By a doorway to the south-west, one entered another rectangular room with a small apsidal pool to its south. This may have been an exit-tepidarium. Further to the south, the two rectangular heated rooms are difficult to reconstruct due to their ruinous state. The caldarium may have had two rectangular alvei. The interpretation of this bathhouse and of the site in general is problematic. Was this bathhouse constructed as part of an ‘ecclesiastical complex’? Or do we have here a large rural estate also boasting a bathhouse (as in Piazza Armerina)?487 Or was the site some sort of rural market, a panegyris, later taken over by the Church?488 In any case, it seems as if the large round room with its 484  Chedé (1882) 275. 485  Chedé (1882) 274 mentioning traces of charred wooden beams. For the tubi, see Yegül (1992) 452, n. 63 possibly alluding to the many fragments of ‘tubes’ found during the excavation. 486  Yegül (1992) 246. 487  Christern (1976) 145–51, n. 72. 488  Yegül (1992) 246.

‘Christian’ decoration and the small bath suite belong to the same construction phase.489 Decoration: The apsidal piscina in the round room was embellished with a mosaic depicting geometric patterns. The colonnade in the round room supported ornamental architraves and a frieze carved with geometric patterns, flowers, fishes and seashells. This room may also have been finished with a parquet floor.490 The piscinae in the bathing rooms, as well as the apse of the frigidarium, were embellished with geometric mosaics. Technology: There is little information on the type of hypocaust tiles or the tubuli. The piscina of the round room was drained by a large sewer, running underneath the room towards the north, while the piscinae in the bath rooms had a separate sewer, also running north. Traces of an aqueduct have been found in the vicinity of the baths, possibly supplying the square enclosure as well. Thamugadi (Timgad, Algeria), Baths of the Donatist Cathedral Excavation history: The ecclesiastical complex was excavated in 1907–1908. Bibliography: Ballu (1909); Thébert (2003) 252–53. Construction: (D) The construction of the ecclesiastical complex should probably be dated in the late 4th c., under the supervision of a certain Optatus mentioned in an inscription, perhaps to be identified with the Optatus known to Augustine.491 Continued use and abandonment: The complex may still have been in use in the Byzantine period.492 489  Christern (1976) 148. 490  Chedé (1882) 274. 491   A E (1939) 79. See however the critical re-examination by Gui et al. (1992) 275–76; also Sears (2007) 62 with anterior bibliography. 492  Gui et al. (1992,) 278 without further specification.

353

Gazetteer

figure 96 Plan of the Basilica of Tipasa (Algeria) with adjacent baptistery and baths in box after Thébert (2003) 647, pl. XCVII.3

Plan: (fig. 40) The large religious complex was built on what seems to be the western outskirts of the city. It consisted of a basilica, several apsidal rooms, open courtyards and a high number of small rooms. The whole complex, covering a surface of around 180 × 120 m, was surrounded by a row of small rooms, interpreted by the excavators as the cells of monks.493 The bathhouse was situated in the north-west corner of the complex, built against a square courtyard. From this courtyard, one entered a large room (9.6 × 8.9 m) with a semicircular cold pool (r: 1.05 m) in its west side. The imprints of two hexagonal structures, maybe pools, were discovered in the marble paved floor of this room. To the south, three small rooms lay on an east-west axis, but did not communicate between themselves. Only the outer rooms were connected to the large cold room. To the north, one doorway led into a heated section, consisting of two rooms on hypocaust, while another door led into a baptistery. The northernmost heated room (4.7 × 2.7 m) may have had a pool above the furnace channel.494 From the southernmost heated room, a doorway also led straight into the baptistery (8.3 × 6.4 m). In the middle of the latter, a hexagonal font (1.9 m wide; 1 m deep) was placed in the centre of the mosaic floor. Through a doorway in the north side, one entered a second paved cold room with two semicircular pools. Continuing north, a third large unheated room had a rectangular piscina in its north-west corner. To the south of this pool, a passage led to a room. To the north, an oblong room with apsidal endings was heated by a furnace in 493  Ballu (1909) 98–99. 494  Thébert (2003) 253.

its north side. Albert Ballu recognized one big alveus here, although Yvon Thébert considers it to be a room with two lateral semicircular alvei.495 To the south lay the praefurnium of the heated rooms next to the baptistery. The baths were only accessible for the users of the complex, as no independent street access existed. The seemingly haphazard layout of the bathing rooms around the baptistery may have been the result of a different use of the southern and northern part. The southern part, better linked to the baptistery, was perhaps reserved for bathing prior to baptism, while the northern part could also be used as ‘normal’ baths. The disposition of the baptistery is remarkable, as if it was a simple room of the bathhouse and the font were just another pool. It stands in contrast to the Baths of the Baptistery in Cuicul (fig. 93), which was a stand-alone architectural entity. It cannot be ruled out that both the north and south part in the ‘Donatist’ complex were used as normal baths, the baptistery being simply closed off from both wings. The presence of a bathhouse in a religious complex, monastery or not, should not surprise us, as literary attestations have made clear that monks bathed and that some convents were equipped with bathhouses (see chapter 2, e.g. Augustine, Zacharias, Cassiodorus). Decoration: The southern courtyard with the apsidal pool was paved with marble slabs. The room with the hexagonal baptismal font had a mosaic floor with geometric and floral patterns. The chrism was laid out in mosaics on the bottom step of the font. A zigzag-pattern and laurel leaves completed the decoration. 495  Ballu (1909) 101; Thébert (2003) 253.

354

Gazetteer

figure 97 Plan of the Baths of the Basilica in Tipasa (Algeria) after Lassus (1930) 225, fig. 1

Technology: There is no information on water supply and disposal. Rainwater collection would have been an option. Tipasa (Tipaza, Algeria), Baths of the Basilica Excavation history: The baths were excavated in the late 19th c. by the Société historique algérienne. Bibliography: Gavault (1883b); Gsell (1894); Lassus (1930); Thébert (2003) Construction: (D) The small bathhouse was discovered adjacent to the baptistery of the large Basilica 3 (fig. 96). The latter is dated to the 4th c. on the basis of construction technique and mosaics.496 The baptistery and baths probably date from the end of the same century.497 Continued use and abandonment: Unknown. Plan: (fig. 97) The remains are badly preserved and the northern part of the complex has disappeared by continuous erosion by the sea. It seems that the bathhouse was only accessible through the baptistery. Due to the bad preservation of the walls, it is difficult to reconstruct the locations of the doorways. However, a mosaic inscription mentioning the power of baptism was found in the corridor north-west of the baptistery.498 Such a text probably welcomed the people entering both baths and baptistery. The corridor gave access to a second corridor with two doorways. The one to the south led to the rectangular baptistery (diameter of font: 3.1 m). The one to the north 496  Gui et al. (1992) 24. 497  Thébert (2003) 226. 498  Gavault (1883b) 402–403.

opened onto the presumed frigidarium of the baths, ending in a semicircular piscina in the north. From this cold room, a succession of three rooms on hypocaust could be reached. The north-west room probably had a heated pool on its west side. All three were covered with vaults made out of tubi fittili. Decoration: The floors were decorated with mosaics: marine fauna and seashells in the frigidarium, polychrome geometric and floral patterns in the heated rooms. A mosaic inscription found in the caldarium was too damaged to interpret.499 The word domi(nus) could point to the Christian context. Technology: The suspensurae were built with standard brick pilae (1 m in height), while the walls were heated through rectangular and ogival openings. There is no information on the water management of the baths or the baptistery.

Fortress Baths Thamugadi (Timgad, Algeria), Baths of the Byzantine Fortress Excavation history: The fortress was excavated in 1938 without any archaeological supervision, but the entire complex was studied in subsequent years by the French archaeological mission.500 Bibliography: Lassus (1981); Thébert (2003) 253–54. 499  Gsell (1894) 370. 500  For a copy of the reports, see Lassus (1981) 27–38.

355

Gazetteer

figure 98 Plan of the Byzantine fortress in Thamugadi (Algeria) with location of the baths in box after Lassus (1981) general plan

Construction: (B) An inscription above the northern entrance gate of the fort dates the inauguration to AD 539– 540.501 The baths are believed to be contemporaneous with the construction of the fort. Continued use and abandonment: The baths were enlarged at least once during their lifespan, but the lack of archaeological data makes it difficult to date this intervention. It is not known when the fort was abandoned. Plan: (fig. 41) Some 380 m south of the old city walls, the fortress was built during the reign of Justinian on top of a sanctuary of the Aqua Septimiana, a sacred source surrounded by temples of the ‘dea Africa’, Serapis and possibly Asclepius (map 18).502 The presence of this sanctuary is of great importance for the fort and the baths. Not only did the pre-existing structures define the layout of the fortress, the presence of water was a crucial factor for its location and for the possibility of building baths. A large pool (13.6 × 3.6 m) that used to collect the sacred water in front of a temple was transformed into a large reservoir. The baths were built to its north-east (fig. 98). The building fitted within a rectangle of 15.4 × 11.6 m. The walls, mainly built out of bricks, were directly placed upon the old opus spicatum floor of the sanctuary. The fact that the baths were accessible from the inner court meant its use was probably not restricted to the high command.503 There were possibly two building

501  Lassus (1981) 13. 502  Pringle (1981) 232–36. 503  Lassus (1981) 123.

phases, as the southern part of the building seems to abut the northern part.504 Upon entering the baths, one found himself in a narrow north-south oriented corridor, ending in a small square room to the north. Jean Lassus interpreted this room as a porter’s lodge, yet a specific location for this person may be redundant in a military context.505 We could suppose that a soldier was assigned the maintenance of the baths. The square room may have been used as a utility room for storing oil and ointments. Thébert interprets this as the original apodyterium.506 Continuing east, a rectangular room had a bench against its southern wall. It was probably the apodyterium, at least in the second phase of the building. A large doorway to the north gave access to a cold room with two lateral pools. The eastern one was badly damaged, but the western one (1.7 × 1.3 m; maximum depth of 1 m) preserved an internal step. Through a narrow doorway (0.65 m in width) that could be closed by a door, one entered the heated section. A first oblong room on a hypocaust had three rectangular niches (0.6 × 0.45 m) in its north wall, possibly for seats.507 A brick-built bench was constructed against the south wall. To the east, a separate small room with a single niche may have been restricted for high ranked personnel508 or for a special (medical) purpose. To 504  Thébert (2003) 254 contra Lassus (1981) 145. 505  Lassus (1981) 128. 506  Thébert (2003) 665, pl. CXV.2. 507  Lassus (1981) 131. 508  Lassus (1981) 132.

356 the west, a rectangular room was directly heated by a furnace to the west. Going south, this room connected to a heated room with two lateral alvei (1.2 × 1.3 m), coated on the inside with a hydraulic mortar. A doorway in the southern side gave access to a narrow rectangular room (4.45 × 0.65 m), covered in a double layer of hydraulic mortar. The function of this narrow space remains unclear, although in a first phase, it lay directly above a furnace in the southern wall. In a second phase, an additional heated room was constructed south of the original heated part. However, there was no communication between the two parts, as it was probably easier to add a new separate room instead of piercing the old walls. This new heated room, with a pool along its western side (1.5 × 1.3 m; maximum depth of 0.9 m), was preceded by the dressing room mentioned earlier. As the new heated room obliterated the southern furnace of the old heated section, a new praefurnium was built along the west side of the building. Lassus wrongly interpreted this massive brick-built construction as a chimney.509 The large rectangular structure at the eastern end of the building, a part of the original construction, is problematic. At a later stage, the structure was heavily altered, but the original function may have been an open-air swimming pool.510 Decoration: The apodyterium of the second-phase baths was paved with bluish limestone slabs. The walls inside the baths had a simple lime mortar coating. Technology: The heating system is well known due to the good preservation of the building. The hypocaust made use of square tiles (s: 18 cm) for the construction of its pillars, while the wall heating was ensured by tubuli. In the room in the north-west corner, a row of tubuli alternated with a row of bricks. In the caldarium with its two alvei, there seems to have been only square openings in the wall, acting as chimneys rather than as heating device. It is questionable whether the small separate room at the eastern end of the heated section may have benefitted from the indirect heat of a furnace that was located 10 m further west. However, this small space may have been heated by a portable brazier, making the room a laconicum. Little is known about the internal piping for the water supply, although the water was provided by the ancient source and stored in the converted pool of the old sanctuary. The cold pools were drained by a channel passing under the frigidarium, turning north under the small square ‘porter’s lodge’ and joining a sewer that left the fort through the north gate. The drainage of the alvei of both the old and new hot room is unknown.

509  Lassus (1981) 145. 510  Thébert (2003) 254.

Gazetteer

Continued Use and Abandonment of Imperial Period Baths during Late Antiquity Banasa (Sidi Ali bou Jenoun, Morocco) The city of Banasa was located in a fertile plain 30 km from the Atlantic coast, on the left bank of the Sebou river. It probably started as an indigenous settlement around the 6th or 5th c. BC and was granted the title of colonia by Octavian in 33–25 BC.511 Four bathhouses have been discovered in Banasa. Due to the poor excavations, none has been dated with certainty. The North Baths (fig. 99) have been dated to the second half of the 2nd c. on the basis of the mosaics.512 There is no reason to reject such a date on the basis of the plan, characterized by the large frigidarium and the multi-lobed heated room. It is interesting that the cold room had two piscinae, lying next to each other but with the northernmost smaller than the southernmost. As the pools seem contemporaneous, this doubling should be explained by a differing use (see above, chapter 4, p. 192). No evidence of fundamental restorations has been identified. The Large West Baths are the largest baths found in Banasa (ca. 800 m2). The excavations did not yield conclusive evidence to date the building. The different types of bricks used in the hypocaust and the walls suggest a long period of use.513 The plan is marked by the long hall preceding the frigidarium, which must have served multiple purposes (dressing, anointing, massages etc.). There is no reason to interpret the semicircular pool as a hot water reservoir.514 The walls were not heated by tubuli, but by a system of terracotta tiles and spacers. With so many restorations, it cannot be ascertained if this system was contemporaneous to the construction. The Small West Baths may have been private, although the main entrance seems to give onto the street (fig. 100). The building has not been dated, but on the basis of the plan with its semicircular and round pools, we could propose a 3rd c. date (or later?). The central panel of the frigidarium mosaic, the bust of a person, was later removed and replaced by a white marble slab. Thouvenot and Luquet interpret this figure as Bacchus, making the removal of his face a deliberate act by Christians.515 Interest in this building lies in the round pool of the frigidarium (diam.: 1.75 m). Its maximum depth was 1.2 m, while the water was supplied through an outlet 2 m above the bottom of the pool. This has led the excavators to believe that the pool was actually a sort of shower.516 However, we can concur with Thébert’s remark that this was a reasonable height for an outlet and that a shower would not have to be so deep.517 The caldarium with a small semicircular and 511  Euzzenat (1976) 140; Girard (1984) 88–93. 512  Thouvenot and Luquet (1951) 39. 513  Thouvenot and Luquet (1951) 16, 21. 514  As proposed by Thouvenot and Luquet (1951) 14. 515  Thouvenot and Luquet (1953) 44–47. 516  Thouvenot and Luquet (1953) 43. 517  Thébert (2003) 256.

357

Gazetteer

figure 99 Plan of the North Baths in Banasa (Morocco) after Thouvenot and Luquet (1951) 35, unnumbered fig.

a rectangular alveus placed perpendicular to each other is reminiscent of Thamugadi’s small neighbourhood baths in (see above, chapter 3, p. 146–148). The Baths of the Frescos (ca. 360 m2) were located south -west of the Large West Baths and were accessible from the street. No construction date has been proposed.518 However, the original design was altered several times, pointing to a long period of use.519 The plan is characterized by semicircular niches and a round room protruding into a courtyard, which could point to a 3rd or 4th c. phase (fig. 101). In the latter, a hexagonal pool (diam.: 1.35 m; 1.20 m deep) was installed, with 5 semicircular niches on its sides. The calcareous deposits on the bottom reveals that the pool was frequently used and that the water was possibly changed often, maybe with each bather that entered it.520 The round room in which the pool was built may have been added later to the frigidarium, as its walls abut the walls of the baths.521 A well was also added next to the existing rectangular reservoir, possibly to

figure 100

Plan of the Small West Baths in Banasa (Morocco) after Yegül (1992) 240, fig. 283

518  Katherine Dunbabin proposes a late 2nd and 3rd-c. date for most of the mosaics in Banasa, although she does not date the separate panels. She reckons that no new mosaics would have been laid after the Roman administration withdrew from the region in the last quarter of the 3rd c. (Dunbabin (1978) 249). 519  Thébert (2003) 258. 520  Thouvenot and Luquet (1951) 24. 521  Indicated on pl. CXX.3 in Thébert (2003) 670.

358

Gazetteer side the baths, commemorates the restoration of the cluaca (sewer; see EA-8). Interventions that can be attributed to the 4th c. or later include the addition of a small apsidal pool in the frigidarium, the reorganisation of the tepidarium by constructing a semicircular pool above a new furnace and the creation of a new round room with two semicircular niches to the south of the frigidarium. The site was probably still in use at the turn of the 5th and 6th c. (late repairs to the mosaics). After the baths had fallen out of use, parts of the baths were reused for habitation and possible industrial activities.523

figure 101

Plan of the Baths of the Frescos in Banasa (Morocco) after Yegül (1992) 240, fig. 284

supply the water needed for the round pool. The caldarium also had a small semicircular pool—maybe a later addition— and another rectangular pool. The geometric black and white mosaics of the first heated room seem rather late, while some marble floors were paved with spolia and pieces of mosaic. The corridor and some of the niches in the frigidarium were decorated with wall paintings imitating marble. The decorative scheme and the alterations made to the plan point to a late phase of these small neighbourhood baths. Belalis Maior (Henchir el-Faour, Tunisia), Forum Baths The site of Belalis Maior has yielded a late antique bathhouse (see above, C35). Just north of these baths, on the south-east corner of the forum, lay the Forum Baths. The first phase of this building has been dated to the 2nd c., while a phase of expansion and restoration took place at the end of the same century.522 In the 4th c., another restoration phase replaced the mosaics of the frigidarium and the hexagonal tepidarium. An inscription dating from the reign of Constantine, found in522  Mahjoubi (1978) 206.

Bulla Regia (Hamam Daradji, Tunisia) The city of Bulla Regia (map 36) was located in one of the most fertile parts of Africa Proconsularis, blessed with several water sources and a water table that was close to the surface.524 It lay on the road connecting Carthage with Hippo Regius. The economy was primarily based on cereals and olives. The city fell under Roman influence from the 1st c. BC, but there are traces of urbanization going back to the 3rd c. BC.525 Under the Flavian dynasty, Bulla Regia was awarded the title of municipium and was then made a colonia under Hadrian.526 The importance of the city for Roman control of North Africa is underlined by many members of the local elite becoming senators in Rome. Its favourable position ensured that the city remained wealthy throughout Late Antiquity. A bishop is attested as late as the 7th c. Even after the Arab invasion, some small-scale habitation must have continued as traces of occupation have been found dating from the 12th c.527 The existence of one or maybe two modest public bathhouses in Bulla Regia during Late Antiquity has been proven (see above, C36 and C37). One other public bath was possibly still in use in the 4th c. The Baths of Iulia Memmia have been dated by archaeological excavations to the Severan period. The ceramics found in test pits put the construction date around AD 220–240.528 Around AD 360, the area west of the baths was transformed (former palaestra?), although the layout of the baths was not altered. The excavations revealed that the baths were still in use during the second half of the 5th c. and perhaps even later.529 The vaults above the pools were constructed using tubi fittili. A large bathhouse some 200 m south-west of the Iulia Memmia baths still awaits excavation.530 The remains are not well preserved.

523  Mahjoubi (1978) 209. 524  Baklouti (1996) 693. 525  Beschaouch et al. (1977) 9–10. 526  Ennabli (1976a) 171. 527  Beschaouch et al. (1977) 12. 528  Broise and Thébert (1993) 106–107. 529  Broise and Thébert (1993) 386. 530  Beschaouch et al.( 1977) 13, fig. 3.

Gazetteer Three houses in Bulla Regia seem to have been equipped with baths in Late Antiquity. The bath of the House of the Venantii has been attributed to Late Antiquity on the basis of its mosaics, but the remains still await further investigation.531 The bath of the House of the Hunt was built during the first half of the 4th c., at the same time as a large basilical hall in a pre-existing 3rd c. insula.532 The plan is characterized by apsidal pools and a lobed caldarium. Unlike the private baths in cities such as Cuicul, those of the Hunters’ House in Bulla Regia do not seem to have been open to the public. Contrary to the claim of several scholars, the apparent doorway between the apodyterium and the street running along the west side of the insula was not an access, but the location of a small semicircular pool of a second phase, protruding onto the street.533 It was not until the second half of the 6th c. or later that the baths fell out of use and the domus was transformed into several smaller houses.534 A third private bath, belonging to House nr. 8 (3rd c.), may still have been in use in the 4th c., when the house was thoroughly restored.535 The baths may have been accessible from the street.536 Iol Caesarea (Cherchel, Algeria) The harbour town of Caesarea is one of the few large cities in North Africa where the old Roman centre has disappeared under medieval and later settlement.537 The city, which started as a Punic settlement called Iol and grew to one of North Africa’s most important cities under the Numidian vassal king Juba II, was made the capital of the new province Mauretania Caesariensis under Claudius (AD 40). The city was sacked during the revolt of Firmus in AD 373.538 It seems to have been embellished under Vandal rule and became capital of Mauretania Secunda under Byzantine rule.539 Our knowledge of late antique Caesarea is necessarily clouded by the limited area of the old town that has come to light. The Large West Baths have not been precisely dated, but the 2nd or 3rd c. has been proposed on the basis of the imperial-type plan.540 The early excavation of the building has not made it possible to recognize any late antique phases. However, on several statue bases recovered from these digs, the text mentions that a transfer of the statues ‘de sordentibus

531  Wiedler (1999) 300. 532  Thébert (2003) 136 quoting his own unpublished excavation report. 533  Ghiotto (2003) 225; de Haan (2010) 248. 534  Thébert (2003) 136. 535  de Haan (2010) 253–56. 536  Ghiotto (2003) 225. 537  Other examples include Carthage, now swallowed up by modern Tunis, and Oea, modern Tripoli (Libya). 538  Amm. Marc. 29.5.18. 539  C  od. Iust. 1.27.2.1; Leveau (1984) 213–14. 540  Krencker et al. (1929) 190–96; Thébert (2003) 191.

359 locis’.541 The relocation of statues implies a late date (4th or 5th c.), but Thébert remarks that similar inscriptions dated to the 3rd c. are known.542 Furthermore, it is not certain if the building still functioned as a bathhouse when the statues were relocated. An inscription of unknown context from Caesarea (EA-39) recalls the building or restoration of thermae by Theodosius I (AD 379–395) or Theodosius II (AD 408–450). It may have referred to the Large West Baths, but as Caesarea was a very large city, it is possible that the inscription belonged to an as yet unidentified building. A second bathhouse, the Small East Baths, has also been discovered in Caesarea. No information about a possible construction date is known and the remains were already damaged at the moment of discovery.543 A third building, the Baths of the City Centre, have disappeared. Only a brief description of the remains survives.544 Philippe Leveau identified three more bath buildings of rather small size that were unearthed in Caesarea, none of which were thoroughly excavated.545 In the so-called Baths of the Muses, a mosaic depicting the busts of the Muses in medallions was dated to the turn of the 3rd and 4th c.546 It is unknown whether the mosaic belonged to the original building phase or to a restoration. At least one private bathhouse is known to have existed in Late Antiquity: the baths of the House of the Iulii or the House of the Graces seems to have been equipped with a private bath in the second half of the 4th c., judging from the mosaics in an apsidal hall.547 Lambaesis (Tazoult / Lambèse, Algeria) The city of Lambaesis started as a military camp (AD 81) on the slope of Mons Aurianus, controlling a mountain pass leading to the Algerian desert. During the reign of Trajan or Hadrian, a military fortress was built for Legio III Augusta. To the south-east of the fortress, a city was created. Under Septimius Severus, it became the capital of Numidia. The legion was disbanded in AD 238, but subsequently restored in AD 253. Under Byzantine rule, the fortress was consolidated.548 Besides the Bath of the Hunters (C49) and the South Baths of the sanctuary of Asclepius (C50), three other bathhouses have been discovered in Lambaesis. The Baths of the Legionary Fortress (Lagerthermen / Thermes du camp) may have been built shortly after the construction of the fortress itself, although archaeological evidence is lacking. The baths 541  ‘from decrepit places’; Waille (1887) 54. 542  Lepelley (1994) 13; Thébert (2003) 192. 543  Krencker et al. (1929) 196–97; Leveau (1984) 53–54. 544  Berbrugger (1864) 477–79. 545  Leveau (1984) 54–55. 546  Wiedler (1999) 221. 547  Waille (1886) 304–305; Wiedler (1999) 222–24. 548   A E (1954) 137. Marcillet-Jaubert (1976) 478; Wilson (1996) 812; Huß (2006a).

360 probably suffered during the disbandment of the legion. The plan that can be seen today reflects the phase after the important reconstructions and reorganisations of the late 3rd c. (dated by brick stamps).549 The oblong hall preceding the heated ‘dressing rooms’ may date from this late phase.550 The apsidal warm water pools that were added to the rooms connecting the frigidaria to the caldarium also belonged to a later phase.551 It is unknown if these fortress baths remained in use after the legion was permanently withdrawn in the beginning of the 4th c., especially since Lambaesis had a second large public bathhouse, the Large Baths (thermes du « palais du légat »). These large imperial thermae may date to the end of the 2nd c., if the statue series of Hercules were part of the original decorative scheme.552 The reuse of building materials would point to a long period of use, but there is no evidence supporting a late antique phase.553 The sanctuary of Asclepius south-east of the fortress had several bathing facilities and was in use for a long period (fig. 87). There are inscriptions dating to the 4th c.554 Leptis Magna (Lebdah, Libya) Leptis Magna was another important Punic harbour that came under Roman influence. From the time of Augustus, the city was remodelled after classic Roman standards.555 The city reached its peak under Septimius Severus, who was born in Leptis. It was equipped with a splendid new forum and adjacent basilica, a large colonnaded street leading up to the restored harbour, a monumental nymphaeum and other public buildings. Insecurity due to raiding tribes in the 4th c., especially the Austoriani who sacked the city, halted further expansion of Leptis. From AD 455 onwards, the city was under nominal Vandal control, although the extent of this rule is not clear. When Justinian walled a part of the old centre, restored the ‘palace’ and built a shrine to the Virgin Mary, 4 churches and a bathhouse, the city was already partially covered by sand.556 Excavations have revealed Byzantine building activities, especially in the area around the Severan forum.557 Six bathhouses have been discovered in Leptis Magna, although none can be identified with the Byzantine baths mentioned by Procopius. The Baths of Hadrian are the first baths with an imperial plan that were built on the African continent (fig. 102). A monumental inscription commemorated the dedication of the baths under Hadrian in AD 127.558 Some resto549  Thébert (2003) 209–210. 550  Yegül (1992) 216. 551  Krencker et al. (1929) 211. 552  Thébert (2003) 209–211. 553  Gsell (1901) 220. 554  Cagnat (1923) 88. Ward-Perkins (1981) 465. 555   556  Procop., Aed. 6.4.1–12. 557  Leone (2007) 272–75. I RT 361. 558  

Gazetteer rations, in particular of the frigidarium, and embellishments with marble took place under Septimius Severus.559 Perhaps the inscription referred to the construction of two small piscinae in the long corridor that originally surrounded the cold room. The glass mosaics covering the walls and the vaults of some of the heated rooms can also be dated to the turn of the 2nd and 3rd c.560 The use of limestone monoliths as hypocaust pillars in the lateral rooms leading up to the caldarium could point to a late antique restoration. In the same rooms, a small heated pool was added just above the furnace. The apsidal alveus of the caldarium was also a later addition, reminding us of the new apsidal shapes of the Constantinian alvei in the Baths of Caracalla in Rome and the Baths of the Forum and of Porta Marina in Ostia. A recent re-examination of the late antique inscriptions of Leptis assembled several fragments of an inscription pertaining to the restoration of a thermas in AD 306 and linked this text to the Baths of Hadrian.561 Several statue bases seem to have been reused in the frigidarium and palaestra in the late 3rd or early 4th c.562 At an unknown date, the cold part of the baths may have been reused as an industrial zone. A small basin (part of a press?) in the palaestra and the circular wear marks of a wheel in the pavement of the eastern latrine can still be seen. The Hunting Baths are arguably one of the best preserved Roman baths in the whole empire (fig. 103). These modest neighbourhood baths were built in a quarter of town that developed during the 2nd c. AD.563 It is believed that the baths also date from this century. The long use of the baths is confirmed by the floor mosaics with large tesserae dating to the 4th c.564 The last phase of mural paintings imitated marble designs. The most important changes to the plan concern the alteration of the bathing itinerary. By blocking two doors, the circular plan was transformed into a parallel row type with reversing itinerary. Furthermore, a rectangular pool seems to have been added to the octagonal room in the west, while a basin was added to the one the east. It is worth noting that a suite of small bath rooms was added to the plan. It was accessible through the western octagonal room and consisted of an unheated room, an indirectly heated room and a directly heated room with a single-person alveus above the furnace. To the east, the baths were expanded with several rooms of

559   I RT 396; Bartoccini (1929b) 79. 560  Dunbabin (1978) 264. 561  Tantillo (2010) 179; nr. 70 and possibly also nr. 75 of AD 355–361 in the catalogue of Tantillo and Bigi (2010). 562  Bigi and Tantillo (2010) 273. 563  Another small bathhouse in a good state of preservation was discovered not far from the Hunting baths, within the Late Roman defense wall (Musso and Matug (2012) 35). As the site has never been excavated, these baths will not be included in the following overview. 564   Ward-Perkins and Toynbee (1949) 192.

361

Gazetteer

figure 102

Plan of Hadrian’s Baths in Leptis Magna (Libya) after Nielsen (1993b) 166, fig. 183

362

figure 103

Gazetteer

Plan of the Hunting Baths in Leptis Magna (Libya) after Nielsen (1993b) 167, fig. 184

363

Gazetteer

figure 104

Plan of the so-called ‘Unfinished Baths’ in Leptis Magna (Libya) with hypothetical reconstruction (right) of the intended layout by Yvon Thébert in Thébert (2003) 699, pl. CXLIX.1–2; with permission by the publisher

unknown function, while to the west, a new entrance corridor covered with a tubi fittili-vault was added.565 The Eastern Baths are the latest baths to be discovered in Leptis.566 This modest public bathhouse was built in a new quarter of town, east of the Wadi Caam, during the first half of the 2nd c. The period of use seems rather limited, however, as the baths were abandoned in the second half of the 3rd c., when the piscina of the frigidarium was filled with debris.567 The building was transformed in an industrial area during the 4th c., when a bakery was installed in the heated section and warehouses and workshops in the northern rooms.568 At the beginning of the 5th c., the site was abandoned and submerged by the sand dunes. In the Islamic period, a lime kiln was constructed on top of the sand layers.569 The Early Baths probably date from the 1st c. AD, as it is built with a type of sandstone blocks that has not been attested after this century.570 It is not known when these baths fell out of use. In a later phase, these baths were incorporated into a large building that remained unfinished. Goodchild interpreted this as a large bathhouse, the so-called ‘unfinished baths’ (fig. 104).571 The reuse of an inscription dating from the reign of Gallienus gives a terminus post quem for the construction 565   Ward-Perkins and Toynbee (1949) 168, 171–72. 566  Laronde (1997); Laronde (1998). 567  Paulin et al. (2012) 131–132. 568  Paulin (2011) 26: Paulin et al. (2012) 132. 569  Paulin et al. (2012) 123. 570  Goodchild (1965) 18. 571  Goodchild (1965) 19–26.

of the hall in the west.572 The reuse of several statue bases with late 3rd or 4th c. inscriptions also seems to confirm this chronology.573 The hexagonal hall to the east, according to Goodchild, was meant to be the new caldarium.574 However, no traces of a heating system were ever found. The cold section, strangely oriented to the south, was never connected to the reused heated rooms of the older baths. The piscinae did not connect to a water supply system and were not coated with hydraulic mortar. If the building was indeed intended to be a bathhouse, it never functioned as such.575 The bathhouse was located outside the Byzantine walls, meaning that it is unlikely to identify this building project with the baths of Justinian mentioned by Procopius (see above chapter 2, p. 64–65). However, it cannot be excluded that a part of the bathhouse, probably the reorganised heated section and perhaps the hall with columns north of it, was used as a basic neighbourhood bath after the ‘imperial bath project’ had been abandoned.576 The Baths of the Schola (fig. 105) were small public baths located along the decumanus maximus, 80 m north-west of the quadrifons dedicated to Septimius Severus. The baths were not internally connected to the large peristyle courtyard that has been defined as a schola. The different types of walls and the different orientation of part of the baths make clear 572  Pentiricci (2010) 163. 573  Bigi and Tantillo (2010) 274. 574  Goodchild (1965) 20. 575  Maréchal (2013) 206–208. 576  Pentiricci (2010) 163–64.

364

Gazetteer

figure 105 Plan of the Baths of the Schola in Leptis Magna (Libya) after Bianchi Bandinelli et al. (1967) 103, fig. 245

figure 106

Baths of the Schola in Leptis Magna (Libya), southern small apsidal pool in the frigidarium seen from the southwest photo: author, June 2012

365

Gazetteer

figure 107

Plan of the church in the frigidarium of the West Baths in Mactaris (Tunisia) after Duval (1971) 306, fig. 8

that this building underwent several changes.577 The frigidarium was equipped with two semicircular single-person pools (fig. 106), while a small rectangular pool was added to the one of the adjacent rooms.578 A new set of rooms was created in the south-west corner of the building, probably replacing an older heated section. The first room had a small basin in a protruding niche, resembling the basin in the east octagonal room of the Hunting Baths (see above). The two other rooms had an alveus on their southern side. The layout, especially the addition of the small semicircular pools, could point to a late antique phase or perhaps even construction date of the building.579 After the baths had fallen out of use, the pools were possibly reused for an industrial function. The southern piscina of the frigidarium and the southern alveus of the first new heated room were subdivided into smaller basins. Mactaris (Maktar, Tunisia) As the city of an indigenous kingdom, Mactaris (Maktar, Tunisia) was heavily influenced by Carthage and Punic culture before it fell under Roman rule. It became a colonia under Antoninus Pius or Marcus Aurelius and subsequently witnessed its greatest expansion. The 4th c. saw some important restoration works and ecclesiastical building activity. The

577  Vergara Caffarelli (1957) 352. 578  The excavation of the bath was never published. The description of the building is based on the author’s own observations on site, in combination with the plan provided by Bianchi Bandinelli et al. (1967) 103, fig. 425. 579  The several ‘late’ restoration phases are also mentioned by Bianchi Bandinelli et al. (1967) 104. The ‘schola’ itself also seems to have been restored in the first half of the 4th c. (Pentiricci (2010) 161–62).

Byzantine period was characterized by reduced habitation and fortifications.580 The dedication of the Large East Baths is commemorated on an inscription of AD 199, found in situ in the vestibule.581 The imperial type plan fits well within this chronological framework. The aqueduct that supposedly supplied water to these thermae was cut around AD 400, forcing the baths to rely on newly dug wells. The large pools probably went out of use and were replaced by smaller ones. In the Byzantine period, the baths were converted into a fortress by building a wall around them.582 These later reorganisations are poorly understood, as the building has not been entirely excavated. The West Baths have not been dated, but a 2nd or 3rd c. date would fit with their plan.583 In a later phase, the frigidarium was transformed into a church (fig. 107). The southern part of the building had probably already collapsed by this time. The layout of the church was influenced by the pre-existing walls of the baths, with the presbytery built on top of the filled-up northern piscina. To the south, a small chapel was built directly onto the mosaic floor of the baths.584 Several graves and inscriptions dating from the Byzantine period were found.585 An old funerary stele was even reused as a sort of shrine, possibly to display the bones of a martyr. The church itself is dated to the 4th or the 5th c., as several coins have been found in a layer of ash inside the baths, pointing to the destruction of the building.586

580  Ennabli (1976b) 540; Leone (2007) 261–63. 581  Thébert (2003) 144. 582  Picard (1974) 24. 583  Thébert (2003) 146. 584  Duval (1971b) 305, 312–14. 585  Picard (1955) 176–79. 586  Duval (1971b) 315–17.

366

Gazetteer

figure 108 Plan of the Large Baths in Madauros (Algeria) after Yegül (1992) 224, fig. 255

Madauros (M’Daourouch, Algeria) Madauros became a colonia under Nerva and grew out to be an important ‘intellectual’ centre in Numidia. Apuleius was born here and Augustine studied in Madauros during his youth. The plan of the Imperial period city is difficult to reconstruct due to the building activity in the Byzantine period (mainly an Amazigh occupation at this point).587 During the Byzantine period, a fortress was built, incorporating the theatre.588 The Large Baths must probably be dated to the beginning of the 3rd c. Several statue bases seem to have been a part of the original decorative scheme and none pre-date the Severan period. An inscription mentioning a restoration of a bathhouse was found in several fragments (EA-19). It dates to the reign of Julian (AD 361–363) and probably referred to the Large Baths.589 The same can be said of a text commemorating the restoration of the thermae aestivae (EA-20) and dated to AD 364. It was found reused in a late wall of the Large Baths. Four other inscriptions were recovered from the excavations of the baths (EA-24, 42, 43, 44). While we cannot be certain that these have been found in their original location and hence refer to the 587  Gsell and Joly (1922) 9; Souville (1976) 541. 588  Gsell and Joly (1922) 126–32. 589  Thébert (2003) 215.

Large Baths, it seems very plausible that these thermae were still in use in the 5th c. The archaeological evidence confirms the many restorations and reorganisations. The most obvious intervention affected the heated section (fig. 108). The original caldarium was closed off, as was the heated room in the south-west corner.590 The heated part now consisted only of two rooms: the small rectangular room that used to connect the caldarium to the frigidarium, and the apsidal room. A new doorway was pierced in their communal wall to connect them.591 The small pool in the rectangular room was obliterated, while the pool in the apse was divided into two small single-person pools.592 Perhaps this reorganisation of the heated section can be identified with the interventions mentioned the inscription of AD 407–408 (EA-42). The lavacra in the text could be the heated rooms or the pools of the heated rooms.593 It is important to notice that, according to the inscription, the baths had fallen out of use (desolatam) before the repairs. The other works mentioned in the inscriptions are more difficult to identify with the archaeological evidence, as restorations of the pools (EA-19; 20), the patinas (cauldrons?), (EA-20), the 590  Thébert (2003) 215–16. 591  Krencker et al. (1929) 220. 592  Gsell and Joly (1922) 99–100. 593  Maréchal (2015) 159.

367

Gazetteer

figure 109 Plan of the church in the frigidarium of the Small Baths of Madauros (Algeria) after Duval (1971) 299, fig. 2

suspensurae (EA-24) and the decoration (EA-19; 24) are difficult to detect. The Small Baths of Madauros were located just north of the Large Baths. The similarity of building techniques seduced scholars to believe that the buildings were contemporaneous and perhaps built by the same architect.594 One of the inscriptions found in the Large Baths (EA-19) may actually refer to these Small Baths.595 If we consider the Large Baths to be the ‘thermas aestivas’ of EA-20, the Small Baths might well have been Madauros’ ‘winter baths’. The late antique restorations are less striking and perhaps did not alter the organisation of the rooms. At an unknown date in Late Antiquity, the frigidarium was converted into a church.596 Post-holes were pierced in the floor, delimiting a chancel. The north-western piscina was filled up and used as an entrance, while the north-eastern one was raised to form the presbytery (fig. 109). The vault that covered the church (and the baths?) was made out of tubi 594  Yegül (1992) 222; Thébert (2003) 216. 595  Thébert (2003) 217. 596  Duval (1971b) 298–304.

fittili.597 If we assume that the restored Large Baths and the church in the Small Baths existed at the same time, Madauros would have had an attractive social hub. It cannot be ruled out that the persons responsible for restoring the Large Baths also constructed the church.598 Nagàzza (Libya) In 1924 a bathhouse was discovered along the road leading from Homs to Tripoli. The building was built on a terrace partially cut out of the rock (fig. 110). On the basis of the mosaics, the baths were dated to the (late) 3rd c.599 Possibly the bath was still in use in the early 4th. The plan was characterized by an oblong hall with an east-west orientation and a cold pool preceded by a low labrum (or rather pediluvium?) on its eastern end.600 The piscina was fed by a reservoir lying just behind 597  Gsell and Joly (1922) 103. 598  Both baths had probably been city property during the High Empire. What happened to each building in Late Antiquity was probably the responsibility of the same people. 599  Bartoccini (1927) 248. 600  Bartoccini (1927) 242.

368

Gazetteer

figure 110 Plan of the Baths of En-Nagazza (Libya) after Bartoccini (1927) 242, unnumbered fig.

it, cut out of the rock and fed by rainwater. In the southern side of the hall, which probably served both as frigidarium and apodyterium, a doorway gave access to another cold room and 4 heated rooms, the third of which had an alveus built directly onto the suspensura and thus possibly a later addition. Through the fourth warm room, probably the exit-tepidarium, the bather could rejoin two separate cold rooms that connected to the hall. The polychrome mosaics found in the heated rooms depicted marine and nilotic scenes, as well as a kantharos with vines.601 Numluli (Henchir Matria, Tunisia) The small site of Numluli was monumentalized with a capitolium by a rich decurion from Carthage in AD 170.602 The local bathhouse, 100 m east of the latter, had the same orientation and may well have been constructed around the same time.603 The use of opus vittatum in the apsidal room in the west may point to a late antique phase of the building. In the Byzantine period, the baths were dismantled and the large blocks of the opus africanum walls were reused for the construction of a

601  Bartoccini (1927) 245–47, unnumbered figs. 602   C IL 8.26121. 603  de Vos Raaijmakers and Attaoui (2013) 152, 384–85, pl. 144.

fortress. A church with a quatrefoil plan 50 m north-west of the bathhouse predates the Byzantine period.604 Pupput (Souk el-Abiod, Tunisia) At least three baths have been discovered so far in Pupput, but none have been properly excavated. The Baths of the Crater, located in the Roman city centre, is rather small (45 × 35 m) and may have been built at a late date in the High Empire, judging from the surviving mosaics.605 Several restorations phases could point to continued use in Late Antiquity. The presence of a small semicircular pool, which seems to have been inserted at a later date in a rectangular niche, could be a late antique addition. When the baths had fallen out of use, an olive oil production unit was created in the hollowed-out hypocaust of the heated rooms.606 The Baths of the Monetary Treasure to the south and the Baths of the Octagon to the north still await investigation.

604  de Vos Raaijmakers and Attaoui (2013) 152–53. 605  Ben Abed and Hanoune (2010) 987–89. 606  Ben Abed and Hanoune (2010) 989–90.

369

Gazetteer Rusguniae (Tametfoust, Algeria) The small bathhouse was partially excavated in 1964.607 Construction probably occurred after the Severan period.608 The building was still in use in the 4th c., when a hypocaust pillar was restored using clay containing a coin of Constantius II.609 Furthermore, the sherds of ceramic found in the preparation layer of a new mosaic also dated to the same century.610 After the bath fell out of use at an unknown date, the shell of the building was reused for housing and a workshop.611 Sabratha (Libya) Sabratha was a Punic trade colony that came under Roman influence from the 1st c. AD onwards and was granted the title of colonia in the 2nd c. AD. During the reign of the Antonines, the city reached its largest size. Some of the city’s most important buildings, such as the temple on the forum, the theatre, and most of the baths, were built in this period.612 Several restorations of monumental buildings, including the curia, the basilica, temples and the forum, were carried out during the 4th c.613 Procopius tells us that Justinian ordered the construction of walls and a new church.614 Archaeological data points to building activity, especially in relation to ecclesiastical architecture.615 At present, 6 bathhouses have come to light in Sabratha (map 37). The Seaward Baths are the largest discovered so far and are probably the oldest (map 37, nr. 3). A Flavian date has been proposed, as they seem to have been part of the original layout of one of the oldest quarters of the town.616 A long use of the baths, at least until the 4th c., can be deduced from the different levels of mosaics laid in the oblong corridor.617 The heated part of the baths may even have been in use in the Byzantine period, when new water pipes and decorative elements were added to the original design.618 The northern section has been destroyed by sea erosion, making it difficult to reconstruct the plan. The Baths under Basilica 3, the largest in Sabratha, were razed to the ground when the new basilica was erected in the 4th c.619 It is not certain that the bathhouse was destroyed to make way for the ecclesiastical building. If they were already out of use, this meant that a considerable building lot in the 607  Guéry (1962–1965). 608  Thébert (2003) 221. 609  Février (1967) 100. 610  Germain (1969) 117. 611  Thébert (2003) 221. 612  Kenrick (2009) 38–39; for the baths, see Bonacasa (2010). 613  Leone (2007) 93; Sears (2007) 74–75. 614  Procop., Aed. 6.4.13. 615  Leone (2007) 276–79. 616  Haynes (1965) 121. 617  Bonacasa (1997) 171. 618  Bonacasa (2010) 42. 619  Bonacasa and Carra (2003) 406.

city centre was available for reuse. The presence of water supply systems (drains, supply channels, reservoirs) may have been a good reason to construct a basilica and adjacent baptistery on this location. It seems that a part of the hypocaust system was even reused.620 The Baths of the Theatre (fig. 111) were constructed in the Severan part of town, within an insula of 26 × 34 m (map 37, nr. 2). A recent re-examination of the building by the Italian archaeological mission in Libya placed the construction date in the 2nd c., with important reorganisations in the 3rd. During the 4th c., the baths seem to have been severely damaged, although the bathing rooms were still partially used until the end of the Byzantine period, be it in a reduced form.621 The west alveus of the caldarium was transformed into a service room paved with spoliated slabs of marble. The large apodyterium lost its function as a new water pipe was laid across the benches and the light well of the frigidarium was transformed into a small water basin, supplied by the same water channel as the latrine. Several windows were also blocked and the mosaic floor of the apodyterium was restored with irregular slabs of spoliated marble. The Bath of Oceanus is a small building (23 × 25 m) close to the ancient harbour and which may have belonged to an association (map 37, nr. 1).622 The baths were built on top of an earlier bathhouse of larger dimensions and dating from the second half of the 1st c. AD, judging from the mosaics.623 The bathhouse which survives today has mosaics dating to the early 2nd c., while restorations were carried out in the Severan era.624 The baths were damaged at the start of the 4th c., but probably continued to function on a reduced scale until the second half of the 4th c.625 The baths near the Fountain of the Serapides (Reg. VI) may have been a private bath. Possibly built in the 1st c. AD, the baths were in use until the Byzantine period.626 The Bath of Region VII is the only one that has been subject to systematic archaeological excavations (map 37, nr. 4). The style of the mosaics and the ceramics found in test pits both point to a construction date in the second half of the 2nd c.627 Subsequent changes to the original plan were minimal. Only a new heated room was added, perhaps in the early 4th c. The most recent mosaics date from the beginning of this century. The baths probably never recovered from the damage of the earthquake in AD 365.628

620  Bonacasa and Carra (2003) 407. 621  Bonacasa Carra and Scirè (2012) 371–78. 622  Bonacasa (2004) 328. 623  Bonacasa (2004) 330. 624  Dunbabin (1978) 267; Bonacasa (2004) 327. 625  Joly (2000) 146; Bonacasa (2004) 328. 626  Bonacasa (2010) 37, 42. 627  Brecciaroli Taborelli (1974–1975) 141–44. 628  Brecciaroli Taborelli (1974–1975) 144.

370

map 37

Gazetteer

Sabratha (Libya), plan of the city with the location of public baths and the most important buildings after Sears (2007) 168, fig. 22 7: Temple of Hercules 1: Baths of Oceanus 8: Antonine Temple 2: Theatre Baths 9: South Forum Temple 3: Seaward Baths 10: Forum 4: Bath of Regio VII 11: Temple of Serapis 5: Temple of Isis 6: Theatre

371

Gazetteer

figure 111 Plan of the Baths of the Theatre in Sabratha (Libya) after Bonacasa Carra and Scirè (2012) 367, fig. 21

Seguia (Tunisia) During prospection by the Italian archaeological mission, a bathhouse was discovered in the Cap Bon region.629 The surface finds of the site revealed some 3rd and early 4th c. material, although nothing connects these sherds to the construction of the baths.630 Hence there is no reliable evidence to attribute a late antique construction date or even a late antique phase of use for these baths. 629  Acquaro et al. (1973) 79–80. 630  C  ontra Leone (2007) 90.

Thenae (Thyna, Tunisia) The port city of Thenae gained some importance during the 2nd c. when it was awarded the status of colonia. The city probably prospered in the 4th c. with two bathhouses built around this time (see above, C60 and C61). The third bathhouse found at Thenae was the Large Bathhouse. It has been discussed in most modern research on Roman baths because of its original plan (fig. 112). The intricate articulation of curvilinear forms is an exceptional example of a ‘grammatical structure’—two concentric rings around a central round room—presented

372

Gazetteer

figure 112 Plan of the Large Baths in Thenae (Tunisia) after Nielsen (1993b) 171, 190

in chaotic design.631 The mosaic of the frigidarium has been dated to the end of the 3rd c.632 The early excavations did not identify late antique interventions.633 However, a small separate bathhouse south of the main complex is reminiscent of the suite of small bathing rooms added at a later stage to the Hunting Baths in Leptis Magna. The small baths consisted of a cold, a tepid and two hot rooms. The frigidarium had two small pools.634 It is difficult to interpret these baths without any archaeological evidence. They may have been used as the women’s section or as baths with a special (medical?) use. Thuburbo Maius (Henchir Kasbat, near El Fahs,Tunisia) Some 35 km south-west of Uthina, in the same fertile Meliana valley, lay the city of Thuburbo Maius. It became a municipium under Hadrian and a colonia under Commodus.635 The town flourished during the 4th c., when a vast building program restored and expanded the existing town fabric.636

631  Yegül (1992) 246. 632  Dunbabin (1978) 273. 633  Drappier (1911–1913). 634  Thébert (2003) 160–61. 635  Ennabli (1976c) 916–17. 636  Leone (2007) 91; Sears (2007) 48.

Five bath complexes have been discovered in the city (map 38). The Summer Baths (fig. 113), mentioned in an inscription found on site (see below) were the largest with a surfaced covering around 2,800 m2 (map 38, nr. 4). They were erected at the end of the 2nd or beginning of the 3rd c., when the town centre was embellished during an important construction programme. The exact date has been placed around AD 200–225 on the basis of ceramics, recovered from a small test pit in the frigidarium, and on the style of mosaics.637 An inscription found in several fragments inside the baths (EA-17) is dated to the reign of Constantius II (AD 361) and commemorates the completion of the baths after a period of 8 years. It specifies that lavacra were restored, because ‘they wept rather than they washed’. Looking at the archaeological remains, we can observe that the caldarium and more specifically the alvei were reorganised. The original cruciform plan was changed into a three-apsidal room, with the alvei changing shape from rectangular to semicircular forms.638 The reorganisation of the pools probably entailed the transformation of a room in the north corner of the complex into a water distribution centre. A well (17 m deep) and two reservoirs were installed 637  Thébert (2003) 167. 638  Drappier (1920) 72.

Gazetteer

map 38

Thuburbo Maius (Tunisia), plan of the city with the location of public baths and the most important buildings after Sears (2007) 152, fig. 6 7: Capitol 1: Baths of the Stars 8: Temple of Mercury 2: Baths of the Capitol 9: Small Temple 3: Winter Baths 10: Market 4: Summer Baths 11: Market 5: Temple 12: Temple of Balaat 6: Forum

373

374

figure 113

Gazetteer

Plan of the Summer Baths in Thuburbo Maius (Tunisia) with late antique reorganisation (left) in Thébert (2003) 609, pl. LIX.3–4; with permission by the publisher

here, while the doorways to the other bathing rooms were blocked.639 We cannot be certain that these interventions are the same as those recorded in the inscription, but as no other important works seem to have altered the plan to such an extent, it is quite possible that this was the reorganisation immortalized in stone. The lavacra should then be interpreted as the alvei.640 The long use of the baths can also be deduced from the mosaics in the apodyterium and corridors of the end of the 4th or beginning of the 5th c.641 A reservoir was not filled up until the middle of the 5th c.642 The resemblance between these ‘summer baths’ and the ‘summer baths’ or Large Baths of Madauros is remarkable, especially the presence of a large semicircular latrine (compare figs. 108 and 113). Against the latrines in Thuburbo Maius, a small complex comprising three rooms on hypocaust and two unheated rooms has been discovered.643 The available data is too scarce to make any conclusions, especially about a possible construction date. However, such a simple succession of heated rooms is reminiscent of heated rooms found in the palaestra of the north baths in Volubilis (see below, p. 377). New investigations are

639  Drappier (1920) 67. 640  Maréchal (2015) 158–59. 641  Ben Abed-Ben Khader et al. (1985) 29–31. 642  Ben Abed-Ben Khader et al. (1985) 2, n. 5. 643  Thébert (2003) 169.

needed to decide whether these rooms functioned as a small bathhouse or had a different function. The Winter Baths (fig. 114; map 38, nr. 4) have been identified by an inscription found in the curia (EA-34). Only one building was large enough to compete with the Summer Baths.644 The construction date is placed at the end of the 2nd or the beginning of the 3rd c., as it is believed that the Winter Baths were built at the same time as the Summer Baths. The inscription of the curia commemorates restoration works carried out under the reign of Arcadius and Honorius (AD 395– 408). A second inscription (EA-36), found inside the baths and dated at the beginning of the 5th c., is more specific. It mentions the restoration of the cella soliaris, the restoration of the copper and lead implements and the addition of a new piscina. Archaeological data corroborates the restorations in the text. An oval piscina (1.8 × 2.5 m; 0.4 m deep) covered with a nerved dome and with a plain white mosaic floor must have been the cochleam (‘shell’) that was added to the frigidarium.645 Probably at the same time, the two large rectangular pools were filled in and concealed under a new mosaic. The interventions in the caldarium—possibly the cella soliaris mentioned in the inscription—concerned the reduction of the rectangular alveus and the construction of a new apsidal pool. The hypocaust pillars of this last phase were made out

644  Thébert (2003) 170. 645  Thébert (2003) 171.

375

Gazetteer

figure 114 Plan of the Winter Baths in Thuburbo Maius (Tunisia) after Thébert (2003) 611, pl. LXI.3–4

of greyish limestone monoliths.646 At the end of the 5th or the beginning of the 6th c., a new mosaic was laid in the frigidarium (black and white geometric patterns). The construction of Baths of the Labyrinth (fig. 115) cannot be posterior to the 3rd c., as the well-dated Temple of Mercurius (AD 211) and adjacent buildings were partially built against the baths.647 However, important restorations were still carried out at the end of the 3rd or beginning of the 4th c. The mosaics that are contemporaneous with these reorganisations date from this period.648 The rectangular pool of the caldarium was reduced in size, while a new round basin was created. As the latter was quite shallow, it has been interpreted as a pediluvium, although a type of shower-installation cannot be ruled out.649 In the 5th c., a small basin was added in the apodyterium, but fell out of use later that century, when a new mosaic covered its traces. A coin dated between AD 410 and

439 was found in the blockage of a furnace. This could mean that the baths, or at least the caldarium, fell out of use in the first half of the 5th c. Small basins in the frigidarium also date to this period. Perhaps these basins were part of an industrial structure that was installed after the disaffection of the baths.650 Or the baths continued under a simplified layout. The Baths of the Capitolium (fig. 116; map 38, nr. 2) were constructed during the second half of the 2nd or the first half of the 3rd c. Both the style of the mosaics and the ceramics found in small test pits investigating the foundation trenches of the walls suggest this date.651 The original plan was altered several times. The last datable mosaics were laid in the 4th c.652 In the last phase of the baths, a new caldarium was created by transforming a room with a rectangular alveus. The door to the original caldarium was blocked and an apsidal pool was added.

646  Drappier (1920) 60. 647  Thébert (2003) 172. 648  Alexander et al. (1980) 30. 649  Thébert (2003) 173.

650  Alexander et al. (1980) 23. 651  Alexander et al. (1980) 80. 652  Alexander et al. (1980) 87.

376

figure 115

Gazetteer

Plan of the Baths of the Labyrinth in Thuburbo Maius (Tunisia) after Alexander et al. (1980) plan 8

figure 116 Plan of the Baths of the Capitolium in Thuburbo Maius (Tunisia), with its late antique reorganisation (right) in Thébert (2003) 612, pl. LXII.4–5; with permission by the publisher

Thugga (Dougga, Tunisia) Thugga had a long pre-Roman history before it became a pagus civium Romanorum connected to Carthage in 46 BC. The Punic constitutional structures remained in use and lasted until AD 205 before Thugga became a municipium and only in the reign of Gallienus (AD 253–268) it was granted the title of colonia.654 Thugga’s survival into Byzantine times is attested by the presence of a Byzantine fortress.655

The Baths of Aïn Doura probably date from the beginning of the 3rd c.656 In the 4th c., the baths were probably restored and embellished, as can be deduced from the mosaics.657 The remains are too badly preserved to make conclusions about the nature of the restorations or the outlook of the baths in Late Antiquity. The second large public baths, the Licinian Baths, would date to the reign of Valerian (AD 253– 260), rather than to the reign of his son Gallienus (AD 253– 268).658 Indeed, an inscription found in three fragments, one in the baths, commemorated the restoration of the atrium therma(arum)(Lic)iniarum (EA-26). The text is dated to the years AD 367–383, meaning the baths were still in use some 100 years after their dedication. The restoration of the ‘shabbily built atrium’ is difficult to recognize in the archaeological

653  Alexander et al. (1994) 1, 15–19. 654  Ennabli (1976d) 917; Huß (2006b). 655  Pringle (1981) 244–46.

656  Thébert (2003) 175 on the basis of mosaics. 657  Wiedler (1999) 362–63; Thébert (2003) 175; both after an unpublished dissertation. 658  Thébert (2003) 177.

The Baths of the Stars or Northern Baths have not been published. The mosaics found throughout the baths date from the late 2nd or early 3rd c. At the end of the 4th or the beginning of the 5th c., new drains cut through some mosaic pavements, revealing a late restoration phase.653

Gazetteer record. The reorganisation of the heated section is better understood. One of the small pools of the exit-tepidarium (between caldarium and frigidarium) was destroyed and covered with a floor of marble slabs.659 At an unknown date, an oil press was installed in one of the heated rooms.660 The Baths of the Cyclops were dated to end of the 3rd c. on the basis of mosaics.661 In a second phase (4th c.?), the tepidarium, probably with a single-person pool in an apse, was closed off from the frigidarium and equipped with a second furnace.662 All of the mosaics, except the one of the cold room, were carefully cut out and removed at an unknown date.663 Near the forum, a small bathhouse was recently re-examined.664 The simple linear layout, the proximity to the Byzantine fortress and the similarities with other Byzantine fortress baths (see, chapter 4) points to a Byzantine or even Early Islamic construction date. As can be concluded from the evidence above, Thugga still had several functioning baths in the 4th c. The fact that an inscription commemorates the construction or restoration of a water reservoir during the reign of Gratian and Valentinian II, confirms that the water supply of the city was still an important concern for local authorities.665 The situation in the 5th c. and later is less clear, as the archaeological data pertaining to this period was neglected in earlier research. Thysdrus (el-Djem, Tunisia) A small bathhouse was discovered in the south-eastern quarter of the Roman city of Thysdrus (El-Djem). The mosaic of the frigidarium, accompanied by an inscription about invidia, has been dated to the end of the 3rd c.666 It is likely that the bath was still in use in the early 4th c. Volubilis (Wālila, Morocco) The site of Volubilis, located in a fertile plain and encircled by the river Fertessa and Khoumane, had a long Amazigh and Punic history, before it became a municipium under Claudius (around AD 40).667 As in many North African towns, the Severan period was its heyday. All too often, the withdrawal of Roman troops and magistrates to northern Morocco (around Tingi, modern Tangier) in AD 285 has been used a terminus ante quem for all building activity inside the city. However, the occupation of the site did not cease, as the local popula659  Poinssot and Lantier (1925) xxx–xxxi. 660  Thébert (2003) 177. 661  Dunbabin (1978) 40, n. 15. 662  Thébert (2003) 180. 663  Gauckler (1902) ccxvi; Merlin (1903) 39. 664  The excavations had not been published at the time of writing, but some preliminary results were presented at the ‘Byzantine and Medieval Islamic baths and their use AD 600–1200’ conference in Rome (5–6 June 2014) by C. Touihri. 665  Merlin (1903) 95. 666  Wiedler (1999) 364. 667  Risse (2001) 3.

377 tion—forming a large part of the total number of citizens— remained. Archaeological excavations have revealed important evidence of late habitation, production and building activity.668 Even under Arab rule, the occupation of the site continued, until the new city of Moulay Idriss was created in the 8th c.669 There were 4 large and 7 small baths inside the city walls (map 39). The North Baths were the largest in town (ca. 2,000 m2) and have been dated by test trenches to the period between AD 60 and AD 80 (map 39, nr. 1).670 The most important restorations have been dated at the end of the 2nd or the beginning of the 3rd c., while a separate small bathhouse was created inside the former palaestra (fig. 58).671 The construction of the latter probably happened after the principal baths fell out of use during the 3rd c. and were transformed into habitations. The small baths were also transformed and probably remained in use until the end of the 3rd c.672 It is interesting to notice that both the frigidarium and the caldarium were equipped with an additional small pool in the later phases.673 Apparently, these single-person pools were already built before the end of the 3rd c., when the baths fell out of use. The small baths in the palaestra are very basic, only consisting of a single heated room, possibly without an alveus.674 Some baths in Volubilis had already gone out of use by the 4th c. The Bath under the House of the Cistern was razed to the ground when the house that gave the baths its name was built in the 3rd c.675 The Baths of the Capitolium, built at the end of the 1st c. AD and enlarged around the end of the 2nd c., probably went out of use during the 3rd c. (map 39, nr. 3).676 Neither the plan nor the decoration reveals clear traces of late antique interventions. The same can be said for the Baths of Gallienus, named after a building inscription that possibly did not come from the baths (map 39, nr. 2).677 The baths of the Palace of Gordian can be approximately dated by an inscription to the period AD 190–240.678 The baths were part of a large domus, but must have had a (semi-) public use as there was a separate street entrance (map 39, nr. 4). 668  Lenoir (1985) 425; Es-Sadra (2012) 644–51. 669  Lenoir (1985) 426. 670  Lenoir (1991) 156. 671  Thébert (2003) 275. 672  Thébert (2003) 273–74. 673  Thouvenot (1945) 160–61. Thébert (2003) 274–75 sees these small pools as original features of the plan. However, the excavators mention that the walls of these pools abutted the outer walls of the baths (which is also clearly visible on pl. X.2 in Thouvenot (1945)) and were constructed using a different technique (spoliated stones; Thouvenot (1945) 160–61). Both pools also protrude from the core of the building, in the case of the cold pool even onto the street. 674  Thébert (2003) 275. 675  Zehnacker and Hallier (1965) 115–17. 676  Akerraz (1986) 112. 677  Thébert (2003) 278. 678  ILAfr 614.

378

Gazetteer

map 39 Volubilis (Morocco), plan of the city with location of public baths after Risse (2001) 112, fig. 165 1: North Baths 2: Baths of Gallienus 3: Baths of the Capitolium 4: Baths of the ‘Palace of Gordian’ 5: Baths of the House of Hercules 6: Baths of the Western Quarter 7: Baths of the House of Venus 8: Baths of the House of the Sundial 9: Baths of the House of Orpheus 10: Islamic suburban baths

As it was the inhabitants of this type of residence that left in AD 285, it is possible that the baths fell out of use at the end of the 3rd c. In any case, there are no traces of important reorganisations after the initial building phase. During the medieval period, a lime kiln was constructed here.679 In some cases, there is evidence that a bathhouse had at least a late antique phase. The Bath of the House of Venus, possibly semi-public with an entrance from the street, remains undated (fig. 117; map 39, nr. 7). The house itself may have been constructed in the second half of the 3rd c., but the baths seem to have been a later addition.680 The use of tubi fittili for the vaults and terracotta tiles with spacers for the wall heating could point to a late phase.681 The Baths of the House of Orpheus were also (semi-) public, judging from its street entrance (map 39, nr. 9).682 A large hall, similar to the one in the Baths of the House of Venus, preceded the frigidarium. 679  Thouvenot (1958) 31. 680  Étienne (1960) 79; Thébert (2003) 282. 681  Thouvenot (1958) 59. 682  C  ontra Thouvenot (1941) 52.

It was a scaled-down version of the halls found in the earlier baths (see for example the Baths of Gallienus, the North Baths). The piscina was divided by a wall into two pools of unequal size.683 The small pool must have had a separate function, perhaps for medical use or a private pool for the owners of the house. The Baths of the House of the Works of Hercules post-date the reorganisation of the north-western quarter of the city in AD 170, as the bathhouse encroaches upon the street (fig. 118; map 39, nr. 5).684 The street access and the large room preceding the frigidarium identify the complex as the same type of baths belonging to the House of Venus and the House of Orpheus. The addition of a small semicircular pool in the first heated room, the use of terracotta tiles and spacers for wall heating, the use of stone monoliths as pilae and the use of spolia in the walls point to a late phase of use (middle or end of the 3rd c.?).685 The Baths of the Western Quarter were also constructed after AD 170, blocking a street running 683  Thouvenot (1941) 58–60. 684  Thébert (2003) 280. 685  For the spolia, see Thouvenot (1948a) 82–84.

379

Gazetteer

figure 117 Plan of the Baths in the House of Venus in Volubilis (Morocco) after Etienne (1960) pl. XVII

figure 118

Plan of the Baths of the House of the Works of Hercules in Volubilis (Morocco) after Etienne (1960) pl. IV

380

figure 119

Gazetteer

Plan of the Baths of the Western Quarter in Volubilis (Morocco) after Etienne (1960) pl. XIV

figure 120 Plan of the Baths of the House of the Sundial in Volubilis (Morocco) after Etienne (1960) pl. XII

381

Gazetteer along the aqueduct and abutting the House of the Nereïds (fig. 119; map 39, nr. 6). Test pits under the street pavement delivered ceramics of the 3rd c.686 The facilities are similar to those of the semi-public houses described above, yet the layout is simpler. The vestibulum / apodyterium, the frigidarium and three heated rooms succeeded each other in a linear reversing route. The caldarium had two alvei, one small semicircular and one rectangular.687 Very similar in plan, was the Bath of the House of the Sundial (‘thermes de la maison au cadran solaire’ / ‘thermes du cardo’; fig. 120; map 39, nr. 8). Robert Étienne proposes a late 3rd c. date on the basis of the poor construction technique and the encroachment on a cardo, but an even later date cannot be ruled out.688 Outside the city walls, near the banks of the river Khoumane, the

686  Étienne (1960) 143. 687  Thouvenot (1948b) 130. 688  Étienne (1960) 102.

remains of a bathhouse dating from the Islamic period have been discovered (see chapter 4, p. 175). The picture that emerges of late antique Volubilis is one of a town without strict Roman organisation. The monumental heritage was kept in use if possible, but private buildings encroached on the roads and former public buildings were turned into habitations.689 By the 4th c., the large public baths in town, the North Baths and the Baths of the Capitolium, had fallen out of use. The small neighbourhood baths that had been constructed during the 3rd c. probably cushioned this loss and were kept in use, some of them maybe for a long time. The Baths of the Western Quarter and the Baths of the House of the Sundial may have been constructed in a very late phase. The type of semi-public bathhouse, built inside an existing domus, is reminiscent of the semi-public late antique baths in Cuicul (C40, C41, C42).

689  Lenoir (1985) 425.

Baths in Cyrenaica New Construction C65. Apollonia (Marsa Susa, Libya), Late Roman Baths

Excavation history: (D) The so-called ‘Byzantine baths’ were partially excavated by the University of Michigan in the 1960s. It seems, however, that the complex should

figure 121

actually be interpreted as two separate baths: the unfinished ‘Byzantine’ baths (to the east, see below) and an earlier ‘late antique’ part (to the south), (fig. 121). Bibliography: Pedley (1967); Pedley (1976). Construction: (D) The excavations did not permit dating of the different phases of the complex. On the basis of the

Plan of the Late Antique and Byzantine Baths in Apollonia (Libya) in Pedley (1976) 226, fig. 1; with permission

383

Gazetteer

C66. Apollonia (Marsa Susa, Libya), Byzantine Baths

plan, we could propose a late antique date. These baths may have been built when the ‘Roman baths’, located 20 m to the east, fell out of use sometime in the 4th c.690 Continued use and abandonment: Unknown. Plan: The north-south oriented baths were built in opus africanum. To the north, a paved zone with a semicircular apse opened onto an east-west oriented portico. To the south, a paved courtyard had two pools against its southern side. Between these two pools, a narrow passage led to a small square room with a single-person pool. To the east lay a reservoir.691 The area to the south has not been excavated. A branch of an aqueduct reached this southern zone, making it likely that more water-related structures could be found here. Decoration: Unknown. Technology: The late antique bath was perhaps built close to the ‘Roman bath’ so that the water supply of the aqueduct could be reused. The water supply system still functioned, as a small fountain and a latrine were constructed in the disused Roman baths at a late period, perhaps in the 6th c.692 Construction technique: Opus caementicium Bathhouse surface: Unknown GPS coordinates: 32°54’8.24”N, 21°58’5.62”E

Excavation history: (D) The so-called ‘Byzantine Baths’ were only partially excavated by the University of Michigan in the 1960s. Bibliography: Pedley (1967); Pedley (1976). Construction: The excavation did not yield datable material in context. A terminus post quem of AD 365 has been proposed, linking the construction of the baths to the earthquake that hit the Libyan coast and allegedly destroyed the Roman baths of Apollonia.693 However, the type of plan, closely resembling the baths in Taucheira (see below), points to a 6th-c. date. Continued use and abandonment: (B) The absence of hypocaust tiles and wall heating (tubuli or tegulae) in the destruction layer and the unfinished state of the pools make it unlikely that this building ever functioned as a bathhouse.694 In the first quarter of the 7th c., the hypocaust was filled up with debris, so that the building could serve a different function. Plan: In contrast to rubble concrete of the ‘Late Antique Baths’ (C65, fig. 121), the east wing of the complex was built in opus quadratum. Just as in Taucheira, the function of the frigidarium is ensured by the (pre-existing) courtyard with its pools. An entrance then led into the first room, which was divided into two unequal parts by an east-west partition wall. The second and third room to the east were constructed on a lower level (1.85 m below threshold), presumably to receive a suspended floor. The central room, the ‘tepidarium’, had two ‘flues’ in the lower part of the wall connecting to the outside. According to the excavator, these were chimneys, but they are more likely to be traditional furnace mouths.695 To the east of the ‘caldarium’, there are three almost square ‘cubicles’, interpreted as pools. The north and south ‘pools’ were connected to the central one by terracotta pipes immediately above the floor. This strange arrangement could be explained by looking at Taucheira: the middle ‘alveus’ would actually be the boiler room. This would explain why the east wall of the middle pool was pierced, as a stoke hole to fire up the furnace underneath the boiler must have been present. For unknown reasons, the bathhouse was never finished. The suspended floors were instead filled with a homogenous layer of debris, containing material of the first quarter of the 7th c., after which some poorly-built walls seem to point to squatter occupation.696 Decoration: As the bathhouse was never finished, there is no evidence for any intended decoration.

690  Rebuffat et al. (1987) 277. 691  As indicated on fig. 1 in Pedley (1976) 226. 692  Goodchild (1976) 180–83.

693  Pedley (1967) 146. 694  Pedley (1967) 145. 695  Pedley (1967) 145. 696  Pedley (1967) 146.

C65 Apollonia (Marsa Susa, Libya), Late Roman Baths (After Pedley (1976) 226–27, fig. 1 and 2)

384

Gazetteer

figure 122 Plan of the Baths of Trajan in Cyrene (Libya), with the insertion of the Byzantine Baths (in orange) in Kenrick (2013) 200, fig. 139; with permission

Technology: Neither the hypocaust pillars nor the wall heating were put into place. The comparison with the baths of Taucheira points to a furnace with a surmounted boiler, flanked on either side by an alveus. For the water supply, the builders might have envisaged a branch of the aqueduct that also fed the Roman baths and the late antique baths. Construction technique: Opus quadratum Bathhouse surface: ca. 360 m2 GPS coordinates: 32°54’8.57”N, 21°58’6.21”E

C67. Cyrene (Shahat, Libya), Byzantine Baths

C66 Apollonia (Marsa Susa, Libya), Byzantine Baths (After Pedley (1976) 226–27, fig. 1 and 2)

Excavation history: The baths near the sanctuary of Apollo were excavated in the 1920s by the Italian colonial government. Inside the large hall of the Imperial period baths (Baths of Trajan or the Myrthusa baths), a smaller late antique bathhouse, the ‘Byzantine Baths’, was discovered (fig. 122). Bibliography: Anti (1927); Oliverio (1931); Goodchild (1971) 177–87.

385

Gazetteer

A doorway, marked by two columns, gave access to two successive cold rooms, each with a piscina in an apse on the eastern side. Three small rooms lying on east-west axis are of unknown function, but they must have acted as a heat lock between the cold and heated sections. The following room, built on a suspended floor, was probably a tepidarium, while the northernmost room, also with a hypocaust and with two lateral alvei, must have been a caldarium.699 Decoration: The cold rooms had plain white or black and white geometric mosaic floors. Technology: Little is known about the technical aspects of these baths. Construction technique: Opus quadratum Bathhouse surface: ca. 625 m2 GPS coordinates: 32°49’22.94”N, 21°51’12.95”E

C67 Cyrene (Shahat, Libya), Byzantine Baths (After Goodchild (1971) 133, fig. 17)

Construction: (D) The excavations did not permit a reconstruction of the chronology either building. The later baths were probably constructed when the imperial baths fell out of use, although it is not impossible that a part of the latter still functioned.697 Continued use and abandonment: The lack of excavation data and detailed analyses of the remains make it difficult to interpret the layout and the several building phases of these baths. It seems that the baths remained in use until the Early Islamic period, as some Arabic inscriptions inside the rooms have been discovered.698 It is not known if the building still functioned as baths, or as something else. Plan: The Byzantine Baths, built entirely from reused limestone blocks, were entered by a vestibule with two lateral colonnades protruding onto the street. Continuing north, a large polyvalent room was flanked on the east by a latrine. 697  Stucchi (1975) 469–70 without presenting any evidence. 698  Goodchild (1971) 132.

Cyrene (Shahat, Libya), Late Baths Excavation history: Unknown. Bibliography: Goodchild (1971) 143; Stucchi (1975) 471. Construction: According to Richard Goodchild, the baths abutted and therefore post-dated the 6th c. basilica. Sandro Stucchi, however, claimed that baths were partially covered by the church, and were therefore an earlier construction, possibly dating between the 1st and 3rd c. A reexamination of the remains should give a decisive answer. The small water basins in the room nearest to the furnace support Goodchild’s hypothesis, as such small basins are often found in late antique baths. Continued use and abandonment: Unknown. Plan: The bathhouse had at least three rooms with suspended floors. One of these had a small semicircular basin or pool. Decoration: Unknown. Technology: The hypocaust had normal pillars made out of square tiles.

C68. Gasr Khuraybah (Libya)

Excavation history: The building has yet to be excavated. The Italian archaeological mission in Libya documented the site and drew a preliminary plan. Bibliography: Gambini and Catani (1976). Construction: (D) Scattered surface finds in the surrounding area point to an occupation of the site between the second half of the 5th and the 7th c. Continued use and abandonment: Unknown. Plan: The bath was partially cut out of the rocks. A large rectangular space was divided into three rooms by walls made of reused ashlar stones. The first room (5 × 2.5 m) had a single entrance and must have served as a dressing room. Two doorways connected it to an oblong room (5 × 699  Goodchild (1971) 132.

386

Gazetteer

C69. Gasr Mismar (Libya)

C68 Gasr Khuraybah (Libya), (After Gambini and Catani (1976) 454, fig. 7)

1.3 m) with a single basin (1.1 × 0.4 m). The next room was slightly larger (5 × 2 m), had two square basins in its east wall (s: 0.67 m) and a rectangular one in its south wall (1.6 × 0.6 m). The last room was circular in plan (diam.: 2.8 m) and entirely cut out of the rock. It was only 1.8 m high. A large tub, cut out in the south-east wall (1.8 × 2.2 m), had a niche above it. Next to it was a smaller square basin (s: 0.6 m). As the building was not been excavated, one should exercise caution when interpreting this as a Romanstyle bathhouse. Decoration: All traces of a decorative scheme, if there ever was one, have vanished. Technology: There is no information available about heating systems or water management. To the west of each of the two northernmost rooms, a doorway led into a corridor that bent south towards the round ‘caldarium’. However, it was only connected to it by a narrow opening. Perhaps hot steam from a boiler in the corridor was introduced into the round room, creating a steam room as found in the later hammams. Except for the larger pool in the round room, all the basins are too narrow to serve as tubs. Nevertheless, they seem contemporary with the construction of the baths, as their measurements are in Roman feet (e.g. 0.67 m = 2.25 Rf). Hence, they were probably used for ablutions rather than immersion. This would again point to a bathing habit that was closer to the hammam than to the Romanstyle immersion bath. Construction technique: Cut out of natural rock; opus quadratum Bathhouse surface: ca. 45 m2 GPS coordinates: Exact coordinates unknown

Excavation history: The building was described by the Italian archaeological mission in the 1970s, but has yet to be excavated. Bibliography: Gambini and Catani (1976). Construction: (D) The small village of Gasr Mismar was dated in the 5th c. by survey finds. The baths are believed to date from the same period or perhaps later. Continued use and abandonment: The building was transformed into a stable in recent times. Plan: The baths were built at the foot of a hill and had a traditional north-south orientation. The first room was presumably added at a later date, as the walls abutted the main building.700 A narrow doorway (0.59 m) led into a cold room, without any fixed water amenities. The second and third rooms had hypocausts and must be identified with a tepidarium and a caldarium. The latter had three rectangular recesses. The largest one in the south (1.7 × 1.4 m) might have been a small sweat chamber, while the east and west recesses might have been equipped with pools. Decoration: All traces of a possible decorative scheme have vanished. Technology: As the baths were transformed into a stable or a fold in more recent times, the interior amenities have disappeared. The cocciopesto on the floors nevertheless confirm that water-related features were present.701 The limited size of the caldarium leaves little room for pools, which could mean only basins were present. If this was the case, the small steamy room with hot water basins would have more in common with Islamic-style baths. 30 m north-west of the building, a reservoir for rainwater collection was identified cut out of the rock. The connection to the baths is unclear.

C69 Gasr Mismar (Libya), (After Gambini and Catani (1976) 451, fig. 4) 700  Gambini and Catani (1976) 450. 701  Gambini and Catani (1976) 452.

387

Gazetteer Construction technique: Opus quadratum Bathhouse surface: ca. 75 m2 GPS coordinates: Exact coordinates unknown

C70. Mghernes (Mgarnes / Mqayrnis, Libya)

Excavation history: The baths have only recently been studied by the University of Benghazi, but a publication about this study is yet to appear. Bibliography: No publication available. Construction: (D) The type of plan and the construction techniques point to a late antique date.702 Continued use and abandonment: Unknown. Plan: The small bathhouse is in excellent state of preservation and stands 30 m north of the citadel. One entered the

baths by a rectangular polyvalent room that was divided into two parts by a north-south oriented arch supporting the two barrel vaults. Originally, this room was accessible by three large arches in the western side or a door to the north. A narrow door in the south wall led to an oblong room covered with a barrel vault. The bathing itinerary then turned east, to presumably the first heated room. It was domed and had a semicircular apse which probably contained a pool. To the south of this room, the caldarium had three similar semicircular apses, presumably for alvei. Decoration: Unknown. Technology: As the bathhouse is still partially buried, no information about the heating techniques or water management is available at this point. Construction technique: Opus quadratum Bathhouse surface: ca. 140 m2 GPS coordinates: 32°49’1.53”N, 21°59’34.30”E

C71. Ptolemais (Tolmeita, Libya), City Baths

C70 Mghernes (Mgarnes / Mqayrnis, Libya), (After a sketch-plan by Kenrick (2013) 305, fig. 201)

Excavation history: The city’s only known large public bathhouse was excavated by the Oriental Institute of Chicago between 1956 and 1958. Bibliography: Kraeling (1962) 160–75. Construction: (B) The building was built on top of an earlier imperial structure, perhaps also a bathhouse.703 A copy of the Prices Edict of Diocletian was found in the lower paving of the raised porch, giving a terminus post quem of AD 301.704 The architectural elements, especially the spirally fluted columns and the Corinthian capitals that seem contemporaneous with the construction phase of the building, also point to a 4th c. date.705 Continued use and abandonment: Coins of Heraclius (AD 610–641) reveal that certain activities, still took place on the site at the beginning of the 7th c.706 It seems that the bathing function—at least partially—stopped in the 6th c. Coins from this period were found in the fill of the hypocausts and the water channels. The destruction of the aqueduct during the 5th c. was perhaps to blame.707 After the abandonment, two lime kilns were constructed in the ruins, profiting from the large quantities of marble available on location. Plan: The bathhouse was located in the centre of the Roman town (map 25), along the Street of Monuments. It was raised 1.2 m above Roman street level and accessible by a portico. The small stairs to the north that connected

702  Stucchi (1975) 475. In the absence of a published excavation report, the following description is based on Philip Kenrick’s archaeological guide to Cyrenaica (Kenrick (2013) 305) and the Ward-Perkins Photo-archive of the British School at Rome (esp. wplib-10262 to wplib-10269 and wplib-G25–041a/b, wplib-G25–042a/b), available at http://www.bsrdigitalcollec tions.it.

703  Kraeling (1962) 175. 704  Kraeling (1962) 208. As the edict was obviously no longer in force, the construction date of the baths should be placed later in the 4th c. 705  Kraeling (1962) 174. 706  Kraeling (1962) 267–68. 707  Procop. Aed. 6.2.9–11.

388

Gazetteer

C71 Ptolemais (Tolmeta, Libya), City Baths (After Kraeling (1962) plan XXI)

the street level to this portico were flanked by two small fountains. The baths were originally accessed by a door in the north-western corner. A small vestibule led into the frigidarium to the east or into a small latrine to the south. In a later phase, a second entrance was created in the north-east corner at the expense of part of the eastern frigidarium wall. The cold room itself was a large rectangular space (24.3 × 16.7 m) with a central octagonal pool (max. 4.45 m in width; 0.8 m deep) and a bench along its northern wall. An inner portico made out of reused bluish marble columns with Corinthian capitals probably supported a roof. The absence of a gutter in the marble paving suggests that the entire room was covered.708 The east wall had 6 semicircular niches to hold statues, although one of them was destroyed to create the additional entrance porch. In the north-west corner, an almost square room, perhaps a porter’s lodge, was constructed with mud bricks at a later date. The size of the cold room in comparison to the entire building is impressive. It must have served different functions. Besides the obvious combination of frigidarium and apodyterium, the room may also have been used as some kind of audience hall or meeting room. A podium was built between two columns in the eastern corner, but seems to have been part of the original design. The excavator suggests that it may have been used by a βῆμα 708  Kenrick (2013) 73.

(a tribune) to speak to an assembly.709 Perhaps this is archaeological evidence for the literary attestations of baths as audience halls (see chapter 2, esp. Augustine, Socrates Scholasticus, Malalas). In the middle of the south wall, a narrow doorway led into the heated parts. It was flanked by two semicircular, small single-person pools that were added in a later phase. The first room (3.5 × 4.6 m), unheated, had a small tub (s: 1.25 m) in its west corner. In a later phase, a single column was placed in the middle of the room, probably to support the ceiling.710 To the east, a square room (s: 4.6 m) had two tubs built against its northern wall, one rectangular (0.9 × 2.5 m) and one square (s: 1.1 m). In the south wall, a doorway led into the cruciform caldarium (14.2 × 4.7 m), which was never excavated. The protruding bays north and south—the arms of the cross—contained heated pools (both 2.8 × 2.5 m). In a second phase, a new room was added on the east side of the heated rooms. It had an irregular shape (ca. 11 × 5.5m) causing it to be interpreted as two different rooms. On the hypocaust level, a north-west / south-east oriented wall divided the room into two unequal parts. The excavators, however, interpreted this wall as the foundation for three columns that supported the vault.711 Two semicircular alvei were constructed in the eastern 709  Kraeling (1962) 166. 710  Kraeling (1962) 168. 711  Kraeling (1962) 169.

389

Gazetteer wall (r: 1.75 m). In between, a single furnace surmounted by a boiler heated the room and a small basin constructed just above it. In the northern wall, a rectangular tub could accommodate a single person (1.4 × 1.7 m; 0.65 m deep). The doorway between this room and the frigidarium was blocked at a later stage. Decoration: The decorative scheme was of a high standard. The floors of both heated and unheated rooms were paved with reused marble slabs. In the two small rooms preceding the heated rooms, these slabs were laid out in geometric patterns, creating an opus sectile. The piscina was finished with slabs of marble, but the small pools were only coated with waterproof mortar. The wall decoration has not survived, but there is ample evidence of freestanding sculptures embellishing the frigidarium. The niches in the cold room’s east wall may have been built to accommodate such statues. Pedestals have also been found in front of the portico’s columns. Several life-size or even larger marble statues were found throughout the baths, including a group of the Three Graces, a portrait of Cleopatra I, a head of Asclepius and several Roman copies of Hellenistic originals. Most of the statues pre-date the baths and were thus imported from other locations. The frigidarium’s spiral-fluted columns were also reused elements, although the Corinthian capitals are more uniform and may have been new creations.712 Technology: The hypocaust system, only found in the caldarium and the newly built room, consisted of normal pillars made out of square tiles. Against the walls, these pillars were slightly larger. Some later restorations of pillars were made out of round tiles. The walls were heated by oval or rectangular tubuli. Both warm rooms were heated by a single furnace, surmounted by a boiler.713 The water supply of the cold pools was ensured by a large reservoir at the western side of the frigidarium. It fed the octagonal piscina by a conduit running under the floor of the frigidarium. It probably also fed the boiler of the caldarium. The latrines had their own reservoir. The alvei of the newly built heated room may have had a separate water reservoir, surmounting the narrow corridor in the masonry of the south-east corner. A water reservoir also existed above the rectangular pool in the north wall of this room. It served not only this pool, but also the pools in the square room, the easternmost semicircular cold pool and the fountains in front of 712  Kenrick (2013) 73. 713  However, both rooms also had openings in the southern wall at the hypocaust level. These have been interpreted by the excavator as ventilation openings to draw the heated air under the suspended floor (Kraeling (1962) 171). Yet, the draft is usually ensured by the tubuli and chimneys in the walls. Evidence for such chimneys was indeed found in the caldarium. Furthermore, in the service corridors south of the heated rooms, thick layers of ash were found (Kraeling (1962) 171), reinforcing the idea that the openings were furnaces.

the main entrance to the baths. The water tank may have been fed by run-off rainwater from the roof. A well adjacent to the furnace of the caldarium seems to pre-date the construction of the baths, but was kept is use.714 The drainage of the pools is less clear. The octagonal piscina had an outlet in its north-eastern corner, but the sewage channel could not be followed. How the other pools were drained is unknown. Construction technique: Opus quadratum Bathhouse surface: ca. 1,350 m2 GPS coordinates: 32°42’35.32”N, 20°57’5.78”E

C72. Ptolemais (Tolmeita, Libya), Baths of the House of Paulus

Excavation history: Just as the City Baths, the House of Paulus was excavated between 1956 and 1958 by the Oriental Institute of Chicago. Bibliography: Kraeling (1962) 140–60. Construction: (C) The building had two distinctive phases: an Early Roman phase (1st c.) and a late antique / Byzantine phase (6th c.). It seems that in the first phase, the apartment block or large house already had a private bath in its north-east corner. It was only in the second phase that the baths were made public by constructing a street entrance. The inscription of Paulus, ὑπαθικός (‘of consular rank’), who took credit for the reorganisation of the building, is used as evidence to date the new bathhouse in the late 5th or early 6th c.715 Continued use and abandonment: Small finds dating to the reign of Phocas (AD 602–610) suggest that the building was still in use in the early 7th c. Plan: The House of Paulus was located at the western end of the Street of Monuments, near the Arch of Constantine (map 25). The original private baths of the insula consisted

C72 Ptolemais (Tolmeta, Libya), Baths of the House of Paulus (After Kraeling (1962) plan XVI) 714  Kraeling (1962) 172, esp. fig. 60. 715  Kraeling (1962) 211.

390

Gazetteer

of only three rooms: frigidarium-tepidarium-caldarium, the basic scheme of every Roman-style bath. It is, however, unclear how these rooms articulated to one another and how they connected to the living quarters. When the baths were opened to the public, a new entrance porch with a small latrine was created in the east corner. Continuing north, the large cold room (9.3 × 6.4 m) was paved and had a large square piscina against its western wall, dating from the first phase. A small sitz-bath was added in the northern corner and benches lined the east and south walls. An angular passage led to the courtyard of the insula, while a doorway in the north gave access to a room with two small bath tubs made out of rubble mortar and plaster. The room immediately to the west had a similar bath tub against its south wall. Both rooms had a hypocaust system. The following room, also with a suspended floor, had its own furnace to the north, but no pools. The last room of the bathing route, the presumed caldarium, had two apses in its north wall. In the north-west apse, there was a small pool, while in the north-east one there was a furnace surmounted by a boiler. It is difficult to find out if the different pools belong to the same phase. At a certain point, the doorway between the first two and the last two heated rooms was blocked. The small sitz-baths in rubble concrete may belong to this last phase. The development of this bathhouse from a Roman type with hypocausts to a Greek type with individual bath tubs is remarkable. Perhaps the change to smaller individual tubs was the result of problems of water supply, with a water scarcity in the city explicitly mentioned by Procopius.716 Decoration: The bad state of preservation of the walls and the collapse of the hypocausts make it impossible to reconstruct the decorative scheme. Technology: The suspended floors of the heated rooms were constructed on pillars made of round tiles. No traces of wall heating were found. The water for the piscina was collected in a reservoir in the courtyard of the insula. The small sitzbaths must have been filled manually, as no water conduits have been recognized within the walls. In the small room with several alcoves, north of the piscina, some traces of firing have been found against the walls. Perhaps the water was heated in cauldrons and then carried in buckets to the small pools.717 The drains of the small pools seem to pass under the frigidarium and presumably on to the latrine.718 Construction technique: Opus quadratum Bathhouse surface: ca. 310 m2 GPS coordinates: 32°42’34.49”N, 20°57’2.23”E

Siret Ain Relles (Libya) Excavation history: The building has never been excavated. The remains were documented in the 1970s by the Italian archaeological mission in Libya. Bibliography: Gambini and Catani (1976). Construction: (D) Surface finds collected in the vicinity of the baths point to an occupation phase in the Byzantine period. Continued use and abandonment: Unknown. Plan: Only part of two heated rooms are preserved. The first room measures approximately 2 × 3 m. The second room had three apses, probably for alvei, as was the case in Mghernes (see above, C70). The plan was probably cruciform, with a simple linear succession of the basic rooms ( frigidarium / apodyterium-tepidarium-caldarium). Resemblances to the baths of Mghernes and Gasr Mismar (see above, C69) are striking. Decoration: The walls were covered with cocciopesto, but no traces of paint were identified. Technology: As the baths remain unexcavated, there is no information on the heating system. In the vicinity, but on a higher level, a water reservoir was identified.

716  Procop. Aed. 6.2.9–11. 717  Kraeling (1962) 152. 718  Indicated on plan XVI in Kraeling (1962).

C73 Taucheira (Tocra, Libya), Byzantine Baths (After Jones (1984) 109, fig. 2)

C73. Taucheira (Tocra, Libya), Byzantine Baths

Excavation history: The bathhouse was excavated in the 1960s, but the results were not published. The data was partially studied in the 1980s. Bibliography: Goodchild (1964); Jones (1984). Construction: (D) The bath could be dated in the Byz­ antine period on the basis of the plan and construction techniques. Continued use and abandonment: The lack of archaeological information makes it difficult to reconstruct the relative chronology of the numerous restoration phases. An Arabic inscription praising Allah was found in the threshold of the

391

Gazetteer main entrance.719 While this does not necessarily mean that the baths were still used for bathing, it does confirm that the town was inhabited after the Arab invasion. Plan: The baths were built on the site of the former gymnasium, just inside the city walls near the east gate. The walls were constructed in limestone blocks and rubble mortar. In a first phase, one entered the baths by a large courtyard to the west. An almost square atrium preceded an oblong corridor, where two lateral piscinae were constructed in a second phase. The following two cold rooms were transformed. The northernmost was equipped with a small basin in its north-west corner in the third phase, while a new entrance was pierced through the old gymnasium wall. The southern cold room originally had a hypocaust and a bench against it western wall. A door was pierced through this wall in phase 2. In phase 3, a large brick bench was constructed in the south -west corner. The central heated room had a single furnace in its southern side, but had no pools. The eastern heated room was the caldarium. The two rectangular alvei in the recesses of the eastern wall belonged to the original design. In a later phase, a smaller tub was added to the south side by piercing the wall and protruding onto the service corridor. The architectural design of these baths resembles the ‘Byzantine’ baths found in the East, in Egypt and in the Levant (see above, chapter 4, p. 187–188). A courtyard with two cold pools flanking the entrance to the actual building is often encountered (see already Apollonia, C65). Decoration: The brief report does not mention the forms of decoration found in the baths. The most luxurious building materials were probably recuperated after the baths fell out of use. Technology: The heated section was serviced by three furnaces: one for the central room and two for the caldarium. The suspended floor rested on square pillars. The only remaining tubuli were found along the eastern end of the northern alveus. The small added alveus to the south was not constructed on a hypocaust. Two reservoirs were found in connection to the baths. A large one lay south of the second cold room and probably supplied the cold pools to its west. A smaller reservoir was found south of the caldarium, constructed above the service corridor. It must have supplied a boiler in the furnace room, lodged between the recesses with the alvei. The waste water disposal of the alvei was ensured by drains leading east of the baths. The third phase piscina in the first cold room had an outlet channel piercing the northern wall. Little is known about the water disposal of the second phase piscinae in the oblong room. Construction technique: Opus quadratum Bathhouse surface: ca. 290 m2 GPS coordinates: 32°32’13.26”N, 20°34’7.46”E

C74. Wadi Senab (Libya)

719  Jones (1984) 111.

720  Luni (1975) 271, fig. 3.

Excavation history: The baths have only been briefly investigated by the Italian archaeological mission in Libya in the 1970s, without any excavations. Bibliography: Luni (1975). Construction: (D) On the basis of plan and similarities with other ‘Byzantine’ baths in the region (C68, C69, C73), their construction is dated between the 4th and 6th c. Continued use and abandonment: Unknown. At some point, the rooms without a specific bathing function were used as a dwelling. A large press was found in the middle room. Plan: In the river valley of the Senab, south of Balagrae (modern Al-Bayda), several tombs were carved out in the karst caves. In one of these tombs, a small bathhouse was constructed (34 × 25 m) with limestone blocks coming from the cave and piled up a secco (without mortar). The site of the baths comprises a part in the cave and a part that was constructed against the outer cave wall. It is in this latter part that the actual bathing rooms were built. The part of the site that was cut out into the rock may not have been a functional part of the baths. A first room was unheated and only had a small water basin carved out in the northern rock wall. A narrow passage (0.65 m) to the west connected it to a small room (2 × 2.6 m) on hypocausts with wall heating. The third room (3.2 × 2.8 m) was also equipped with a hypocaust and wall heating, but also had a single-person pool in its south wall and two small basins in the west wall. The three rooms were all covered with a barrel vault, reaching a maximum height of 4 m. Decoration: The decoration was limited to the walls of the heated rooms. In the caldarium, some simple garlands were painted onto the plaster coating of the walls.720 Technology: The heating system was very basic. A single furnace west of the caldarium heated both warm rooms. The walls were heated by creating parallel recesses in the rock (10–15 cm deep; 25 cm apart). A draught for the hot air was ensured by chimneys in the 4 corners of each room, made with terracotta tubes (diam.: 10 cm). The water supply for the pool and basins relied on the collection of rainwater, channelled by gullies cut out in the rock to cisterns and a reservoir. The two small basins in the caldarium must have had the function of the ancient labrum. Construction technique: Opus quadratum Bathhouse surface: ca. 60 m2 (850 m2 rock chambers included) GPS coordinates: 32°40’46.33”N, 21°36’30.69”E

392

Gazetteer

C74 Wadi Senab (Libya), (After Luni (1975) 267, fig. 1)

Possible Late Antique Baths Other small bathhouses with similar plans and features as those described above have been discovered in Cyrenaica. While there is little doubt that these buildings were bathhouses, none of them have been excavated. Their lifespan and their accessibility (public or private) are difficult to reconstruct. The baths of Gabu Iunes probably date from the second half of the 5th c. The simple linear succession of small rooms ended with two warm rooms oriented towards the south. There seems to have been a second floor.721 The baths of Limnias (Lamluda) were built in large blocks. The simple succession of three rooms may point to a late antique date.722 Two churches were also found in the vicinity. The baths of Zawiet el Argub, located in a valley, are similar to the baths of Mghernes.723 The fortified building, interpreted as a church or an official building, located in the vicinity of the baths was dated to 6th c.724 The small baths at Gasr Bandis were part of a cluster of buildings centred on a late antique qasr and at least 7 cisterns. The mosaics depicting wild animals and male and female benefactors might have belonged to a church. The ceramics and small finds recovered during small-scale excavations and surveys during the 1970s point to a late antique phase of the site.725 The baths at Gasr Sherbin, 16 km north-west of Cyrene, may be contemporaneous with the small church found in its vicinity.726 The baths of Mechili only consisted of three rooms. Besides the type of plan, there 721  Stucchi (1975) 474. 722  Stucchi (1975) 473. 723  Reynolds (2003) 373. 724  Reynolds (2003) 381. 725  Reynolds (2003) 393–95. 726  Reynolds (2003) 400.

is no evidence for dating this building.727 The measurements are still in Roman feet, although the structure could also date to the Early Islamic phase.728

Fortress Baths Taucheira (Tocra, Libya) Excavation history: The bath of the Byzantine fort in Taucheira was excavated in 1969 by the Libyan Department of Antiquities. The results were summarily described by Jones in the 1980s. Bibliography: Jones (1983). Construction: (D) While the baths could not be securely dated by stratigraphic finds, it is clear that they were constructed contemporaneously with the Byzantine fort in the 7th c. Continued use and abandonment: Unknown. Plan: (fig. 123) One entered the bathhouse by a porch and a door in the north side of the cold room.729 The door to the south might have been a second entrance or an exit. The frigidarium was equipped with stone benches and a small rectangular piscina. The first heated room should be interpreted as the tepidarium. The caldarium, lying at the eastern end of the building, had two small alvei, each only large enough to seat one person. They flanked the only furnace, 727  Stucchi (1975) 476. 728  Kenrick (2013) 303. 729  The baths have only been briefly described by Jones (1983). The following description is based on the plan provided by this article (Jones (1983) 118, fig. 10) and the photographs in the Ward-Perkins archive of the British School at Rome (wplib-09917).

Gazetteer

393

figure 123 Plan of the Fortress Baths in Taucheira (Libya) after Jones (1983) 118, fig. 10

which, judging from the mortar platform, was surmounted by a boiler. The simple linear layout and basic commodities fit well within the type of functional buildings that a fortress required. Decoration: There is no information available about possible decoration. Technology: The two heated rooms had a hypocaust, but no wall heating. The flow of air was guaranteed by several chimneys that were located in recesses in the walls. The hypocaust pillars were made out of stone monoliths instead of terracotta tiles. The water conduit was identified along the northern wall of the baths. Each alveus was drained by a channel heading east.

Continued Use and Abandonment of Imperial Period Baths during Late Antiquity Erythron (al-Athrun / Latrun, Libya) The ancient town of Erythron was a newly founded Roman settlement. It was still an important harbour in Late Antiquity, when it became the seat of a bishop.730 Most of the town 730  Michel (2011a) 205–207.

fabric remains to be explored. Since 2006, the French archaeological mission has excavated a Roman bathhouse on the highest point of the city. The plan is characterized by its octagonal cold room with a round pool at its centre.731 The small heated rooms are dispersed around the octagon. Such a scheme is also found in Late Antiquity.732 The baths were probably built in the early 3rd c. and continued to be used until the late 4th c.733 At this point, the pools were filled with debris and the surviving rooms were transformed into a villa.734 In the Byzantine period, a baker’s shop and kiln were inserted in the north-western part of the building.735 Some of the building materials of the baths were reused in the nearby East Church.

731  Laronde and Michel (2009) 181. 732  The resemblance to the 5th-c. Baths of the Blues in Tyrus (Lebanon) is remarkable (Kahwagi-Janho (2014) 400–401). The late 4th- or early 5th- c. Baths of St-Philip in Hierapolis (Turkey) show a similar dispersion of the heated rooms around a central cold room (Caggia (2014)). 733  Michel (2014) 376. 734  Michel (2011b) 31. 735  Michel (2011a) 217.

Baths in Egypt New Construction C75A. Abu Mena, South Baths

Excavation history: The bathhouse of the Christian pilgrimage site of Abu Mena was discovered in 1905–1907 and excavated in the 1960s by the German Archaeological Institute. Bibliography: Müller-Wiener (1966); Fournet et al. (2017) 451–55. Construction: (A) The pilgrimage city of Abu Mena was famous for the tomb of Menas of Alexandria, an officer in the army of Diocletian who refused to kill Christian captives and openly proclaimed his Christian faith. According to the legend, his remains were carried back from Phrygia, where he died in AD 296, to Egypt by camel. His tomb was built where the animal collapsed. Subsequently, a spring arose on the spot and the patch of desert turned into a fertile oasis.736 The city was known for its curative thermal spring, which was bottled in small ampullae as a souvenir and distributed widely across the Mediterranean. Some 150 m north of the main sanctuary and adjoining monastery, a large bath complex was built.737 Test pits and stratigraphic excavations made clear that these baths were built in the 5th c.738 Continued use and abandonment: (A) The bathhouse was enlarged twice, at the turn of the 5th and 6th c., and during the 6th c. The stratigraphy pointed to abandonment in the second half of the 8th c.739 Plan: In the first phase (early 5th c.), the bathhouse comprised only the southern part of the later double baths. It was directly accessible from the street running south of the complex. A large basilical hall preceded 4 bathing rooms in an angular row articulation (C75A). The foundations of these rooms, as far as these were not reused in the later building, were found underneath the later baths. The small frigidarium, possibly with a small semicircular piscina in its western wall, gave access to the tepidarium to the east and perhaps a sudatorium to the north. The tepidarium led into the caldarium, which had a semicircular and rectangular alveus. The water was supplied by a saqiyah with an annex reservoir to the east of the baths. 736  Jaritz (1993). 737   The excavations were concisely described by Wolfgang Müller-Wiener (1966), without exploring the separate rooms in much detail. The plan of the complex (fig. 1) is, however, rather detailed and informative. The following description is based primarily on Müller-Wiener’s description and on the information indicated on the plan. 738   Müller-Wiener (1966) 175. Müller-Wiener (1966) 180. 739  

In a second phase, tentatively dated to the late 5th / early 6th c., the bathhouse was enlarged with a new northern wing. These new rooms formed a separate bathhouse and resembled the south wing in both plan and size (C75B). A basilical hall, interpreted by Daniel Krencker as an actual Christian church, with an east-west orientation preceded the bathing rooms.740 The frigidarium was an oblong room with a single semicircular piscina. The heated rooms lay in a simple linear articulation. The caldarium had two semicircular alvei and the tepidarium, a single rectangular pool. The southern wing was also enlarged, partially at the expense of the public road. The hall was expanded to the south and equipped with a latrine and a small semicircular pool. The sudatorium was abandoned and the caldarium was transformed into the tepidarium, losing the larger alvei, but gaining a small single-person pool in its western side. A new caldarium with two semicircular alvei in its east wall was constructed east of the old caldarium.741 A larger saqiyah was constructed north of the old one to increase the water supply. In the third phase (C75C), dated to the 6th c. by ceramic finds in test pits, both wings were further enlarged, adding more small pools. In the southern wing, a new entrance porch was created, probably with adjacent porter’s lodge. The hall received a second semicircular pool, while a new apodyterium was constructed to the east of it. The caldarium was equipped with an additional single-person alveus in its northern wall. In the north wing, the basilical hall was transformed into an apsidal hall with two small rectangular basins. The frigidarium was created directly south of this hall. It received four semicircular singleperson pools in its southern wall. The tepidarium was enlarged with a small pool in its north-western corner and the caldarium with a semicircular alveus in its south-western corner. A new room with separate entrance from the cold section and with two small rectangular pools was created east of the heated rooms. It must have had a special function, perhaps for medical treatments or for a special group of bathers (clergy or elite?). The enlargement of this important public bath has been linked to the growth of the pilgrimage centre from the 4th to 6th c.742 The size of the city and the production of local ceramic souvenirs declined in the 5th and especially 6th c., the period of the largest expansion of the 740  Krencker et al. (1929) 235. 741  It is not clear whether the new caldarium belonged to this second phase or was added at a slightly later date. However, it clearly predates phase 3 of the baths (Müller-Wiener (1966) 177). 742   Müller-Wiener (1966) 177.

395

Gazetteer

C75A Abu Mena (Egypt), South Baths (phase 1) (After Müller-Wiener (1969) 176, fig. 2a)

C75C Abu Mena (Egypt), South Baths (phase 3) (After Müller-Wiener (1966) 172, fig. 1)

C75B Abu Mena (Egypt), South Baths (phase 2) (After Müller-Wiener (1966) 176, fig. 2b)

baths. The public baths played an important role in the accommodation of the pilgrims. It is even possible that the bath complex and the surrounding structures offered a sleeping-place for the visitors, just as the North African hammams used to up until the beginning of the 20th c.743 The ‘doubling’ of the facility has been interpreted as the creation of a separate men’s and women’s bathhouse. The 743  Carlier (2000) 1311.

excavators found glass beads and fragments of glass oil or perfume flasks in the sewage channels of the southern wing, encouraging the interpretation of these parts as the women’s baths.744 The numerous similarities between the two buildings—especially considering the size of the pools and the number of heated rooms—makes a differentiation between winter and summer baths less convincing. The same can be said for an interpretation of ‘normal’ and ‘therapeutic’ baths. Such a differentiation was more likely to appear within a bathhouse. This might explain the high number of single-person pools, as medical baths require a personal treatment of the patient with the appropriate ointments (see chapter 2, Oribasius, Cassius Felix and other medical writers) and a frequent change of the bathing water. Decoration: Most of the cold rooms, including the large halls, were paved with limestone slabs. In the southern wing, marble decorative elements have also been found. The capitals of the columns were of the Corinthian style. Technology: The excavation report gives little information on the heating system. It seems that the furnaces and boilers were lodged in two small rooms in between the two bathing wings. In the last phase of the baths, the water storage capacity of the saqiyah was augmented by a new 744   Müller-Wiener (1966) 179, n. 3.

396

Gazetteer

reservoir. 100 m north of the baths, a natural source with adjacent cisterns may have provided additional water.745 Waste water disposal was optimized by creating a single sewer that started in the north section and passed west of the hot rooms to the south section, flushing the latrines and continuing west towards the sewers in the street.746 A second sewer passed east of the baths, collecting the waste water of the south wing caldarium. Construction technique: Opus quadratum Bathhouse surface: A: ca. 170 m2; B: ca. 920 m2; C: ca. 1,300 m2 (1,500 m2 reservoirs included) GPS coordinates: 30°50’32.09”N, 29°39’44.24”E

C76. Abu Mena (Egypt), North Baths

Excavation history: The bathhouse was discovered in 1905 by C. Kaufmann and partially excavated during the large excavation campaign by the German Institute in the 1960s. Bibliography: Müller-Wiener (1967) 209–16; Müller-Wiener and Grossmann (1976) 458–59; Grossmann (1973) 43–44; Fournet et al. (2017) 456–57. Construction: (A) The excavations point to a construction date in the early 6th c. Continued use and abandonment: (A) By the second half of the 7th c., a pottery workshop seems to have been implanted in the abandoned baths. Plan: Just as the South Baths in Abu Mena, the North Baths consisted of two separate, independent wings. The northern wing was only partially excavated. It had a street entrance through two rooms, giving into a large open-air courtyard. This multipurpose space had a rectangular exedra on its northern and southern end and a covered portico on its eastern and western side. A doorway in the eastern portico led into the bathing area (heated rooms), which remain unexcavated. In the portico, there was one rectangular and one semicircular pool, large enough for one person, on each side of this doorway. A fifth rectangular pool for two or three persons stood at the southern end of the portico. At least one of the heated rooms had a rectangular, single-person alveus. By a narrow passage, the north wing was connected to the south wing. This part of the bathhouse was also accessible from the same street, through a somewhat larger vestibule. From a second room, one could possibly enter the latrines to the north, or proceed through an L-shaped cloakroom with benches to the south. From the latter, the bather entered a rectangular courtyard similar to that of the north wing (the exedrae on the east and west sides). In addition to the single-person pools flanking the entrance to the heated section and the fifth larger cold pool in the north corner of the portico, two single-person pools also stood in the courtyard itself (see also the east wing of Marea 2, below). Once inside the 745   Müller-Wiener (1966) 183. 746  Indicated on fig. 1 in Müller-Wiener (1966).

C76 Abu Mena (Egypt), North Baths (After Fournet et al. (2017) 457, unnumbered fig.)

bathing block, a small square room to the west permitted a cold-water dip in a small rectangular pool. A corridor-like transitional room (tepidarium) gave access to two rectangular heated rooms, each equipped with pools. The first room (sudatorium?) had two rectangular single-person pools built in its south-western wall. The second heated room (caldarium) had three larger alvei built into its walls (north, east and west). Decoration: Wall plasters with human figures and geometric patterns were found. Technology: The service corridors ran under the south side of the heated sections and were accessible by separate entrances north and east of the bathing block. A central furnace was surmounted by 4 large boilers, providing warm water for both the north and the south wing. Several smaller furnaces may have heated the hypocausts of the heated rooms.747 Chimneys in the corners of the heated rooms ensured the necessary draught. No tubuli are reported. The water seems to have been supplied by a saqiyah to the north of the baths, undoubtedly connected to 6 cisterns nearby. Construction technique: Mud brick on opus quadratum base 747  Fournet et al. (2017) 456.

397

Gazetteer

map 40

Alexandria (Egypt), hypothetical reconstruction of the late antique city grid with the location of public baths after Rodziewicz (2009) 199, fig. 1 1: Baths of the Governmental Hospital

Bathhouse surface: ca. 1,500 m2 (including reservoirs) GPS coordinates: 30°50’36”N, 29°39’44”E

C77. Alexandria (Egypt), Kom al-Dikka Baths

Excavation history: The baths were excavated during the 1960s and 1970s by the Polish archaeological mission. Bibliography: Lipinska (1966); Koɫątaj (1972); Rodziewicz and Rodziewicz (1983); Koɫątaj (1992); Fournet et al. (2017) 459. Construction: (A) Test pits beneath the floor of the cold rooms yielded pottery from the 4th c., placing the terminus post quem around AD 330.748 Continued use and abandonment: (A) Several important reorganisations could be discerned thanks to the architectural analyses of the remains. The most radical changes 748  Koɫątaj (1972) 167.

occurred during the second half of the 6th c. New floors, one yielding a coin of the reign of Maurice (reigned AD 582–602), were laid out and the caldarium was thoroughly transformed.749 The destruction layers contained building material, wall paintings, marble slabs and ceramics of the late 6th / early 7th c.750 Plan: The Kom al-Dikka archaeological site is located near the modern central station, along the ancient Via Canopica (map 40). It offers a good archaeological section of the ancient city. It comprises two insulae bordered by the cardines R5 and R4. In the southern insula, the remains of a theatre and lecture halls bordering the colonnade of the R5 street have been found. Along the R4 street, a late antique reservoir has been identified. East of this street, some 749  Koɫątaj (1992) 50. 750  Koɫątaj (1972) 160; Koɫątaj (1976).

398

C77 Alexandria (Egypt), Kom al-Dikka Baths (After Koɫątaj (1992) plan IV)

late antique houses, occupied from the 5th to 7th c. came to light.751 In the northern insula, the remains of the large brick-built bath complex were excavated.752 Up until the 3rd c., the insula consisted of sumptuous domus. The large complex was of the imperial type, covering over 3,300 m2. The north-east part was destroyed by an explosion of gunpowder at the end of the 19th c., but thanks to the symmetrical layout, the plan can be reconstructed with a high degree of certainty. A large basilica-type frigidarium was stripped down to its foundations in later phases. Only the semicircular (diam.: 3.3 m) and rectangular pools (4 × 3.6 m) in its west wall can be reconstructed. The basis of a labrum was also found imbedded in the floor. The original design of the cold room probably included 4 large piscinae in the corners, which were filled in at a later stage.753 From the cold room, a central large bay, flanked on each side by two small doorways, gave into the heated rooms. On the central east-west axis, a succession of three heated rooms ended with the caldarium in the west. On both the north and south side of this main axis, two interconnected large halls enabled the bathers to follow a circular itinerary. They had small pools (2.7 × 1.6 m) in the large bays of the 751  Majcherek (1990). 752  Lipinska (1966). 753  Koɫątaj (1992) 114–17.

Gazetteer outer walls. Above these pools, there were large windows. The westernmost of these rooms had a semicircular alveus (diam.: 6.3 m) in its western side. The caldarium (12 × 28 m) had an oblong shape with apsidal endings in the north and south. In the last reorganisation of the baths, they may have been covered by a dome resembling that of Hagia Sophia in Constantinople.754 In the centre of the western side, there was a large round communal alveus (diam.: 8.3 m; 1.5 m deep), flanked on either side by a smaller square pool (s.: 2.5 m). Furthermore, the room was equipped with 4 semicircular alvei (diam.: 4 m) in the corners of the room, and two round single-person pools (diam.: 3 m) lodged in the bays connecting to the central tepidarium. It is possible that the latter consisted of three smaller rectangular rooms lying on a north-south axis. In the first phase of the building, a round room without pools probably served as a dry sweat room and was accessible by the south-eastern heated room. Decoration: The baths were embellished with precious construction materials such as cipollino-marble columns with capitals of the Corinthian order. Most of the architectural elements seem to have been reused from earlier structures.755 The floor of the frigidarium was paved with marble slabs, while the passageways running along the northern and southern side of the baths were paved with limestone. None of the floors of the heated section have been found. Presumably, they were paved with marble and then dismantled after the baths fell out of use. The walls of the heated rooms had a marble plinth at the bottom and wall paintings on the upper parts. Technology: The heated rooms had a suspensura resting on large brick-built pillars connected by barrel vaults. The walls were heated by flues lodged in the walls instead of the traditional tubuli. The furnaces were constructed as pits in the hypocaust floors (fig. 124), just as in the baths of Kom el-Ahmar. This was probably to reduce heat loss and to adapt the furnace to the high flames when burning reeds and straw.756 They were fired from a network of underground service corridors running along the southern and western side of the heated part and centrally under the tepidarium and to the eastern end of the caldarium. At the western end of the building, 5 boiler rooms were discovered. These boilers supplied the pools of the heated rectangular halls and the westernmost pools of the caldarium with hot water. Analysis of the calcareous deposits in the drains and on the inside of the pools revealed that the eastern pools of the caldarium contained cold water (fig. 125).757 The circular pool of the caldarium was not supplied by the 754  Koɫątaj (1992) 164. 755  Koɫątaj (1992) 93. 756  Koɫątaj (1992) 174. 757  Koɫątaj (1992) 75, esp. n. 69 and 183–97.

399

Gazetteer

figure 124 Kom al-Dikka Baths in Alexandria (Egypt), cross-section (top) and plan (bottom) of the ‘pit-hypocaust’ after Koɫątaj (1992) 141, figs. 47–48

boilers, but perhaps by a type of testudo that was constructed above the furnace pit in the hypocaust (fig. 126).758 The water supply was probably ensured by the large reservoirs located 40 m south of the baths. Excavations revealed that these were contemporaneous with the construction of the bathhouse.759 A direct connection with the baths, however, could not be recognized. The pools of the frigidarium were drained by sewage channels continuing north under the cold room and probably joining the sewers of the street.760 The pools of the heated rooms were drained by a sewage system that passed in the underground service corridors along the southern and western sides of the baths. These flushed the northern and southern latrines that lay to the west of the baths. Construction technique: Opus testaceum, opus quadratum Bathhouse surface: Unknown (min. 3,300 m2) GPS coordinates: 31°11’45.02”N, 29°54’14.29”E

758  Such an adapted testudo was found in the unpublished baths of Tell Atrib (Koɫątaj (1992) 174). The testudo did not stand in direct contact with the pool, but was placed above the furnace. The water passed through lead pipes into the pool, permitting the exchange of cold and hot water through convectional movement and joined vessels. 759  Majcherek (1992) 5–7. 760  Koɫątaj (1972) 161.

Alexandria, Baths under the Governmental Hospital Excavation history: During construction works on the premises of the Governmental Hospital in 1987, a small bathhouse was discovered (map 40). The site was not excavated. Bibliography: No publication available. Construction: (C) The construction technique of the walls and the level at which the remains were found suggest that these baths were built between the 4th and 6th c.761 Continued use and abandonment: Unknown. Plan: Unknown. Decoration: Unknown. Technology: Unknown.

C78. Alexandria (Egypt), Baths in Anubis Street

Excavation history: In 1986, remains of a large bathhouse were identified during construction works in Anubis Street (map 40).762 The site was subsequently looted by locals searching for artefacts and construction material. Bibliography: No publication available. Construction: (C) The baths were probably built in the 4th or 5th c., judging from construction techniques.763 761  Rodziewicz (2009) 197. 762  The site was never properly excavated. Only some sketches of the latrine survive. 763  Rodziewics (2009) 198, n. 35.

400

figure 125

Gazetteer

Plan of the Kom al-Dikka Baths in Alexandria (Egypt), with pools of different temperatures (blue = cold; orange = tepid; red = warm) after Koɫątaj (1992) plan IV

401

Gazetteer

figure 126 Kom al-Dikka Baths in Alexandria (Egypt), cross-section of a possible type of testudo in the caldarium after Koɫątaj (1992) 145, fig. 49

C78 Alexandria (Egypt), Baths in Anubis Street (After Rodziewicz (2009) 201, fig. 14)

Continued use and abandonment: Unknown. Plan: The best-preserved room was a large latrine, flushed by the waste waters of the baths. It is believed that this bathhouse covered a large area in the north-west quarter of Alexandria, delimited by the ancient streets L1 (Via Canopica)—L2 and R2 bis—R3 bis. The walls were made out of dressed small limestone blocks alternated with layers of brick. The size of the Baths in Anubis Street is thought to equal that of the Kom al-Dikka baths. Decoration: Unknown. Technology: Unknown. Construction technique: Opus mixtum Bathhouse surface: Unknown GPS coordinates: 31°12’34.95”N, 29°54’49.79”E

C79. Alexandria, Baths of the Cinema Majestic Excavation history: The Centres d’Études Alexandrines was entrusted with rescue excavations when the old cinema Majestic was demolished in 1992–1993 (map 40). Bibliography: Empereur (1994). Construction: In the layers underneath the bath, thick walls of what seems to be a temple were found. The structure was destroyed in the second half of the 4th c., giving a terminus post quem for the construction of the baths. Continued use and abandonment: The site of the baths was subsequently occupied from the 5th to the 7th c. Only in the Mamlūk period (13th c.) the remains were buried by layers of waste. Plan: A small bath complex was identified by a room with hypocaust, two semicircular alvei and three cisterns. It is worth noting that the bathhouse stood on the grounds of

C79 Alexandria (Egypt), Baths of the Cinema Majestic (After Empereur (1994) 507, fig. 5; no scale)

402 the main church of Alexandria (Kyriakon), constructed in the 4th c. under Bishop Athanasius. Perhaps these baths belonged to an ecclesiastical complex or a private residence of the clergy.764 Decoration: Unknown. Technology: The hypocaust had standard pillars made of square tiles. There is no information on the wall heating. The pools were filled from the cisterns, yet it is unknown how the latter were supplied. Construction technique: Opus caementicium? Bathhouse surface: Unknown GPS coordinates: 31°11’57.73”N, 29°54’3.97”E

C80. Alexandria (Egypt), Baths behind the Water Company

Excavation history: During construction works on the building of the Water Company in 1978, a foundation trench enabled the Polish archaeological team to investigate the late antique ruins that came to light (map 40). The remains of a large domus with a bathhouse were identified. Bibliography: Rodziewicz (1984). Construction: A preliminary study of the ceramics pointed to a construction date in the 5th c.

Gazetteer Continued use and abandonment: The complex was abandoned in the first half of the 6th c., perhaps after an earthquake (AD 535?). One of the walls of the building caved in, crushing several individuals.765 In the late 6th or early 7th c., a church was built on this location. Plan: The bathhouse covered a surface of at least 300 m2. To the north, two identical rectangular rooms (ca. 20 m2) opened directly onto the Via Canopica, Alexandria’s most important road. These rooms belonged to the baths, as their floors and walls were coated with hydraulic mortar. This direct street access points to a public use of these baths.766 The frigidarium consisted of a large room surrounded on three sides by a portico. Against the western side, a semicircular pool was constructed. Bordering the cold room on its western side were the heated rooms. The supposed caldarium was brick-built and was accessible by a doorway to the west, flanked by two semicircular niches. There was a small rectangular pool to the north. The living quarters were south of the baths. Decoration: The decoration is known by the reuse of decorative elements in the structures that were erected on top of the baths. Fragments of marble, limestone and red granite were found. Technology: There is no information on the heating systems and service areas, which must have been situated to the west of the heated rooms. The pool in the caldarium was drained by an outlet at its bottom, giving onto the floor. A sewage channel seems to have crossed the frigidarium. Observations made at the end of the 19th c. mention a cistern in the proximity of these baths.767 Construction technique: Opus testaceum Bathhouse surface: Unknown (at least 300 m2) GPS coordinates: 31°12’9.51”N, 29°55’15.32”E

C81. Alexandria, Baths of the Mehattet Masr Square (Baths near the Central Station)

Excavation history: A rescue excavation on the square in front of Alexandria’s central station, led by a Polish archaeological team, saw the discovery of a modest bathhouse (map 40). Bibliography: Rodziewics (1979). Construction: (B) The remains, lying 5.5 m under modern street level, were dated to the 5th c. by the study of ceramic material and the type of wall construction. Continued use and abandonment: (B) The layers of debris contemporaneous with the abandonment of the baths yielded ceramics of the late 6th and early 7th c. C80 Alexandria (Egypt), Baths behind the Water Company (After Rodziewicz (1984) 318, fig. 329) 764  Rodziewicz (2009) 198.

765  Rodziewicz (1984) 317–18. 766  Rodziewicz (1984) 322. 767  Botti (1899) 23.

403

Gazetteer

The water supply for the pools is unknown, as the standing walls have all disappeared. Waste water was disposed of through a network of brick-built sewers than ran west from the frigidarium, possibly passing a latrine and then turning back east to the heated section. Some cisterns are known from excavations in the vicinity, but cannot be linked directly to these baths. Construction technique: Opus mixtum Bathhouse surface: Unknown (at least 500 m2) GPS coordinates: 31°11’34.21”N, 29°54’17.62”E

C82. Clysma (Suez), South-west Baths

C81 Alexandria (Egypt), Baths of the Mehattet Masr Square (Baths near the Central Station), (After Rodziewicz (1984) 315, fig. 328)

Plan: The complex was located in the centre of the ancient city, delimited by the streets R3–R4 and L’2-L’3, occupying a part of an insula (map 40). The whole building was constructed in brick and courses of dressed limestone blocks. The entrance of the bathhouse could not be located. The cold rooms occupied the south-western part of the site. As the walls were systematically dismantled in the past, the form and dimensions of the frigidarium are difficult to reconstruct. One square pool (s: 2 m), a small oval pool and a network of underground sewage channels is all that can be mapped with certainty. The heated rooms lay east of the cold sector. An underground service corridor ran between the two parts. At least three rooms with a suspended floor could be identified. The rooms lying next to the service corridor were each heated by a single furnace. The southernmost room had three small basins: two oval in the north wall and one semicircular in the east wall. The rest of the building fell beyond the excavation limits and could not be investigated. Decoration: Most of the decoration of the baths was stripped during the dismantling of the walls. The square piscina and small pools were originally paved with marble slabs. The floor of the frigidarium seems to have been embellished with opus sectile. Technology: The suspended floors rested on brick pillars or possibly parallel walls. The walls were heated by tubuli.

Excavation history: The baths were excavated between 1930 and 1932 by the Institut français d’archéologie orientale du Caire. Bibliography: Bruyère (1966) 72–75; Fournet et al. (2017) 468–69. Construction: (D) The building was dated to the RomanoByzantine phase, without presenting further evidence.768 The plan, the type of bricks and the heating system could point to a late antique construction date.769 A re-examination of the ceramics found during the excavations point to time span between the 4th and 5th c.770 Continued use and abandonment: It appears that the baths were still used in Islamic times, when ‘des modifications regrettables’ (‘unfortunate modifications’) were added and later the new invader ‘construit sur les thermes de sordides maisons’ (‘built sordid houses on top of the baths’).771 Plan: The port city of Clysma (modern Suez) is known through the ancient literature, starting with Herodotus until the Islamic period, when it was called Qolmouz. The archaeological remains that were uncovered on the tell confirm on occupation history from the Ptolemaic to the late antique and Early Islamic periods.772 The bathhouse was probably accessed from the north-west. The bathers entered an almost square room (s: 6.75 m) with 4 limestone columns at its centre, possibly to support a dome.773 To the north and west of this central cold room (frigidarium), there was a rectangular recess, perhaps to serve as apodyterium. To the north-east, opposite the presumed entrance, there was a flight of stairs leading into an unexcavated part of the building. To the west of the cold room, there was a square pool, made entirely out of bricks. A corridor along the south side of this pool led to an oval basin, perhaps a pool, but more likely a reservoir due to the absence of internal steps. The heated 768  Bruyère (1966) 72. 769  Already Redon (2012a) 75. 770  Fournet et al. (2017) 468. 771  Bruyère (1966) 72; translation by the author. 772  Bruyère (1966) 26–27, 35, 46. 773  Bruyère (1966) 73.

404

C82 Clysma (Suez, Egypt), South-west Baths (After Bruyère (1966) pl. XXI)

rooms lay south of the cold section. It comprised at least two rooms, although the bad state of preservation makes it difficult to identify different rooms. The excavator proposed two north-south oriented rooms,774 but we could also reconstruct three rooms and a heat lock. We should change the interpretation of the westernmost room as the tepidarium and the easternmost room as the caldarium.775 It would make more sense if the caldarium had the large round pool on its southern side and the sudatorium or tepidarium had a semicircular apse ‘à fontaine centrale’.776 The northernmost part of the eastern room could then be a transitional tepidarium, preceded by an east-west oriented heat lock (north of the caldarium), leading to the cold section. Between the heat lock and the actual caldarium, we could reconstruct a praefurnium. The furnace and platform 774  Bruyère (1966) 73–74. 775  Bruyère (1966) 73–74. 776  c ontra Bruyère (1966) 73. The fountain could itself be an Islamic transformation of a Roman pool (see Epilogue, the Baths of al-Bara).

Gazetteer to support a boiler are clearly indicated on the excavation plan. Decoration: The decoration of the baths seems to have been very modest. The piscina was constructed in brick and mortar without any apparent veneer to embellish the inside. The floors of the heated rooms were paved with white marble slabs, as was the interior of the fountain in the eastern heated room. The round alveus of the western hot room had three marble access steps. In the destruction layer found inside the baths, fragments of mortar that probably held glass window panes were found. The excavator believed these belonged to the Islamic phase of the building, although the Byzantine Baths in Marea had similar windows.777 Technology: The hypocaust consisted of square pillars made out of brick and were connected to one another by arches. The walls were heated by tubuli that were grouped in rows of 4 and lodged in recesses in the walls. Other recesses may have acted as chimney flues. The large furnace in the praefurnium had a large boiler mounted on top of it and heated the eastern room. The excavator identified 6 other ‘boilers’ in the floors of the heated section.778 These round barrel-shaped pits were constructed in terracotta or brick and were approximately 0.7 m deep (diam.: 0.8–1.1 m). However, these ‘boilers’ may well have been later (Islamic?) additions, when the building had already changed function. Their location in connection to the heated rooms seems strange. Fournet only reconstructs two furnaces in connection to the baths.779 The water to the pools was supplied for a castellum west of the baths.780 A lead water pipe fed the water directly to the oval basin and, by a separate branch, to the piscina. The lead inlets for the alveus and the fountain can also be seen in the walls, but it is not specified how these connected to the boiler(s). The pools were emptied by outlets at their bottom. Waste water from the heated section was led west of the building to an unknown sewer. The drainage pipe of the piscina and a sewer running underneath the frigidarium led into a large sewer, which led downwards, through the city and then to the harbour, where it discharged into the sea. Construction technique: Opus testaceum, opus caementicium Bathhouse surface: ca. 300 m2 GPS coordinates: Exact coordinates unknown

777  Bruyère (1966) 73. For Marea, see below. 778  Bruyère (1966) 74. 779  Fournet et al. (2017) 468. 780  Bruyère (1966) 74.

405

Gazetteer

C83. Ezbeth Fath’Allah (Egypt)

Excavation history: The baths were discovered and excavated by the Egyptian Department of Antiquities sometime in the second half of the 20th c. The remains were re-investigated in 2007 by a team of the Greco-Roman Museum of Alexandria. Bibliography: Abd el-Fattah and Seif el-Din (2009); Fournet et al. (2017) 472–73. Construction: (B) Preliminary conclusions deduced from the unpublished excavations point to a lifespan in the 6th and 7th c.781 Continued use and abandonment: A detailed analysis of the pottery should enable a more precise chronology of the building. Plan: The bathhouse was constructed in bricks (bath rooms) and local limestone. An oblong vestibule with two benches led to a round frigidarium (diam.: 6 m) with 4 semicircular niches. The two western ones are equipped with a small pool. A doorway between the northern and western niche led to an irregularly shaped room that gave access to a partially excavated natatio. Continuing west of the frigidarium, a small room acted as a heat lock with the heated section. It had a small rectangular basin in its northern wall, which is reminiscent of the function of the old labrum. A doorway opposite this basin led to the first heated room. It had no pools, only a rectangular recess in its southern wall that could hold a bench. The second heated room was rectangular in shape and had two single-person alvei. The one in the southern wall was semicircular, while the one in the western wall was rectangular, even if it was lodged in a semicircular niche. North-west of this room, a praefurnium was large enough to house the brick platform for a large boiler. Originally, this room seems to have been part of the bathing section, as a door connected this praefurnium to the heat lock.782 It was blocked when the boiler was installed. The furnace that heated both caldarium and boiler was the only one of the entire building. A service corridor was built against the northern side of the heated section. Decoration: The decorative scheme fits within CopticByzantine tradition.783 The walls of the cold and tepid rooms were covered in plaster and painted red at the bottom and left white above. In the heated rooms, the same wall coating was painted with floral motifs such as palm leaves. The floors of the cold rooms were paved with bluish marble slabs. The exterior walls were coated with a white plaster. Technology: The two heated rooms both had a hypocaust and wall heating. The suspended floor rested on walls 781  Abd el-Fattah and Seif el-Din (2009) 269. 782  Abd el-Fattah and Seif el-Din (2009) 267. 783  Abd el-Fattah and Seif el-Din (2009) 269.

C83 Ezbet Fath’Allah (Egypt), (After Abd el-Fattah and Seif el-Din (2009) 272, fig. 3)

with arched openings, rather than on the usual brick pillars, while the walls were heated by rectangular chimneys rather than tubuli. The second heated room must have had a tepid temperature, as it did not have its own furnace and the airflow coming from the furnace of the caldarium was ‘interrupted’ due to the perpendicular position of the tepidarium. The water supply is unknown. A saqiyah was possibly present somewhere in the vicinity of the baths, as evidence of water pots has been found. In a later phase, a reservoir was built against the south side of the building. The channel, however, led away from the bath building. Supply pipes inside the building connected the rectangular basin in the heat lock with the western cold pool and the natatio. Each pool had an outlet that gave into a sewage channel, running from the western alveus around the south of the building to the frigidarium and on to the natatio. The cold pools had an overflow-outlet at floor level. The channel probably discharged in the city’s sewers somewhere north of the presumed natatio. Construction technique: Mud brick, opus testaceum Bathhouse surface: Unknown (not exceeding 500 m2) GPS coordinates: 30°54’27.80”N, 29°34’3.31”E

C84. Gherra / Mohamedia (Egypt)

Excavation history: 15 km east of Pelusium, bordered by the Mediterranean coast on the north and Lake Bardawil on the east, the ruins of a late antique settlement were discovered in the early 20th c. A public bathhouse was unearthed near the sea between 1909 and 1910. Bibliography: Clédat (1909); Fournet et al. (2017), 474–75.

406

Gazetteer

C85. Karanis (Kafr Al Massalat), North Baths

Construction: (C) The construction technique, using alternate layers of brick and local gypsum, was dated by the excavators to the 5th or 6th c.784 A recent re-examination of the artefacts recovered during these excavations also point to a Byzantine phase of use.785 Continued use and abandonment: Unknown. Plan: The description of the remains is rather limited and the articulation of the different rooms is unclear. No plan was drawn to accompany the report. A schematic has recently been made after the description and photos by Clédat.786 The bathhouse was square in plan (s: 20 m) and was approached from the western side. An entrance porch with columns and a semicircular fountain (or perhaps pool?) led into a corridor. From there, one could enter the heated section, comprising three heated rooms. The southeastern room had two pools, one of which was double and one of which was circular. Decoration: The floors and the walls were originally embellished with marble slabs, most of which were restored in later periods. Technology: The rooms were heated by a hypocaust with the traditional pillars supporting the floor. Tubuli were identified on photos of the central heated room.787 The praefurnium is located at the centre of the building. A possible saqiyah, a well and reservoirs were recognized south of the baths.788 Construction technique: Unknown Bathhouse surface: ca. 350 m2 GPS coordinates: Exact coordinates unknown

Excavation history: The town of Karanis was located on the northern fringe of the Fayum oasis. Between 1972 and 1975, the University of Cairo excavated a small bathhouse in the town centre. Bibliography: Nassery et al. (1976); Castel (2009); Fournet et al. (2017) 477–78. Construction: (B) Radiocarbon dating of wooden beams used in the last construction phase of the dressing room were dated between the 4th and 6th c.789 Architecture and decoration point to a 4th or 5th-c. phase.790 Continued use and abandonment: Unknown. Plan: The baths, fitting within a rectangle of only 15 × 8 m, lay along an important east-west oriented street, 40 m north of the Soxis Temple. While the baths were called the ‘Large North Baths’ by the excavators to distinguish it from another bath found in the south-west of Karanis,791 this building was obviously a modest neighbourhood bath. It probably served the rich houses located in this part of town and it can be expected that similar baths were found in other neighbourhoods.792 As with most of the structures in town, the baths were built in mud bricks. Fired bricks were only found in the parts that were directly exposed to moist and heat. In a first phase, the baths were accessed through a door in the north-western corner of the dressing room. In a second phase, this door was blocked and a new one was created in the south-west corner. The apodyterium (4.9 × 4.5 m) had 4 rectangular niches (0.52 m large; 0.72 m high; 0.27 m deep). One of them was obliterated when the new door was created. In the middle of the south wall, a semicircular niche (1.2 m high; 0.6 m wide; 0.3 m deep) may have contained a statue. The eastern wall was constructed in a later phase, probably when the roof over the room was changed.793 Against the east, south and west walls were brick-built benches (0.7 to 0.8 m deep). To the north-west, the dressing room opened up onto the frigidarium (3 × 2.6 m), which had a rectangular pool in a recess of the western wall. The pool (internal dimensions 1.4 × 0.7 m; 0.6 m deep) was built in bricks and coated with mortar. A single step on the outside enabled the bathers to step over the rim. This piscina was covered by a semi-dome resting on spherical triangles. Above the pool, in the western wall, was a small niche (0.25 m deep). By a bent corridor and a narrow doorway (0.47 m wide), one entered the tepidarium (2.1 × 1.5 m). It had a small niche in its northern wall, probably for some type of lighting. To the west of this room,

784  Clédat (1909) 770. 785  Fournet et al. (2017) 474. 786  Fournet et al. (2017) 475, unnumbered fig. 787  Fournet et al. (2017) 474. 788  Fournet et al. (2017) 474.

789  Castel (2009) 231. 790  Fournet et al. (2017) 477. 791  Nassery et al. (1976) 237. 792  Nassery et al. (1976) 234. 793  Nassery et al. (1976) 246.

C84 Gherra / Mohameddia (Egypt), (After Fournet et al. (2017) 475, unnumbered fig.)

Gazetteer

C85 Karanis (Kafr Al Massalat, Egypt), Large Baths (After Nassery et al. (1976) plan 2)

one found a sweat chamber (1.5 × 2.7 m) with two benches (stone slabs on brick pillars) and a large niche of unknown function in the north-east corner. A small doorway gave access to the caldarium (2.8 × 2.1 m). It had one hot tub chiselled out of a single large block of local limestone (internal dimensions 1.7 × 0.6 m). The bottom was raised with stone slabs in a later phase until 34 cm under the rim. A small seat made out of bricks was also added at its western end. The tub, which seemed to have been an ordinary bath tub that was reused, was later reinforced on the outside by a brick wall. It is reminiscent of the massive granite tubs found in the fortress baths of Nag el-Hagar (see below, p. 423). All of the rooms of this bathhouse were covered with domes, except the sweat room, which had a barrel vault. Decoration: The decoration has been preserved in the frigidarium. A layer of white plaster covering the mud bricks was painted with vegetal and floral design. The paintings in the dome above the piscina showed vineyards. The plaster in the heated rooms does not seem to have been painted. Technology: The exceptional state of preservation makes it possible to reconstruct the technical aspects of these baths. The heated section had a single furnace, located in a room east of the caldarium. This room was divided in a lower and an upper part. The first was the stokehole, while the latter seems to have contained two cylindrical boilers (fig. 127). Three chimneys, lodged in the walls of the praefurnium, ensured a draught for the fire. The hypocaust floors under the caldarium and the sweat room were constructed with arches instead of the usual pillars. The wall heating was

407 ensured by square recesses that widened to shafts at room level and narrowed again at the start of the domes, where terracotta tubes slid into one another to form a chimney (fig. 128). This unusual system of wall heating may have been invented to cope with the absence of mass-produced box tiles. It is worth noting that a small square opening in the eastern wall of the caldarium connected the latter to the praefurnium. The excavators suggested that the steam of the boilers was channelled directly into the caldarium (fig. 127).794 This method of heating only became popular in the Islamic period. If this system was part of the original design, it is one of the earliest examples of a steam conduit in a bathhouse. The steam in the small caldarium must have created a sauna-like atmosphere, which may also explain the limited use of wall heating. Radiocarbon dating of charcoal found in the furnace gave results from the 4th c. BC to the 1st c. AD.795 Rather than suggesting a phase of use during this early period, it seems that the baths used old timber as fuel, probably salvaged from surrounding buildings.796 The heating system is an interesting mix of traditional Roman techniques adapted to local conditions and with some inventive alternatives (wall heating and steam vent). The simple design and basic facilities reveal a concern for fuel costs and water use. Furthermore, a production site for glass has been discovered near the furnace. The same type of glass vessels was also found in the shops surrounding the baths. Thus the baths may have been involved in small-scale glass production (availability of water and firewood). The water for the pools was stored in a reservoir north of the caldarium. A channel connected it to both the boiler in the furnace room and to the alveus. It is not clear how the heated water in the boilers got into the pool. An inlet was found just above the rim of the tub, piercing the northern wall of the caldarium, but the connection with the reservoir or the boilers is not clear. Metal conduits were probably salvaged after the baths fell out of use. There was no such inlet for the piscina, meaning the pool had to be filled manually.797 Both cold and heated tubs had an outlet at the bottom, which gave directly onto the floor of the room. The waste water of the frigidarium was led by inclination of the floor to an outlet in front of the bent corridor, while the water of the alveus was led to an outlet next to the doorway to the tepidarium. The collected waters were led by underground channels to a well in the open courtyard north-east of the building. The water may have been reused for other purposes, such as gardening. Construction technique: Mud brick, opus testaceum 794  Nassery et al. (1976) 245. 795  Nassery et al. (1976) 272–73. 796  contra Nassery et al. (1976) 273. This problem is known as the ‘old wood effect’. 797  Nassery et al. (1976) 254.

408

Gazetteer

figure 127 North Baths of Karanis (Egypt), cross-section (west-east) of the caldarium and praefurnium after Castel (2009) 238, fig. 19

figure 128

Drawing of the wall heating in the North Baths of Karanis (Egypt) in Nassery et al. (1976) plan 13

409

Gazetteer Bathhouse surface: ca. 100 m2 GPS coordinates: 29°31’12.76”N, 30°54’10.10”E

C86. Karm Kandara (Egypt)

Excavation history: The site of Karm Kandara is located at the western end of the Nile Delta, along the CairoAlexandria express way. Excavations in the 2000s unveiled part of an agrarian estate or settlement, including olive presses and a bathhouse. Bibliography: Abd el-Fattah and Abd Al-Razeq (2008); Fournet et al. (2017) 479. Construction history: (B) A preliminary study of the ceramic finds suggests a late antique lifespan.798 Continued use and abandonment: Unknown. Plan: The baths were accessed from the south-west. An oblong hall with benches along its east, south and west sides served as a combined apodyterium / frigidarium. In a first phase, a rectangular pool, coated with hydraulic plaster on the inside, was constructed against the southern wall, near the entrance. In the north-east side of the hall, two doors led into two interconnected heated rooms. The westernmost room had a doorway in its north wall, connecting to a third heated room (sudatorium?). To the east of this lay the caldarium. It had a semicircular pool in its north wall and two rectangular single-person pools in its east wall. To the west of this main bathing block, a separate bathing room with a cold-water pool also stood in connection to the hall. Its precise function is unknown, although it clearly had a specific use. South of the baths, a small latrine was found.799 Decoration: The decoration of these baths was rather modest. The walls were plastered with a white coating and the floors of both cold and heated rooms were paved with limestone slabs. In a first phase, the large hall had been paved with irregular marble fragments. Technology: The hypocaust system of the heated rooms was not investigated. The walls were heated by rectangular tubuli lodged in recesses. The praefurnia were not discovered, even if the access to an underground service corridor has been located east of the oblong hall. The water supply could not be detected either. The drainage system of the pools started at the southern rectangular pool of the caldarium and went around the building towards the west, where it collected the waste water of the second rectangular pool. The semicircular pool does not seem to have had an outlet. The sewer was joined by a second branch coming from an unknown location north of the baths and then turned around the north-western wing of the baths, where it collected the water from the cold pool. It then disappeared

798  Abd el-Fattah and Abd Al-Razeq (2008) 210. 799  Abd el-Fattah and Abd Al-Razeq (2008) 220, fig. 20.

C86 Karm Kandara (Egypt), (After Abd el-Fattah and Abd Al-Razeq (2007) 218, fig. 17)

underneath the hall, probably to head towards the latrine located south of the baths. Construction technique: Opus testaceum Bathhouse surface: ca. 200 m2 GPS coordinates: 30°53’47”N, 29°52’24”E

C87. Kom el Ahmar (Egypt)

Excavation history: The remains of Kom el Ahmar lie 17 km north-west of ancient Hermopolis Mikra (modern Damanhur), near the modern hamlet of Tell Bisinthaway. In the 1940s, the Egyptian Department of Antiquities discovered and excavated a site of which the ancient name remains unknown. Recent survey has identified important building activity in the surroundings of the baths, suggesting that the site should be identified with the capital of the nomos of Metelis.800 Two bath complexes were found here, located 100 m apart. The smallest bathhouse, mistakenly called the ‘cold baths’ by the excavators, was constructed during the Ptolemaic period and was a Greek-style bathhouse with hip-baths.801 The large bath complex, called thermae in the excavation report, had a long period of use.802 Bibliography: Khachab (1949); Fournet et al. (2017) 481–82. Construction: (D) Coin finds and ceramics from the site date from the 1st to the 8th c. AD.803 The fact that the 800  Kenawi (2014) 106–112. 801  Khachab (1949) 30. 802  Khachab (1949) 36. 803  The contexts of the coins and the ceramics are not specified; however, a gold dinar dated AD 771 was found inside the baths (Khashab (1949) 30).

410

C87 Kom el-Ahmar (Egypt), (After Khashab (1949) plan 2)

complex was a double bath, with a plan emphasizing small round pools, could point to a late antique layout, even if a High Empire predecessor once stood on the same location. According to Fournet, the double bath should more likely be interpreted as two different phases (the north wing being late antique).804 Continued use and abandonment: (B) A gold dinar (dated AD 771) was found inside the baths, but it is unknown whether the baths were still in use or if the building had changed function. Plan: The north wing of the baths was larger than the south, and could thus been interpreted as the men’s baths. The cold section of this north part was divided into two separate rooms, each with its own large piscina. The eastern cold room was equipped with a smaller rectangular pool (2.3 × 1.3 m; 1.2 m deep) and a semicircular pool (diam.: 1.3 m; 0.7 m deep) in a niche flanking the large piscina (4 × 3.5 m; 1.3 m deep). Both pools were probably added to the original design. A shallow basin (1.1 × 0.5 m; 0.5 m deep) was added in the last phase of use. The heated section consisted of three successive rooms with hypocaust, preceded by an unheated room with a semicircular niche. The first heated room may have had a pool in its apsidal west side, while the third room had a semicircular and an 804  Fournet et al. (2017) 481.

Gazetteer oval alveus.805 In the last phase of use of the building, the east frigidarium may have been converted into a heated, all-purpose bath room.806 The furnace in the room south of the large piscina may have been used to heat bath water, or simply belongs to a post-bathing phase of the building.807 The southern wing of the bath complex consisted of a small frigidarium with a square piscina (s: 1.1 m; 0.7 m deep), which could be reached by a long corridor. Both of the warm rooms were heated by several furnaces, but only the second one had a circular alveus (diam: 1.8 m; 1.3 m deep). At the entrance of the baths, there was a latrine, flushed by the waste waters of the different pools. Decoration: The decorative scheme of the baths is not well known. The west frigidarium of the north wing was paved with limestone slabs. In the cold section of the south wing, the floors were embellished with mosaics, while the heated rooms of this section were paved with marble slabs. One could imagine that the heated rooms of the north wing were decorated in a similar fashion. The walls throughout the baths were coated with lime plaster and then painted, but no details of the motifs are mentioned in the excavation report. Technology: The suspended floors were broken up in a later phase, but at some points the rectangular pillars could still be seen. The wall heating was ensured by wall cavities. The furnaces were constructed as pits in the hypocaust floor, in contrast to the traditional stoke holes at the level of this same floor.808 The excavation report does not mention tubuli or tegulae. The water for the baths was supplied by two large wells (diam.: 6.5 m and 3 m) south of the building. It is not clear how the water was elevated to the different reservoirs found throughout the baths. The water was distributed to the different pools by terracotta water conduits, which were still visible in some of the walls. There were two sewer networks: one for the cold pools of the northern wing and one for the heated pools of the northern wing and all the pools of the southern wing. Both systems flushed a latrine.809 Construction technique: Opus testaceum, opus quadratum Bathhouse surface: ca. 720 m2 GPS coordinates: 31°9’40.20”N, 30°26’51.40”E

C88. Kom el-Dosheh (Egypt)

Excavation history: In the vicinity of the modern city of Jirays, in the Ashmoun district of the Nile Delta, several mounds (Arabic: Kôm) have yielded archaeological sites. 805  Khachab (1949) 37–38. 806  Khachab (1949) 42. 807  Khachab (1949) 37. 808  These ‘high flame’ furnaces seem to become popular in late antique and Byzantine times (see above, the Kom al-Dikka Baths in Alexandria), but were already present in 2nd–3rd c. baths (Fournet and Redon (2017a) 294). 809  Sewers indicated on plan 2 in Khachab (1949).

411

Gazetteer The Kom el-Dosheh was occupied exclusively during the Byzantine period.810 Among the remains, a bathhouse built out of bricks was discovered and excavated in the beginning of the 20th c. Bibliography: Daressy (1912); Fournet et al. (2017) 483. Construction: (D) The lifespan of the village, which seems to have been occupied exclusively in the 4th to 6th c., provides a terminus post quem for the main building. The plan, however, does not point to a late antique date, resembling the design of the 2nd-c. Hunting Baths in Leptis Magna (see above, p. 360). Continued use and abandonment: (B) A terminus post quem for the destruction of the baths (collapse of the vault) has been delivered by a coin of Justin I (AD 518–527).811 After the baths had fallen out of use, the building was reused as a dwelling and a small production unit. Two ovens were found in the room south of the rectangular piscina. Plan: The bathhouse was entered through a door in the north-eastern corner of the building. It led straight into a rectangular room (3.9 × 2.7 m) that could be interpreted as the apodyterium. The oblong frigidarium (10.6 × 3.6 m) to the west had two apsidal endings. The southern one certainly lodged a semicircular piscina (1.3 m deep). There may also have been a pool in the northern apse, as a sewage channel led up to it.812 On the frigidarium’s eastern side, a third pool (1 m deep) took up almost the entire space of a room (2.2 × 1.9 m). The room immediately south of it was originally paved with small marble slabs and may have served as a districtarium or unctorium. In the middle of the western side of the frigidarium, a doorway (1.1 m) led to a central room (2.7 × 2 m) on a hypocaust. It had a semicircular niche to the west and two lateral doors. Choosing the southern one, the bathers entered a second heated room with no special features. A door to the west led into the caldarium (4.7 × 3.4 m) with its two alvei. The southern alveus (1.4 × 1 m) was rectangular and more or less screened off from the rest of the room by a thin wall. The larger semicircular alveus (diam.: 2.5 m) was surrounded by a rim of 30 cm. To the north, a fourth heated room lay just above the baths’ only furnace, and could hence be interpreted as a sweat room.813 The last heated room to its east had a large brick bench against its north wall. It may have been used for massages or simply for resting.814 A doorway south led back into the central room with apse, thus completing the bathing route. From the frigidarium, a doorway in the south-east corner led outside to some annexes that were too damaged to interpret. The doorway was screened off from the room by a thin wall. 810  For the ceramics, see Daressy (1912) 185–91. 811  Daressy (1912) 175, n. 1. 812  Daressy (1912) 178. 813  Daressy (1912) 180. 814  Daressy (1912) 180.

C88 Kom el-Dosheh (Egypt), (After Daressy (1912) 177, fig. 2)

Decoration: The walls of the baths were coated with mortar and a fine layer of plaster, which was then painted. Just as in the baths of Ezbet Fath Allah, the bottom part of the walls was painted plain red, while the upper part was white with polychrome motifs. The floors of the cold part were paved with small marble slabs and those of the heated section with limestone. Most of the marble slabs were reused pieces of marble cut into slices to fit their new purpose. One marble statue of a young man was found just outside the baths. Technology: The heated rooms were built on a brick pillar hypocaust. For extra support and good airflow, the pillars were connected to one another by arches. The wall heating consisted of box tiles placed in recesses within the walls. The tubuli were connected to each other by lateral openings. It is remarkable that the entire heated section had only one furnace, located in the almost square room west of the sweat chamber. This could point to a shortage of fuel. The hypocaust did not continue under the alvei. There are no traces of water supply channels to fill the pools. A possible reservoir was found north of the baths, but it is unclear if this structure was linked to the baths. Water disposal is well understood. Each of the pools was drained by a lead pipe discharging into sewage channels. The semicircular alveus of the caldarium also had an overflow-outlet at floor level. The outlet of the rectangular piscina was still plugged with the bottom of an amphora at the time of excavation.

412 Construction technique: Opus testaceum Bathhouse surface: ca. 300 m2 GPS coordinates: Exact coordinates unknown (destroyed)

C89. Marea (near Al Hawwariyah), Byzantine Baths

Excavation history: The city of Marea was located on the south-west shore of Lake Maryut. It acted as an important transit city between the Nile Delta and Alexandria. It was also known for its fertile hinterland, producing olives, grapes and other types of fruit.815 The city—judging from the ceramics found on site—prospered during the Byzantine period, perhaps as a stop on the way to the pilgrimage site of Abu Mena. The absence of Arab glazed pottery (8th c.) indicates that the city was abandoned shortly before or after the Arab invasion.816 The bathhouse—excavated during the first decade of the 21st c. by the Polish archaeological mission—was situated 160 m south-east of the harbour. Bibliography: Szymanska and Babraj (2008); Szymanska and Babraj (2009); Fournet et al. (2017) 485–86. Construction: (A) On the basis of the ceramics and coins found during the excavations, the construction date of the baths has been placed at the end of the 5th c.817 Continued use and abandonment: (B) The layer of debris that accumulated after the abandonment of the baths contained material of the 8th c.818

Gazetteer Plan: The bathhouse was built entirely out of brick and surrounded by a stone wall. It consisted of separated north and south wings, each accessible through a colonnaded courtyard. Both wings were accessed by a vestibule on the north side, which gave into these courts. The western court gave access to the south wing and was clearly the largest. It had 4 single-person cold pools on its eastern side, with a maximum depth of 1.8 m. Upon entering the main building, one passed two cold rooms before entering the heated section. The first unheated room had a pedestal in its north-eastern corner. The terracotta conduit coming out of the wall above it indicates the presence of a water basin— a type of small labrum.819 An oval room to the east of this first room probably contained a boiler, which could be heated from the underground service corridor. Continuing south, one entered the tepidarium, mildly heated by sets of tubuli in the wall. The first room with a hypocaust was rectangular in shape (3.3 × 5.3 m) and had three small pools (all ca. 1.3 m deep). The semicircular pool (r: 0.63 m) was initially rectangular. To the north of this room, a small square room had a semicircular (r: 1 m) and a round pool (diam.: 1.05 m), both only large enough to accommodate one bather at a time. The northern part of the baths had a similar layout. A colonnaded courtyard (8.3 × 10 m) had two small rectangular pools and access to a cold room with a semicir-

C89 Marea (Al Hawwariyah, Egypt), Byzantine Baths (After Szymanska and Babraj (2008) 48, fig. 14) 815  Szymanska and Babraj (2009) 247. 816  Szymanska and Babraj (2008) 13. 817  Majcherek (2008) 105–127; Lichoka (2008) 145–52; Malarczyk (2008) 153–59. 818  Malarczyk (2008) 153–59.

819  Szymanska and Babraj (2009) 249.

413

Gazetteer

figure 129

Byzantine Baths of Marea (Egypt), drawings of the fragments of claustra found inside the baths after Kucharczyk (2009) 261, fig. 3

cular niche. It is possible that a small labrum stood here.820 A first heated room (3.9 × 1.9 m) had two single-person pools, one semicircular (diam.: 1.6 m) to the north and one rectangular (0.9 × 1 m) to the south-east (both around 1.1 m deep). The second heated room was rectangular and had two pools in its northern side. The original rectangular shapes were later transformed into smaller semicircular forms. The bathing rooms were all covered with barrel vaults, while most of the pools had a semi-dome vault. The rooms lining the north of the baths may have been shops, as glass flasks and small objects were found. The division of these baths into two equal parts is explained by the excavators as a division between a bathhouse for men and for women. Several ‘female objects’ such as bone hair pins and glass pearls have been found in the sewers of the north part of the baths. Such an interpretation is also proposed for the double baths in Abu Mena (see above, C75–C76). An interpretation as winter and summer baths is difficult to support, as both parts are almost equal in size. The south part was slightly larger and hence had more pools than its northern counterpart. The number and size of the heated rooms, however, was the same for both parts. The large courtyards must have functioned as frigidaria. It is not known if or how these areas were roofed. Possibly only the space between the columns and the walls was covered.821 The type of plan, especially the courtyard frigidarium and small heated section, resembles Byzantine and Early Islamic baths in the Near East (see above, chapter 4, p. 187–188). Decoration: The walls of the internal rooms of both parts were covered with thick layers of plaster. Traces of painted vegetal motifs and Maltese crosses were found on them. All the pools were lined and paved with marble slabs, some of them still in situ during the excavation. The columns in the courtyard were limestone monoliths, but the 4 Corinthian capitals found throughout the site were

of a greyish Proconessian marble. The floors of courtyards and rooms were paved with marble slabs. Several fragments of window panes and claustra (grating) were also found (fig. 129). The resemblance with the Early Islamic oculi, especially the inverted heart shapes, is remarkable.822 Technology: The service corridor was accessed from a staircase south of the baths, leading underneath the two unheated rooms of the southern wing. From this corridor, the furnaces for the southern and northern heated rooms were fired, as was the large boiler that was lodged in an oval room east of the room with labrum. Traces of reed and straw were found when analysing the ashes in the praefurnium. It is possible that dried dung was used as fuel. The hypocausts were made out of massive brick pillars forming arcades. The pools in the heated sections were not constructed on hypocausts, meaning the water was only heated in boilers and not in the pools themselves. The wall heating consisted of tubuli lodged in separate recesses inside the walls. The water for the pools was supplied by a large well 5 m north-east of the baths, operated by a saqiyah. It was of rectangular shape (1 × 3.4 m), at least 5 m deep and stood in connection with a reservoir (5.5 × 3 m) and a cistern (2.5 × 1.8 m). From the reservoir, the water was led to the pools by terracotta pipes. Water disposal was ensured by a network of underground drainage channels that connected with the outlets at the bottom of the pools.823 Construction technique: Opus testaceum (heated parts), opus quadratum (courtyard, annexes) Bathhouse surface: ca. 642 m2 GPS coordinates: 30°59’36.65”N, 29°39’26.42”E

820  Szymanska and Babraj (2008) 33. 821  As reconstructed in Szymanska and Babraj (2008) 49, fig. 16.

822  Kucharczyk (2009) 258, 262, fig. 4. 823  Szymanska and Babraj (2008) 50, fig. 17.

C90. Marea (near Al Hawwariyah), Central Baths

Excavation history: The site was excavated between 1979 and 1981 by the University of Alexandria. The heated section was not investigated.

414

Gazetteer

C91 Mergham (Egypt), Baths (After Fournet et al. (2017) 493, unnumbered fig.)

C90 Marea (Al Hawwariyah, Egypt), Central Baths (After Fournet et al. (2017) 488, unnumbered fig.)

Bibliography: Fakharani (1983); Sadek (1992); Fournet et al. (2017) 487–88. Construction: (D) The bathhouse is located in the centre of the Byzantine town, near an economic zone. The bath is hence dated following the lifespan of the surrounding urban fabric, i.e. between 5th and 7th c.824 Continued use and abandonment: Unknown. Plan: The bathhouse consists of two identical wings (east and west), lying next to each other. The east wing was probably accessed from the east, passing two vestibules. Two rooms with benches along the walls should be interpreted as apodyteria. The cold section is a large colonnaded hall with semicircular apses to the east and west. Two small pools abut the southern colonnade. In the portico just south of this colonnade two large piscinae (two to three persons) are located at the east and west end, while two small semicircular pools flank the doorway that gave 824  Fournet et al. (2017) 487.

access to the unexcavated heated section. The west wing was identical in layout, although there were no small pools along the colonnade, but two extra semicircular pools flanking the doorway to the heated section. At least one of the heated rooms had two small rectangular alvei. Both wings were connected through a passage in their shared side, consisting of 4 connecting rooms that impeded a direct insight into the cold halls. Decoration: Marbles tiles were used as flagstones and possibly to adorn the walls. Technology: There is no information on the heating system or water management. Construction technique: Opus testaceum Bathhouse surface: Unknown (at least 1,200 m2) GPS coordinates: 30°59’37”N, 29°39’18”E

C91. Mergham (near Teibat Sweiha, Egypt), Baths

Excavation history: The bathhouse was excavated by the Supreme Council of Antiquities and the Graeco-Roman Museum of Alexandria in 1990, but the results were never published. The unpublished report was recently studied by the Balnéorient project. Bibliography: Fournet et al. (2017) 492–93. Construction: (D) The artefacts that were collected on the surface date between the 5th and 7th c.825 Continued use and abandonment: Unknown. Plan: The bathhouse was enlarged at an unknown date from a simple angular row type to a double bath with an 825  Fournet et al. (2017) 492.

415

Gazetteer east and west wing. Both wings had a similar layout. The west wing was the larger of the two and had a hall with two rectangular pools. The heated section consisted of a tepidarium with possibly a labrum in the semicircular apse, a small heated room (sudatorium?) and a large rectangular caldarium.826 The latter had one larger semicircular alveus protruding out of the south wall and two adjacent singleperson pools built into the west wall. A third alveus possibly existed in the eastern apse during the first phase of the building. The east wing had a similar cold hall, which may have been divided into two rooms (at a later date?). There was one semicircular cold pool in the north-eastern wall and a bench. The heated section comprised only two rooms. The one at the end of the itinerary (caldarium) had two semicircular alvei. Decoration: Corinthian capitals and marble and limestone columns were found during the excavation. Technology: The main furnace was built after the baths had been doubled, at the centre of the building to service both wings, on the location of a presumed alveus of the west wing caldarium. Several boilers may have surmounted this furnace. Both caldaria had a separate furnace as well (the one in the west wing lying directly under the semicircular alveus). The heated rooms have hypocausts, but no wall heating. Chimneys are cut in the walls. The waste waters of all the cold pools, as well as those of the north alveus of the east wing caldarium, were drained by a sewer system heading westward. The alvei of the west wing caldarium had separate drains leading to an unknown sewer. Construction technique: Opus testaceum Bathhouse surface: Unknown (at least 230 m2) GPS coordinates: 31°05’37”N, 29°51’07”E

C92. Mit Abul Kum (Egypt), Baths

Excavation history: The site lies 3 km south-west of Marea and close to the remains of a possibly Byzantine settlement. The baths were excavated by the Supreme Council of Antiquities and the Graeco-Roman Museum of Alexandria in 1998–1999, but the results were never published. The remains have been re-examined by the Balnéorient project in 2010. Bibliography: Fournet et al. (2017) 494–95. Construction: (D) The unpublished inventory of finds from the excavations gives a time span between the 5th and the 7th c.827 Continued use and abandonment: Unknown. Plan: At an unknown point, the original layout was changed from a simple ring type bath into a double bath with a north and south wing of the angular row type. The north wing had a cruciform courtyard with a large rectangular piscina on its north side and 4 smaller pools on the opposite south 826  Fournet et al. (2017) 492. 827  Fournet et al. (2017) 494.

C92 Mit Abul Kum (Egypt), Baths (After Fournet et al. (2017) 495, unnumbered fig.)

side. A small doorway in the latter led into the heated section, consisting of a transitional room without hypocaust (tepidarium), a first heated room with a small semicircular basin in its east wall and a third larger caldarium with three alvei (the one in the west wall was large enough for two or three people). The south wing had a large courtyard (only partially excavated) to the east, with at least three piscinae (the southern one being the largest). Continuing west, the bather entered the bathing block by an oblong transitional room (tepidarium), a heated room without pools (sudatorium) and a caldarium with three alvei. The largest alveus was semicircular and enabled up to three people to bathe together. Decoration: Some capitals and column drums in marble and granite were recovered. Marble slabs adorned some of the walls of the pools. Technology: The main furnace was located at the intersection of the two wings and could have heated several boilers. Additional furnaces to heat the rooms may have existed. The heated rooms had a hypocaust, but no wall heating (only chimneys were found in the walls). The alvei of both caldaria were drained by a sewer system that collected waste water west of the building. The sewage system of the cold pools could not be found. Construction technique: Opus testaceum Bathhouse surface: Unknown (at least 370 m2) GPS coordinates: 30°58’25”N, 29°39’32”E

C93. Psenemphaia? (Kom Trougah)

Excavation history: During the 1950s the Egyptian Anti­ quities Organization discovered and excavated the remains of a small site in a hamlet called Kom Trougah, 36 km south-east of Alexandria. Besides a Greek-style private bathroom in a private residence, a Roman-style bathhouse was discovered. Bibliography: Khachab (1957); Fournet et al. (2017) 506–507.

416

C93 Psenemphaia? (Kom Trougah, Egypt), (After Khashab (1957) plan I)

Construction: (D) The resemblance to the Baths of Kom alDosheh (see above, C88), itself not securely dated, however, and the type of heating system (see above, the Byzantine Baths of Marea; Baths of Kom al Dikka in Alexandria) could point to construction in Late Antiquity.828 Fournet also prefers a late antique date on the basis of architectural and technological parallels with other late antique baths.829 Continued use and abandonment: Unknown. Plan: The baths were entered by a vestibule in the north, in which a small basin was built in a later phase. The frigidarium was a large rectangular hall with a piscina (2.75 × 1.5 m) lodged in an adjacent room to the east. The latrine in the south-east corner of the building was accessed by a separate entrance from the outside. The heated section was reached by a doorway on the west side of the frigidarium. A first square room functioned as a heat lock, as it was connected to the entrance and exit-tepidarium (to the north and south respectively). Following the bathing itinerary north, one passed from the tepidarium into the first heated room. It had a semicircular niche in its west wall, which still lodged the base of a statue at the time of excavation. Continuing south, the square room acted as a sudatorium, before entering the caldarium south of it. The latter had two pools, one circular and one rectangular. By a doorway to the east, the bather could rejoin the heat lock 828  Nielsen (1993b) 35 dates the baths to the 1st or 2nd c., probably on the basis of coin finds mentioned by Khashab (1957) 132– 39. However, the context of these finds is not communicated. It is only said that they were found scattered around the site (‘Ces pièces en bronze, dispersées dans le kôm’, p. 132); they do not necessarily date the baths. 829  Fournet et al. (2017) 506.

Gazetteer and then the cold rooms. The baths were probably exited by a separate doorway south of the frigidarium. The interpretation as private baths is not supported by the archaeological evidence, as no obvious remains of a dwelling were found in connection with these baths. Decoration: According to the excavator, the baths were richly decorated with statues, painted plasters and mosaics. A head of Venus was amongst the finds. There is no information available about the revetment of the walls. Technology: The three warm rooms were heated by a single furnace, located in an oval room at the centre of the heated section. This praefurnium was reached by an underground corridor running under the heat lock and the exit-tepidarium. The corridor was accessed by steps located next to the exit of the baths. The furnace itself was surmounted by a brick platform, probably to hold the boiler. The hypocaust consisted of the traditional square pillars, while the walls were heated by flues lodged in the walls, using tegulae mammatae to separate these vertical heat channels from the room. The water for the pools was supplied by a large well just north of the baths and was stored in elevated reservoirs to the west of it, supplying the pools and the boiler through lead pipes. The piscina was drained by a sewer that passed the latrines before heading west to join the drain of the alvei.830 Construction technique: Opus testaceum, opus africanum Bathhouse surface: ca. 170 m2 GPS coordinates: 30°57’47.74”N, 30°10’28.86”E

C94. Sersena (Egypt), Graeco-Roman Baths

Excavation history: In 1966, the Egyptian Antiquities Organization discovered a bathhouse at Tell Sersena near the modern city of el-Shohada. Bibliography: Melek Wasif (1979); Fournet et al. (2017) 508–509. Construction: (D) The excavators attributed the bath to the ‘Graeco-Roman’ period without any further attempts of dating, even if a large quantity of ceramics was found.831 On the basis of the plan, the baths have been tentatively dated to Late Antiquity.832 Continued use and abandonment: Unknown. Plan: The interpretation of the plan as presented by the excavator needs revision.833 The bathhouse was built entirely out of brick and seems to have been altered several times. One probably entered from the north-east or south, accessing a courtyard. According to the excavators, this area was intended for the personnel of the baths and the storage of bathing utensils, fuel, etc., as could be deduced from 830  Indicated on plan 1 in Khachab (1957). 831  Melek Wasif (1979) 177, 181. 832  Redon (2012a) 76; Fournet et al. (2017) 508. 833  For an alternative, yet very similar interpretation of the remains, see also Fournet et al. (2017) 508.

Gazetteer

C94 Sersena (Egypt), Graeco-Roman Baths (After Melek Wasif (1979) pl. I)

the numerous small rooms built out of mud brick and the large amount of sherds, including “amphorae for the wine for the bathers.”834 A small latrine was also found in the north-east corner. There are several reasons to doubt this interpretation. First of all, the excavators seem to overlook the fact that both the mud-brick walls and the sherds could post-date the bath phase of the building. As there is no information on stratigraphy or interrelationship of walls, the contemporaneity of these archaeological features is difficult to assess. Furthermore, the presence of the bath’s main entrance and of a latrine rather indicates that this was the cold section of the baths. Indeed, the plans of the Double Baths of Abu Mena (C75–C76), Marea (C89) or Taposiris Magna (see below, C95) reveal that the cold section was often preceded by a courtyard with adjacent latrines. The courtyard was followed by a cold room, labelled a tepidarium by the excavators. The oblong hall (12.1 × 4.25 m) was bordered on its north, east and south sides by rectangular rooms. These rooms, separated from the hall by a pair of columns, were interpreted as apodyteria.835 It is more likely that the alleged ‘tepidarium’ was in fact the frigidarium. The rectangular recess west of the hall is reminiscent of similar arrangements in the frigidaria of the Baths of Karanis (C85), Taposiris Magna (C95) and especially Teiba (see below, C96). According to the excavator, it was heated by a hypocaust. Two (unheated?) pools—the southern one 834  Melek Wasif (1979) 181. 835  Melek Wasif (1979) 179.

417 rectangular (2.15 × 1.87 m; 1.12 m deep), the northern one square (s: 2.1 m; 1.2 m deep) but later reshaped to an octagon—could hence be interpreted as the single person piscinae. The round basin (diam.: 1.05 m; 0.5 m deep) in the floor of the recess was most likely a pediluvium, rather than a bath for babies.836 The recess with opposite pools had two doorways that gave into an oblong heated room, called the ‘laconicum’ by Melek Wasif. This room ended on its short sides with semicircular basins (1.2 × 0.8 m), “used as hot air bath each, (Greek function); and afterwards for the emission of hot vapour comming (sic) through tubes from the boilers.”837 As both basins could be entered by a step, it seems that these are the small hot-water pools associated with the tepidarium. The rooms west of this oblong room were destroyed and could not be examined. The excavators reconstructed the frigidarium on this location, but it is more likely that these are the actual heated rooms. The vicinity of the water reservoir and sewer support this hypothesis. The two semicircular apses probably lodged the alvei, while the square recess with semicircular interior north of the northernmost room was probably a single-person pool. The southernmost room, benefitting from afternoon sun heat, was likely the caldarium.838 Decoration: The decorative schemes of these baths seem quite modest. The floors of the oblong heated room of the large hall were paved with limestone slabs. The small pools and basins were coated with a layer of waterproof plaster. Only the bottom of the pediluvium had limestone slabs. Technology: The hypocaust pillars of the oblong room were made out of tiles and traces of tubuli have been found inside wall recesses. There were allegedly 5 furnaces for the heated room, although it seems that these ‘arched apertures’ were actually the heat channels that connected the hypocaust of the oblong room with the hypocaust of the frigidarium recess and of the rooms west of the oblong room.839 According to the excavation report, there were 836  As argued by Melek Wasif (1979) 179. For similar pediluvia in the cold room, located before the entrance of the heated section, see Karanis (C85). See also the baths of Legio (Jordan) and Bostra (Palace of Trajan; Piraud-Fournet (2014) 693–94) in the Middle East. 837  Melek Wasif (1979) 178. 838  Note, however, that the single-person pool was often located in the caldarium. If this were also the case for this bath, the north-west room would be the caldarium (compare with the north-oriented caldarium of Karm Kandara, C86). This is also the reconstruction preferred by Fournet et al. (2017) 508. 839  The five openings in the walls of the oblong room in the plan by Melek Wasif (1979) pl. I should hence be interpreted as hypocaust arches and not as doors. However, these arches were often located underneath thresholds, because of their solid structure, and could thus point out the location of doorways. The two small openings in the south-west part of the west wall are less likely to have supported thresholds if we imagine that the south-west room of the complex was the caldarium and if

418 furnaces below each small basin of the oblong heated room. Additional furnaces likely heated the three rooms on the west side of the building. The water for the pools was supplied by a large well (diam.: 7 m) located north-west of the building. According to Melek Wasif, it was operated by a shaduf, a hand-operated system that resembled a seesaw with a rope and a bucket on one end of the horizontal pole and a counterweight on the other end. The water was then led into a reservoir (3.3 × 2.2 m) just east of the well, and then distributed to the different pools and boilers by a system of pipes. The piscina and the pediluvium were drained by outlets that gave into a central sewer running east underneath the pavement of the large hall. It probably joined the large sewer running just south of the bathhouse.840 At some point, the sewer of the latrine must also have joined the latter. Construction technique: Opus testaceum Bathhouse surface: At least 500 m2 GPS coordinates: 30°35’47”N, 30°54’02”E

C96. Taposiris Magna (Abusir on Lake Mareotis), Byzantine Baths

Excavation history: In 2009, the remains of a Roman-style bathhouse were identified at Taposiris Magna.841 Since discovery, several excavation campaigns by the French archaeological mission in Egypt have investigated these baths. Bibliography: Le Bomin (2015); Le Bomin et al. (2016); Fournet et al. (2017) 511. Construction: (A) A test pit along the walls of the baths established a chronology on the basis of the ceramic finds. It seems the baths were built around the middle of the 5th c.842 Continued use and abandonment: (B) The last phase of the baths could be dated to the middle of the 6th c.843 The baths seem to have gone out of use during the second half 7th c., when layers of debris filled the building.844 Plan: The building is located 20 m west of the main road leading to the old Osiris sanctuary, which was later transformed into a military camp. The maximum extent of the baths, built out of bricks and local stone, is some 40 × 18 m, oriented along a north-east / south-west axis. At least during its last phase, the bathhouse was divided into two we assume that the bathing itinerary was reversed (a circular itinerary would suppose a doorway connecting the caldarium to the corridor-like tepidarium). 840  Melek Wasif (1979) 181. 841  The discovery of the site was announced on the Balnéorient website (http://balneorient.hypotheses.org/383). The preliminary results were presented on the same website, see Le Bomin (2015); Le Bomin et al. (2016). 842  Fournet et al. (2017) 511. 843  Fournet et al. (2017) 511. 844  Le Bomin et al. (2016) §3.

Gazetteer

C95 Taposiris Magna (Abusir on Lake Mareotis, Egypt), Byzantine Baths (After Le Bomin (2015) fig. 2)

separate sections, likely for male and female bathers.845 The baths were accessed through a courtyard with a portico on its northern and southern side. At a later date, the latter seems to have been divided into separate rooms. From this communal courtyard, the bathing itinerary split up. To the south, a doorway led into a cold room with benches, with two single-person pools facing one another. A rectangular transition room (tepidarium?) then led into a heated room which had two small single-person pools. The entrance to the northern wing of the baths was flanked by two basins (acting as labra?). The cold room with a bench along its northern and eastern wall had no pools. It gave access to a first heated room, apparently also without a pool. To the west of this room, a second heated room is still under investigation. It had at least one square single-person pool in an apse on its southern end. Decoration: The pools were lined with marble slabs. Some fragments of the decoration have already been found, including some wall paintings depicting human figures. The presences of round glass panes point to the existence of oculi. Technology: There is no information on the hypocaust. The draught for the hot air seems to have been ensured by flues lodged in recesses within the caldaria and tepidaria walls. Along the western side of the bath rooms lay the service corridors, as could be deduced from the thick layers of ash encountered during excavation. The furnace itself was of the ‘high flame type’ and heated the caldaria of both wings.846 It was surmounted by two boilers. On a higher level to the west of the praefurnium, a water reservoir which could contain up to 8 m3 was found. The water was probably supplied by a saqiyah, as numerous fragments of saqiyah jars were recovered during excavation. The alvei 845  Le Bomin (2015) §4. 846  See already the Kom al-Dikka Baths in Alexandria.

419

Gazetteer and piscinae of the south wing were drained by the same sewer, passing underneath the courtyard of the south wing in south-west direction. Construction technique: Opus quadratum, opus testaceum Bathhouse surface: ca. 720 m2 GPS coordinates: 30°56’36.42”N, 29°31’11.17”E

C96. Teiba (Egypt)

Excavation history: During a prospection campaign in 1988 in the area of el-Nahda, 35 km south of Alexandria, several Roman and Early Islamic archaeological sites were discovered. One of these sites, located on a low tell near the modern village of Teiba (or Tiba), was excavated in 1990 by the Department of Antiquities. Bibliography: Choukri (2007); Fournet et al. (2017) 512–13. Construction: (B) The bath was constructed in the 6th c., according to the ceramic finds in test pits. Continued use abandonment: (B) During the 7th c. debris with datable sherds filled the building. Plan: The baths were entered from the north. A vestibule (7.3 × 4.5 m) paved with limestone slabs probably served as apodyterium. Benches built out of compacted mud (including crushed shells and chalk) were located in the south-east and south-west corners of the room. The walls were built out of rubble concrete mixed with clay and covered with coloured plaster. Continuing south, the apodyterium opened up to a rectangular frigidarium (2.7 × 8.4 m) flanked by two round pools. The east pool (diam.: 0.8 m; 1.35 m deep) was slightly larger than the west pool (diam.: 0.7 m; 1.4 m deep). Both were constructed in brick with their interior covered with a layer of hydraulic mortar. A small doorway to the south led the bathers into the first heated room, probably the tepidarium.847 A doorway in its south-west corner connected with the presumed sudatorium. From here, a large passage led to the southernmost room, the caldarium. It was the only heated room to have pools: two lateral round pools and one semicircular pool to the south (no measurements available). The bathers could then return to the frigidarium by a transitional room situated above an underground praefurnium.848 There is no reason to assume that these baths were for private use only, as no villa has been found in the immediate vicinity.849

847  Choukri (2007) 328. 848  The excavator does not consider the space above the praefurnium to have been a room and thus considers a reversed route, see Choukri (2007) 328–30. As only the foundations of the heated rooms have been preserved, there is no reason to assume that there was no room or corridor above the praefurnium (see Pl.I,1 in Choukri (2007)). Similar rooms / corridors above the praefurnium have been found in the Byzantine Baths in Marea (see above). 849  c ontra Choukri (2007) 325.

C96 Teiba (Egypt), (After Choukri (2007) 329, fig. 2)

Decoration: The floors of the cold rooms were paved with rectangular limestone slabs. The floors of the heated rooms, on the other hand, have not been found. The walls were covered with a plain white plaster. Only in the frigidarium could a trace of colouring be recognized. Technology: The limited size of the heated rooms enabled the builders to support the suspensura by a small number of pilae. For the first heated room, only one large brick pillar seems to have sufficed. It is not clear whether the recesses in the walls lodged tubuli or if these were chimney flues. The oblong room east of the heated rooms was filled with ashes when discovered by the archaeologists. It probably

420

Gazetteer

served as a storage room for fuel and as a provisional dump for ashes. The water for the pools may have been stored in a reservoir located 5 m south- west of the baths.850 The waste waters of the piscinae were drained by a T-shaped sewer heading north underneath the vestibule. The pools of the caldarium had outlets leading to the exterior of the building and collecting south of the caldarium. Construction technique: Rubble concrete, opus testaceum Bathhouse surface: ca. 216 m2 GPS coordinates: Exact coordinates unknown Tell el-Kanais Excavation history: A bathhouse in the surroundings of Pelusium was excavated in 1992–1993 by the Egyptian Department of Antiquities, but remains unpublished. Bibliography: No publication available. Construction and abandonment: The site, known as Kanaïs IV, yielded the remains of brick-built rooms and water channels. A preliminary study of the ceramics proposes a lifespan between the 4th and 6th c.851 Plan: Unknown. Decoration: Unknown. Technology: Unknown.

C97. Tell el-Ruhban

Excavation history: In 1993–1994, the Supreme Council of Antiquities of Egypt carried out a small-scale excavation around the mound of Tell el-Ruhban, in the eastern Nile Delta. A bathhouse of around 450 m2 was discovered just outside the modern village. The results were never published, but more than a decade later a brief description of the site was given by A. Ashmawy Ali. Bibliography: Ashmawy Ali (2009); Fournet et al. (2017) 516. Construction: (D) A late antique date could be proposed on the basis of the plan and the resemblance to the baths at Sersena (see above, C94).852 Continued use and abandonment: Unknown. Plan: The walls, built entirely out of brick and covered with lime mortar, were badly damaged, making interpretation of the different rooms difficult. At least two semicircular pools can be recognized. In a second phase of expansion, more rooms seem to have been added on the northern and southern ends of the building. Decoration: Unknown. Technology: There is no information on the heating system. A well (diam.: 4 m) south-west of the building provided the pools with water. Two sewage channels disposed of waste water. Construction technique: Opus testaceum 850  Choukri (2007) 330. 851   Carrez-Maratray and Tabaie (1993) 108 quoting the unpublished excavation report. 852  An opinion shared by Fournet et al. (2017) 516.

C97 Tell el-Rohban (Egypt), (After Ashmawy Ali (2009) 207, fig. 5)

Bathhouse surface: Unknown (not exceeding 500 m2) GPS coordinates: Exact coordinates unknown. Xois (Sakha) Excavation history: The baths were discovered during construction works in 1960 and partially unearthed. The site has never been excavated. Bibliography: el-Khachab (1978); Fournet et al. (2017) 523. Construction: (D) Based on the photos in a book by A. el-Khachab, Fournet proposes an Early Byzantine date (plan, use of tegulae mammatae).853 Continued use and abandonment: Unknown. Plan: No plan was ever published. According to el-Khachab, at least one room had a hypocaust. Two other heated rooms could have been the tepidarium and sudatorium. There seem to have been pools in heated rooms and a natatio (sic)—more probably a piscina—in the cold section. Decoration: Unknown. Technology: The praefurnium was situated underground. There were tegulae mammatae to heat the walls in the sweat room.

Possible Late Antique Baths Some sites have yielded bathhouses which have been preliminarily dated to Late Antiquity, but have never been 853  Fournet et al. (2017) 523.

421

Gazetteer excavated or are in the course of investigation. In Alexandria, excavations on the cricket playground uncovered a late antique building.854 The use of bricks and the presence of sewage channels were seen as possible traces of a bathhouse. However, no other evidence for a bath building—such as pools or a hypocaust—is mentioned in the excavation report. 20 km to the east of Marea, the village of Merghib yielded a bathhouse that probably belonged to the Byzantine period.855 Still in the Nile Delta, the remains of a possible bathhouse were identified at al-Barnugi.856 The type of bricks and the presence of small pools could point to a late antique date. The surface finds collected during the survey do not exclude a late date, as material from the 1st to 8th c. AD was found.857 On the south-eastern fringes of the Delta, the site of Tell Ishnik was briefly surveyed in the 1960s. The unpublished sketches that were made show a room with two opposing semicircular basins.858 The remains of this possible bathhouse could date from the Roman or late antique period.859 At the site of Kellis (Ismant el-Kharab), magnetometry in combination with survey have brought to light a Roman or late antique bathhouse. The surface finds from the site date from the 1st–2nd c. and the 4th–5th c.860 The Baths of Kom Kobbeiz in the Delta are tentatively dated to Late Antiquity.861 Surface pottery found during a visit to the site by B. Redon points to a 5th or 6th c. date.862 In the port city of Clysma, a presumed Islamic-style sweat room was identified on the city tell. The large cistern was thought to double as a hammam, but the evidence presented by the excavator is unconvincing.863 Similarly, the room with hypocaust and the several ‘basins’ and ‘boilers’ found on the tell interpreted as a large public bathhouse could easily be part of a private residence with some small-scale production units.864 The option of a private bath has also been proposed.865 The site of Kom al-Baroud lies 5 km south-west of Dilingat, between two ancient canals. During a survey in 2009, the remains of a building constructed entirely in bricks were discovered.866 The surface finds of the area had a wide chronological range dating between the 1st and 8th c. AD.867 Near the 854  Shenouda (1973) 193. 855  Redon (2012a) 76. 856  Kenawi (2014) 155–56, fig. 284. 857  Kenawi (2014) 155. 858  Ashmawy Ali (2017) 85. 859  Fournet et al. (2017) 514. 860  Fournet et al. (2017) 480. The discovery was published on the website of Monash University, http://artsonline.monash.edu .au/ancient-kellis/bath-house/. 861  Redon (2012a) 75. 862  Fournet et al. (2017) 484. 863  Bruyère (1966) 71. 864  C  ontra Bruyère (1966) 75. 865  Fournet et al. (2017) 470. 866  Kenawi (2014) 137–38, figs. 241 and 242. 867  Kenawi (2014) 137.

Roman fortress of Nicopolis (Ramleh), two small late antique baths may have stood outside the north-west and south-west corner of the walls.868 The remains have not been preserved.

Other Private Baths with a Possible Public Use Abu Mena / Karm el-Barasi, Private Baths Excavation history: In the vicinity of the pilgrimage city of Abu Mena (at Karm el-Barasi), a small bathhouse was discovered in the 1980s by the Egyptian Antiquities Organization. A preliminary study of the architecture is the only publication available. Bibliography: Negm and Kosciúk (1990); Fournet and Redon (2017b) 443. Construction: (B) The pottery seems to indicate an early 4th c. date, although the context of these finds is not communicated.869 The plan, however, points to a Ptolemaic date. Continued use and abandonment: Unknown. Plan: The baths fit within a rectangle of 10 × 11 m. They were built out of sun-dried mud bricks resting on a base of local limestone blocks. The parts of the construction that came into direct contact with water were made out of brick and lime mortar. Several layers of plaster, the outer one made of fine gypsum, coated the walls. The floors and the pools were coated with opus signinum. Upon entering the baths, one found himself in an almost square room that must have functioned as an apodyterium. A first door to the north led into a latrine, while a second gave into a corridor that continued to the bathing parts. The first of these, circular in plan, had a single-person ‘sitz-bath’ in a recess. The second room had a large bath tub in its southern wall (ca. 1.5 × 0.7 m). The small size of these baths, especially of the pools, incited the excavators to interpret this structure as private baths. “In the vicinity, the villa to which the bath building belongs must somewhere exist;”870 however, the bath was clearly not integrated into a house and no traces of nearby occupation have been found. The use of a single-person bath tub is not restricted to private baths (see above in Karanis, C85). The size of the latrine could also point to a public use. The influence of the traditional Greek baths is undeniable: the round ‘tholos’ with a Greek-style sitz-bath. The absence of a Roman-style heating system is surprising. The late antique construction date should probably 868  Fournet et al. (2017) 501–502 with anterior bibliography. 869   Negm and Kosciúk (1990) 445. See the reservations by Rodziewicz (2009) 196, n. 21. The scholar proposes a 1st c. BC or 1st c. AD date, based on the type of plan and personal observations of the pottery during a visit to the excavations. Fournet and Redon (2017b) 443 also consider these baths to belong to the Late Ptolemaic period. 870  Negm and Kosciúk (1990) 445.

422

Gazetteer

be reconsidered. The resemblance to the baths of Asafrah, tentatively dated to the 2nd c. BC, further supports a Hellenistic or Early Roman date.871 However, we should note that a similar plan was found in the bathhouse of the monastery at Bawit (see below). Decoration: The decoration consisted of sober stuccomouldings. Technology: There was no trace of a hypocaust system or any form of wall heating. To the west of the bathing rooms, two service rooms, in no way connected to the baths, had a separate entrance. A furnace was dug into the bedrock and was mounted by a metal boiler, the imprint of which was still visible at the moment of excavation. A water reservoir was constructed at the end of the corridor and connected to an oval reservoir in the round room by means of an overflow-outlet. A separate reservoir embedded in the south-east wall of the round room supplied the sitz-bath. There was no sewer in the baths. The outlet of the tub gave directly onto the floor of the room and was probably led to the latrine.

a single-aisle chapel (to the north of the baths). The bath consisted of a dressing room, a round room with three niches for bath tubs and a third room, possibly a caldarium. The link between this bathhouse and the Apollo monastery is unknown. The use of a khurus partition wall to hide the liturgical actions going on in the sanctuary dates this complex to the second half of the 7th c.876 It is clear that these simple bath rooms, which cannot be connected with any certainty to the actual monks, have little to do with contemporary public bathhouses found elsewhere in Egypt. About 1 km east of the theatre of Pelusium (Tell el-Farama), a monastery of the martyr Epimachus was built on a small mound (Tell el-Makhzan). The oldest church of the complex is dated to the first half of the 4th c., while the large basilica dates from the 6th. The site was occupied until at least the 8th or even 9th c.877 A small bathhouse, comprising a cold room with mosaic floor, a tepid room and a hot room, has been identified.878

Ecclesiastical Baths

Abu Sha’ar Excavation history: The fortress of Abu Sha’ar (fig. 131), situated on the Red Sea coast about 25 km north of modern Hurghada, was partially investigated by the University of Delaware in the 1990s. A bathhouse built just outside the north wall of the fortress was partially excavated in 1993. Bibliography: Sidebotham (1993); Fournet et al. (2017) 458. Construction: (B) The fortress was constructed between AD 309 and 311. A monumental inscription commemorated its inauguration.879 The baths were probably built slightly later than the fortress. Test pits revealed that the site was uninhabited before the camp was built, giving a terminus post quem for the bathhouse. Continued use and abandonment: (B) The army probably withdrew at the end of the 4th or the beginning of the 5th c. according to the study of the debris layers inside the camp. A layer containing pottery and coins of the late 4th or early 5th c. was found obstructing the furnace. It seems that at least the caldarium fell out of use at this time. And yet, evidence for a late occupation of the site has been discovered. Blocked doors, gullies cut in the floor and several hearths point to late housing.880 One of the aims of the excavations was to determine if the baths had been reused in an ecclesiastical context, serving as a baptistery for the

A special category of baths found in Egypt concerns bathhouses that were built near monasteries or pilgrimage centres. As we have seen in the literary attestations (see chapter 2, e.g. Cassiodorus), bathing with women or ‘bathing for pleasure’ was prohibited for monks. Only the therapeutic or medical use of baths was allowed. However, bathhouses were found in and around pilgrimage centres, for the pilgrims and possibly the inhabitants of the surrounding villages, e.g. Abu Mena or Tell el-Makhzan.872 The evidence for monks visiting public baths is limited.873 Only in cases of disease may permission have been granted by the monastery. The archaeological evidence does not reveal classic Roman-style bath buildings inside monasteries. In the monastery of Apu Shenute near modern Suhag, what appears to be a small sitz-bath was discovered in the socalled ‘building with cruciform pillars’.874 It was located in an almost square room on the eastern side of the building. The tub itself seems to have been a reused industrial vat.875 It is unknown who used this small bathroom. The site was inhabited in the 5th and 6th c. A small bathhouse was also found on the monastery site of Bawit (fig. 130). It strongly resembled the ‘private baths’ of Abu Mena (see above), but was part of a larger structure, including 871  Riad (1975) 117. 872  Gatier (2009) 280. 873   Mossakowska-Gaubert (2009) 297–301. 874  Grossman (2009) 291–92. 875  Grossmann et al. (2009) 182.

Fortress Baths

876  Grossmann (2009) 292. 877  Bonnet et al. (2004) 49. 878  Bonnet et al. (2004) 54. 879  Sidebotham (1993) 272. 880  Sidebotham (1993) 271.

423

Gazetteer

figure 130 Plan of the chapel (left) and baths (right) in the monastery of Bawit (Egypt) after Grossmann (2009) 295, fig. 3

church inside the fort. Unfortunately, no definite conclusions could be made. Plan: (fig. 132) Only part of the baths, probably the caldarium, was excavated. Two pools were identified on the north and north-east side. The walls were constructed with cobbled stones on the outside and fired bricks on the inside. Decoration: Some fragments of window panes were found in the destruction layer. The walls were coated with a layer of plaster. Technology: A hypocaust floor was constructed with white gypsum tiles resting on parallel brick-built walls. Several rectangular tubuli were found lining the walls. The room was heated by a single furnace piercing the east wall. No aqueducts to the baths have been discovered; however, a water installation was found 1 km west of the fortress. Nag el-Hagar Excavation history: The Roman fortress of Nag el-Hagar (fig. 133), located 30 km north of modern Aswan, was excavated by the Egyptian Antiquities Organization in the late 1980s. Bibliography: Wareth and Zignani (1992); Fournet et al. (2017) 499. Construction: (D) The layout of the fortress, as well as the architecture and archaeological finds, especially a coin treasure found in one of the towers, point to an occupation during the 4th c. Within the walls, a bathhouse was found adjacent to the supposed palatium. Its construction date is linked to the construction of the fortress.881 881  Wareth and Zignani (1992) 206.

Continued use and abandonment: Large parts of the baths have been destroyed due to the spoliation of the building materials. Most of the walls of the heated section seem to have been dismantled to salvage the fired bricks. Plan: (fig. 134) The baths probably had an independent entrance from the Via praetoria or Via principalis.882 An oblong room with an apsidal ending is supposed to be the frigidarium. A suite of three rooms ended with at least one warm room, heated by a single furnace. The other rooms, which probably reached as far as the Via principalis, are too damaged to identify. One massive granite bath tub with festoon decoration was found in the north-east corner of the building. A similar bath tub with carved handles was found outside the walls.883 The military fortress was not an isolated stronghold in an otherwise deserted landscape. Traces of a village were found to the north, while a church and a small harbour were found just outside the walls. South-west of the baths, in an intra-mural area that seems to have been used as a production site, a church with an adjacent baptistery was built. The architecture and decoration resemble the church in the Temple of Luxor, dated to the 5th or 6th c.884 The large ceramic refuse dump found outside the walls also reveals that the fortress served as a regional production centre. It is possible that the baths were accessible for the inhabitants of the surrounding village. Whether or not the baths still functioned when the church was built is unknown. Decoration: Unknown. Technology: No traces of the hypocaust or of wall heating have been found due to the spoliation of building material after the abandonment of the baths. A sewer ran under the oblong room towards the south, probably collecting the waste water of the pools. The water supply must have been regulated by the castellum aquae south of the baths, which connected to a well with saqiyah in the south-western corner of the fortress. Pelusium (Tell el-Farama) Excavation history: In the 1980s, the Egyptian Antiquities Organisation started the excavations at the site of ancient Pelusium, an important port city at the eastern fringe of the Nile Delta, where Pompey was murdered. On the site, which was heavily damaged by the Israeli army, a small bathhouse was discovered, lying 90 m north of a late antique fortress. Since the initial Egyptian excavations, a Polish team has uncovered several other buildings in the urban fabric, including a theatre and two churches. The city remained an important town until at least the 11th c.885 882  Wareth and Zignani (1992) 200. 883  Wareth and Zignani (1992) 200, n. 56. 884  Wareth and Zignani (1992) 202. 885   Jansen-Winkeln (2006b) §1.

424

figure 131

Gazetteer

Plan of the late antique fortress of Abu Sha’ar (Egypt) with a bathhouse outside its walls (5) in Sidebotham (1994) general plan; with permission

Gazetteer

425

figure 132 Plan of the Fortress Baths of Abu Sha’ar (Egypt) after Sidebotham (1994) fig. 20

figure 133 Plan of the late antique fortress at Nag el-Hagar (Egypt) in Wareth and Zignani (1992) 189, fig. 1; with permission

426

Gazetteer

figure 134 Plan of the Fortress Baths of Nag el-Hagar (Egypt) after Wareth and Zignani (1992) pl. 22

Bibliography: Abd el-Maqsoud (1984–1985); Fournet et al. (2017) 503. Construction: (D) The bronze coins, glass and ceramics that were found during the excavations have never been studied to propose a chronology. If we assume that the baths were built at the same time as the fortress, a late 3rd or early 4th c. date could be proposed. Continued use and abandonment: The site was used as a burial ground at an unknown date. No less than 140 skeletons buried “in threads of stuff looking like jute,”886 heavily disturbed the site. Plan: The bathhouse has been briefly described as a brickbuilt building comprising 4 rooms and a reservoir. The most important room was a large hall. Decoration: The large hall had a polychrome mosaic floor, in which a Greek inscription read ‘Good luck to the builder’.887 Technology: A brick water channel ran under the tiled floors of the rooms. The reservoir was square (s: 9.5 m) and was divided into two parts, one of which was heated by a furnace.

Continued Use and Abandonment of Imperial Period Baths during Late Antiquity

886  Abd el-Maqsoud (1984–1985) 5. 887  Abd el-Maqsoud (1984–1985) 5 and pl. IIa.

888  See Redon (2012a) 75–76, table 2. More recently Redon (2017b). 889  Fournet et al. (2017) 510.

The archaeological evidence for public bathhouses of the High Empire is surprisingly limited in Egypt.888 Furthermore, the majority of the remains are poorly known, as excavation in the past tended to focus on necropoleis and temple complexes. The baths in important cities such as Boubastis (Tell Basta), Xois (Sakha) or Diospolis Magna (ancient Thebes, modern Luxor) have been discovered, but were never properly investigated. Therefore, it is difficult to pinpoint their construction date and even harder to reconstruct their lifespan. Bitylion? (Sheikh Zawyet) A small bathhouse near the seashore was discovered and excavated in 1913. The unpublished excavation reports have recently been examined by the Balnéorient project. The artefacts recovered during the excavations allegedly belonged to the second half of the 4th c., but the plan of the baths shows similarities with 2nd to 3rd c. examples.889 The artefacts from the excavations may point to the last phase of use. The different phases of remodelling are difficult to date, but it seems

427

Gazetteer that in the last phase of the baths, the caldarium had a protruding round alveus much like those found in the baths of Gherra (C84) or Clysma (C82). Pelusium (Tell el-Farama) A second bathhouse in Pelusium may have been erected in the 2nd c., if an inscription mentioning a ΓΥΜΝΑΣΙΑΡΧΟΣ (gymnasiarchos) can be related to a bathhouse.890 It seems that these baths were still in use in the Byzantine and even Islamic period. The latest phase of use is preliminarily placed in the Abbasid period (8th c. and later), but more research is needed to identify the exact layout of the building and its chronology.891 Thebes (Karnak), Imperial Baths (or Baths north of the first pylon) This bathhouse was recently excavated by the Supreme Council of Antiquities (2012–2014).892 The baths were in use between the 2nd and 4th c. and were altered many times during its lifespan. At least 6 construction phases have been identified, pointing to the popularity of this building during the Roman period.893 During the fourth and fifth phase, dated to the late 3rd or early 4th c., the layout underwent important changes: after an addition of two horseshoe-shaped alvei 890  Tabaie and Carrez-Maratray (2009) 224. Contrary to the interpretation of Tabaie and Carrez-Maratray, we can assume that gymnasium in a 2nd- c. context would rather refer to a bathhouse than to a Greek-style gymnasium, see also Meyer (1997) 691. 891  Tabaie and Carrez-Maratray (2009) 223. 892  Boraik and el-Masekh (2012); Boraik (2013); Masekh et al. (2017). 893  Masekh et al. (2017) 221, 231–41.

and an apse with a labrum in the heated rooms, a number of single-person pools were added to the warm section, without sacrificing the existing larger alvei.894 The hypocaust systems were also restored and the boilers were refurbished. The double ring type was transformed into a double row type (with reversing itinerary). Trimithis (Amheida) The bathhouse at Trimithis has a complex building history. The excavations, started in 2007, are ongoing. The bath was first built in the 1st or 2nd c., but fell out of use at the end of the 3rd and was used as a dump. Shortly after AD 360, the baths were again in use, with two phases of restoration that can be identified. By the end of the 4th c., however, the baths were abandoned again, possibly during a phase of restoration.895 It seems that the entire city was abandoned at this point. During the 4th c. phase, the heated rooms were equipped with a small single-person pool (southern room, the original sudatorium) and a round pool (northern room, caldarium). The most important addition during this phase was the mudbrick central hall. It acted as a multipurpose cold room and is reminiscent of other late antique baths such as those in Marea and Abu Mena. The imprints of wooden beams, planks and panels (some apparently decorated with iron studs), were found here.896 The presence of a cold section made of mudbrick and wood is not exceptional, as it was also found in the baths of Karanis (C85) and Sersena (C94).

894  Masekh et al. (2017) 236–40. 895  Davoli (2017) 194–95. 896  Davoli (2017) 195.

Baths in Palestina New Construction Ashkelon (Israel), Baths on the City Tell Excavation history: From 1985 onwards, the tell of Ashkelon was excavated by the Oriental Institute of Chicago and the Semitic Museum of Harvard University. Part of a small bathhouse was discovered. Bibliography: Stager and Esse (1986); Hoss (2005) 125–26. Construction: (C) Construction techniques point to the 4th c. Continued use and abandonment: (C) An opus sectile floor was dated to the 5th or 6th c.897 Plan: The remains of a large exedra (diam.: 8 m), a hypocaust floor and a cold-water pool were found near the slope of the tell. To the west, a nearly square pool (2.8 × 2.5 m; 1.8 m deep) was coated with plaster and had three steps to facilitate descent. Decoration: On the western exterior side of the pool, a text was painted reading ‘ἐισελτήρ ἀπολαύσον και […’ (‘Enter, enjoy and …’). An opus sectile floor was also found. Technology: There is no information on the type of hypocaust or wall heating.

C98. Beth Yerah (Israel)

Excavation history: The baths in Beth Yerah were excavated by the Hebrew University of Jerusalem in collaboration with the Israel Department of Antiquities between 1944 and 1946. Bibliography: Maisler et al. (1952); Hoss (2005) 132. Construction: (B) The bath was built on top of a Roman fortress which fell out of use in the late 3rd c. Excavations yielded late antique pottery and a fragmentary Greek inscription dated to the 6th c.898 Continued use and abandonment: (B) The debris layer in the baths contained material dating to the late 6th or early 7th c. The shell of the building was reused during the Islamic period. Plan: The bathers entered an almost square hall (11 × 10.5 m) by entrances to the west (1.5 m wide) and to the south-west (1.1 m wide). In the centre of this hall, there was a round pool (diam.: 2 m; 0.75 m deep) surrounded by pillars forming a square and probably supporting a dome. Along the western side of the pool, a low partition wall probably protected the bathers from the draught entering the doorway and from the glances of the curious passer-by.899 Along the west and south-east wall, there was a bench. 897  Stager and Esse (1986) 6. 898  Maisler et al. (1952) 222–23. 899  Maisler et al. (1952) 220.

C98 Beth Yerah (Israel), (After Maisler et al. (1952) 219, fig. 1)

The hall thus functioned as both frigidarium and apodyterium. In the south-east wall, two apsidal pools enabled the bathers to perform personal ablutions. The articulation of the heated rooms is unclear. The small rectangular room with lateral recesses was probably the original caldarium, as it was located near the only furnace. The recesses may have contained small alvei. The oblong south room, with an east-west orientation, may have been divided into two rooms, judging from the reinforced piers at hypocaust level in the centre of the room. A separate furnace, blocked up in a later phase, used to heat this room from the south-west. In the south-west corner, a brick platform may have supported a pool. The original entrance to the heated sector could not be identified. It may have been located in the south-west corner and was then blocked up by one of the apsidal pools of the hall. Decoration: The hall was paved with marble slabs, while the dome above the round pool was embellished with glass and gilt tesserae. The round pool itself was also adorned with marble slabs. The floors of the heated rooms were paved with marble, but there is no evidence for wall decoration. Technology: The hypocaust pillars were made out of stone monoliths or square ceramic tiles and were coated with a waterproof mortar. Square recesses in the walls point to the location of the chimneys. The round ceramic chimney pipes were still in situ when discovered. The walls were heated by tubuli. The water supply was ensured by a terracotta conduit coming from the south-west and splitting

429

Gazetteer up to feed the round pool and apsidal tubs. It was supplied by a branch of the aqueduct heading towards Tiberias.900 A sewer ran from the round pool underneath the south wall of the hall and along the west side of the oblong heated room towards an unknown destination. Construction technique: Opus quadratum, opus caementicium Bathhouse surface: ca. 320 m2 GPS coordinates: Exact coordinates unknown

C99. Caesarea Maritima (Har Qesari, Israel), Area 1 Baths

Excavation history: The Phoenician portal town of Stratonos Pyrgos was transformed by Herod the Great into an important Roman city in Judaea. The city, now known as Caesarea Maritima, became the seat of the Roman procurator from AD 6 onwards. During the early 1990s, a large public bathhouse was discovered by the Israel Antiquities Authority in the southern part of an insula. Bibliography: Porath et al. (1998); Hoss (2005) 134–35; Piraud-Fournet (2014). Construction: (D) Its construction was dated to the 4th c.901 According to some scholars, the baths were part of a large private residence dating to the 6th c.902 Continued use and abandonment: (D) The baths seem to have fallen out of use shortly before the Arab conquest.903

Plan: The internal organisation of the baths within the insula is not quite clear. From the cardo east of the insula, a door led into a succession of three vestibules that may already have been part of the bathhouse. A T-shaped hall with an apsidal ending to the south may have served as an apodyterium. A narrow passage to the west led into the frigidarium. It had at least one small semicircular piscina and benches along its north and west walls. To the south, there was a small latrine that was flushed with the waste waters of the pools.904 The heated rooms, at least 5 of which had a hypocaust, were located to the west of the cold room. To the north of the bathing rooms, a large paved courtyard (11.5 × 13 m) surrounded by columns has sometimes been interpreted as a palaestra.905 Its western end lay at a higher level, creating a sort of terrace overlooking the courtyard. Alternatively, scholars considered this to be the courtyard of a sumptuous residence.906 This would fit with the dichotomy between the northern (residential) and southern (bathing) part of the insula. Decoration: A mosaic with the depiction of a woman had a Greek inscription reading ‘beautiful woman’. The different parts of the baths were decorated with plain white mosaics and white marble slabs. Fragments of painted wall plaster and wall mosaics were also found during the excavations. Technology: The heated section was divided into different sections, perhaps with different rooms for the male and

C99 Caesarea Maritima (Sdot Yam, Israel), Area 1 Baths (After Porath et al. (1998) fig. 4) 900  Hestrin (1993) 259. 901  Porath et al. (1998) 42–43 without further specifications. 902  Patrich (1999) 106; Piraud-Fournet (2014) 695. 903  Porath et al. (1998) 42–43 without further specifications.

904  Indicated on Porath et al. (1998) 42, fig. 4. 905  Porath et al. (1998) 43. 906   Piraud-Fournet (2014) 708, fig. 16.

430

Gazetteer

female bathers. The water supply channels were found in situ, coming from the branch of an aqueduct running underneath the cardo. Construction technique: Opus quadratum Bathhouse surface: ca. 1,000 m2 GPS coordinates: 32°29’56.48”N, 34°53’28.34”E

C100. Caesarea Maritima (Har Qesari Israel), Villa Suburbana Baths

Excavation history: About 1 km north of Caesarea’s city wall, the remains of a small bathhouse were found in 1964 by an Italian archaeological mission. The building was re-examined between 1976 and 1978 by the Joint Expedition to Caesarea. Bibliography: Amore et al. (1966) 294–304; Holum et al. (1988); Horton (1996); Hoss (2005) 135. Construction: (A) The re-examination of the building in the 1970s produced stratigraphic evidence pointing to a construction phase around AD 550.907 Continued use and abandonment: (A) The debris layer that was linked to the destruction of the baths gives a terminus ante quem of AD 640 for the abandonment of the building.908 Plan: The small size of the building led archaeologists to believe that these were the baths of an undiscovered suburban villa.909 However, there is no reason to discard the possibility of a small privately owned public bath.910 The bathers probably accessed the building from the south, passing a paved courtyard into the frigidarium. A small basin in the south-east corner may have served as a pediluvium.911 The large round pool in the centre had a bench running along its sides. Ceramic pots lodged in the sides of this pool and the presence of fish bones in the drains suggest that the pool was (also) used as a fish pond.912 West of the frigidarium lay the presumed apodyterium, with a bench against its eastern wall. To the north of the cold room, a square heat lock had a small water basin, interpreted as a labrum. East of this room was another room with a bench, perhaps an unctorium.913 Continuing north, the bathers entered a first heated room, before entering the caldarium to the west. The latter had two apsidal single-person alvei. Decoration: The bathhouse was luxuriously decorated, with opus sectile floors in the apodyterium, the room east of the frigidarium, the heat lock and the presumed unctorium. In the first heated room, fragments of painted wall plaster were found. Fragment depicted a ‘tree of life’ and 907  Holum et al. (1988) 182; Horton (1996) 184–87. 908  Holum et al. (1988) 182; Horton (1996) 184–87. 909  Holum et al. (1988) 182. 910  Horton (1996) 188–89. 911  Horton (1996) 179. 912  Holum et al. (1988) 182; Horton (1996) 177, esp. n. 6. 913  Horton (1996) 179.

C100 Caesarea Maritima (Sdot Yam, Israel), Villa Suburbana Baths (After Horton (1996) 178, fig. 1)

a jewelled cross.914 Most of the marble slabs seem to have been broken out during the Early Islamic period. Technology: The hypocaust consisted of standard square pillars made out of terracotta tiles. There was only one furnace, west of the caldarium. The draught for the hot air was ensured by flues, lodged in recesses in the caldarium walls.915 This probably meant that the second room, in which tubuli were found, was only mildly heated. The water for the baths was supplied by an aqueduct running 300 m east of the baths. The water was stored in a large reservoir to the north-east of the baths and distributed to the different pools by terracotta conduits underneath the floors. The small basin in the heat lock had its own reservoir. The drains of the round pool and of the southern alveus flushed a latrine, located just north of the apodyterium. Construction technique: Opus quadratum Bathhouse surface: ca. 525 m2 GPS coordinates: 32°30’44.35”N, 34°53’56.22”E

C101. Dharih ( Jordan), Late Baths

Excavation history: The Late Baths at Dharih were discovered and excavated in 2004 by the French-Jordanian archaeological mission. Bibliography: Sartori (2015). Construction: (D) The baths seem contemporary with the 5th or 6th c. occupation of the site. Continued use and abandonment: (A) Samples of charcoal found in the furnace of the baths were radiocarbon dated

914  Horton (1996) 179. 915  Furnaces and flues are indicated on the plan provided by Horton (1996) 180, fig. 1.

431

Gazetteer

C101 Dharih (Jordan), Byzantine Baths (After Sartori (2015) 126, fig. 2)

to between AD 540 and AD 640, the last time the furnace was lit. Plan: The site of Dharih was a regional sanctuary during the High Empire. After the earthquake of AD 363, the site fell into decline, only to be restored during the 5th or 6th c., when a Christian community transformed the temple into a church. Late antique occupation of the site was limited to the fortified sanctuary. In the south-east corner of the sanctuary, the small bathhouse (ca. 40 m2) was discovered. The building was constructed out of local limestone blocks and consisted only of three spaces. The first seems to have been unroofed. It only had a small basin, hewn out in a column base, lodged in a recess in the south-west wall. By a small doorway, one entered into the first heated room (tepidarium). This room was heated indirectly by the furnace of the second room. A low and narrow bench made out of limestone blocks ran along the walls. In the southeast and north-west corners of this bench, a shallow circular hollow may have been used to hold jugs for dousing.916 A narrow doorway, which could be closed off, led into the 916  Sartori (2015) 128.

second heated room (caldarium). A similar low bench with a hollow in the north-east corner ran along the walls. Against the south wall stood two rectangular single-person pools (west: 1.25 × 0.8 × 0.57 m and east: 1.1 × 0.8 × 0.57 m). Decoration: The floors of the rooms were paved with flagstones of local limestone. The cold section reused the floor level of the sanctuary. No evidence for wall decoration has come to light. Several architectural elements (column drums) of the Roman sanctuary were reused as building material in the walls of the baths. Technology: A single furnace heated the caldarium and tepidarium. The hypocaust consisted of three arched tunnels, made of tiles and supporting opus caementicium, running north-south underneath both rooms. Tubuli, circular and oval, were only found against the east, south and west sides of the baths. Three chimney flues in the caldarium and two more in the tepidarium ensured a draught of hot gasses in the hypocaust. The fact that the furnace lay directly underneath the west alveus probably meant that the water in this pool was warmer. The alvei were drained by two orifices giving directly onto the caldarium floor. A hole in the west wall of the tepidarium probably channelled waste water out of the building. In the south wall of the caldarium, a small basin acted as a distribution point for supplying water to the two pools. The actual pipes have not been found. The basin in the cold section was probably filled manually. Construction technique: Opus quadratum (spolia) Bathhouse surface: ca. 40 m2 GPS coordinates: 30°54’26”N, 35°42’10”E Diocaesarea / Sepphoris (Israel), Western Bathhouse Excavation history: The settlement of Sepphoris was located on a hill and boasted an orthogonal street grid and a 2nd-c. colonnaded street. In the 5th c., there were few fundamental changes to the urban fabric. A church with a baptistery and a new bathhouse were constructed in the western part of town.917 The site was excavated between 1985 and 1990 by Duke University. Bibliography: Weiss and Netzer (1995); Hoss (2005) 172. Construction: (D) The baths were dated to the end of the 3rd or the early 4th c. Continued use and abandonment: (D) The baths may have been in use until the Early Islamic period.918 Plan: The Western Baths were probably accessible from a cardo running south of the insula. Two rooms were heated by a hypocaust: one octagonal to the north-west and one rectangular to the south-west. The octagonal room was later equipped with a semicircular pool, divided into two equally large quadrants by a small partition wall. The layout of the cold section is unclear: “Groups of pools of unclear function were uncovered next to the hot rooms, as 917  Weiss and Netzer (1995) 41–42. 918  Weiss and Netzer (1995) 43 without any further evidence.

432

Gazetteer

well as additional rooms, some of them paved with partially preserved mosaics in geometric designs.”919 Structures with plastered pools, some of which interconnected, were also found south of the heated rooms. Decoration: The octagonal room was paved with plain white mosaics. Technology: Water supply pipes, frequently restored, were found south of the two heated rooms. El Burj (Israel) Excavation history: The construction of a highway from Shomeriyya to Moshav Sheqef in 2006 enabled a salvage excavation to discover the remains of a Roman-style bathhouse. Bibliography: Peretz and Nahshoni (2009). Construction: (D) The bath was dated to the late antique period.920 Late antique industrial facilities and reservoirs were also found, pointing to a small settlement making use of the baths. Continued use and abandonment: (D) The baths fell out of use at the end of the Byzantine period.921 Plan: Two rooms on a hypocaust and a pool were identified. Decoration: Unknown. Technology: Fragments of hypocaust pillars, terracotta pipes and hydraulic mortar were found throughout the site. Water was supplied by a saqiyah found in the vicinity of the pool. Elusa (Israel) Excavation history: During a survey campaign in the Negev desert, 20 km south-west of Be’er Sheva, the remains of a bathhouse were discovered. Bibliography: Negev (1993); Hoss (2005) 138. Construction: (D) Surface finds suggest that the baths were in use during the Byzantine period.922 Continued use and abandonment: Unknown. Plan: Unknown. Decoration: Unknown. Technology: The water was supplied by the nearby wadi, filling a reservoir connected to the baths by means of lead and terracotta pipes.

C102. Gadara (Umm-Qeis, Jordan), Byzantine Baths

Excavation history: The baths were excavated between 1978 and 1983 by a Danish-German team on a steep slope of the acropolis of the city, near the theatre.923

919  Weiss and Netzer (1995) 43. 920  Peretz and Nahshoni (2009) §3 without further evidence. 921  Peretz and Nahshoni (2009) §3 without further evidence. 922  Negev (1993) 380. 923   Holm-Nielsen (1993) 1–10.

C102 Gadara (Umm Qais, Jordan), City Centre Baths (After Nielsen et al. (1993) Plan II and III)

Bibliography: Holm-Nielsen et al. (1986); Nielsen et al. (1993); Hoss (2005) 141–42. Construction: (A) The construction phase was placed in the early 4th c., judging from the pottery recovered from the stratigraphy.924 Continued use and abandonment: (B) After an earthquake, probably in AD 447, the baths were restored with spolia. They may have been used until the Early Islamic period, but were later transformed into a house with a mirhab and a stable. Another earthquake in AD 746 probably ended all occupation of the site.925 Plan: The bathhouse covered an entire insula and was partially built on an artificial terrace. Due to the later reuse of the building, the original entrances are difficult to identify. These should probably be reconstructed in the north-east or north-west sides.926 The bathers started their route in the cold section, where the oblong frigidarium may have been preceded by an apodyterium. Two small semicircular piscinae, reduced in size at a later date, were identified in the south wall. A first rectangular heated room was flanked on its west side by two smaller rooms, both with their own 924  Nielsen (1993d) 132. 925   Holm-Nielsen et al. (1986) 229; Nielsen (1993d) 145. 926   Holm-Nielsen et al. (1986) 220.

433

Gazetteer furnaces. The southernmost had a small semicircular basin and could be interpreted as a sudatorium.927 Returning to the central north-south axis, the next rectangular room (14.5 × 9 m) had a rectangular alveus on its east end. On the west side, a square room with its own furnace may also have been a sudatorium.928 The almost square room (9.5 × 10.4 m) at the end of the bathing itinerary must be interpreted as the caldarium. Its south half was taken in by a large semicircular alveus. A doorway in its west side connected to a heated room with two rectangular alvei. The service corridors lay along the east and west sides of the main bathing rooms. In the second phase of the baths, possibly after AD 447, the original caldarium was abandoned and transformed into a praefurnium for the middle room, which was now the new caldarium. In the doorway between the ‘old’ and the ‘new’ caldarium, an apsidal alveus was constructed. The heated room in the south- west corner of the building lost its alvei, getting a heated recess instead. In the Early Umayyad period, new water pipes were laid within the bathing rooms, presumably to supply the dwelling with water.929 When the baths fell out of use, the rooms were transformed into living units. A mirhab seems to have been created in the small north-west room.930 In the last phase of the building, several rooms were used as stables or storerooms. Decoration: The decoration of these baths suffered from the subsequent phases of reuse as living units. The floor of the frigidarium and of the original tepidarium was paved with large slabs of limestone. The caldarium was the most luxurious room, with a large grey granite column in each corner. The capitals were of the Corinthian order, in a style that was common in the 2nd c. AD, meaning the columns were reused from an older building.931 Several fragments of marble, slabs and architectural elements, as well as window panes have been found in the destruction layers of the baths and can be attributed to the original building phase. Mortar fragments with imprints of wooden window frames have also been found. Technology: The reuse of the baths as a house caused important interventions to the original infrastructure. The hypocaust floors were broken up and pools may have been destroyed. The pillars of the hypocausts were made out of basalt monoliths or square terracotta tiles. Only the first and second rectangular heated rooms had wall heating, while the caldarium had only flues in the corners acting as chimneys. The tubuli were square in cross-section and box-like in shape. The high number of furnaces ensured 927   Holm-Nielsen et al. (1986) 225. 928   Holm-Nielsen et al. (1986) 226; Nielsen (1993e) 99. 929  Nielsen (1993e) 86. 930  Nielsen (1993d) 143. 931  Gorm Andersen (1993) 162.

that the small rooms on the western side may have acted as sudatoria or unctoria and districtaria.932 It is possible that the alvei had a testudo.933 A lot of olive pits were recovered from the ashes found in the furnaces, meaning that the baths were heated—at least in their last phase of use—with garbage and waste of agricultural production. The water was presumably supplied by an aqueduct, as the total volume of water contained in the pools must have been quite high. Only a small reservoir was found to the north-east of the original caldarium. Construction technique: Opus quadratum Bathhouse surface: ca. 2,340 m2 GPS coordinates: 32°39’23.71”N, 35°40’36.29”E

C103. Gadara (Umm-Qeis, Jordan), Herakleides Baths

Excavation history: In 1960, part of a bathhouse was discovered in the ancient Gadara. The Deutsche Palästine Verein excavated the site. Bibliography: Lux (1966); Hoss (2005) 142. Construction: (C) The mosaics were dated between the 4th and 6th c.934

C103 Gadara (Umm Qais, Jordan), Herakleides Baths (After Lux (1966) plan 1)

932  Nielsen (1993f) 123. 933  Nielsen (1993f) 124. 934  Lux (1966) 69–70.

434

Gazetteer

Continued use and abandonment: Unknown. Plan: The remains comprised three rooms, none of them heated. The building was interpreted as a bathhouse due to the mosaic inscription wishing good health to the commissioner and to the bathers.935 The only room that was fully excavated was rectangular in plan (20.9 × 6.5 / 6.6 m). A second room to the east was partially destroyed when a pool was constructed north of it. To the south, three stairs led to an unexcavated room. Decoration: A mosaic wished good health to the commissioner and to the bathers. The floors of the rectangular room and of the room east of it were embellished with polychrome geometric mosaics. The walls of the three rooms were adorned with marble slabs. Technology: Narrow water channels were found in the floor of the rectangular room. Construction technique: Opus quadratum Bathhouse surface: Unknown GPS coordinates: Exact coordinates unknown

C104. Ha’on (Israel)

Excavation history: Excavations by the Israel Antiquities Authority prior to the construction of a holiday village in Ha’on in the summer of 2009 discovered the remains of a late Byzantine bathhouse. Bibliography: Cinamon (2013). Construction: (A) The pottery found in the foundation trenches suggest a construction date between the end of the 6th and the beginning of the 7th c. Continued use and abandonment: (A) The filling of the pool was dated to the second half of the 6th c.936 The baths may have remained in use in the Early Umayyad period until they were transformed into living units at the end of the 7th c. Plan: Only two rooms of the baths were identified. The southernmost room had a hypocaust and tubuli for wall heating. The northern room was lined with a waterproof mortar and had plastered benches along its walls. It probably functioned as a pool.937 In a later phase, the pool was filled up and a new floor level was created. Near the bathhouse, a church with an adjacent monastery has been found. It is thought that the baths might have been part of the monastery or at least belonged to the church.938 Decoration: No traces of the decorative scheme were identified. Technology: The hypocaust pillars were made with square and round bricks. The walls were heated by tubuli. 935  Mittmann (1966) 71. 936  Cinamon (2013) §6, fig. 11. 937  Cinamon (2013) §6. 938  Cinamon (2013) §8.

C104 Ha’on (Israel), (After Cinamon (2013) fig. 6)

Construction technique: Ashlar Bathhouse surface: Unknown (not exceeding 500 m2) GPS coordinates: Exact coordinates unknown Horbat Zikhrin (Israel), Area C Baths Excavation history: The site of Horbat Zikhrin yielded the remains of three Byzantine bathhouses, two of which were public. Large parts of the site were excavated in the late 1980s by the Israel Antiquities Authority in collaboration with the University of Tel Aviv. Bibliography: Fischer (1986); Fischer (1994); Hoss (2005) 176. Construction: (C) The entire village went through a period of wealth and expansion between the 4th and 7th c. The baths have been assigned to Late Antiquity on the basis of the ceramic assemblages found during the excavations and on the basis of the Byzantine construction techniques.939 Continued use and abandonment: Unknown. Plan: The baths in area C were located next to a taberna and had two heated rooms. The presumed caldarium had apsidal and rectangular niches in its walls. The praefurnium lay to the north, while a house in area C may also have had a bath, although only one heated room (with hypocaust and tubuli) was found. Decoration: Unknown.

939  Fischer (1986) 271, 273.

435

Gazetteer GPS coordinates: Exact coordinates unknown

C106. Jerusalem / Al-Quds (Israel / Palestine), Baths under the Notre Dame monastery

C105 Horbat Zikhrin (Israel), Area F Baths (After Fischer (1994) plan)

Technology: The hypocausts consisted of brick-built pillars. The walls were heated by rectangular tubuli. The water was stored in three cisterns. In total, some 53 cisterns were found in Zikhrin, 24 of which can be dated to the Byzantine period.940

Excavation history: Salvage excavations on the ground of the Notre Dame Monastery in the 1980s by the Israel Antiquities Authority yielded the remains of a small bathhouse. Bibliography: Chambon (1989); Hoss (2005) 156. Construction: (D) The baths were possibly late antique, judging from the construction techniques.942 Continued use and abandonment: Unknown. Plan: Only three rooms could be identified, laid out on a north-south axis. The cold room was located on the north end and was not entirely excavated. To the south, two rooms on a hypocaust may be identified as the tepidarium and the caldarium. A small basin was located against the exterior eastern wall. The niche in the north-west corner of caldarium may have lodged a basin. Decoration: The caldarium had a floor paved with marble slabs.

C105. Horbat Zikhrin (Israel), Area F Baths

Excavation history: The site of Horbat Zikhrin yielded the remains of three Byzantine bathhouses, two of which were public. Large parts of the site were excavated in the late 1980s by the Israel Antiquities Authority in collaboration with the University of Tel Aviv. Bibliography: Fischer (1986); Fischer (1994); Hoss (2005) 177. Construction: (C) The entire village went through a period of wealth and expansion between the 4th and 7th c. The baths have been assigned to Late Antiquity on the basis of the ceramic assemblages found during the excavations and on the basis of the Byzantine construction techniques.941 Continued use and abandonment: Unknown. Plan: The baths lay in an administrative quarter, close to the church, houses and workshops. It had an apodyterium, a frigidarium with a plastered piscina (1 × 1 m; 0.6 m deep), a tepidarium (8 × 4 m) and a cruciform caldarium. Decoration: The frigidarium had a mosaic floor. Technology: The hypocaust of the caldarium had pilae made out of bricks and stone monoliths, rising to an exceptional 1.5 m. Some of the cylindrical tubuli were still in situ. The water supply of these baths was ensured by a large cistern, hewn out of the bedrock. Construction technique: Unknown Bathhouse surface: ca. 150 m2 940  Fischer (1994) 45. 941  Fischer (1986) 271, 273.

C106 Jerusalem / Aelia Capitolina (Jerusalem / Al-Quds, Israel / Palestine), Notre Dame Monastery Baths (After Chambon (1989) 146, fig. 132)

942  Chambon (1989) 146.

436 Technology: The basin was drained by a channel heading north-east. A sewage channel also ran along the western side of the baths, although the actual connection with the baths could not be determined. The hypocaust was partially hewn out of the bedrock and had pillars (ca. 1 m high) made out of terracotta tiles. The pillars of the so-called tepidarium were reinforced by double arches. Construction technique: Unknown Bathhouse surface: Unknown (at least 50 m2, not exceeding 500 m2) GPS coordinates: 31°46’48.38”N, 35°13’32.89”E Jerusalem (Israel / Palestine), Baths in the Jewish Quarter (Area C) Excavation history: Excavations in the Plugat haKotel Street in 1970 revealed the remains of a bathhouse (5.6 × 17 m). Bibliography: Avigad (1970); Hoss (2005) 155. Construction: (B) The baths were built on layers with Roman material and are dated in the late antique period.943 Continued use and abandonment: (B) During the later phase (6th c.?), the pools were filled and covered with a plaster floor. The building was reused in the Early Islamic period.944 Plan: In the east, a large courtyard probably combined the different functions of the cold rooms (dressing room, anointing, etc.). Three rooms—oriented on a north-south axis—were heated by a hypocaust. The southernmost was identified as the caldarium. It had brick benches running along the walls and a round single-person alveus. Decoration: Little can be said about the decoration of the building. The courtyard was paved with flagstones, while the remains of marble slabs were found on the caldarium floor and on the internal sides of the alveus. Technology: The suspensura was constructed on brick pillars and the walls were heated by square tubuli. The water supply could not be identified, but a single sewage channel running underneath the courtyard was found. Jerusalem / Al-Quds (Israel / Palestine), Mamilla Baths Excavation history: In the early 1990s, a bathhouse was discovered between the Jaffa Gate and the Jaffa Road. It was excavated by the Israel Antiquities Authority. Bibliography: Reich and Shukron (1994); Hoss (2005) 155–56. Construction: (B) The only material recovered during the excavations was late antique in date.945 Continued use and abandonment: Unknown.

943  Avigad (1970) 134 without further evidence. 944  Avigad (1970) 136. 945  Reich and Shukron (1994) 95.

Gazetteer

C107 Jerusalem / Aelia Capitolina (Jerusalem / Al-Quds, Israel / Palestine), Old City Baths (After Baruch (2002) 91, fig. 116)

Plan: At least two rooms had a hypocaust, heated by a single furnace. Only the remains of a single pool could be identified. Decoration: To the north, a room was paved with a mosaic of simple geometric and floral design. Technology: A reservoir was found to the west of the baths. It may have been fed by an aqueduct found near the site.

C107. Jerusalem / Al-Quds (Israel / Palestine), Old City Baths

Excavation history: In 2000, the remains of a modest bathhouse were discovered in the old city centre of Jerusalem. The excavations were carried out by the Israel Antiquities Authority. Bibliography: Baruch (2002); Hoss (2005) 155. Construction: (A) The ceramics and coins found during in the destruction layer and in the mortar of the hypocaust floor give us a lifespan between the early and late 7th c.946 Continued use and abandonment: (A) In the Early Islamic period, a new building with a colonnade was constructed on top of the remains.947 Plan: The building was oriented on the Byzantine cardo maximus and had a direct entrance from this street. The remains of two heated rooms were discovered, each heated by at least one furnace. The southern room had a rectangular pool, while the northern room had a small singleperson pool ending in an apse. Decoration: The small south alveus was embellished with marble slabs. Technology: The hypocaust pillars were made with large terracotta tiles and ended in arches. Tubuli were found in the destruction layer. Construction technique: Ashlar Bathhouse surface: Unknown GPS coordinates: Exact coordinates unknown

946  Baruch (2002) 76. 947  Baruch (2002) 76.

437

Gazetteer Jerusalem / Al-Quds (Israel / Palestine), Ophel Baths Excavation history: North-west of Robinson’s Arch, the remains of a bathhouse were discovered during excavations in the centre of Jerusalem between 1968 and 1978. Bibliography: Avi-Yonah (1993); Hoss (2005) 157. Construction: (C) The construction technique points to a late antique date.948 Continued use and abandonment: (D) The entire Ophel quarter seems to have suffered severe damage in the Persian raid of AD 614, with most of buildings left in ruins afterwards.949 Plan: A room on a hypocaust and two small pools could be identified. Several tiles with the mark of Legio X were reused for the construction of these baths. Decoration: Unknown. Technology: The hypocaust pillars were made with standard square tiles.

Excavation history: During salvage excavations in 2009 by the Israel Antiquities Authority, the remains of a bathhouse came to light. Bibliography: Abu Raya (2013a).

Construction: (B) Ceramics and small finds—including glass kohl flasks—in and around the baths point to a Byzantine construction phase. Continued use and abandonment: (B) The ceramics found inside the sewers of the baths, date to the Early Umayyad period and reveal the last phase of use of the building.950 Plan: Two heated rooms have been identified, both with a hypocaust and tubuli. There were two square basins, one in each room. A furnace found 2 m south of the remains reveals the presence of other heated rooms. Decoration: The floor of the western room was paved with slabs of white marble. Fragments of wall paintings with geometric figures were also found. Technology: The hypocaust pillars were made of round tiles on a square brick base and ending in arches. The basin in the western room was drained by a lead fistula. Along the northern side of the bath rooms, a plastered sewer must have collected waste water from the pools. No information is available concerning the water supply. Construction technique: Ashlar Bathhouse surface: Unknown (at least 60 m2, not exceeding 500 m2) GPS coordinates: 32°52’1.27”N, 35°12’57.45”E

C108 Kabul (Israel), (After Abu Raya (2013a) fig. 2)

Khirbet Lasan (Israel) Excavation history: The site of Khirbet (or Horbat) Lasan was partially excavated by the Israel Antiquities Authority when a railroad was constructed between Ashkelon and Be’er Sheva in 2009. Among the identified remains were Byzantine olive presses, a pottery kiln, a lime kiln, a cistern and several buildings, including a bathhouse. Bibliography: Seriy (2011). Construction: (A) The stratigraphic excavation pointed to a construction phase in the late 4th c. Continued use and abandonment: (A) The stratigraphy points to a continued occupation until the 6th c. During the 6th c., the baths were reused as a pottery workshop. A lime kiln was also lodged in its north-west corner. During the Islamic period, the structure was used as a garbage pit. Plan: Inside the building (ca. 20 × 20 m), a room heated by hypocaust and a cold room with two small pools could be identified. Decoration: The floor of the heated room was paved with large white marble slabs. Technology: The hypocaust was partially dug out of the natural soil. The pillars were made out of square or round tiles. No information is available about the water supply or waste water disposal.

948   Avi-Yonah (1993) 774. 949   Avi-Yonah (1993) 774.

950  Abu Raya (2013a) §3, fig. 9.

C108. Kabul (Israel)

438

figure 135

Gazetteer

Plan of the Baths and cistern (right) at Mampsis (Israel) after Negev (1988) 168, 35; Courtesy of the Institute of Archaeology, The Hebrew University of Jerusalem; with permission

C109. Mampsis (Kurnub, Israel)

Excavation history: The baths were excavated during the late 1960s and early 1970s by the Hebrew University of Jerusalem. Bibliography: Negev (1988); Hoss (2005) 157. Construction: (B) Few contexts provide information for the dating of construction. The oldest coins found on the site date from the late 4th c.951 However, an earlier construction date cannot be ruled out.952 Continued use and abandonment: The baths were in use in the late antique period, but it is not clear when they were abandoned. The praefurnium was possibly reused as a part of a house.953 Plan: The building consisted of a courtyard or hall (6.5 × 8.1 m) that was accessible from the street. It was paved with stone slabs and had benches along its 4 sides. To the north-east, a subterranean channel could point to the presence of a latrine.954 From this hall, the bathers proceeded through a broad doorway (2.1 m) into the frigidarium. The two single-person pools, one octagonal and one semicircular (3.5 × 2.3 m), were probably added in a second phase.955 The tepidarium (2.9 × 2.7 m) to the west had benches along its 4 walls. The door towards the frigidarium could be closed. The two heated rooms used a single furnace. No pools were identified by the excavators, although the reinforcement of the hypocaust pillars in the 951  Negev (1988) 180–81. 952  Hoss (2005) 157. 953  Negev (1988) 173. 954  Negev (1988) 170. 955  Negev (1988) 171.

C109 Mampsis (Kurnub, Israel), (After Negev (1988) 175, plan 36)

eastern part of the last room could point to the presence of an alveus (fig. 135). Decoration: The decorative scheme was rather modest. The pillars that supported the roof of the hall were possibly made out of wood, as well as the roof itself. The pavement of this hall was in slabs of local stone. Inside the actual bathing section, the walls were covered with plaster. The pools had a coating of water-resistant mortar. Technology: The hypocaust was made by brick pillars forming pointed arches. Rectangular recesses in the walls lodged the flues that ensured the necessary draught. Some tubuli were found in the destruction layers of the second heated room. The walls of the hypocaust and the furnace were blackened, probably as a result of burning juniper bushes, a practise continued by the Bedouins of the

439

Gazetteer surrounding area at the time of excavations.956 The water for the pools was supplied by a reservoir east of the baths (fig. 135), but the distribution within the complex cannot be reconstructed. However, the semicircular piscina was fed directly by a separate conduit coming from the reservoir. It arrived at around 1.55 m above the bottom of the pool. The piscinae were drained by outlets giving directly onto the frigidarium floor. The waste waters were collected by a drain, running underneath the threshold of the hall, passing the presumed latrine and ending in a collector. Construction technique: Opus caementicium Bathhouse surface: ca. 100 m2 GPS coordinates: 31° 1’33.20”N, 35° 3’54.04”E

C110. Nicopolis (Emmaus or Hammat, Palestine)

Excavation history: The site of Nicopolis was known for its thermal springs. A well-preserved bathhouse was excavated during the 1970s by Tel Aviv University. Bibliography: Gichon (1979); Gichon (1993); Hoss (2005) 138–39. Construction: (B) The stratigraphy inside the building was heavily disturbed, which made dating the building difficult. In the flat parts of the original roof, 4th c. sherds were added to the mortar.957 Continued use and abandonment: The baths may still have been in use at the beginning of the Umayyad period. At an unknown date during the Early Islamic period, the baths were transformed into a warehouse and later into a shrine for Abu Ubaida, a Muslim conqueror.958 A cemetery developed around this shrine, impeding further archaeological research in the vicinity. Plan: The bathing block, built in opus quadratum of local limestone, included 4 rooms. The easternmost room probably functioned as a combined apodyterium-frigidarium. It was covered by a dome flanked by two barrel vaults (north and south). No traces of a cold pool or a labrum could be identified, although these may have been lodged in the rectangular recesses under the barrel vaults.959 To the west, a first room (3.2 × 3.5 m) might have been heated. A labrum stood in the southern apse. The second heated room had a similar layout as the previous room, but had two alvei in its apsidal endings. The westernmost room had a square plan and low ceiling. It might have been a sudatorium. Decoration: The phases of reuse had a negative impact on decoration. Traces of wall plaster were still visible in the frigidarium. An opus sectile out of marble slabs also embellished the floor. 956  Negev (1988) 173. 957  Gichon (1979) 108. 958  Gichon (1979) 109. 959  Gichon (1979) 104.

C110 Nicopolis (Emmaus, Palestine), (After Gichon (1979) 103, fig. 1)

Technology: The heating system of the first room did not survive the phase of reuse, although a furnace mouth can still be seen in the southern apse. The walls were slightly set back, probably pointing to the use of wall heating.960 The second heated room still retained its hypocaust, consisting of parallel rows of arcaded heating channels. The water for the pools was stored in a small reservoir outside the north-east corner of the building. It was undoubtedly fed by a nearby source. Three parallel aqueducts conducted water from the ‘Eqed source to the settlement. One of these aqueducts was repaired and used until the Middle Ages. Construction technique: Opus quadratum Bathhouse surface: ca. 105 m2 GPS coordinates: 31°50’22.25”N, 34°59’22.41”E

C111. Oboda (Avedat, Israel)

Excavation history: The site of Oboda, which was occupied since the 4th or 3rd c. BC, was a Nabataean stronghold in the Negev desert that came under Roman control in the first half of the 3rd c. In the Byzantine period, the small settlement seems to have focused on wine production.961 The bathhouse was first described by Alois Musil. The Hebrew University of Jerusalem carried out excavations in the 1970s and in 1989. Bibliography: Musil (1908); Negev (1997); Hoss (2005) 164–65. Construction: (D) The excavations were not able to determine the construction date of the building. However, the pottery that was found in and around the bathhouse dated to the Byzantine period.962 The layout of the baths also shows strong similarities with the late antique baths in Horbat Zikhrin (C105), Legio (Jordan) and Zikhron Ya’aqov (see below, C117).

960  Gichon (1979) 105. 961  Negev (1997) 9. 962  Negev (1997) 176.

440

C111 Oboda (Avedat, Israel), (After Negev (1997) 172, fig. 26)

Continued use and abandonment: Unknown. Plan: The bathers entered the baths by a spacious hall (9.8 × 6.3 m). Two square pillars must have supported a roof of perishable materials.963 The remains of a staircase along its eastern wall gave access to the roof, possibly to maintain the gutters in the vaults and to suspend wet towels, as is often the case in (modern) hammams. In the south-west corner of the room, a doorway gave access a square room (s: 4.1 m) divided by a north-south wall into two pools. In the south-east corner of the hall, a doorway gave access to the heated sections. The first room had a small basin (1 × 1.1 m) in its north-east corner. By a doorway in the south-east corner, the bather entered an oblong room (3.8 × 2.0 m) with stone-built benches along its walls. According to the excavators, the small recess to the north-west (ca. 0.5 × 2 m) was a urinal, its drain running underneath the threshold of the room with benches.964 A doorway in the centre of the southern wall of the latter gave access to the first room on a hypocaust. The doorway could be closed by a door to prevent a loss of heat. This room, interpreted as a laconicum by the excavators, had a fully preserved barrel vault. By a doorway to the west, the bather finally arrived in the cruciform caldarium with alvei in the northern, southern and western arms. Only the northern tub was semicircular in shape. All three were heated by the hypocaust, which formed a cross-shaped channel— or H-shaped in the case of the western tub—underneath. The northern and southern tub had a row of tubuli against their respective north and south walls. The central part of the caldarium was covered by a dome made out of ashlar, 963  Negev (1997) 173. 964  Negev (1997) 173.

Gazetteer resting on 4 squinches in the corners. In the apex of the dome was an oculus. The praefurnium of the caldarium was located on its western side. To the west of the hall, there were two small service rooms, the westernmost having a water reservoir. Decoration: All of the rooms seem to have been coated with a simple undecorated plaster.965 The piscinae of the square room were coated in a waterproof plaster. The two unheated rooms preceding the ‘laconicum’ were paved with flagstones. The floors of both heated rooms had disappeared. Technology: The hypocaust pillars were built out of square terracotta tiles, which may have ended in arches, judging from the arched heating channels that connected both heated rooms and the furnaces with the hypocaust.966 The walls seem to have been heated by tubuli, which were circular in cross-section. The double heat channel underneath the western alveus may point to a different temperature inside this pool. The recesses in the walls probably lodged the chimney flues. The recesses in the unheated room with benches might have contained water pipes, collecting water from the roof (see below, Rehovot C113). The water for the pools and basins was provided by a well located south-east of the bathhouse and probably also by rainwater catchment. The projecting stones along the southern external wall of the baths, of unknown purpose according to the excavator, probably carried a water conduit from the well to the praefurnium and further towards the reservoir.967 The water for the cold pools was collected in a reservoir west of the hall and then piped to the piscinae and probably to the basin in the room west of these. A sewer ran along the south side of the hall and then further west. Construction technique: Ashlar, opus testaceum (hypocausts) Bathhouse surface: ca. 500 m2 GPS coordinates: 30°47’43.72”N, 34°46’14.31”E

C112. Rama (Palestine)

Excavation history: The settlement of Rama, located along the Acco-Tiberias road, flourished during the late antique period. The construction works on a road in 1972 uncovered parts of the antique settlement, including olive presses, a Byzantine church and a public bathhouse. Stratigraphic excavations were carried out on behalf of the Department of Antiquities. Bibliography: Tzaferis (1980); Hoss (2005) 167–68. 965  Negev (1997) 170 quoting an older description of 1904. 966  Negev (1997) 175, photo 279. 967  See photo 274 and 275 in Negev (1997) 171. Such a solution is known in the Baths of Neptune in Ostia (Ricciardi and Scrinari (1996a) 158–59), or closer to Oboda, in the late antique baths of Serdjilla in Syria (Charpentier (1994) 125).

441

Gazetteer

C112 Rama (Israel), (After Tsaferis (1980) 67, fig. 1)

Construction: (B) The stratigraphy dated the baths to the 3rd or 4th c. Continued use and abandonment: (B) The building fell out of use in the 5th or 6th c., when a building of unknown function was built on top of it. Plan: The main entrance to the baths could not be located, but should probably be reconstructed on the east side.968 A rectangular hall or courtyard (17 × 8 m) had a large pool (4.5 × 7.5 m; 1.2 m deep) on its northern end. A broad doorway in the north-west gave access to the frigidarium. This room had plastered benches along its walls and two small apsidal piscinae (both 0.7 m deep) lodged in its north wall. To the south of the cold room, a dressing room (3 × 5 m) had a bench along its west wall. To the east, two doorways led into the large hall, while to the west, a doorway led into the first heated room (tepidarium). The latter seems to have had a hypocaust only in its south part.969 A small semicircular pool was lodged in the southern wall, although it had to be accessed by a room south of the tepidarium. To the west, the caldarium (10 × 6.5 m) may have been divided into two separate rooms, perhaps creating a caldarium and sudatorium. The rectangular apse in the north-eastern corner likely lodged an alveus. The reinforced rectangular hypocaust pillars along the north wall of the ‘western’ section of the caldarium may also point to the presence of an alveus. North of the caldarium, the praefurnia were recognized by furnace mouths and large quantities of ash and soot. Decoration: The decorative scheme of the baths was rather modest. The large hall was paved with plain white mosaics, consisting of large tesserae. The floor of the large pool was paved with the exact same mosaics. Three semicircular niches adorning the north wall of the pool probably contained statues, none of which were found. The frigidarium 968  Tzaferis (1980) 66. 969  Tzaferis (1980) 69.

was arguably the most lavishly decorated room. The floor was paved with a geometric polychrome mosaic. The apodyterium was the only room not to have a mosaic floor. A simple plastered floor was laid instead. The tepidarium and caldarium both had mosaic floors, although the caving in of the suspensurae meant little survived. Some of the walls inside the baths still had a plaster coating, but no traces of wall painting were found.970 Technology: The heated rooms had a standard hypocaust consisting of square and rectangular pillars made out of tiles. Only the caldarium had walls heated by tubuli. If the reconstruction of alvei upon the reinforced hypocaust pillars is correct, there were possibly water boilers on top of the furnaces. The water for the baths was probably supplied by natural sources springing up 100 m north of the building.971 Terracotta conduits were found in the south walls of the piscinae. The latter were drained by ceramic conduits leading waste water to a sewer underneath the frigidarium and continuing towards the hall. An outlet in front of the pool of the hall drained the water when overflowing. Construction technique: Ashlar Bathhouse surface: ca. 690 m2 GPS coordinates: Exact coordinates unknown

C113. Rehovot-in-the-Desert (Israel)

Excavation history: Rehovot was one of the sites that lay along the ancient road crossing the Negev Desert, serving as a stopover for traders and surviving through agricultural production in the small loess valley of the nearby streams. The archaeological remains found on site point to a continued occupation between the 1st and 8th c.972 A bathhouse was first identified in 1902, when Alois Musil published a

C113 Rehovot (Israel), (After Musil (1908) 75, fig. 46)

970  contra Hoss (2005) 168. 971  Tzaferis (1980) 68, n. 5. 972  Tsafrir (1988) 3–8.

442 description, photographs and a drawing of the standing remains. The baths, however, were destroyed in Turkish times, impeding new research. Bibliography: Musil (1907) 73–85; Tsafrir (1988); Hoss (2005) 170. Construction: (D) Late Antiquity seems to have been an exceptionally prosperous period for Rehovot. The first church was built in the late 4th or early 5th c.973 The baths probably date from this golden age. The plan shows strong similarities with some Umayyad baths in Syria, such as Gabal Says.974 Continued use and abandonment: Unknown. Plan: The bathhouse was located 50 m south of the city walls. One entered the baths by a door of 1.5 m wide, accessing a domed room with two lateral square exedrae with benches on three sides. Directly opposite the main entrance, a doorway gave access to a rectangular room, once cross-vaulted, without specific features. The same can be said for the next room to the east. The northernmost room had three semicircular pools (0.7 m deep), their bottoms lying lower than walking level. According to Musil’s sketch, the first room was the only one to be heated by a hypocaust. This would be surprising, as it clearly functioned as an apodyterium / frigidarium in its last phase of use, while the supposed caldarium was unheated. This strange layout could be explained by a complete reorganisation of the building at some point during its existence. The first room with it two square exedrae may have been the original caldarium, with alvei in the exedrae. This would fit with the popularity of cruciform caldaria in the region, but would also explain why this was the only room to have a hypocaust. It would also account for the strange orientation of the baths, with the frigidarium in the south and the caldarium in the north. The caldarium described by Musil was clearly added at a later date, as its wall abutted what must have originally been the apodyterium / frigidarium. Decoration: The decorative scheme seems to have been quite modest, although the floor and the walls of the southernmost room had a marble veneer. The small rooms were lit by small square windows in the upper parts of the wall and in the vaults and dome. Technology: To the east of the bathhouse, a circular well was found. A circular groove around it was probably made by the animals turning around the well. Water was transport by a bucket, still on site when Musil visited the ruins, to a reservoir on the roof of the caldarium. A system of gullies on the roof seems to have distributed the water from this reservoir to the different rooms of the bath, possibly by terracotta pipes in the walls (fig. 136). Construction technique: Unknown 973  Tsafrir (1988) 4. 974  Bloch (2014).

Gazetteer

figure 136

Baths in Rehovot (Israel), drawing of the roof of the baths with the system of a cistern, gullies, windows and orifices in Musil (1908) 81, fig. 52

Bathhouse surface: ca. 80 m2 GPS coordinates: 31° 1’35.94”N, 34°33’56.40”E

C114. Scythopolis (Beth She’an / Baysan, Israel / Palestine), Southern Baths

Excavation history: During the 1980s, a bathhouse (map 27) was excavated in the area between the theatre and the amphitheatre by the Israel Antiquities Authority. Bibliography: Peleg 1987; Peleg (2004); Hoss (2005) 130–31. Construction: (B) On the basis of the ceramics found during the excavations, the construction date is placed somewhere between the end of the 3rd and the beginning of the 4th c.975 Continued use and abandonment: (C) At the beginning of the 6th c., a new mosaic was laid over the pools, indicating that the building had lost its original function.976 Plan: The main entrance of the baths was not found, but it should possibly be located to the west. Here a room of uncertain dimensions may have served as the apodyterium.977 A central room (4.5 × 5.5 m) was probably a tepidarium. It gave access to the frigidarium to the north and to the first heated room to the south. The former was a large hall (3.3 × 4.2 m) with two ‘sitz-baths’ (semicircular pool: 0.75 × 1 m; 0.4 m deep) flanking a semicircular niche in the 975  Peleg (2004) 78. 976  Peleg (2004) 80. 977  Peleg (2004) 57.

443

Gazetteer Construction technique: Opus quadratum Bathhouse surface: ca. 265 m2 GPS coordinates: Exact coordinates unknown

C115. Scythopolis (Beth She’an / Baysan, Israel / Palestine), Western Baths

C114 Scythopolis (Beth She’an / Baysan, Israel), Southern Baths (After Peleg (2004) 56, plan 1)

northern wall. The latter may have contained a labrum.978 The frigidarium was flanked by two rectangular pools (eastern pool: 1.6 × 2.2 m; western pool: 2.9 × 4.8 m), both only 0.6 m deep. The central ‘tepidarium’ was heated by a hypocaust and tubuli. To the south, a second heated room with its own furnace may have functioned as a sudatorium. A semicircular wall that encircled the furnace at a later phase might actually have been the foundations of a semicircular alveus added to the room. The fact that this furnace was blocked at a later date reinforces such a hypothesis. The southernmost heated room was cruciform and had two rectangular ‘sitz-baths’ (both 1 × 1.5 m; 0.6 m deep) in the eastern arm. The southern and western arms may also have lodged larger alvei. Decoration: The walls of the rooms were plastered and, at least in the apodyterium, embellished with marble slabs. The floors of the cold rooms were paved with mosaics of large tesserae. The apodyterium and frigidarium had black and white geometric motifs (meander and net patterns), while the large pools had a plain white mosaic. The floors of the heated rooms were paved with flagstones. Technology: The pillars of the hypocausts were made out of round or square tiles, reaching to a height of 0.4 m. The tubuli were rectangular in section. The so-called ‘air pipes’ found in the frigidarium should rather be interpreted as tubi fittili used in the construction of the vaults.979 The water for the pools was partially supplied by the collection of rainwater from the roofs, as gutters were found throughout the site. Along the east side of the baths, a sewer drained the eastern piscina and probably the alvei. A large plastered basin—perhaps a reservoir—was identified 30 m west of the building.980 978  Peleg (2004) 58, fig. 8. 979  Peleg (2004) 59, fig. 11–12 and 60, fig. 13. 980  Vitto (1987) 45; Peleg (2004) 78.

Excavation history: The complex was excavated between 1986 and 1991 by the Israel Antiquities Authority. Bibliography: Bar-Nathan and Mazor (1993); Mazor and BarNathan (1998); Hoss (2005) 131 Construction: (B) Excavations revealed that the earliest phase of the baths was 4th c.981 Continued use and abandonment: (B) Several restoration phases in the 5th and 6th c. altered the original layout of the baths. The baths fell out of use in the late 6th or early 7th c., when the walls were dismantled for their building material and the site became an Early Islamic burial ground.982 Plan: Some 20 m to the north of the theatre, a bathhouse (map 27) was built in Late Antiquity, along the colonnaded street and on top of an earlier, unidentified building.983 The complex (90 × 60 m) consisted of a T-shaped bathing block enclosed by a rectangular portico. The small rooms

C115 Scythopolis (Beth She’an / Baysan, Israel), Western Baths (After Tsafrir and Förster (1997) fig. D)

981  Mazor and Bar-Nathan (1998) 23. On the 4th-c. date of the building, see Mazor and Bar-Nathan (1994) 124 (in Hebrew). 982   Bar-Nathan and Mazor (1993) 44. 983  Mazor and Bar-Nathan (1998) 24.

444 along the portico’s north, south and west sides were often remodelled and probably changed function several times. In the original layout of the baths, the westernmost room of the bathing block was a natatio (6.5 × 13 m), accessible from the pillared hall to its east. The latter probably functioned as a multipurpose cold room. To the east, a double doorway led into an oval room, heated by a hypocaust. No traces of pools were found in this room, even if the apsidal endings would make for a logic location. By two narrow doorways in the east wall, the bathers continued the bathing itinerary into an elongated heated room with a round alveus (diam.: 2 m) on its eastern side. From this room, one could access a rectangular heated room to the south or a succession of two heated rooms to the north. The northern rooms had a rectangular alveus along their eastern walls, while the northernmost room even had an additional apsidal alveus in its northern side. In a later phase, 4 cold pools (ca. 3.7 × 9 m) were added to the eastern side of the portico, two to the north of the main bathing block and two to the south. All were made out of brick and coated with a layer of hydraulic mortar. They were surrounded by columns. Perhaps at the same time, the natatio was filled in and paved with a polychrome mosaic floor. The doorway to the east was also blocked up. During the 6th c., an oblong hall with an apsidal ending to the north was constructed on the western side of the portico. The western rooms of the portico were reduced, while the two in the north-west corner were filled with rubble. Also during the 6th c., a ‘propylaeum’ was created on the east side of the portico to serve as a new monumental entrance along the colonnaded street. A new latrine was built north of this new entrance. In the last phase of the baths, the pillared hall lost its function and was transformed into a praefurnium for the oval room. Decoration: The natatio was originally paved with shale, while the smaller piscinae were decorated with marble slabs. Some of the small rooms of the portico had a plain white mosaic floor. The mosaic of the 6th c. apsidal hall had a simple geometric pattern with a (now illegible) inscription in its centre. The wall of the apse was embellished with a colourful glass mosaic. The heated rooms also showed traces of mosaic, but most had been destroyed when the suspensura collapsed. Technology: The pillars of the hypocaust were made out of round tiles on a conical stone base. Wall heating with tubuli was found in all of the heated rooms. At least 9 furnaces have been identified for the heated section, although the numerous restoration phases make it difficult to understand which furnaces functioned at the same time. There is no information available about the water supply. Inside the baths, water was distributed by terracotta conduits. The 4 piscinae in the courtyard had shallow drains running along their sides.

Gazetteer Construction technique: Opus quadratum Bathhouse surface: ca. 3,400 m2 GPS coordinates: 32°30’7.87”N, 35°30’3.01”E Tel Sera (Israel) Excavation history: The site of Sera was located in the western Negev Desert, on the southern banks of the Gerar river. The remains of a bathhouse were discovered by the Ben Gurion University of the Negev and the Israel Exploration Society in the 1970s. Bibliography: Oren (1993a); Hoss (2005) 173. Construction: (D) The baths were labelled ‘Byzantine’ without further specification.984 Continued use and abandonment: (A) The excavations dated the last phase of the baths at the turn of the 5th and 6th c. (pottery and coins).985 Plan: The description of this building provided by the excavators is very brief. The bathhouse had a room with hypocaust heating and tubuli. Decoration: Unknown. Technology: The pillars were made out of bricks and ended in arches. Plastered channels for water supply and drainage were also identified. Tiberias (Israel) Excavation history: In the 1950s and again in the 1990s, salvage excavations by the Israel Antiquities Authority in the modern city centre uncovered the remains of the central bathhouse. It was located along the cardo maximus and next to the market place. Bibliography: Hirschfeld (1993); Hirschfeld (1997); Hoss (2005) 174. Construction: (A) The building was constructed in the 4th c., as the ceramics found in test pits underneath the floors have confirmed.986 Continued use and abandonment: (A) The building was restored several times, with an especially important intervention in the 6th c., and continued to be used until the 11th c. The baths probably fell out of use after an earthquake (AD 1033?).987 Plan: The bathhouse consisted of two parts: an east wing for the heated rooms and a west wing for the cold rooms and auxiliary halls. The bathers entered the building by a long corridor (19 × 5.8 m) in the north. Two doorways in the southern side of this corridor gave access to two halls with a possible piscina in between.988 A subterranean pool was discovered 3 m beneath the floor level of the western hall. 984  Oren (1993a) 1335. 985  Oren (1993a) 1335. 986  Hirschfeld (1993) 1466. 987  Hirschfeld (1993) 1466. 988  Hirschfeld (1993) 1466.

445

Gazetteer It was about 1 m deep and had 24 short columns (each 1.4 m in height) which supported a ceiling of basalt beams. In the western wing of the baths, several warm water pools were discovered. The caldarium lay at the centre. Decoration: The pools in the west wing were lined with marble slabs. Technology: The hypocaust pillars were made with square tiles.

C116. Umm el-Amr (Palestine), Monastery Baths

Excavation history: The bathhouse was discovered in 1997 during excavations by the Gaza Antiquities Department. The excavations took place in 1999 and in 2002–2003 with the help of the French archaeological mission. Bibliography: Elter and Hassoune (2008). Construction: (B) The ceramics and coins found during the excavation point to a phase of use between the 6th and 8th c. Continued use and abandonment: (B) The building may well have been used after the 8th c., but by then it had probably lost its bath function. Plan: The bathhouse lay just outside the walls of a monastery. This complex was presumably built around the tomb of the monk Hilarion, who returned from Egypt to his native Gaza in AD 310 and founded a cenobitic society.989 The monastery—some 14,000 m2 large—had an ecclesiastical part focused on the church and a residential part with living quarters, a possible xenodochium and a bathhouse. The baths date from the ‘monumentalisation’ phase of the site. At least three distinct phases could be discerned. According to the excavators, the heated section of phase 2 was abandoned when the heated section of phase 3 was built.990 However, Fournet and Redon believe both heated sections could have functioned simultaneously, making this a double bath with a shared cold section.991 In the final phase, the bathhouse had a large cold room with two semicircular single person basins (showers?). In the east wall, two doorways gave access to the heated sections. The north wing (presumably phase 3) had two tepid rooms (without hypocaust) and a square caldarium with three rectangular recesses. The northern and eastern recess lodged pools, while the southern recess had a bench running along its sides. The eastern pool was located above the only furnace. The southern heated wing (presumably phase 2) also had two tepid rooms (one with apsidal ending and benches) and two rooms with hypocaust. The caldarium was square and had three alvei lodged in the recesses. The eastern alveus lay above the only furnace. Perhaps in a 4th phase, a new cold section with two rectangular piscinae was built 989  Chitty (1966) 13–16. 990  Elter and Hassoune (2008) 138. 991  Fournet and Redon (2017a) 309.

C116 Umm el-Amr (Palestine), Monastery Baths (After Elter and Hassoune (2008) 136, fig. 7)

south of the original cold section. Two latrines abutted the northern side of the baths, but could only be accessed from the outside, not from within the bathhouse. Decoration: The imprint of an opus sectile was visible in the large hall, although the actual (marble?) slabs had been torn out. The walls were covered in white plaster. Some of the benches and some of the floors were clad with white marble slabs. Technology: The hypocausts pillars were made with square tiles. The wall heating was ensured by tubuli (round or rectangular in section). The northern wing praefurnium had an earthen floor and was accessible by a staircase. The ashes seem to have been dumped just east of the service corridors, forming a layer over 1 m thick, spread out over several hundred metres. The baths were built near an existing well with noria. The waste water of the heated wings was collected in a basin south of the baths, which was interpreted as a fish tank because of the half open pottery jars embedded in the floor. Construction technique: Opus quadratum Bathhouse surface: ca. 700 m2 GPS coordinates: 31°26’49”N, 34°21’58”E

C117. Zikhron Ya’aqov (Israel)

Excavation history: In 2007, excavations prior to construction by a private contractor exposed a bathhouse of around 20 × 20 m. The site was excavated by the Israel Antiquities Authority.992 992  The results of the excavations had not been published at the time of writing. The discovery was announced on the website of the Israel Antiquities Authority (http://www.antiquities .org.il/article_eng.aspx?id=1447&sec_id=25&subj_id=240).

446

Gazetteer identified. In the northernmost, a simple mosaic of white and red tesserae was found. The middle room had a heating channel in its eastern wall, although no hypocaust floor was present. The southernmost room may have been a frigidarium, as its walls and floor were coated with hydraulic mortar. Several fragments of marble and ceramic tiles were found during the excavation. The lack of water-related elements such as pools or conduits makes the identification of this building as a bathhouse problematic. Barqa / Gan Yavne (Israel) During the construction of a road in 1998, the remains of a hypocaust were found near modern Gan Yavne. Besides mudbrick pillars, the salvage excavations also found tubuli and plastered floors. The ceramics found in connection to these digs dated from the late antique period.994

C117 Zikhron Ya’aqov (Israel), (Sketch on the basis of aerial photography, no scale)

Bibliography: No publication available. Construction: (B) The preliminary study of the ceramics recovered during the excavations point to a 4th c. construction date. Continued use and abandonment: Unknown. Plan: (C117) The bathhouse included piscinae, single person bath tubs, a tepidarium and a caldarium with two small semicircular alvei. According to the excavators, the baths may have been part of a large villa estate. Decoration: The baths were decorated with imported marble elements, including Corinthian capitals, the head of a statue, pavement slabs and part of a fountain (a sculptured shell with bird motifs). Technology: The water was supplied by the nearby spring of Ein Hanzir. Construction technique: Opus quadratum, opus caementicium Bathhouse surface: ca. 400 m2 GPS coordinates: Exact coordinates unknown.

Possible Late Antique Baths Ashkelon (Israel) The construction of a pipeline in 2003 uncovered the remains of a possible late antique bathhouse in the el-Qabu area of Ashkelon. The pottery and a coin recovered during excavations point to use between the 5th and 7th c. After abandonment, the building was reused for residential purposes.993 The remains were only partially excavated. Three rooms were 993  Sion (2009) §5–6.

Baqa el Gharbia (Israel) The remains of a heated room with a praefurnium dated to the 5th c. were found in the courtyard of a private house.995 The hypocaust was partially hewn into the bedrock, while the pillars were made out of square tiles forming arcades. The floor of the room was paved with marble slabs and benches ran along the walls. The tubuli were still in situ along three of the walls. The building was out of use by the 6th c. and a building of unknown function was built on top of it in the Islamic period.996 Beth Nehemya (Israel) A small test pit in Beth Nehemya confirmed that construction works had hit the archaeological layers during a building project in 2001. In the square test pit, the remains of a hypocaust and a mosaic floor were found. Fragments of hydraulic mortar and terracotta pipes—possibly water conduits, tubuli or tubi fittili—reinforce the interpretation of a bathhouse. A furnace was found near the hypocaust. The hypocaust itself was covered by destruction layers containing ceramics from the 5th and 6th c.997 Deir el-Balah (Palestine) A salvage excavation in 1973 identified parts of an alleged bathhouse in a Byzantine town.998 No further information is available.

994  Avner (2000) 60–61. 995  Badhi (2001) 325. Dating on the basis of small finds and ceramics, see Badhi (2001) 51*-71* (in Hebrew). 996  Badhi (2001) 325. 997  Nagorski (2009) §3. 998  Oren (1993b) 1396.

447

Gazetteer En Qobi / Ein el-Qabu (Israel) In 2001, a subterranean pool connected to a spring was discovered. The excavations revealed several fragments of tubuli, bricks and tesserae pointing to the existence of a bathhouse in the vicinity of the pool. The associated ceramics date to the late antique and Early Islamic period.999 (Tel) Hefer / Tell el-Ifshar (Israel) The remains of a bathhouse were identified in the late antique quarter of the settlement, north-west of the tell.1000 No further information is available. Horbat Hanot / Khirbet el-Khan (Israel) Several hypocaust tiles were found during salvage excavations in 1985. The site was probably a late antique road station.1001 Horbat Minim / Khirbet el-Minya (Israel) The site is located on the north-west shore of the Sea of Galilee and is known for the Umayyad palace that was built there. Some 200 m south-east of this palace, a late antique bathhouse was once identified.1002 The remains have now disappeared. Horbat Tarbenet (Israel) In the summer of 2011, a white marble statue of Hercules was discovered near a large pool.1003 This pool was flanked by benches on two of its sides and was fed by a conduit with the water of a saqiyah (diam.: 2.9 m; 4 m deep), found in the vicinity. The pool was in use until the Early Islamic period, when it was filled with debris. Jaffa / Yafo (Israel) The remains of a possible late antique bathhouse were discovered in 2006 near the Ganor Compound in Jaffa.1004 The complex was still in use in the Early Islamic period, but no further information or a description of the baths is given. Jerusalem / Al-Quds (Israel / Palestine), Rosary Sisters school compound in Beit Hanina Prior to developments in the Beit Hanina neighbourhood in 2007–2008, excavations uncovered the remains of a presumed late antique bathhouse underneath a schoolyard. Only parts of a heated room were found, with brick pillars ending in 999  Billig (2005) §5. 1000  Paley and Porath (1993) 612–13. 1001  Shenhau (1986); Hoss (2005) 146. 1002  Grabar (1993) 1050–51. 1003  The results of the excavation were not published at the time of writing. The find was reported on the website of the Israel Antiquities Authorities (http://www.antiquities.org.il/Article_ eng.aspx?sec_id=25&subj_id=240&id=1868&hist=1). 1004  Peilstöcker and Burke (2009) §2.

arches. The ceramics found associated with the structure date from the late 6th and early 7th c.1005 Kefar Uriyya (Israel) Archaeological test trenches in Kefar Uriyya uncovered the remains of a possible late antique bathhouse.1006 The site awaits excavation. Khirbat Burin (Israel) Prior to the construction of a road south of the Qalansuwa Junction on Highway 57, a salvage excavation in 1999–2000 uncovered a possible late antique bathhouse. The walls were built of mudbricks on a preparation layer of river pebbles bonded by a pinkish mortar. The L-shaped heated room had a hypocaust made out of bricks and reused roof tiles. The finds in between the pillars were dated to the ‘Byzantine period’.1007 Khirbat Ta’ena (Israel) The settlement near modern Khirbat Ta’ena was dated by surface finds between the late antique and Mamluk periods. The remains of a pool and possibly another bathhouse were discovered near a spring.1008 Lydda / Diospolis (Lod, Israel) The remains of a possible bathhouse included stone paving, terracotta conduits and bricks. The ceramics found on site date to the end of the late antique period.1009 Na’aran (Israel) Remains of tubuli and hypocaust tiles, both round and square, were found near a Roman road. The structure may have belonged to a road station of the late antique period or earlier.1010 Nahal Ashan forest (Israel) Heavy floods along the banks of the Nahal Karkur river revealed part of a heated room. Besides the hypocaust, a mosaic dated to the late antique period was also found. The possible baths may have belonged to the Byzantine settlement around modern Horbat Karkur.1011 Nahf (Israel) In the old city centre of Nahf, several excavations have revealed a long occupation history of the Tell, starting from the Early Bronze Age and continuing to this day. A small salvage excavation in 2010 revealed part of a heated room with an 1005  Adawi (2012) §9. 1006  Kisilevitz (2014). 1007  Ad (2005) §6. 1008  Zingboym (2011) §2. 1009  Anonymous (1982) 67. 1010  Dauphin (1989) 6–7; Hoss (2005) 164. 1011  Aladjem (2013) § 11.

448

figure 137

Gazetteer

Plan of the possible baths in Nahf (Israel) after Abu Raya (2013b) fig. 2

associated cistern (fig. 137). The remains were identified as part of a bathhouse. The ceramics and small finds associated with the functioning of the building were dated to the late 6th and early 7th c. It was filled up with waste at the beginning of the 8th c.1012 The remains of a heated room consisted of arcaded pillars made out of square bricks. The cistern was bellshaped (5 m deep) and coated on its interior with a hydraulic plaster. The large quantities of glass, glass debris and lumps of raw glass found during the excavation point to a Byzantine or Umayyad glass workshop in the vicinity.1013 Neve Shalom (Israel) A building with a polychrome mosaic floor was found during a community-educational excavation in the settlement of Neve Shalom in 2006–2007. The square tiles (s.: 20 cm) and the use of a waterproof mortar on the walls could point to a bathhouse. The mosaics have Christian motifs, such as crosses and fish. The ceramics found during the excavation are dated at the turn of 6th and 7th c.1014 Ptolemais (Acco, Israel) In 1956, a 15 m long pool was excavated in the modern city of Acco. Its walls were built in fieldstones and traces of pinkish plaster were also found. The remains were dated to the late antique or Early Islamic period.1015 Qalandia (Palestine) The Baths of Qalandia, 12 km north of Jerusalem, were excavated in 1931 by the Department of Antiquities in Palestine. The late antique construction date was probably deduced from the type of wall construction and mosaics.1016 There is no further information about the site’s abandonment. The building (fig. 138) was constructed in rubble concrete with the walls plastered on both the inside and outside. The building was converted to an unknown function at a later 1012  Abu Raya (2013b) §4. 1013  Abu Raya (2013b) §8. 1014  Solimany (2012) §4. 1015  Goldmann (1993) 25 without further specifications. 1016  In the excavation report, the baths are called ‘Byzantine’ without any further specification, see Baramki (1933).

date, making the interpretation of the original plan difficult.1017 The main room had a floor paved with limestone slabs. Several niches were lodged within its walls. To the south, there was a room with two basins, one round and one rectangular (1.7 m deep). The room was paved with large white tesserae. In a first phase, this room formed a continuous space with the main room. Instead of the two small basins, it had a larger pool, accessible by steps in the east and west. A connection to the rooms east of the main block could not be identified. The easternmost room had two small rectangular pools on its eastern side. No room heated by a hypocaust could be identified, although a layer of ash was found on the mosaic floor of the rooms north and south of the main room.1018 On the eastern end of the northern room, a furnace was discovered. The excavation report is not clear whether it heated the main room or not. There is no mention of a hypocaust underneath the latter. To the north of the building, a large round furnace coated with plaster was found. It was identified as a furnace for heating water.1019 The basins south of the main room were fed by water channels running underneath the floor. The northeast pool in the eastern room had a slump in its north-eastern corner, possibly to facilitate scooping out the water. The east pool in the same room was drained by a channel leading into a round collector, partially cut out into the bedrock and with two steps. It was drained by a channel heading south-east. The evidence for this bathhouse is incomplete. The absence of (traces of) a hypocaust is problematic. The later reuse of the site, possibly for industrial purposes, may account for the layers of ash. The basins may also have served industrial— rather than bathing—purposes. The small pool with a slump in a corner is reminiscent of a Jewish ritual bath (miqveh), rather than a Roman-style pool.1020 Ramat Ha-Sharon (Israel) The remains of a possible bathhouse were discovered during a survey campaign near Ramat Ha-Sharon in 2007. The surface finds in the area suggest an occupation between the late antique and Ottoman period.1021 No further excavations were carried out, as the site was not under threat. Ramla (Israel) Salvage excavations in 2008 east of the city of Ramla discovered large quantities of brick building material, possibly from a bathhouse.1022 The associated ceramics date to the Early 1017  Hoss (2005) 167. 1018  Baramki (1933) 106. 1019  Baramki (1933) 107. 1020  However, the basin is not recognized as a miqveh by Hoss (2005) 167. 1021  Ad et al. (2009) §2. 1022  Sion (2010) §7.

449

Gazetteer

figure 138 Plan of the possible baths in Qalandia (Palestine) after Baramki (1933) 108, fig. 95

Islamic period. South of the city, two separate buildings with a hypocaust floor dating from the Early Islamic period were also found. In the Area K1 baths, the hypocaust pillars were made out of reused round bricks, while in the Area I2 baths, the pillars consisted of limestone monoliths. The stratigraphy of the site points to a phase of use in the 7th and 8th c. for both buildings.1023 Tel Aviv (Israel) Salvage excavations on the site of the future HaAretz museum in 1984 uncovered the remains of a possible bathhouse. Fragments of hypocaust pillars, ceramic floor tiles and terracotta water conduits were recovered. Among the finds, glass and ceramics dated the contexts to the late antique period.1024 Tel Qasile (Israel) Located on the fringes of modern Tel Aviv, the site yielded an apsidal room heated by a hypocaust. The brick pillars ended in arches. On the basis of the stratigraphic excavations in the 1970s and 1980s, the use of the building was dated to the 6th and 7th c.1025

1023  Gorzalczany (2009) §6. 1024  Ayalon (1985) 107–108. 1025  Mazar (1993) 1212.

Private Baths Several bathhouses of private domus or villas have also been found in Palestina. The limited sizes of the excavations often do not permit answers on whether these baths were also accessible to a larger public or not. For the same reasons, some of the complexes described above (see above ‘Possible baths’) could also have been private baths. In Ashkelon, salvage excavations in the Barne’a quarter in 1991 uncovered a late antique villa estate with a pars rustica and a bathhouse. The baths consisted of a frigidarium with two semicircular ‘sitz-baths’, a tepidarium and a caldarium with possibly two semicircular single person alvei.1026 The baths in the palace at Caesarea Philippi (Banias, Israel) were obviously private. The installation of the baths inside the palace has been dated to the late antique phase (3rd to 6th c.), although little remains except for some rooms with hypocausts.1027 In Pella (Tabaqat Fahl, Jordan), a villa dated between the 3rd and 7th c. had a private bathhouse.1028 The remains were destroyed before excavation could take place. Recent excavations in Mishmar David (Israel) revealed the existence of a possible bathhouse in an agricultural estate dating from the late antique or Early Islamic period.1029 In the late antique villa found in Yehud (Israel), a mosaic of sandals could also point 1026  Israel (1993) 100–105. 1027  Tzaferis and Israeli (1999) 1–2. 1028  Watson (1996) 73. 1029  Yannai (2014) §4.

450

Gazetteer

to an unexcavated private bath.1030 A bathhouse was added to the ‘Palace of the Governor’ in Caesarea Maritima, possibly in the early 4th c.1031 The bathhouse was accessible from the ‘public part’ of the palace and may thus have had a public use. The plan, with several bathing rooms arranged around a large frigidarium with an octagonal pool at its centre, resembles the City Baths of Ptolemais (C71).

Ecclesiastical Baths The bathhouses that were found within the walls of a monastery (or attached to it) constitute a separate category of baths, as it is often unclear whether they were also accessible for a larger public than the monks or not. In contrast to the monasteries in Egypt, where Roman-style baths were rather small, the complexes in Palestina often had larger bath suites within the walls. The monastery on Mount Scopus in Jerusalem was enlarged to the south during Late Antiquity. A small bathhouse was the main addition to the existing infrastructure (fig. 139). It had a cold room (9.2 × 4.3 m) with a small pool (1.1 × 0.5 m) in its south-west corner. A doorway in the western side gave access to a small cubicle (s: 2.4 m) with a bench running along the walls. There were three heated rooms to the south of the cold room. All three had a hypocaust consisting of brick arches. No alvei have been identified and no wall heating is mentioned in the excavation report. The floors of the cold rooms were paved in mosaics. A plain white field covered the floor of the cubicle, while a polychrome mosaic with rhomboid motifs covered the large cold room’s floor.1032 Still in Jerusalem, at the end of the modern Prophet’s Street, the remains of a 6th c. ecclesiastical complex were uncovered in the 1991–1992 campaign of the Third Wall Excavation Project. Among the identified buildings was a church, parts of an Armenian monastery and a small bathhouse.1033 A Greek mosaic inscription of an abbot named Silvanus was dated to the 6th c., the presumed construction phase of the building.1034 The complex was still in use in the Early Islamic period and was even expanded in the second half of the 7th c., judging from coins found underneath the new mosaics. The bathing part was located 15 m south of the church, possibly in the living quarters of the monks. It consisted of a paved courtyard giving access to a heated pool (1.6 × 3.5 m).1035 A furnace was found to the east. The water for the baths and for other purposes was stored in a reservoir that could contain up to 1030  Korenfeld and Bar-Nathan (2014) §7. 1031  Piraud-Fournet (2014) 695. 1032  Amit et al. (2000) 76–78. 1033  Amit et al. (1993) 80; Amit and Wolff (2000) 293. 1034  Amit et al. (1993) 81; Amit and Wolff (2000) 296. 1035  Amit et al. (1993) 82.

figure 139

Plan of the Mount Scopus monastery in Jerusalem (Israel / Palestine) with the location of the bathhouse in box, scale is 10 m after Amit et al. (2000) fig. 137

700 m3. In a later salvage excavation, plastered channels were found west of the baths, interpreted as either the aqueduct for the baths or the sewer for waste water.1036 The monastery of Martyrios was located 5 km east of Jerusalem, on the road connecting Aelia Capitolina with Jericho. It was built over the cave in which the presbyter Paulus is supposed to have lived. The monastery was named after its founder Martyrios, who lived in the second half of the 5th c. The excavations between 1982 and 1985 revealed a walled compound, centred on a courtyard and including a chapel, several storerooms, cells for the monks, and a bathhouse.1037 The latter was located in the centre of the west wing and consisted of three rooms, one of which was heated by a hypocaust. This caldarium (2 × 6 m) had a semicircular niche in its east side, perhaps for individual use by the abbot. The heated room was preceded by an oblong hall with a polychrome mosaic floor with leaf pattern. A pool was found to the east of this hall (no dimensions given). Two large cisterns were found close to the baths. It is estimated that the total capacity of all of the monastery’s cisterns could amount to 20,000 or 30,000 m3. The system was entirely depended on the catchment of rainwater, which averaged some 300 mm per year.1038

1036  Adawi (2005) §7. 1037  Magen (1993) 969. 1038  Magen (1993) 972.

451

Gazetteer

Fortress Baths Ayn Gharandal (Jordan) Excavation history: The Roman fort at Ain Gharandal has been known since the first European travel accounts from the early 20th c. The presence of a spring probably encouraged the Roman army to set up camp on this location. Some 50 m south-east of the fort, the remains of a bathhouse were revealed by illegal digs in the late 2000s. Bibliography: Darby and Darby (2010); Darby and Darby (2012); Darby and Darby (2015). Construction: (D) Surface finds in the area were dated to the 4th and 5th c.1039 If the baths are contemporaneous with the fort, which is plausible, we could propose a late antique date for the baths. The fort was dated to the reign of the Tetrarchs by an inscription found during the 2013 campaign. Continued use and abandonment: (D) Modern disturbance of the site does not permit any conclusions about the different phases of use. Plan: The presence of large amounts of tubuli and burnt hypocaust tiles identified the building as baths. The tepidarium and caldarium have been identified. A large room without heating system was interpreted as the frigidarium. To the south-west of the heated rooms, a latrina was uncovered. As the excavations are on-going, the floor levels of the frigidarium and caldarium still have to be reached. Decoration: Two rooms with plastered walls could be recognized. The latrina was paved with large flagstones, while the seats were probably wooden.1040 A graffito with Greek text and an image of a seated camel were discovered on the caldarium wall.1041 Technology: The tepidarium still had most of its tubuli wall heating in place (round in section). During the 2009 survey, the remains of what appears to be an aqueduct, supplying both baths and fort, were also identified. Be’er Shema (Israel) Excavation history: During a salvage excavation in the summer of 2006, the remains of a Byzantine military installation with a possible adjacent bathhouse were identified in Horbat Be’er Shema, 20 km east of Be’er Sheva. Bibliography: Erickson-Gini (2011) Construction: (B) The preliminary results of the pottery recovered during the excavation point to a ‘Byzantine’ construction phase. Continued use and abandonment: (B) The site was abandoned in the Early Islamic period.

1039  Darby and Darby (2010) 535. 1040  Darby and Darby (2015) 466. 1041  Darby and Darby (2015) 465.

Plan: The results of the excavations were not published at the time of writing. Decoration: Unknown. Technology: Unknown. Be’er Sheva (Israel), Bate Hamalkha street Excavation history: The illegal widening of a drainage channel on the outskirts of Be’er Sheva uncovered the remains of a hypocaust. Bibliography: Negev (1994). Construction: (D) The late antique ceramics that were found could not be linked directly to the structures, although the late antique city centre did reach the location of the remains. Continued use and abandonment: (D) Unknown. Plan: A single room with hypocaust could be identified. A second possible bathhouse, probably belonging to a late antique fortress, was recognized on an aerial photograph of 1918.1042 It has now been destroyed by the modern city. Decoration: Unknown. Technology: Unknown. Bir Madhkur (Jordan) Excavation history: Some 34 m south-east of an early 4th c. fort, the remains of a possible bathhouse (ca. 30 × 25 m) were identified due to the large quantities of hypocaust tiles found. The site of Bir Madhkur may be identified with the ancient road station Moa, known from the Be’er Sheba Edicts (6th cent.) and the Madaba mosaic map.1043 Bibliography: Smith (2005); Smith (2010); Darby (2015). Construction: (D) The bathhouse seems contemporary with the late antique military fort. Continued use and abandonment: (B) Unknown. Plan: At the present state of knowledge, two (possibly three) rooms have been identified. One of these was a latrina. It is likely that the frigidarium was located in the immediate vicinity.1044 Decoration: Unknown. Technology: A large quantity of hypocaust tiles was found on site. Legio (el-Lejun, Jordan) Excavation history: The bathhouse was excavated during the Limes Arabicus Project between 1980 and 1989. Bibliography: Hoss (2005) 159–69; de Vries and Lain (2006). Construction: (B) The excavations revealed that the baths were constructed around AD 300 and may have fallen out of use as early as AD 363, when an earthquake hit the region.1045 1042  Fabian (1995) 236–37, fig. 2A. 1043  Darby (2015) 73. 1044  Darby (2015) 75. 1045  de Vries and Lain (2006) 220–21.

452 Continued use and abandonment: (B) The building was reused as a house in the beginning of the 5th c. and was transformed into a stable during the 6th.1046 Plan: The baths were constructed against the northern curtain wall of the fort as a compact rectangular unit (23 × 9.7 m). The bathers entered by a small door in the southeast, giving into an apodyterium with niches in its walls and a bench. The north-west corner of the room was delimited by a thin wall. Perhaps it was the only covered space in an otherwise roofless room?1047 By a large bay, one entered a room with a large rectangular apse (2.3 × 2.1 m), interpreted as a lounge area.1048 Alternatively, a piscina might have been located here.1049 It could also have served as an unctorium or districtarium. The three heated rooms were laid out following an angular row pattern. The first room (4.4 × 2.7 m) was preceded by a shallow basin (1.5 × 3.5 m), possibly a footbath (pediluvium).1050 This first heated room only had a hypocaust in its west part. Neither tubuli nor chimneys were identified. The second room had the same dimensions as the previous one, but had a hypocaust underneath its entire floor. The excavations in this part of the building were not finished. The last heated room had a cruciform shape (each apse around 2.3 × 1.3 m). In each apse there was probably an alveus.1051 The praefurnia must have been located south and west of the heated section. Decoration: The decoration of the building was modest. The walls seem to have been coated with a layer of plaster, while the delimited area in the north-west corner of the apodyterium and the pediluvium were paved with limestone slabs. Technology: The hypocaust was constructed with pillars made out of round tiles with a total height of around 0.75 m. The draught of the hot gasses was ensured by flues in the corners of the second and third heated room, but no tubuli were found. A small test pit in the hypocaust of the first heated room revealed ashes with traces of dung.1052 Evidence of a water adduction pipe was found in the wall of the first heated room, but other information on the water management of these baths is lacking.

1046  de Vries and Lain (2006) 222–23. 1047  de Vries and Lain (2006) 214. 1048  de Vries and Lain (2006) 215. 1049  Hoss (2005) 158. The frigidarium / apodyterium with adjacent recesses with pool and pediluvium is reminiscent of similar layouts in the baths of Karanis (C85) and Sersena (C94). 1050  de Vries and Lain (2006) 215. 1051  de Vries and Lain (2006) 217. 1052  de Vries and Lain (2006) 220.

Gazetteer Yotvata (Israel) Excavation history: The baths were first discovered in 1974 and partially excavated by Tel Aviv University. A new study of the fortress in 2006 and 2007 by Florida International University and the University of North Carolina at Chapel Hill included a restoration and an additional excavation. Bibliography: Meshel (1989); Davies and Magness (2007); Davies and Magness (2008); Davies and Magness (2011). Construction: (A) The baths are contemporaneous with the fort, which was dated by an inscription to the reign of Diocletian.1053 Continued use and abandonment: (A) The fort and baths seem to have been abandoned before the end of the 4th c.1054 Recent radiocarbon dating confirms that fire swept through the camp around AD 360.1055 Plan: (fig. 140) The baths were found 60 m north of the fortress. From a paved courtyard, the bathers entered a cold room (9 × 11 m) that functioned as dressing room and frigidarium. Six stone piers in two rows of three supported the roof. In the north-east corner, a rectangular room housed a square pool (at least 1 m deep). The tepidarium and small caldarium lie on the same axis. The latter has one alveus lying directly above the furnace. The size of the baths could point to a public use for the local population and travellers.1056 Decoration: The walls above the pool were coated with a thick layer of plaster. It seems that no traces of decoration were found. Technology: The heated section comprised two rooms on a hypocaust, heated by a single furnace. The latter was probably surmounted by a boiler, supplying a small alveus in the caldarium. The pilae of the hypocaust were made out of stone monoliths and bricks, while the walls were heated by tubuli.

Continued Use and Abandonment of Imperial Period Baths during Late Antiquity Aila (Tell Abila, Jordan) The large bathhouse in Aila was located in the city centre, next to a large nymphaeum. The baths were built during the High Empire, but were still in use in Late Antiquity.1057 The bad state of preservation does not allow reliable reconstructions of the different phases of use.

1053  Meshel (1989) 229. 1054  Meshel (1989) 237–38. 1055  Davies and Magness (2008) §8. 1056  Davies and Magness (2011) 474. 1057  Mare (1994).

Gazetteer

453 cavated at the time of writing, it was too early to identify late antique interventions. Humeima (Hawara, Jordan) The small bathhouse lay next to a Roman fortress and was built in the shell of a former Nabataean house. In the last phase of baths, dated to Late Antiquity, the caldarium had a small single-person pool.1063 A latrine or possibly a fuel storage area may also have been added during this last phase.1064

figure 140 Plan of the Fortress Bath of Yotvata (Israel) after Davies and Magness (2007) fig. 7

Eleutheropolis (Beth Guvrin, Israel) The city of Eleutheropolis was elevated to the status of a polis during the reign of Septimius Severus and equipped with a variety of public buildings, including a bathhouse (Area 750). It was located at the north-west edge of the city, near the amphitheatre. Important restorations took place in the 4th c., probably to repair damage caused by an earthquake (AD 363?).1058 The palaestra was transformed into a peristyle courtyard and the number of heated rooms, with hypocaust pillars made of round tiles, was reduced.1059 A new room, square in plan, was created at the centre of the bathhouse. It had a more or less oval piscina in its centre, paved with a plain white mosaic. The pool was drained by a sewer which also flushed the latrines. In the last phase of the baths, the pool was transformed into two smaller pools. In the Umayyad period, the building was reused as a fortification.1060 Hippos (Sussita, Israel) The city of Hippos, which was occupied from the Hellenistic until the Umayyad period, had two bathhouses. The large central bathhouse has been identified along the decumanus maximus leading up from the east towards the forum, but the building still awaits excavation.1061 The excavations of the southern baths started in 2007. Preliminary results of the ceramics and small finds from the baths point to an early 3rd-c. date with subsequent alterations later in the course of this century. The baths probably fell out of use in the late 4th or early 5th c.1062 As only the apodyterium had been ex-

1058  Kloner and Cohen (1999) 79*. 1059  Hoss (2005) 129 after an unpublished M.A. thesis. 1060  Kloner and Cohen (1999) 79*. 1061  Segal (2013) 73. 1062  Eisenberg and Jastrzebska (2010) 48.

Horbat Mesar (Israel) A bathhouse dating from the High Empire was discovered during salvage excavations in Horbat Mesar in 2003.1065 The stratigraphy placed the construction date somewhere at the beginning of the 2nd c. During Late Antiquity, the apse of the original caldarium was blocked, a new alveus was constructed and the floor was paved with flagstones. The baths fell out of use in the Early Islamic period (7th or 8th c.), when the building was used as a garbage dump.1066 Jerusalem / Aelia Capitolina / Aelia Commodiana (Israel / Palestine) The site of modern-day Jerusalem / Al-Quds has been inhabited ever since the 2nd millennium BC. The Romans started to interfere with local power as early as the 1st c. BC, when Pompey settled a dispute between two Hasmonaean brothers and plundered the Temple treasury for services rendered. When the Parthians took the city in 40 BC, Caius Sosius besieged Jerusalem and installed Herod—later nicknamed ‘the Great’—as a Roman client king. Under the latter’s rule (37–4 BC), the city expanded and took a more ‘Roman appearance’. In AD 66, the First Jewish Revolt ended with a 5-month siege and the eventual destruction of large parts of the city and the Temple.1067 The Legio X Fretensis was henceforth stationed in the city. Hadrian re-founded Jerusalem as a military colony (Colonia Aelia Capitolina), which has sometimes been interpreted as the cause for the Second Jewish Revolt under Bar Kochba (AD 132–135).1068 We are poorly informed by the ancient sources about the period between the foundation of Aelia and the reign of Constantine. Probably under Septimius Severus, the name of the provincial capital was changed to Aelia Commodiana. Under Diocletian, the Legio X Fretensis was probably stationed elsewhere, meaning a large part of the city became available for civilian settlement.1069 1063  Oleson (1990). 1064  Hoss (2005) 149 quoting an unpublished M.A. thesis. 1065   Tabar and Gendelman (2013) in Hebrew with English summary. 1066  Tabar and Gendelman (2013) 26. 1067  Bieberstein (2007) 134–315. 1068  Bieberstein (2007) 146 with anterior bibliography. 1069  Bieberstein (2007) 159.

454 The accession of Constantine was an important turning point in the history of the city. Jerusalem became an empirewide symbol for Christians and was equipped with several ecclesiastical buildings. It was made a patriarchate by the Council of Chalcedon (AD 451). The bishop became the most powerful leader in the city, having a hand in both ecclesiastical and civil affairs. In contrast, little is known about the civil government of Jerusalem. Overall, Jerusalem and its hinterland enjoyed a peaceful 4th and 5th c., escaping the threats of a Hun invasion in AD 395 and the Isaurian rampage of Syria and Phoenicia in AD 405. The most important revolts were linked to religious groups such as the Monophysites or Origenist monks that came to blows with their rivals. In AD 614, The Sasanians raided the city for a first time, leaving a path of destruction. The Byzantine army briefly retook control of the city, until the Islamic troops arrived at its doors in AD 635.1070 The number of public bathhouses dating from the High Empire is rather low. No large thermae have been found to this day. Unsurprisingly, most of the baths found in Jerusalem date from the Herodian period and were part of private residences. Seven private baths that had been destroyed by the fire of AD 70–71 have been identified.1071 These baths could be labelled as Greek-Hellenistic type bathrooms with singleperson tubs and often complemented by a ritual pool (miqveh). They are characterized by the absence of a Romanstyle hypocaust. The evidence for neighbourhood baths dating from the High Empire is meagre. Near the so-called Wilson’s Arch, at the foot of the Temple Mount, a latrine that may have belonged to a bathhouse was discovered.1072 Just outside the city wall, on the modern Ma’alot Ir David Street, a fragment of round mudbrick with the stamp of the Legio X was retrieved during an excavation.1073 This type of building material was often used to construct the hypocaust pillars of a heated room. The stamp points to a pre-Diocletianic fabrication date of the tile (see above). On the outskirts of the modern city, some better examples have come to light. A bath along the modern Ramallah road (Shu’fat district) was in use in the 1st and 2nd c.1074 On a hill along the Jerusalem-Bethlehem road, illicit excavation in 2009 revealed a Roman bathhouse (Mar Elias site). Some of the tiles used for the construction of the hypocaust were marked with the stamp ‘LEXFR’ (Legio X Fretensis), hinting at a terminus ante quem of the reign of Diocletian. The building had at least two heated rooms and a plastered pool. A coin hoard dating from the turn of the 4th and 5th c. may provide a terminus ante quem for the abandonment of the 1070  Schick (2007) 170–73, 179–80. 1071  Hoss (2005) 152–54, Cat. Nrs. 74–80. 1072  Onn et al. (2011) §24. 1073  Avner (2011) §4. 1074  Sklar-Parnes (2006) §2.

Gazetteer bathhouse.1075 In the late 19th c., a Roman-style bathhouse was excavated in Kidron Valley, south of the ancient city gate.1076 No information is available on the date of construction or its abandonment. A bathhouse on Mount Zion, also excavated in the late 1890s, was built with tiles stamped with the Legio X mark.1077 Pella (Tabaqat Fahl, Jordan) Pella was located at the foot of the hills bordering the western banks of the Jordan, 12 km south-east of Scythopolis. A bathhouse—probably built in the 1st c. AD—was located in the city centre, close to a nymphaeum. A coin from the period AD 582–641 was found under the destruction layers of the building, giving a terminus post quem for the abandonment of the building.1078 However, it may have changed function long before this date. North of the city, the hot springs now known as Hammamat Abu Dhabla have been in use since antiquity. The remains of a bathhouse, probably belonging to a villa, were discovered during construction works in the 1990s. The associated ceramics are dated between the 3rd and 7th c.1079 Petra (Jordan) The Temple Baths in Petra were constructed in the 1st c. AD, and only fell out of use after an earthquake in AD 551. Excavations in 2006 identified several phases, including an important restoration after an earthquake in AD 363. Some rooms were abandoned, while others were paved over. The ornamental pool was filled in and paved, which might point to issues with water supply.1080 A large bathhouse has recently been discovered along the Colonnaded Street in the centre of ancient Petra. Preliminary results of the excavations point to a phase of use between the Nabataean and Byzantine period.1081 The results of the excavations will shed light on the different phases of use and of possible late antique changes made to the original plan. In the outskirts of the city, two baths have recently come to light. The baths on the Jabal Khubthah may still have been in use during the 4th c., but the interpretation of the different phases of the building remains difficult.1082 The baths of Sabra may date from the 1st or 2nd c. In the absence of archaeological data, the plan seems to indicate a construction date during the High Empire, but the large peristyle courtyard may point to a late antique phase.1083 1075  Baruch and Ganor (2014) §5. 1076  Bliss and Dickie (1898) 225–30. 1077  Hoss (2005) 157 with anterior bibliography. 1078  Hoss (2005) 165. 1079  Watson (1996) 73. 1080  Joukowski (2006). 1081  Shiyyab (2014). 1082  Tholbecq et al. (2015) 31. 1083  Fournet and Tholbecq (2015) 42.

455

Gazetteer Ramat Rachel (Israel) On the southern outskirts of modern Jerusalem / Al-Quds, a bathhouse perhaps belonging to a suburban villa was partially excavated in 1960.1084 It was built in the 3rd c. on top of a Herodian gate and remained in use in Late Antiquity, as several small finds on the floors of the rooms and on the bottom of the pools seem to suggest. No important structural changes seem to have been carried out during this long period of use. The floor decoration of this last phase included simple white flagstones. Some 15 m to the north, a small church was built in the 5th or 6th c. The baths were probably destroyed at the same time as the church at the beginning of the 7th c.1085

m south-east of the late antique fort (late 3rd-early 4th c.). The excavations revealed two phases, one in the 3rd and one in the 4th c.1087 The baths were equipped with an apodyterium, a tepidarium, a sudatorium and a caldarium. The earthquake of AD 363 possibly caused significant damage, although the baths seem to have remained in use.1088

Sahir al-Baqar (Jordan) The unexcavated bathhouse might have been in use between the 1st and 4th c., if the association with the phase of use of the Nabataean temple is correct.1086

Umm el-Hajar (between Tell er-Ruqeis and Tell er-Rihan, Palestine) A large bathhouse was discovered during salvage excavations in 1973. At least 5 rooms had a mosaic floor, one of which had a Greek inscription dated to the 5th or 6th c.1091 Several pools and terracotta supply channels were also found, but no information about their location was published.

Tamara (En Hazeva, Israel) The site of ancient Tamara was located on the crossroads of a north-south and an east-west trade route crossing the Negev Desert, near an important spring. A bathhouse was found 50 1084  Aharoni (1962) 24–27; Cascia (1962) 69–72. 1085  Aharoni (1962) 26. 1086  Hoss (2005) 171.

Thana / Thornia (at-Tuwanah, Jordan) The settlement lay on the ancient Via Nova Traiana and was discovered during a survey in 1992.1089 On the basis of ceramic finds, the building recognized as a bathhouse may date to 1st–4th c.1090

1087  Cohen and Israel (1996) 111. 1088  Darby (2015) 70–71. 1089  Fiema (1993). 1090  Hoss (2005) 175. 1091  Biran (1974) 141.

Glossary The following glossary contains Latin and Greek terms that are frequently used in this book. These mainly pertain to baths and bathing habits, although some historical and architectural terms are also explained. The Greek terms are transcribed in Latin for reasons of convenience, the original Greek spelling being added in brackets. Frequently used plural forms are also given for a select number of terms. It should be stressed that the definitions presented for the bath-related terminology are mainly the result of modern scholarship that standardised existing Latin and Greek terms. In the ancient texts, the significance of terms could vary according to author, geographical or chronological contexts of the document (see Rebuffat (1991) 7; Nielsen (1993c)). For more detailed explanations of bath-related terms, see the appendix in Inge Nielsen’s seminal work Thermae et Balnea ((1993a) 153–66), and, more generally, John-Bryan Ward-Perkins’ Roman Imperial Architecture (1981) and René Ginouvès’ Dictionnaire méthodique de l’architecture grecque et romaine. Tome III (1998). For the terminology on bath attendants and staff, see the article by Michael Wissemann (1984). For the terminology of baths and bathing habits in the east, including the Islamic period, see the glossary by Valentine Denizeau and Bérangère Redon in Le bain collectif en Égypte (2009) 399–407. Alousia (ἀλουσία) A term designating literally the state of being unwashed. In the ascetic movement of Early Christianity, the refusal to bathe was considered a virtue, as it avoided confrontation with nudity and underlined the sacrament of baptism as the only necessary washing. Alveus A hot-water pool, mainly found in the heated rooms of Roman-style baths. Most of these pools could accommodate several persons at a time. The hot water was supplied by a boiler in the service rooms, mounted above the furnace of the hypocaust. It was kept warm by means of the hypocaust running under the pool and / or through a testudo. Alternatively, the alveus was heated only by an underlying furnace or by the hypocaust and wall heating. Apodyterium The changing room of a Roman-style bathhouse. These were often equipped with benches, sometimes made out of bricks or stones, but often in wood. Especially in early bathhouses, niches to store clothes were incorporated into the walls. Sometimes, a small fountain or labrum is found in this room. In more luxurious bathhouses, the dressing room was also heated by a hypocaust. Artabè (ἀρτάβη) An originally Persian measure of capacity used in Egypt from the Ptolemaic period onwards. Depending on the region, it equals a capacity between 28 and 60 litres. Balaneion (βαλανεῖον) A term that designates a Greek-style bathhouse. In ancient literature, it could pertain to both

publicly and privately owned baths. The building characteristically consisted of one or more round rooms (tholoi) with single-person bathtubs placed against the walls. These tubs could be hip-baths (or sitz-baths) with an integrated seat or long baths resembling modern tubs. In some Hellenistic examples, large heated pools for communal use have been found. These could be heated by a type of protohypocaust, consisting of a single underground channel that was heated by a furnace. Bal(i)neum (pl. balnea) The original Latin term to denote a bathhouse, both public and private. The meaning of the word shifted during the Early Imperial period, when the term thermae was introduced. The difference between the two terms is still the subject of scholarly debate. Textual and epigraphic evidence does not permit a clear distinction. Possibly, balnea were privately owned and had to buy a permit to use the public water supply. In modern research, balnea is mainly used to designate modest neighbourhood baths. Balneator A vague title that could be used for anyone working in the baths. In smaller baths, there was probably one such balneator who was kind of a jack-of-all-trades. His duties could include anointing, massaging and pouring water over the bathers, or even firing the furnace. In larger baths, these different tasks were performed by numerous specialized staff. Basilica thermarum Large rectangular halls, often with a peristylium and hence resembling a secular basilica. Such halls were usually found in large baths of the imperial type, flanking the frigidarium. The walls often had recesses or apsidal exedrae in which statues were displayed. The basilica stood in connection with the apodyteria and the frigidarium, and was probably used as an exercise area, and more generally, as a meeting place. Bipedales A measuring unit of two Roman feet, equalling 59.2 cm. This measurement was used for large square terracotta tiles, which could be placed on top of the hypocaust pillars. The distance between two pillars was hence restricted to ca. 45 cm. Caldarium The hot room of a Roman-style bathhouse. It could have different forms: in early bathhouses, it often had a rectangular plan with one apsidal ending for the labrum. In later times, it was often cruciform or round. The room was heated by a hypocaust and tubuli, and had at least one warm water pool (alveus) for communal use. The caldarium was oriented to the south to take maximum advantage of the afternoon sun. For this reason, it often projected out from the main building. Calida piscina A large swimming pool heated by a hypocaust. In the centre of the pool, a metal container

Glossary connected to the hypocaust was sunk into the bottom to ensure the continuous heating of the water. These pools are uncommon and mainly found in more luxurious bathhouses (both public and private). They were always confined to a separate room in connection to the tepidarium or the caldarium. Capitolium A temple complex for the divine triad Jupiter, Juno and Minerva, which was located in the centre of a Roman city on the forum. It was named after the Capitoline hill in Rome, where a temple for the trias capitolina had stood since Republican times. Capsarius A cloakroom attendant in the baths. His prime concern was to guard clothes and personal belongings of the bathers. Cardo Maximus The main north-south street of a Roman town. It often intersected with the decumanus maximus at the forum. Claustrum A Latin term literally meaning barrier. In an architectural sense, it can mean a lock or a grate in a window opening. Crusta Meaning literally ‘crust’ or hard shell, the word was also used to denote the pieces of marble or other stone that embellished the surface of floors and walls. Decumanus Maximus The main east-west street of a Roman town. It often intersected with the cardo maximus at the forum. Destrictarium A room in which the bathers could scrape (destringere) the mixture of oil, sweat and dirt off their skin with a strigilis (sickle-shaped scraper). The room had no specific features. The term itself is only attested in an inscription at the Stabian baths in Pompeii (CIL 10.829), but has found its way into the vocabulary of modern research. Exedra An extension of a wall, often semicircular in shape, which could contain benches or niches for statues. The exedrae could be used for a variety of purposes, including use as a library or lecture hall. In baths, they are often found in connection to the palaestra or the perimeter walls (of the thermae). Frigidarium The cold room of a Roman-style bathhouse. It was often the largest room in the baths and contained one or more cold-water pools (piscinae). In smaller baths, the frigidarium could also serve as a dressing room. It was normally rectangular in shape and connected to the apodyterium and the tepidarium. Gymnasium In the Greek-Hellenistic world, the gymnasium was an athletic training ground where the male youth was also educated. It evolved from an open space to an enclosed exercise area with surrounding classrooms and fountain houses. In some larger cities, the gymnasium also had basic washing facilities. During the Roman period, the gymnasium merged with the public bathhouse in Asia Minor, forming a specific type of gymnasium-bath. In Egypt, the term is also used to designate a bathhouse.

457 Heliocaminus The term is used by scholars to denote a room for sunbathing, although the few textual attestations do not mention this word in direct connection to a bathhouse. Alternatively, the word solarium is sometimes used, even if this term normally designates a sundial. Other than being oriented towards the south for maximum sunshine, such a room is also supposed to have had large windows. Hypocaustum The heating system used in Roman-style bathhouses to heat rooms and pools. It consisted of several pillars (pilae) supporting the floor (suspensura). The hollow space underneath the walking-level was heated by one or more furnaces. The hot gasses of these furnaces heated the floor and hence also the room above. The term is a Latinized form of the Greek ὑπόκαυστον, meaning ‘heated from below’. Insula Literally meaning an island, it was also used to denote a tenement for poor people. In modern Roman archaeology, it is used to denote an urban area delimited by streets on all its sides. Labrum (pl. labra) A shallow basin elevated on a pedestal and containing water. The labrum is mainly found in the caldarium or sometimes in the apodyterium. Until the end of the 1st c. AD, it was usual to place the labrum in an apsidal recess (schola labri) to enable several bathers to gather around the washing stand at the same time. The water was supplied by a spout in the centre of the basin or by a spout in the wall behind the labrum. In some cases the water was heated. The diameter of the labrum usually varied from 1 m to 2.5 m. The Greek term is λουτήρ or λουτήριον. Laconicum A round sweat room that was heated by charcoal braziers. This domed room often had seats or benches to accommodate bathers. It usually stood in connection to the apodyterium or tepidarium. After the 1st c. AD, the laconicum disappeared in favour of the sudatorium, which was heated by hypocaust and wall heating. The etymology of the term goes back to the Greek Λακωνικός (‘Laconian’), pointing to the link with the round sweat rooms found in Greek gymnasia. Lavacrum A term that probably denoted ‘a place where one could submerge in water’, but later took on the meaning of bathhouse or pool. In a Christian context, it was often used to designate the sacrament of baptism (lavacrum sacrum and lavacrum regenerationis). It was mainly used during Late Antiquity. Lavatrina A washroom or bath room in a private residence. It was mainly used during the Republican period. Louter (λουτήρ) A shallow basin elevated on a pedestal containing water. It is the Greek equivalent of the Latin labrum. Loutron (λουτρόν) Originally the washroom of a Greek gymnasium, the word gradually took on the meaning of bathhouse during Late Antiquity. It is closely related to verb λουεῖν: ‘to let water flow over something, to wash’.

458 Natatio The large unheated swimming pool which was located in the cold section of baths. It was often unroofed and surrounded by porticoes, with a depth that could reach 1.5 m. Such pools were mainly installed in the larger baths and demanded a considerable supply of water. Nymphaeum Originally a cave in which water surged and which was dedicated to nymphs. In architecture, it denotes a monumental fountain. These could be freestanding or part of another structure. It often had an elaborately decorated façade with columns, niches and statues. In Roman baths, these could form the monumental frontage of the natatio or, in a more modest form, the background of piscinae. Alternatively, they were incorporated into the peri­ meter wall of larger bathhouses. Opus africanum A modern technical term denoting the masonry technique in which walls are constructed by alternating ashlar or orthostatic blocks with spaces of rubble concrete. The technique is mainly found in North Africa, as it was probably invented in the Phoenician and Punic building tradition. Opus caementicium A modern technical term denoting Roman concrete. It consisted of mortar and fragments of rock. The mixture was packed in between temporary timber formworks to create walls or it was used for the infill of brick built walls (opus testaceum and opus listatum). Opus listatum A modern technical term denoting the masonry technique in which the facing of a wall consists of alternating layers of brick with layers of tufa stone. The infill of the wall consists of opus caementicium. This technique is also called opus vittatum in modern literature. Opus quadratum An ancient technical term to denote a construction technique in which walls are constructed with cut blocks of stone laid in regular courses, without the use of mortar. Opus reticulatum An ancient technical term to denote a construction technique in which the facing of a wall was constructed by placing pyramidal tufa blocks in an opus caementicium core, hence creating a reticulated pattern. The corners of the walls were often reinforced with brick pillars (opus mixtum). The technique was mainly used in the Italian Peninsula during the 1st c. AD. Opus sectile A type of mosaic work consisting of slabs of cut stone forming figural patterns. Geometric and floral patterns were very popular. It was often used to embellish pavements. Opus signinum A type of pavement consisting of mortar, small pieces of stone and crushed terracotta, which gave it a pinkish colour. Its surface was often polished. Opus spicatum A technical term denoting a type of pavement, consisting of thin terracotta tiles laid in a herring bone pattern. It was introduced in the Early Imperial period.

Glossary Opus testaceum A modern technical term denoting the masonry technique in which the facing of a wall consists of regular layers of flat (triangular) bricks, imbedded in an opus caementicium core. Otium Latin term to denote vacant time, leisure and freedom of political activity. It was a privilege of the Roman nobility. Palaestra An open courtyard which was attached to the bathhouse, often in front of the frigidarium. The palaestra was rarely paved, but was sometimes laid out as a garden. It was often rectangular in shape and was surrounded by a portico with exedrae. The palaestra was used as an exercise yard and could be equipped with running tracks and sports equipment. The natatio could also be located here. Pila The pillar of a Roman-style hypocaust system that supported the floor of the heated rooms. These pillars had an average height between 60 and 100 cm, and stood some 40 to 60 cm apart. The pillars were often constructed of round or square terracotta tiles bonded by mortar. Alternatively, stone monoliths or hollow terracotta pillars were used. Arches connecting the pillars could offer extra support. Piscina A Latin term for a pool. In modern research, it mainly denotes the cold pools of the frigidarium. The piscinae had internal steps for easy access and could be used by several bathers at the same time. Praefurnium The service room that preceded a furnace. In modern research, it is often erroneously used to designate a furnace. These rooms lay on a lower level than the floors of the bathhouse, as the furnaces had to heat the cavity underneath the suspensura. Schola labri The apse in which a labrum was placed. It enabled several bathers to stand around the labrum at the same time. A window in its semi-dome could give additional light. Solium An alternative term to denote a warm water pool (see alveus) in a Roman-style bathhouse. The word probably derived from the verb sedere (to sit), linking its origin to the Greek-style hip bath tubs that incorporated a seat. Strigilis (pl. strigiles) Sickle-shaped hollow blade made out of metal, wood or bone, used to scrape off dirt, oil and sweat after physical exercises or massages. Sudatorium A sweat room in a Roman-style bathhouse. It was derived from the verb sudare (to sweat). In modern research, sudatorium is used for a room with a hypocaust and wall heating, heated by its own furnace(s). It is often located between the tepidarium and the caldarium. In this room, the bathers sat on benches and sweated out the bad bodily fluids. In most cases, there was no pool in this room. Suspensura The floor supported by the pillars of the hypocaust. This ‘suspended floor’ usually consisted of (bottom to top): bipedales tiles, a thick layer of waterproof mortar, a fine layer of mortar and a marble or mosaic revetment. The total thickness could vary between 30 and 50 cm.

Glossary Taberna Latin term to denote a shop, shed or booth. Tabernae were often built against the outer walls of the palaestra and consisted of one or two rooms open to the street. The tabernae around baths usually sold food, beverages and possibly bath-related items such as oil and nitrum. Tegulae mammatae Meaning literally ‘nipple tiles’, these square or rectangular terracotta tiles had projecting bosses in the corners of one side, so that a cavity was created when placing these tiles vertically against a wall. The tiles were attached with iron T-shaped clamps. The open space between wall and tiles stood in connection with the hypocaust, enabling the hot gasses to rise ‘inside’ the wall. This system of wall heating is mainly found in Republican and Early Imperial times. It was gradually replaced by tubuli. Tepidarium The tepid room of a Roman-style bathhouse. It could have different shapes and was located between the frigidarium and the caldarium or between the frigidarium and the sudatorium. In this room, the bathers acclimatized to the heat before entering the hot rooms. Early examples were heated with charcoal braziers, but from the imperial period onwards, they were heated by a hypocaust. Sometimes, the tepidarium had no wall heating in an effort to reduce the temperature. For the same reason, it was often indirectly heated by the furnace of the sudatorium or caldarium. From the 2nd c. onwards, the tepidarium sometimes had a small pool to enable bathers to have a dip in tepid water, before plunging into the piscina. Tessera A small cube, tag or token. In an architectural context, the term is used to denote the small cubes that were used to form mosaics. The tesserae were often made of stone, glass or terracotta. Testudo (alvei) Latin term meaning literally a tortoise shell, it is also used as a technical term to denote the halfcylindrical metal container that was placed with its flat side above a furnace. It was open at one end and was in direct contact with the bottom of the alveus. As cold water sank, it was heated inside the testudo to rise again. This cyclical process ensured that the water inside the alveus stayed warm. Thermae Latin term to denote a Roman-style bathhouse, it came into use during the 1st c. AD, as a way of demarcating a new type of bathhouse from the type called balneum. There is still uncertainty about the exact difference between both terms. The most recurring hypotheses interpret thermae as the more luxurious bath buildings, often donated by the emperor, or as bathhouses that were run by the civic authorities, enjoying a free water supply. Parameters such as size and the presence of a palaestra do not seem to have been distinguishing features. Thermae aestivales (or aestivae) Latin term to denote Roman-style summer baths, as opposed to thermae hiemales. Both types of baths are often found close to each other or are sometimes structurally linked. Alternative

459 terms are balneum or lavacrum aestivum (see above, the difference between thermae, balnea and lavacra). There is still uncertainty about the distinctive characteristics of these baths. It seems that the summer baths had more and larger pools than the winter baths. In known North African examples, the summer baths also had an open semicircular latrine. Dual summer-winter baths seem to have been popular in the warmer regions of the Mediterranean, such as the Italian Peninsula, the Iberian Peninsula, North Africa and the eastern provinces. Thermae hiemales Latin term to denote Roman-style winter baths, as opposed to thermae aestivales. Both types of baths are often found close to each other or are sometimes structurally linked. Alternative terms are balneum hiemale or balnearia hiberna (see above, the difference between thermae and balnea). There is still uncertainty about the distinctive characteristics of these baths. It seems that the winter baths were smaller and had fewer and smaller pools than the summer baths. The heated section was also smaller, as large rooms meant more heat loss and hence higher fuel costs in winter. In known North African examples, winter baths had closed rectangular latrines, as opposed to the large semicircular latrines of summer baths. Dual summer-winter baths seem to have been popular in the warmer regions of the Mediterranean, such as the Italian Peninsula, the Iberian Peninsula, North Africa and the eastern provinces. Tholos (θόλος) A round room that was covered by a dome. This dome could have a circular opening in its centre to permit air and light to enter. In Greek-style baths, such rooms were often equipped with single-person hip-baths. In Roman-style baths, especially early examples, such a round room was first used as a sweat room, heated by a brazier. In the Pompeian and Herculanean baths, this sweat room was later transformed into a cold pool. Tubus fittilus (pl. tubi fittli) Cylindrical or bottle-shaped ceramic building material used to construct walls and especially vaults. They were open at both ends, but wide at the bottom and narrow at the top, so that they could be set upon each other. The oldest examples date from the 3rd c. BC, but they were mainly used from the 2nd c. AD onwards. Often found in North Africa. Tubulus (pl. tubuli) Hollow terracotta tile, often box-like or cylindrical. In Roman-style baths, rows of superimposed tubuli were placed against the wall and connected to the hypocaust to create a heated wall. Chimney flues that were connected with the top row of the tubuli ensured that the hot gasses in the hypocaust were drawn into the tubuli. If the tubuli were not connected to a hypocaust, they served a mere insulating function for the walls. Unctorium The room for anointing, and possibly massaging bathers. It had no special characteristics and could be connected to the palaestra, to the frigidarium or to one of

460 the heated rooms. In some baths, the unctorium may have been heated. Vestibulum An entrance court. The first room when one entered a building. In a bathhouse, it was often used as a lounge to relax before or after bathing.

Glossary Vir clarissimus Literally meaning a ‘highly regarded man’ (i.e. a man of noble rank), it was a title used for members of the senatorial class. After the reforms of Diocletian and Constantine, it became the only remaining title for higher officials. In the later 4th c., the titles of vir illustris and vir spectabilis were considered to be higher titles.

Bibliography Primary Sources Agnellus Lib. Pont. Eccl. Rav. = D. Mauskopf-Deliyannis ed. and transl., Liber pontificalis Ecclesiae Ravennatis (Corpus Christianorum Continuatio Mediaevalis 199) (Turnhout 2006). Alexander of Tralles = T. Puschmann ed. and transl., Alexander von Tralles: Original-Text und Übersetzung nebst einer einleitenden Abhandlung: ein Beitrag zur Geschichte der Medizin (Vienna 1878–1879). Amb. Ep. 18 = PL 16. 971–82 Amb. In Luc. = PL 15. 1527–1850 Amm. Marc. = J. C. Rolfe ed. and transl., Ammianus Marcellinus. History, 3 volumes (Loeb Classical Library 300, 315, 331) (Cambridge, Mass. 1939–1950). Ananias of Shirak = F. C. Conybeare ed. and transl., “Ananias of Shirak (A.D. 600–650), On Easter”, Byzantinische Zeitschrift 6 (1897) 574–84. Anon. Vales. = J. C. Rolfe ed. and transl., Ammianus Marcellinus. History, Volume III, Excerpta Valesiana (Loeb Classical Library 331) (Cambridge, Mass. 1939). Antidotarium Bruxellense = V. Rose ed., Theodori Prisciani Euporiston libri III cum Physicorum fragmento et additamentis pseudo-Theodoreis (Leipzig 1894). AL = N. Kay ed. and transl., Epigrams from the Anthologia latina: text, translation and commentary (London 2006). Anth. Pal. = W. R. Paton. Revised by Michael A. Tueller ed. and transl., Greek Anthology, 5 volumes, (Loeb Classical Library 67, 68, 84, 85, 86) (Cambridge, Mass. 1916–1918 (rev. 2014)). Apophth. Patr. = J. Wortley ed. and transl., The Anonymous Sayings of the Desert Fathers: A Select Edition and Complete English Translation (in Greek and English). (Cambridge 2013). Apul. Met. = J. Arthur Hanson ed. and transl., Apuleius. Metamorphoses (The Golden Ass), 2 volumes (Loeb Classical Library 44) (Cambridge, Mass. 1996). Asterius Amas. Hom. = G. Anderson and E. Goodspeed, revised by R. Pearse transl. Ancient Sermons for Modern Times by Asterius, Bishop of Amasia Ca. 375–405, A.D. (New York 1904, rev. 2003). Ath. V. Ant. = PG 26, 837–976. T. Vivian et al. transl., The Life of Antony: The Greek life of Antony, The Coptic life of Antony, and An encomium on Saint Anthony by John of Shmūn, and A letter to the disciples of Antony by Serapion of Thmuis, by Athanasius of Alexandria (Kalamazoo, Mich. 2003). August. acta seu disputatio contra fortunatum manichaeum = PL 42. 111–30. ad Don. post Coll. = PL 43. 651–90. Conf. = J. Pilkington ed. and transl., The Confessions of St. Augustine (New York 1943).

c. Faust. = PL 42. 207–518 De catechizandis rudibus = PL 40. 309–348 De civ. D. = P. Levine ed. and transl., Augustine. City of God, Volume IV: Books 12–15. (Loeb Classical Library 414) (Cambridge, Mass. 1966). De doctrina Christiana = PL 34. 15–122. De moribus manichaeorum = PL 32. 1309–1378 De Natura Boni Contra Manichaeos = PL 42. 551–72 En. in Ps. = PL 36 Ep.= PL 33 In Evang. Iohan. = PL 35. 1375–2062 Soliloquiorum = PL 32. 867–904 Aul. Gell. Noct. Att. = J. C. Rolfe ed. and transl., Gellius. Attic Nights, 3 volumes (Loeb Classical Library 195, 200, 212) (Cambridge, Mass. 1927). Aur. Vict. Caes. = H. W. Bird ed. and transl., Liber de Caesaribus of Sextus Aurelius Victor (Liverpool 1994). Auson. Ordo nob. urb. = H. G. Evelyn-White ed. and transl., Ausonius. Volume I: Books 1–17 (Loeb Classical Library 96) (Cambridge, Mass. 1919). Auson. Mos. = H. G. Evelyn-White ed. and transl., Ausonius. Volume I: Books 1–17 (Loeb Classical Library 96) (Cambridge, Mass. 1919). Basil, Hex. = PG 29. 1–208 Basil, Hom. = PG 31. 164–617 Caelius Aurelianus = I. E. Drabkin ed. and transl., On acute diseases and On chronic diseases (Chicago 1950). Cassiod., Chron. = PL 69. 1213–1248 Cassiod., Inst. = PL 70. 1105–1150 Cassiod., Var. = Centre Traditio Litterarum Occidentalium ed., Variarum libri xii / Cassiodorus [electronic resource] (Turnhout 2010). Cass. Dio = E. Cary and H. B. Foster ed. and transl., Dio Cassius. Roman History, Volume VIII: Books 61–70 (Loeb Classical Library 176) (Cambridge, Mass. 1925). Cas. Fel. = A. Fraisse ed. and transl., Cassius Felix. De la médicine (Collection des universités de France. Série latine 366) (Paris 2002). Celsus = W. G. Spencer ed. and transl., Celsus. On Medicine, Volume III: Books 7–8 (Loeb Classical Library 336) (Cambridge, Mass. 1938). Chronicle of Edessa = B. Cowper ed. and transl., “The Chronicle of Edessa”, Journal of Sacred Literature (N.S.) 5 (1865) 28–45. Cic. Att. = D. R. Shackleton Bailey ed. and transl., Cicero. Letters to Atticus, Volume I (Loeb Classical Library 7) (Cambridge, Mass. 1999). Cic. Hortensius = L. Straume-Zimmermann ed. and transl., Cicero’s Hortensius (Frankfurt 1976). Cic. Off. = W. Miller ed. and transl., Cicero. On Duties (Loeb Classical Library 30) (Cambridge, Mass. 1913).

462 Claud. Carm. Min. = M. Platnauer ed. and transl., Claudian. On Stilicho’s Consulship 2–3. Panegyric on the Sixth Consulship of Honorius. The Gothic War. Shorter Poems. Rape of Proserpina (Loeb Classical Library 136) (Cambridge, Mass. 1922). Clem. Al., Paid. = PG 8. 247–684. Cod. Iust. = P. Krüger ed. Corpus Iuris Civilis 2: Codex Iustinianus (Berlin 1906). Cod. Theod. = C. Pharr ed. and transl., The Theodosian code and novels, and the Sirmondian constitutions (New York 1952– 1969; rev. 2001). Colloquia Monacensia = G. Goetz ed., Corpus Glossariorum Latinorum III (Leipzig 1893). Commodianus = J. Martin ed., Commodiani carmina (Corpus christianorum. Series latina 128) (Turnhout 1960). Const. Apost. = M. Metzger ed. and transl., Les Constitutions Apostoliques, 3 volumes (Sources chrétiennes 320, 329, 336) (Paris 1985–1987). Cosm. Ind. = PG 88. 51–476. Council in Trullo = P. Schaff, H. Wace and H. Percival ed. and transl., Nicene and Post-Nicene Fathers, Second Series, Vol. 14 (Edinburgh 1900). Cypr. De habitu virginum = PL 4. 439–464. Dig. = J. E. Spruit, R. Feenstra and F. Wubbe ed. and transl., Corpus iuris civilis: tekst en vertaling, 6 volumes (Zutphen 1994–2001). Epiph. Adv. Haeres. = F. Williams ed. and transl., The Panarion of Epiphanius of Salamis. De Fide, 2 volumes (Nag Hamadi and Manichean Studies 63–79) (Leiden 2009–2012). Epiph. De Fide = F. Williams ed. and transl., The Panarion of Epiphanius of Salamis. De Fide, 2 volumes (Nag Hamadi and Manichean Studies 63, 79) (Leiden 2009–2012). Epit. de Caes. = F. Pichlmayr ed., Sexti Aurelii Victoris Liber de Caesaribus: praecedunt Origo gentis romanae et Liber de viris illustribus urbis Romae, subsequitur Epitome de Caesaribus (Leipzig 1966). Eunap. = W. C. Wright ed. and transl., Philostratus, Eunapius. Lives of the Sophists. Eunapius: Lives of the Philosophers and Sophists (Loeb Classical Library 134) (Cambridge, Mass. 1921). Eutr. = F. L. Müller ed. and transl., Eutropii Breviarium ab Urbe Condita: Eutropius, Kurze Geschichte Roms seit Gründung (753 v Chr-364 n Chr) (Stuttgart 1995). Euseb. De Mart. Palest. = PG 20. 1457–1518. Euseb. Hist. eccl. = K. Lake ed. and transl., Eusebius. Ecclesiastical History, 2 volumes (Loeb Classical Library 153 and 265) (Cambridge, Mass. 1926–1932). Euseb. Onomasticon = S. Timm ed. and transl., Eusebius und die Heilige Schrift: die Schriftvorlagen des Onomastikons der biblischen Ortsnamen (Texte und Untersuchungen zur Geschichte der altchristlichen Literatur 166) (Berlin 2010). Euseb. Vit. Const. = A. Cameron and S. Hall ed. and transl., Life of Constantine (Clarendon ancient history series) (Oxford 1999).

Bibliography Evagr. = A. Hübner ed. and transl., Historia ecclesiastica: Kirchengeschichte (Fontes Christiani 57) (Turnhout 2007). Faventinus = H. Plommer ed. and transl., Vitruvius and later Roman building manuals (Cambridge Classical Studies) (Cambridge 1973). Gal., De meth. med. = I. Johnston and G. H. R. Horsley ed. and transl., Galen. Method of Medicine, 3 volumes (Loeb Classical Library 516, 517, 518) (Cambridge, Mass. 2011). Greg. Naz. Or. 28 = PG 36. 213–38. Greg. Naz. Ep. = PG 37. 21–386. Greg. Naz. Or. 43 = PG 36. 493–606. Greg. Naz. Ep. = PG 36. 27–72. Greg. Nyss. Answer to Eunomius’ Second Book = P. Schaff, H. Wace and H. Percival ed. and transl. Nicene and Post-Nicene Fathers, Second Series, Vol. 5 (Edinburgh 1892). Greg. Nyss. Ep. = A. M. Silvas ed. and transl., Gregory of Nyssa: the letters (Supplements to Vigiliae christianae 83) (Leiden 2007). infant. = PG 46. Greg. Tur. Hist. = MGH I. E. Brehaut transl., History of the Franks by Gregory Bishop of Tours (New York 1965). Greg. Dial. = PL 77. 149–430. Ep. = PL 77. 441–1328. Hdt. = A. D. Godley ed. and transl., Herodotus. The Persian Wars, 4 volumes (Loeb Classical Library 117, 118, 119, 120) (Cambridge, Mass. 1920–1925). Inst. Iust. = J. E. Spruit ed. and transl., Corpus iuris civilis: tekst en vertaling, 1: Instituten (Zutphen 1993). Isid. Chron. Mai. = PL 83. 1017–1058. Isid. Etym. = PL 82. 73–728. Itin. Burd. = PL 8 .333. Jer. Ab Abr. = R. Helm and B. Jeanjean, ed. and transl., Chronique: continuation de la Chronique d’Eusèbe années 326–378 suivi de quatre études sur les Chroniques et chronographies dans l’Antiquité tardive (IV e–VI siècles) (Rennes 2004). Jer. Ep. = F. A. Wright ed. and transl., Jerome. Select Letters (Loeb Classical Library 262) (Cambridge, Mass. 1933). Cassian = M. Petschenig and E. Pichery ed. and transl., Jean Cassien. Conférences (Sources chrétiennes 42, 54 bis) (Paris 1881, rev. 2008–2009). Joh. Chrys. Hom. in Act. = PG 60. 13–384. Joh. Chrys. Ad Ill. Cat. = PG 49. 223–42. Joh. Chrys. Hom. ad pop. Ant. = PG 49. 15–222. Joh. Chrys. Consolatio ad Stagyram = PG 47. 423–93. Joh. Chrys. De Chananaea = CPG 4529 Joh. Chrys. De Diab. Tent. = PG 49. 241–76. Joh. Chrys. De inani gloria et de educandis libris = M. Malingrey ed. and transl., Jean Chrysostome. Sur la vaine gloire et l’éducation des enfants (Sources chrétiennes 188) (Paris 1972). Joh. Chrys. De mutat. Nom. Hom. = PG 51. 113–56.

Bibliography Joh. Chrys. De Sanctis Martyribus Sermo = PG 50. 629–40. Joh. Chrys. Hom. in Ep. ad Col. = PG 62. 299–390. Joh. Chrys. Hom. in Eph. = PG 62. 9–176. Joh. Chrys. Hom. in Hebr. = PG 63. 9–236. Joh. Chrys. Ep. ad Olymp. = A.-M. Malingrey ed. and transl., Lettres à Olympias (Sources chrétiennes 13 bis) (Paris 1968). Joh. Chrys. Hom. in Ep. ad Rom. = PG 60. 583–680. Joh. Chrys. Ep. ad Theodor. = PG 47. 277–316. Joh. Chrys. Hom. in Tit. = PG 62. 663–700. Joh. Chrys. Hom. in Ep. I ad Cor. = PG 61. 9–382. Joh. Chrys. Hom. in Ep. I ad Tim. = PG 61. 501–600. Joh. Chrys. Hom. in Ep. II ad Cor. = PG 61. 383–609. Joh. Chrys. Hom. in Ep. II ad Tim. = PG 61. 600–62. Joh. Chrys. Hom. in Matth. = PG 57 Joh. Chrys. Hom. in Joh. = PG 59. 23–482. Joh. Moschus Prat. = PG 87.3, 2851–3112. Joh. Eph. Hist. eccl. = J. B. Chabot, E. W. Brooks and R. Hespel ed., Corpus Scriptorum Christianorum Orientalium, Scriptores Syri 53–54–213 (Leuven 1933–1935, 1989). Jo. Nic. = H. Zottenberg ed. and transl., Chronique de Jean, évêque de Nikiou. Texte éthiopien publiés et traduit (Paris 1883). Jord. Get. = Centre Traditio Litterarum Occidentalium ed., Getica [electronic resource ‘Monumenta Germaniae Historica’] (Turnhout 2010). Josh. Styl. = J. B. Chabot ed., Corpus Scriptorum Christianorum Orientalium, Scriptores Syri 43–66 (Leuven 1927–1949). Julian, Ep. = W. C. Wright ed. and transl., Julian. Letters. Epigrams. Against the Galilaeans. Fragments (Loeb Classical Library 157) (Cambridge, Mass. 1923). Lib., Or. = A. F. Norman ed. and transl., Libanius. Selected Orations, Volume I: Julianic Orations (Loeb Classical Library 451) (Cambridge, Mass. 1969). Lib. Pont. = Centre Traditio Litterarum Occidentalium ed., Liber Pontificalis [electronic resource ‘Monumenta Germaniae Historica’] (Turnhout 2010). Livy = B. O. Foster and F. G. Moore ed. and transl., Livy. History of Rome, 14 volumes (relevant volumes I and VIII) (Loeb Classical Library 114 and 381) (Cambridge, Mass. 1919 and 1949). Lucian, Hippias = A. M. Harmon ed. and transl., Lucian. Phalaris. Hippias or The Bath. Dionysus. Heracles. Amber or The Swans. The Fly. Nigrinus. Demonax. The Hall. My Native Land. Octogenarians. A True Story. Slander. The Consonants at Law. The Carousal (Symposium) or The Lapiths (Loeb Classical Library 14) (Cambridge, Mass. 1913). Malalas, Chron. = J. Thurn ed., Ioannis Malalae Chronographia (Corpus fontium historiae Byzantinae. Series Berolinensis 35) (Berlin 2000). Jeffreys E., M. Jeffreys and R. Scott transl., The Chronicle of John Malalas (Byzantina Australiensa 4) (Melbourne 1986).

463 Marcell. Comes, Chron. = PL 51. 917–48. M. Aur. Med. = C. R. Haines ed. and transl., Marcus Aurelius (Loeb Classical Library 58) (Cambridge, Mass. 1916). Mart. = D. R. Shackleton Bailey ed. and transl., Martial. Epigrams, 3 volumes (Loeb Classical Library 94, 95, 480) (Cambridge, Mass. 1993). Nil., epp. = PG 79 Nov. = J. E. Spruit et al. ed. and transl., Corpus iuris civilis: tekst en vertaling, 11 & 12: Novellen (Zutphen 2011). Olympiodorus (see Photius). Opt. = PL 11. 323–387. Or. Coll. Med. = U. C. Bussemaker and C. V. Daremberg ed. and transl., Oeuvres d’Oribase, 6 volumes (Collection des médecins grecs et latins) (Paris 1851–1876). Pall., Dial. = P. R. Coleman-Norton ed., Palladii Dialogvs de Vita S. Joannis Chrysostomi (Cambridge 1928). Hist. Laus. = PG 34. 991–1278. Palladius, Opus agriculturae = R. Martin ed., Traité d’agriculture (Paris 1976). Philostorgius, Hist eccl. = PG 65. 459–637. Philox. = R. Kitchen transl., The Discourses of Philoxenos of Mabbug: A New Translation and Introduction (Trappist 2013). Phot., Bibl. = PG 103 Plin. HN = H. Rackham, W. H. S. Jones and D. E. Eichholz ed. and transl., Pliny. Natural History, 10 volumes (relevant volumes III, VIII & X) (Loeb Classical Library 353, 418, 394) (Cambridge, Mass. 1938–1962). Plin., Ep. = B. Radice ed. and transl., Pliny the Younger. Letters, Volume I: Books 1–7 (Loeb Classical Library 55) (Cambridge, Mass. 1969). Plut. Vit. = B. Perrin ed. and transl., Plutarch. Lives, 11 volumes (relevant volumes II, III & IV) (Loeb Classical Library 47, 65, 80) (Cambridge, Mass. 1914–1916). Polemius Silvius = Centre Traditio Litterarum Occidentalium ed., Laterculus [electronic resource ‘Monumenta Germaniae Historica’] (Turnhout 2010). Procop., Aed. = H. B. Dewing and G. Downey ed. and transl., Procopius. On Buildings (Loeb Classical Library 343) (Cambridge, Mass. 1940). Procop., Bell. = H. B. Dewing ed. and transl., Procopius. History of the Wars, 5 volumes (Loeb Classical Library 48, 81, 107, 173, 217) (Cambridge, Mass. 1914–1928). Procop., HA. = H. B. Dewing ed. and transl., Procopius. The Anecdota or Secret History (Loeb Classical Library 290) (Cambridge, Mass. 1935). Pseudo-Theodorus = V. Rose ed., Theodori Prisciani Euporiston libri III cum Physicorum fragmento et additamentis pseudoTheodoreis (Leipzig 1894). Quint., Inst. = Quintilian. D. A. Russell ed. and transl., The Orator’s Education, 5 volumes (Loeb Classical Library 124, 125, 126, 127, 494) (Cambridge, Mass. 2002).

464 Sen., Ep. = R. M. Gummere ed. and transl., Seneca. Epistles, 3 volumes (Loeb Classical Library 75, 76, 77) (Cambridge, Mass. 1917–1925). SHA = D. Magie ed. and transl., Historia Augusta, 3 volumes (Loeb Classical Library 139, 140, 263) (Cambridge, Mass. 1921–1932). Sid. Apoll., Ep. = W. B. Anderson ed. and transl., Sidonius. Poems. Letters, 2 volumes (Loeb Classical Library 296, 420) (Cambridge, Mass. 1936–1965). Sid. Apoll., Carm. = W. B. Anderson ed. and transl., Sidonius. Poems. Letters, 2 volumes (Loeb Classical Library 296, 420) (Cambridge, Mass. 1936–1965). Socrates Hist. eccl. = PG 67. 30–842. Sophron. Narr. de mir. Cyr. et Joh. = PG 87.3. 3373–3696. Sozom. Hist. eccl. = PG 67. 843–1130. Stat., Silv. = D. R. Shackleton Bailey, revised by Christopher A. Parrott ed. and transl., Statius. Silvae (Loeb Classical Library 206) (Cambridge, Mass. 2015). Strab, Geogr. = H. L. Jones ed. and transl., Strabo. Geography, 8 volumes (Loeb Classical Library 49, 50, 182, 196, 211, 223, 241, 267) (Cambridge, Mass. 1917–1932). Suet., Galb. / Dom. = J. C. Rolfe ed. and transl., Suetonius. Lives of the Caesars, 2 volumes (Loeb Classical Library 31, 38) (Cambridge, Mass. 1914). Synod of Laodicea = P. Schaff, H. Wace and H. Percival ed. and transl., Nicene and Post-Nicene Fathers, Second Series, Vol. 14 (Edinburgh 1900). Tac. Agr. = M. Hutton and W. Peterson, revised by R. M. Ogilvie, E. H. Warmington and M. Winterbottom ed. and transl., Tacitus. Agricola. Germania. Dialogue on Oratory (Loeb Classical Library 35) (Cambridge, Mass. 1914). Tert., Ad ux. = PL 1. 1273–1304. Tert., De Corona = PL 2. 73–102. Theod., Ep. = PG 83. 1171–1492. Theod., Hist. eccl. = PG 82. 879–1278. Theod., Hist. Rel. = PG 82. 1279–1497. Theodorus Priscianus = V. Rose ed., Theodori Prisciani Euporiston libri III cum Physicorum fragmento et additamentis pseudo-Theodoreis (Leipzig 1894). Theoph., Chron. = C. De Boor ed., Theophanis Chronographia (Leipzig 1883). Veg. Mil. = L. F. Stelten ed. and transl., Epitoma rei militaris (American University studies ser. 17, classical languages and literature XI) (New York 1990). Venantius Fortunatus Carm. = PL 88. 51–596. Victor Vit. = PL 58. 179–260. Vitr. De arch. = F. Granger ed. and transl., Vitruvius. On Architecture, 2 volumes (Loeb Classical Library 251, 280) (Cambridge, Mass. 1931–1934). Zach. Myt. Hist eccl. = E. W. Brooks ed., Corpus Scriptorum Christianorum Orientalium 83 (Leuven 1919). Zosimus = PG 78. 1679–1702.

Bibliography

Modern Sources Abd el-Fattah A. and Abd Al-Razeq M. (2008) “Rapport préliminaire sur le site de Karm Kandara” in Alexandrie médiévale 3, edd. J.-Y. Empereur and C. Décobert (Cairo 2008) 207–228. Abd el-Fattah A. and ed-Din M. S. (2009) “Les bains de ‘Ezbet Fath Allah (Maréotide). Rapport préliminaire—novembre 2007” in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études Urbaines 7) (Cairo 2009) 263–74. Abd el-Maqsoud M. (1984–1985) “Preliminary report on the excavations at Tell el-Farama (Pelusium). First two seasons (1983/4 and 1984/5)”, ASAE 70 (1984–1985) 3–8. Abu Raya R. (2013a) “Kabul (A)”, Excavations and Surveys in Israel 125 [Online]. Available: http://www.hadashot-esi.org .il/report_detail_eng.aspx?id=2207&mag_id=120 [Accessed on: 28/04/2015]. Abu Raya R. (2013b) “Nahf”, Excavations and Surveys in Israel 125 [Online]. Available: http://www.hadashot-esi.org.il/ report_detail_eng.aspx?id=2265&mag_id=120 [Accessed on: 28/04/2015]. Acquaro E., P. Bartolini, A. Ciasca et al. (1973) Prospezione archeologica a Capo Bon I (Collezione di studi fenici 2) (Rome 1973). Ad U. (2005) “Khirbat Burin”, Excavations and Surveys in Israel 117 [Online]. Available: http://www.hadashot-esi.org.il/ report_detail_eng.aspx?id=235&mag_id=110 [Accessed on: 20/05/2015]. Ad U., Barkan D. and Jakoel E. (2009) “Ramat Ha-Sharon”, Excavations and Surveys in Israel 121 [Online]. Available: http://www.hadashot-esi.org.il/report_detail_eng.aspx? id=1008&mag_id=115 [Accessed on: 19/05/2015]. Adam J.-P. (1984) La construction romaine: matériaux et techniques (Paris 1984). Adawi Z. (2005) “Jerusalem, near the Third Wall”, Excavations and Surveys in Israel 117 [Online]. Available: http://www .hadashot-esi.org.il/report_detail_eng.aspx?id=199&mag_ id=110 [Accessed on: 20/05/2015]. Adawi Z. (2012) “Jerusalem, Beit Hanina”, Excavations and Surveys in Israel 124 [Online]. Available: http://www .hadashot-esi.org.il/report_detail_eng.aspx?id=2160&mag_ id=119 [Accessed on: 06/05/2015]. Aharoni Y. (1960) “The 1961 season” in Excavations at Ramat Rahel. Season 1959 and 1960, ed. Y. Aharoni (Rome 1960) 1–25. Aicher P. J. (1995) Guide to the aqueducts of ancient Rome (Wauconda 1995). Akerraz A. (1986) “Les thermes du capitole (Volubilis)”, BAMaroc 16 (1985–1986) 101–120. Aladjem E. (2013) “Nahal Pattish, Survey”, Excavations and Surveys in Israel 125 [Online]. Available: http://www

Bibliography .hadashot-esi.org.il/report_detail_eng.aspx?id=4341&mag_ id=120 [Accessed on: 30/04/2015]. Albertoni M. (1992) “Lastrine di rivestimento dall’antica via di Porta San Lorenzo”, BullCom 94 (Rome 1992) 341–92. Alexander, M. A., Abed A. B., Mansour S. B.-B., et al. (1980) Thuburbo Maius. Les mosaïques de la région du forum (Corpus des mosaïques de Tunisie 1) (Tunis 1980). Alexander M. A., Ben Abed Ben Kader A., Soren D., et al. (1994) Thuburbo Majus. Les mosaïques de la région est. Mise à jour du catalogue de Thuburbo Majus et les environs. Les mosaïques de Aïn Mziger, Bir Chana, Draa Ben Jouder et Zaghouan (Corpus des mosaïques de Tunisie, 2, 4) (Tunis 1994). Allais Y. (1938) Djemilla (Paris 1938). Allais Y. (1953) “Djemila. Le quartier à l’est du forum des Sévères”, RAfr 97 (1953) 48–65. Allen T. J. (2009) “Baiae in Africa. Attitudes to Health and Bathing in the Maghreb” in Contrôle et distribution de l’eau dans le Maghreb antique et médiéval, ed. V. Bridoux (CÉFR 426) (Rome 2009) 165–75. Almagro M., Caballero L. and Zozaya J. (1975) Qusayr ‘Amra. Tesidencia y Banos Omeyas en el desierto de Jordania (Madrid 1975). Alquier J. and Alquier P. (1930) “Les Thermes Romains du Val-d’Or (près de Oued-Athménia)”, Receuil des notices et mémoires de la société archéologique, historique et géographique du département de Constantine 59 (1930) 298–318. Amit D., Seligman J. and Zilberbod I. (2000) “Jerusalem, Mount Scopus (East)”, Excavations and Surveys in Israel 112 (2000) 94–98. Amit D. and Wolff S. (2000) “An Armenian Monastery in the Morasha Neighborhood, Jerusalem” in Ancient Jerusalem Revealed, ed. H. Geva (Jerusalem 2000) 293–98. Amit D., Wolff S. and Gorzalezany A. (1993) “Jerusalem, the Third Wall (Area D)”, Excavations and Surveys in Israel 13 (1993) 80–83. Amitai-Preiss N. (1997) “Arab inscriptions, graffiti and games” in The Roman baths of Hammat Gader: final report, edd. Y. Hirschfeld and N. Amitai-Preiss (Jerusalem 1997) 267–78. Amore G., Adamesteanu D., Borroni V., et al. (1966) Scavi di Caesarea Maritima (Rome 1966). Ampolo C., Carandini A., Pucci G., et al. (1971) “La villa del Casale a Piazza Armerina. Problemi, saggi stratigrafici ed altre ricerche”, MÉFRA 83 (1971) 141–281. Anderson G. and Goodspeed E. (1904) Asterius of Amasea: Sermons (New York 1904). Andreussi M. (1993) “Aventinus Mons”, LTUR 1 (1993) 147–50. Angelis d’Ossat G. (1943) Tecnica costruttiva e impianti delle terme (Civiltà romana 23) (Rome 1943). Anonymous (1982) “Lod”, Excavations and Surveys in Israel 1 (1982) 67.

465 Anonymous (1990) ed. Milano capitale dell’ Impero Romano 286–402 d.C. (Milano, Palazzo Reale, 24 gennaio-22 aprile 1990) (Milan 1990). Anonymous (1991) ed. Les thermes romains: actes de la table ronde (Rome, 11–12 novembre 1988) (CÉFR 142)(Rome 1991). Anonymous (2013) Terme romane di Curinga [online]. Available: http://www.beniculturalicalabria.it/bandi.php? id=38&action=VIEW [Accessed on: 13/07/2016]. Anti C. (1927) “Campagna di scavi a Cyrene nell’estate del 1926. I. I monumenti architettonici e figurati”, AfrIt 1 (1927) 296–316. Arce I. (2007) “Umayyad building techniques and the merging of Roman-Byzantine and Partho-Sassanian traditions: continuity and change” in Technology in Transition A.D. 300–650, edd. L. Lavan, E. Zanini and A. C. Sarantis (late antique Archaeology 4) (Leiden 2007) 491–537. Arce I. (2015) “The Umayyad baths at Amman citadel and Hammam al-Sarah analysis and interpretation: the social and political value of the Umayyad baths”, Syria 92 (2015) 133–68. Arce J. (2005) “Un grupo de situlas decoradas de la Antigüedad tardía: función, cronología, significado”, AnTard 13 (2005) 141–58. Arslan E. (1966) “L’edificio termale romano detto “tempio di Castore e Polluce” presso Curinga (Catanzaro)”, Klearchos 8 (1966) 23–47. Arslan E. and Bossi M. G. (1968) “La Chiesa di S. Tommaso in Pavia nella sua ambiente urbanistica” in Atti del Convegno di Studio sul centro storico di Pavia (4–5 Iuglio 1964), edd. Canepari, L., A. Peroni and G. Testa (Pavia 1968) 305–311. Arthur C. (1975) “The Aqueduct of Ptolemais”, LibAnt 11 (1974– 1975) 243–49. Aryamontri D. C. (2009) “Running water: advances in urban water supply during the Roman Empire” in The nature and function of water, baths, bathing, and hygiene from antiquity through the Renaissance, edd. C. Kosso and A. Scott (Technology and change in history 11) (Leiden 2009) 319–36. Ashmawy Ali A. (2009) “The Public Bath of Tell Gomaimah and other Graeco-Roman Baths from the Eastern Delta” in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études urbaines 7) (Cairo 2009) 203–208. Ashmawy Ali A. (2017) “The Lost Graeco-Roman Baths of the Eastern Delta” in Collective Baths in Egypt 2. New Discoveries and Perspectives, ed. B. Redon (Études urbaines 10) (Cairo 2017) 83–98. Augenti A. (1999) “Palatium (età tardoantica)”, LTUR 4 (1999) 39–40. Augenti A. (2000) “Continuity and discontinuity of a seat of power: the Palatine Hill from the fifth to the tenth century” Early medieval Rome and the Christian West: essays in honour of Donald A. Bullough (Leiden 2000) 43–57.

466 Avigad N. (1970) “Excavations in the Jewish Quarter of the Old City of Jerusalem, 1970 (Second Preliminary Report)”, IEJ 20 (1970) 129–140. Avi-Yonah M. (1963) “The Bath of the Lepers at Scythopolis”, IEJ 13 (1963) 325–26. Avi-Yonah M. (1993) “Jeruzalem. The Byzantine period” in The New Encyclopedia of Archaeological Excavations in the Holy Land II, edd. E. Stern, A. Levinzon-Golbo’a and J. Aviram (New York 1993) 768–82. Avner R. (2000) “Barqa”, Excavations and Surveys in Israel 111 (2000) 60–61. Avner R. (2011) “Jerusalem, City of David”, Excavations and Surveys in Israel 123 [Online]. Available: http://www .hadashot-esi.org.il/report_detail_eng.aspx?id=1741&mag_ id=118 [Accessed on: 07/05/2015]. Ayalon E. (1985) “Tel Aviv, HaAretz Museum”, Excavations and Surveys in Israel 4 (1985) 107–108. Baatz D. (1979) “Heizversuch an einer rekonstruierten Kanalheizung in der Saalburg”, SaalbJb 36 (1979) 31–44. Badhi R. (2001) “Remains of a Bathhouse and Other Structures at Baqa-el-Gharbiya”, ‘Atiquot 42 (2001) 325. Bady G. and Foschia L. (2014) “Chrétiens, rabbins, païens dans le même bain? Les bains dans l’Orient romain (IV e– VIIe s.) ou comment s’en accommoder” in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.4) (Cairo 2014) 985–1000. Baker-Brian N. (2011) Manichaeism: an ancient faith rediscovered (London 2011). Baklouti H. (1996) “L’hydraulique de Bulla Regia: les citernes publiques “BR3””, AfrRom 11 (1996) 693–709. Baldassari P. (2008) “Indagini archeologiche a Palazzo Valentini: domus di età imperiale ai margini del foro Traiano”, RendPontAcc 81 (2008) 343–84. Baldini Lippolis I. (2001) La Domus tardoantica: forme e rappresentazioni dello spazio domestico nelle città del Mediterraneo (Studi e scavi 17) (Imola 2001). Ballu A. (1921) “Ruines de Djemila”, RAfr 62 (1921) 200–274. Ballu A. (1917) “Rapport sur les fouilles exécutées en 1916 par le service des monuments historiques de l’Algérie, BAC (1917) 206–300. Ballu A. (1916) “Rapport sur les travaux de fouilles exécutés en 1915 par le service des monuments historiques en Algérie”, BAC (1916) 165–242. Ballu A. (1915) “Rapport sur les fouilles exécutées en 1914 par le service des monuments historiques de l’Algérie”, BAC (1915) 100–144. Ballu A. (1912) “Rapport sur les fouilles exécutées en 1911 par le service des monuments historiques de l’Algérie”, BAC (1912) 467–98.

Bibliography Ballu A. (1910) “Rapport sur les fouilles exécutées en 1909 par le service des monuments historiques de l’Algérie”, BAC (1910) 100–126. Ballu A. (1909) “Rapport sur les fouilles exécutées en 1908 par le service des monuments historiques de l’Algérie”, BAC (1909) 75–111. Ballu A. (1908) “Rapport sur les fouilles éxécutées en 1907 par le service des monuments historiques en Algérie”, BAC (1908) 230–54. Ballu A. (1906) “Fouilles archéologiques d’Algérie en 1905”, BAC (1906) 182–222. Ballu A. (1905) “Rapport sur les travaux de fouilles opérés en 1904 par le service des monuments historiques en Algérie”, BAC (1905) 75–103. Ballu A. (1903a) “Séance de la commission de l’Afrique du Nord”, BAC (1903) CLXXIII–CLXXX. Ballu A. (1903b) Les ruines de Timgad, antique Thamugadi. Nouvelles découvertes (Paris 1903). Bansal N. K. (1999) “Characteristic parameters of hypocaust construction”, Building and Environment 34 (1999), 305–318. Baramki D. C. (1933) “A Byzantine bath at Qalandia”, QDAP 2 (1933) 105–109. Barbera M., Di Pasquale S. and Palazzo P. (2007) “Roma, studi e indagini sul cd. Tempio di Minerva Medica”, The Journal of Fasti Online [Online]. Available: http://www .fastionline.org/docs/FOLDER-it-2007-91.pdf [Accessed on: 27/05/2015]. Barbera M., Magnani Cianetti M. and Barrano S. (2014) “Da Massenzio a Costantino: le indagini in corso nel c.d. tempio di Minerva Medica”, in La villa restaurata e i nuovi studi sull’edilizia residenziale tardoantica: atti del Convegno internazionale del Centro interuniversitario di studi sull’edilizia abitativa tardoantica nel Mediterraneo (CISEM), Piazza Armerina 7–10 novembre 2012, edd. P. Pensabene and C. Sfameni (Bari 2014) 255–66. Barbera M. and Paris R. (1996) “I Balnea” in Antiche Stanze. Un quartiere di Roma imperiale nella zona di Termini, edd. M. Barbera and R. Paris (Milan 1996) 117–71. Barker G. and Mattingly D. J. (1996) Farming the desert: the UNESCO Libyan Valleys Archaeological Survey (Paris/ Tripoli/London 1996). Barmby J. (1895) The Book of Pastoral Rule and Selected Epistles of Gregory the Great, Bishop of Rome (A select library of the Nicene and Post-Nicene Fathers of the Christian Church, Second Series 12) (Edinburgh 1895). Barmby J. (1898) Selected Epistles of Gregory the Great, Bishop of Rome (Books IX.–XIV.) (A select library of the Nicene and Post-Nicene Fathers of the Christian Church, Second Series 13) (Edinburgh 1898). Bar-Nathan R. and Mazor G. (1993) “The Beth She’an Excavation Project (1989–1991). City Center (South) and Tel Iztabba

Bibliography Area. Excavations of the Antiquities Authority Expedition”, Excavations and Surveys in Israel 11 (1993) 33–51. Barney S. A., Lewis W. J., Beach J. A., et al. (2006) The Etymologies of Isidore of Seville (Cambridge 2006). Barnish S. (1996a) “Cassiodorus (Magnus Aurelius Cassiodorus Senator)”, in OCD (1996) 298–99. Barnish S. (1996b) “Gregory” in OCD (1996) 656–57. Barnish S. (1992) Cassiodorus. Selected Variae (Translated Texts for Historians 12) (Liverpool 1992). Barrano S. and Martines M. (2006) “Roma. Via Ariosto. Un contribuito per la conoscenza della topografia antica dell’Esquilino”, The Journal of Fasti Online [Online]. Available: http://www.fastionline.org/docs/FOLDER-it -2006-55.pdf [Accessed on: 09/09/2015]. Bartoccini R. (1929a) “Scavi e rinvenimenti in Tripolitania negli anni 1926–1927”, AfrIt 2 (1929) 77–110. Bartoccini R. (1929b) “Le Terme di Lepcis (Leptis Magna)” (AfrIt IV) (Bergamo 1929). Bartoccini R. (1927) “Rinvenimenti vari di interesse archeologico in Tripolitania (1920–1925)”, AfrIt 1 (1927) 213–48. Baruch Y. (2002) “Jerusalem, the Old City”, Hadashot Arkheologiyot 114 (2002) 75*-76*. Baruch Y. and Ganor A. (2014) “Jerusalem, Mar Elias”, Excavations and Surveys in Israel 126 [Online]. Available: http://www.hadashot-esi.org.il/report_detail_eng.aspx? id=8488&mag_id=121 [Accessed on: 07/05/2015]. Basset S. G. (1996) “Historiae custos: Sculpture and Tradition in the Baths of Zeuxippos”, AJA 100 (1996) 491–506. Bauer F. A. and Heinzelmann M. (2001) “L’eglise episcopale d’Ostie” in Ostia: port et porte de la Rome antique, ed. J.-P. Descœudres (Geneva 2001) 278–82. Becatti G. (1953) “Ostia, Lo sviluppo urbanistico” in Scavi di Ostia I: Topografia Generale, ed. G. Calza (Rome 1953) 91–178. Becatti G. (1961) Scavi di Ostia IV. Mosaici e pavimenti marmorei (Rome 1961). Béjaoui F. (2004) Sbeitla. Ancient Sufetula (Sites and Monuments of Tunisia) (Tunis 2004). Bellini delle Stelle F., Mannari A. and Sabelli R. (1984) Le Terme Romane di Fiesole (Fiesole 1984). Ben Abed Ben Kader A. and Hanoune R. (2010) “Des bains à l’huilerie: la fin des thermes du Cratère à Pupput”, AfrRom 18 (2010) 987–94. Ben Abed-Ben Khader A., Ennaïfer M., Spiro M., et al. (1985) Thuburbo Maius. Les mosaïques de la région des grands thermes (Corpus des Mosaïques de Tunisie II.2) (Tunis 1985). Benkheira M. H. (2003) “‘La Maison de Satan’ [Le hammâm en débat dans l’islam médiéval]”, Revue de l’histoire des religions 220 (2003) 391–443. Benkheira M. H. (2007) “Hammam, nudité et ordre moral dans l’islam médiéval (I)”, Revue de l’histoire des religions 224 (2007) 319–71.

467 Benseddik N. (1989) “La pratique médicale en Afrique au temps d’Augustin”, AfrRom 6 (1989) 663–82. Berbrugger A. (1864) “Thermes du centre à Cherchel”, RAfr 8 (1864) 477–79. Berger A. (1982) Das Bad in der byzantinischen Zeit (Miscellanea Byzantina Monacensia 27) (Munich 1982). Berger A. (2011) “Baths in the Byzantine Age” in Bathing culture of Anatolian civilizations: architecture, history, and imagination, ed. N. Ergin (Ancient Near Eastern Studies. Supplement 37) (Leuven 2011) 49–63. Bermond Montanari G. (1972) “S. Maria di Palazzolo (Ravenna)”, ArhVest 23 (1972) 212–18. Bermond Montanari G. (1984–1985) “Ravenna—1980—lo scavo della Banca Popolare. Relazione Preliminare (1)”, FelRav 127 (1984–1985) 21–36. Bertacchi L. (1981) “Contributo alla conoscenza delle grandi terme di Aquileia”, AquilNost 52 (1981) 37–64. Bertacchi L. (1990) “Aquileia, l’organizzazione urbanistica” in Milano capitale dell’ Impero Romano 286–402 d.C. (Milano, Palazzo Reale, 24 gennaio–22 aprile 1990), ed. Anonymous (Milan 1990) 209–212. Bertherand A. (1856) “Mosaïque romaine d’Aïn Temouchent (près de Sétif)”, RAfr 1 (1856) 122–23. Bertocchi F. and Bianchini M. (1992) “Terme Lomuscio” in Principi imperatori vescovi: duemila anni di storia a Canosa, ed. R. Cassano (Venice 1992) 736–40. Beschaouch A., Hanoune R. and Thébert Y. (1977) Les ruines de Bulla Regia (CÉFR 28) (Rome 1977). Bessard F. (2008) “Pratiques sanitaires, produits d’hygiène et de soin dans les bains médiévaux (VIIIe–IXe siècles)”, Bulletin d’études orientales 57 (2008) 111–25. Biaggio Simona S. and Janke R. (2008) “Muralto (Cantone Ticino): Aspetti dell’attività produttiva in un vicus ai piedi delle Alpi”, ZSchwArch 65 (2008) 103–104. Bianchi-Bandinelli R., Caffarelli E. V. and Caputo G. (1966) The buried city: excavations at Leptis Magna (London 1966). Bieberstein K. (2007) “Aelia Capitolina” in Jerusalem before Islam, edd. Z. Kafafi and R. Schick (BAR-IS 1699) (Oxford 2007) 134–68. Bigi F. and Tantillo I. (2010) “Il riempiego: le molte vite delle pietre di Leptis” in Leptis Magna. Una città e le sue iscrizioni in epoca tardoromana, edd. I. Tantillo and F. Bigi (Cassino 2010) 253–302. Billig Y. (2005) “‘En Qobi”, Excavations and Surveys in Israel 117 [Online]. Available: http://www.hadashot-esi.org.il/ report_detail_eng.aspx?id=274&mag_id=110 [Accessed on: 20/05/2015]. Biran A. (1974) “Tell er-Ruqeish to tell er-Ridan”, IEJ 34 (1974) 141–42. Biscop J.-L. and Blanc P.-M. (2014) “Les bains de Télanissos, entre village et sanctuaire” in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international

468 Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 413–32. Blanchard-Lemée M. (1975) Maisons à mosaïques du quartier central de Djemila (Cuicul) (Études d’AntAfr) (Aix-en-Provence 1975). Blas de Roblès J.-M. (1999) Libye: grecque, romaine et byzantine (Aix-en-Provence 1999). Bliss F. and Dickie A. (1898) Excavations at Jerusalem 1894–1897 (London 1898). Bloch F. (2014) “The Bath at Gabal Says. Brief Reflections on Function and Water Use” in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 585–94. Bloch H. (1953) “I bolli laterizi nella storia edilizia di Ostia” in Scavi di Ostia I: Topografia generale, ed. G. Calza (Rome 1953) 215–27. Blonski M. (2014a) Se nettoyer à Rome. IIe avant J.-C. -IIe s. après J.-C. Pratiques et enjeux (Études Anciennes 77) (Paris 2014). Blonski M. (2014b) “Comment définir et utiliser des détergents dans le balneum? L’example du ‘nitre’ à l’époque romaine” in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.4) (Cairo 2014) 883–97. Blyth H. (1995) “Economics of Public Baths”, Balnearia 3 (1995) 2–4. Blyth H. (1999) “The consumption and cost of fuel in hypocaust baths” in Roman baths and bathing: proceedings of the first International Conference on Roman Baths held at Bath, England, 30 March–4 April 1992, edd. J. DeLaine and D. E. Johnston (JRA Supplementary Series 37) (Portsmouth 1999) 87–98. Bockmann R. (2013) Capital continuous. A study of Vandal Carthage and Central North Africa from an Archaeological Perspective (Spätantike—Frühes Christentum—Byzanz. Kunst im ersten Jahrtausend) (Wiesbaden 2013). Boersma J. (1985) Amoenissima civitas: Block V:ii at Ostia: description and analysis of the visible remains (Scrinium 1) (Assen 1985). Boersma J. (1991) “The Valesio project. Final interim report. Campaign of 1990”, BABesch 66 (1991) 115–31. Boersma J. (1995) Mutatio Valentia: the late Roman baths at Valesio, Salento (Scrinium 9) (Amsterdam 1995). Boersma J. and Yntema D. (1987) Valesio. Storia di un insediamento apulo dall’Età del Ferro all’epoca tardoromana. Bilanvio delle ricerche dopo tre campagne di scavo (Fasano di Puglia 1987). Boëtsch G. and Tamarozzi F. (2014) “L’eau, entre imaginaire et pratiques sociales autour du Bassin méditerranéen” in

Bibliography 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.4) (Cairo 2014) 1159–170. Boin D. (2013) Ostia in Late Antiquity (Cambridge 2013). Bonacasa Carra R. M. and Scirè F. (2012) “Sabratha: le fasi dell’edificio termale a NO del Teatro attraverso l’analisi delle strutture”, AfrRom 19 (2012) 365–82. Bonacasa N. (1997) “Missione Archeologica a Sabratha e Leptis Magna” in Missioni archeologiche italiane: la ricerca archeological, antropologica, etnologica, ed. R. Marcucci (Rome 1997) 169–78. Bonacasa N. (2004) “Sabratha—Terme di Oceano. Restauro dei mosaici -1999–2002” in Apparati musivi antichi nell’area del Mediterraneo conservazione programmata e recupero: contributi analitici alla carta del rischio: atti del I Convegno internazionale di studi “La materia e i segni della storia”, Piazza Armerina 9–13 aprile 2003, ed. Anonymous (I Quaderni di Palazzo Montalbo 4) (Palermo 2004) 328–31. Bonacasa N. (2010) “Roma in area punica: le terme di Sabratha”, Bollettino di archeologia on line (Volume Speciale— Congresso di Archeologia A.I.A.C. 2008), A9, 6 [Online]. Available: http://www.bollettinodiarcheologiaonline .beniculturali.it/documenti/generale/6_BONACASA.pdf [Accessed on: 12/05/2014]. Bonacasa N. and Carra R. M. (2003) “Gli edifici termali di Sabratha. Nota preliminare”, QAL 18 (2003) 403–419. Bonnet C., el-Samie M. A. and el-Hofez M. A. (2004) “L’ensemble religieux de Tell el-Makhzan. Les campagnes de fouilles de 2003 et 2004”, Cahier de recherches de l’Institut de papyrologie et d’égyptologie de Lille 24 (2004) 47–60. Boraik M. (2013) “A Roman Bath at the Karnak Temples. A Preliminary Report”, Cahiers de Karnak 14 (2013) 33–46. Boraik M. and el-Masekh S. (2012) “A Roman Bath at Karnak. A Preliminary Report”, AncEg 12/6 (2012) 34–49. Botti G. (1899) “Les cisternes d’Alexandrie”, BSRAA 2 (1899) 15–26. Bouchaud C. (2014) “Gestion et utilisation des combustibles végétaux dans les structures thermales. Études carpologiques et anthracologiques de cinq thermes d’époque byzantine et ommeyyade au Proche-Orient” in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 595–610. Boucher J.-P. (1954) “Nouvelles recherches à Tébessa Khalia”, MÉFR 66 (1954) 165–87. Bouet A. (1997) “Les thermes romains: de gros consommateurs d’eau ? Quelques exemples de Narbonnaise et des provinces voisines” in Les aqueducs de la Gaule romaine et des

Bibliography régions voisines, Actes du colloque de Limoges, 16–1 mai 1996, ed. R. Bedon (Limoges 1997) 133–60. Bouet A. (2003a) Les thermes privés et publics en Gaule Narbonnaise. I. Synthèse (CÉFR 320) (Rome 2003). Bouet A. (2003b) Les thermes privés et publics en Gaule Narbonnaise. II. Catalogue (CÉFR 320) (Rome 2003). Bouillot J. (2008) “The physical and climatic dimensions of the Mediterranean Hammāms”, International Journal of Architectural Research 2 (2008) 122–35. Bourdy F. (1992) “Du bon usage des bains d’après Oribase”, in Les Eaux thermales et les cultes des eaux en Gaule et dans les provinces voisines: actes du colloque, 28–30 septembre 1990, Aix-les-Bains, ed. R. Chevalier (Caesarodunum 26) (Tours 1992) 31–38. Boussac M.-F., Fournet T. and Redon B. (2009) edd. Le bain collectif en Egypte. Balaneia, thermae, hammam (Études Urbaines 7) (Cairo 2009). Boussac M.-F., Denoix S., Fournet T. and Redon B. (2014) “Introduction” in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.1) (Cairo 2014) 1–40. Boussac M.-F., Denoix S., Fournet T. and Redon B. (2014) edd. 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009) (Études urbaines 9) (Cairo 2014). Bowes K. (2010) Houses and Society in the Later Roman Empire (London 2010). Branciforti M. G. (2013) Le Terme dell’Indirizzo. Ricerca archeologica e valorizzazione (Palermo 2013). Branciforti M. G. and Guastella, C. (2008) Le Terme della Rotonda di Catania (Palermo 2008). Brandenburg H. (1999) “S. Stefano”, LTUR 4 (1999) 373–77. Brecciaroli Taborelli L. (1974) “Le terme della “Regio VII” a Sabratha”, LibAnt 11–12 (1974) 113–46. Brehaut E. (1965) History of the Franks by Gregory Bishop of Tours (New York 1965). Brödner E. (1983) Die römischen Thermen und das antike Badewesen: eine kulturhistorische Betrachtung (Darmstadt 1983). Brödner E. (1977) “Zur Entwicklungsgeschichte des Bades, 1”, AntW 8 (1977) 45–56. Brödner E. (1978) “Zur Entwicklungsgeschichte des Bades, 2”, AntW 9 (1978) 47–57. Brogiolo G. P. (1993) Brescia altomedievale: urbanistica ed edilizia dal IV al IX secolo (Documenti di archeologia 2) (Mantova 1993). Brogiolo G. P. and Cantino Wataghin G. (1998) edd. Sepolture tra 4 e 8 secolo. 7° Seminario sul Tardo Antico e l’Alto Medioevo in Italia Centro Settentrionale, Gardone Riviera, 24–26 ottobre 1996 (Mantua 1998).

469 Broise H. (1991) “Vitrages et volets des fenêtres thermales à l’époque impériale” in Les thermes romains: actes de la table ronde (Rome, 11–12 novembre 1988), ed. Anonymous (CÉFR 142) (Rome 1991) 61–78. Broise H. (1994) “La pratique du bain chaud par immersion en Sicile et dans la péninsule italique à l’époque hellénistique”, Xenia Antiqua 3 (1994) 17–32. Broise H. (2009) “Entre continuité et rupture. L’introduction des pratiques balnéaires grecques et romaines en Égypte” in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études urbaines 7) (Cairo 2009) 9–15. Broise H. (2015) “Thermes ‘classiques’ et thermes ‘curatifs’. Réflexions sur l’architecture et l’organisation interne des thermes utilisant l’eau des sources thermales chaudes durant l’Antiquité” in Le thermalisme. Approches historiques et archéologiques d’un phénomène culturel et médical, edd. J. Scheid, M. Nicoud, D. Boisseuil and J. Coste (Paris 2015) 45–63. Broise H. and Curie J. (2014) “L’étude de travertins carbonatés du sanctuaire de Jebel Oust (Tunisie)” in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 573–83. Broise H. and Jolivet V. (1991) “Le bain en Etrurie à l’époque hellénistique” in Les thermes romains: actes de la table ronde, Rome, 11.11.1988–12.11.1988, ed. Anonymous (CÉFR 142) (Rome 1991) 79–95. Broise H. and Thébert Y. (1993) Les thermes memmiens, étude architecturale et histoire urbaine (CÉFR 28) (Rome 1993). Brown P. (1988) The body and society: men, women, and sexual renunciation in early Christianity (Lectures on the history of religions. New series no 13) (New York 1988). Browne C. G. (1893) Select Orations of Saint Gregory Nazianzen (A select library of the Nicene and Post-Nicene Fathers of the Christian Church, Second Series 7) (Edinburgh 1893). Browning R. (1996) “Procopius”, OCD (1996) 1251. Brundrett N. and Simpson G. (1997) “Innovation and the baths of Agrippa”, Athenaeum (N.S.) 85 (1997) 220–27. Brusin G. (1922) “Aquileia. Scavi”, NSc (1922) 187–88. Brusin G. (1923) “Aquileia. Scavi di un grande edificio pubblico”, NSc (1923) 224–31. Bruyère B. (1966) Fouilles de Clysma—Qolzoum (Suez) 1930– 1932 (Fouilles de l’Institut français d’archéologie orientale du Caire 27) (Cairo 1966). Bukowiecki E., Dessales H. and Dubouloz J. (2008) Ostie, l’eau dans la ville. Chateaux d’eau et réseau d’adduction (CÉFR 402) (Rome 2008). Bullough D. A. (1966) “Urban Change in Early Medieval Italy: The Example of Pavia”, PBSR 34 (1966) 82–130. Busana M. S. (1995) Oderzo. Forma Urbis. Saggio di topografia antica (Rome 1995).

470 Busch S. (1999) Versus Balnearum: die antike Dichtung über Bäder und Baden im römischen Reich (Stuttgart 1999). Bussemaker U. C. and Daremberg C. V. (1854) Oeuvres d’Oribase. Tome deuxième (Collection des médecins grecs et latins) (Paris 1854). Butler H. C. (1919) Ancient Architecture in Syria. Southern Syria (Leiden 1919). Buzzetti C. (1993) “Castra Equitum Singularium”, LTUR 1 (1993) 246–48. Caggia M. P. (2014) “The Thermùal Building of the Sanctuary of Saint Philip in Hierapolis (Phrygia, Turkey)”, in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 433–52. Cagnat R. (1923) “L’Asclepieium de Lambèse (Numidie)”, AttiPontAcc (Serie III). Memorie I.1 (1923) 81–88. Calza G. (1953) “La Storia di Ostia” in Scavi di Ostia I: Topografia Generale, edd. G. Calza (Rome 1953) 9–90. Campos Carrasco, J. M., Bermejo Meléndez J. and Fernández Sutilo L. (2012) “El balneum del castellum de Tamuda. Análisis arqueoarquitectónico y arqueológico”, AfrRom 19 (2012) 2429–41. Candilio D. (1985) “Terme di Diocleziano—Museo Nazionale Romano: scavo nella palestra nord-occidentale” in Roma Archeologia nel Centro II: la ‘città murata’, ed. A. M. Bietti Sestieri (Rome 1985) 525–32. Candilio D. (1989) “Terme di Diocleziano”, BullCom 92 (1989) 339–40. Candilio D. (1994) “Roma—Indagini archeologiche nell’aula ottagona delle Terme di Diocleziano”, NSc I–II (1994) 165–83. Cantino Wataghin G. and Lambert C. (1998) “Sepolture e città. L’Italia settentrionale tra IV e VIII secolo” in Sepolture tra 4 e 8 secolo. 7° Seminario sul Tardo Antico e l’Alto Medioevo in Italia Centro Settentrionale, Gardone Riviera, 24–26 ottobre 1996, edd. G. P. Brogiolo and G. Cantino Wataghin (Documenti di archeologia 13) (Mantua 1998) 89–114. Caporusso D. (1990) “Milano (Mediolanum). La situazione idrografica di Milano romana” in Milano capitale dell’ Impero Romano 286–402 d.c. (Milano, Palazzo Reale, 24 gennaio-22 aprile 1990), ed. Anonymous (Milan 1990) 94–96. Carboni F. (2003) “Scavi all’esedra nord-orientale delle Terme di Traiano”, BullCom 104 (2003) 65–80. Carboni R., Cicu E., Corrias F., et al. (2012) “Turris Libisonis, Terme Pallottino: nuovi scavi e ricerche”, AfrRom 19 (2012) 2625–44. Cardin L. (2008) Epigrafia a Roma nel primo medioevo (secoli IV–X). Modelli graficie tipologie d’uso (Rome 2008). Carignani A. (1993) “Un piccolo balneum privato: ancora una domus sul Celio?”, RM 100 (1993) 502–505.

Bibliography Carinci F. (1974) “Saggi di scavo nelle Terme dei Cisiarii (Reg. II, Is. II, 3)”, FA 24–25 (1974) 563 (nr. 8341). Carlier O. (2000) “Les enjeux sociaux du corps. Le hammam maghrébin (XIX e–XX e siècle), lieu pérenne, menacé ou recréé”, Annales. Histoire, Sciences Sociales 55 (2000) 1303–333. Carloni C., Puzzo F. and Ventura M. (2014) “Le terme meridionali: il caldarium e le fasi i riutilizzo in èta medievale” in La villa restaurata e i nuovi studi sull’edilizia residenziale tardoantica atti del Convegno internazionale del Centro interuniversitario di studi sull’edilizia abitativa tardoantica nel Mediterraneo (CISEM), Piazza Armerina 7–10 novembre 2012, edd. P. Pensabene and C. Sfameni (Bari 2014) 575–82. Carrez-Maratray J.-Y. and Wagner G. (1993) “Tell el-Kanaïs”, Cahier de recherches de l’Institut de papyrologie et d’égyptologie de Lille 15 (1993) 105–110. Cartocci M. C. (1993) “S. Agatha Gothorum”, LTUR 1 (1993) 24–25. Carton L. (1922) “Les fouilles de Bulla Regia au printemps de 1922”, CRAI 66 (1922) 326–33. Carucci M. (2007) The Romano-African Domus. Studies in space, decoration, and function (BAR-IS 1731) (Oxford 2007). Caruso G. (1993a) “Aqua Alexandrina”, LTUR 1 (1993) 60–61. Caruso G. (1993b) “Terme di Tito”, BullCom 94 (1993) 81–82. Caruso G., Ceccherelli A., Giusbert P., et al. (1990) “Scavi alle terme di Tito”, Archeologia Laziale 10 (1990) 58–67. Carver M. O. H. (1996) “Transitions to Islam: Rôles in the East and South Mediterranean, Fifth to Tenth Centuries A.D.” in Towns in Transition. Urban Evolution in Late Antiquity and the Early Middle Ages, edd. N. Christie and S. T. Loseby (Aldershot 1996) 184–212. Cascia A. (1962) “A hypocaust at Ramat Rachel” in Excavations at Ramat Rahel. Season 1959 and 1960, ed. Y. Aharoni (Rome 1962) 69–72. Caskey J. (1999) “Steam and ‘Sanitas’ in the domestic Realm: Baths and Bathing in Southern Italy in the Middle Ages”, Journal of the Society of Architectural Historians 58 (1999) 170–95. Cassano R. and Bianchini M. (1992) “Le terme Ferrara” in Principi imperatori vescovi: duemila anni di storia a Canosa, ed. R. Cassano (Venice 1992) 730–35. Cassatella A. and Iacopi I. (1990) “Il balneum presso le scalae caci sul Palatino” in Les thermes romains: actes de la table ronde (Rome, 11–12 novembre 1988), ed. Anonymous (CÉFR 142) (Rome 1990) 129–38. Castel G. (2009) “Bain nord de Karanis” in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études Urbaines 7) (Cairo 2009) 229–45. Cattalini D. (1993) “Aqua Iulia”, LTUR 1 (1993) 66–67. Cavalieri M. and Lenzi S. (2015) “Rivestimenti parietali e pavimentali dalle terme d’età imperiale di Cures Sabini: nuovi

Bibliography dati alla luce della ripresa delle ricerche archeologiche” in Atti del XX Colloquio dell’Associazione italiana per lo studio e la conservazione del mosaico (Roma, 19–22 marzo 2014), edd. C. Angelelli, A. Lugari and F. Sposito (Tivoli 2015) 239–46. Cavalieri Manasse G. and Bruno B. (2003) “Edilizia abitativa a Verona” in Abitare in città—Leben in der Stadt. La Cisalpina tra impero e medioevo—Obertitalien zwischen römischer Kaiserzeit und Mittelalter. Convegno tenuto a Roma il quattro e cinque novembre 1999—Kolloquium am vierten und fünften November 1999 in Rom, edd. J. Ortalli and M. Heinzelmann (Palilia Band 12) (Wiesbaden 2003) 47–64. Cecchelli M. (1999) “Ss. Philippus et Iacobus, Basilica (Ss. Apostoli)”, LTUR 4 (1999) 84–86. Cecchelli M. (1996) “S. Maria Maior”, LTUR 3 (1996) 217–18. Cecchini M. G. (1985) “Terme di Carcalla. Campagna di scavo 1982/1983 lungo il lato orientale” in Roma Archeologia nel Centro II: la ‘città murata’, ed. A. M. Bietti Sestieri (Rome 1985) 588–94. Cera C. (2008) “Herdonia nel contesto della società dell’Italia meridionale tardoantica: ruolo e funzione dei complessi termali” in Ordona XI: ricerche archeologiche a Herdonia, edd. G. Volpe and D. Leone (Bari 2008) 539–78. Ceresa Mori A. (1987) “Milano Corso Europa 16. Sondaggio nell’area delle terme di Massimiano”, Notiziario della Soprintendenza archeologica della Lombardia 1986 (1987) 148–49. Ceresa Mori A. (1990a) “Milano (Mediolanum). L’organiz­ zazione urbanistica. Le terme” in Milano capitale dell’ Impero Romano 286–402 d.c. (Milano, Palazzo Reale, 24 gennaio–22 aprile 1990), ed. Anonymous (Milan 1990) 100–101. Ceresa Mori A. (1990b) “Milano. Corso Europa. Largo Corsia dei Servi. Scavi nell’area delle Terme Erculee”, Bolletino di Archeologia 1–2 (1990) 102–104. Ceresa Mori A. and Sartori A. (1997) “Le mura e le terme” in La città e la sua memoria. Milano e la tradizione di sant’Ambrogio, ed. Anonymous (Milan 1997) 28–32. Chambers T. (1889) The Homilies of St. John Chrysostom, Archbishop of Constantinople, on the Epistles of Paul to the Corinthians (A select library of the Nicene and Post-Nicene Fathers of the Christian Church 12) (Edinburgh 1889). Chambon A. (1989) “Notre Dame Monastery”, Excavations and Surveys in Israel 9 (1989) 145–47. Charpentier G. (1994) “Les bains de Sergilla”, Syria 71 (1994) 113–42. Charpentier G. (1995) “Les petits bains proto-byzantins de la Syrie du Nord”, Topoi 5 (1995) 219–47. Charpentier G. (2014) “Les bains d’al-Bāra (2008–2010)” in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 465–93.

471 Charpentier G. and Denoix S. (2011) “‘Bains antiques et médiévaux dans la méditerranée orientale’: un programme de recherche interdisciplinaire” in Hammam: le terme nell’Islam: convegno internazionale di studi, Santa Cesarea Terme, 15–16 maggio 2008, edd. R. D’Amora and S. Pagani (Aquae 9) (Florence 2011) 69–100. Charpentier G. and Duvette C. (2014) “Le complexe monumental de Tyr et les grands thermes de la fin due IVe s.” in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 385–97. Chedé (1882) “Fouilles exécutées à Tébessa-Khella”, Receuil des notices et mémoires de la société archéologique, historique et géographique du département de Constantine 22 (1882) 269–79. Chehab M. (1963) “The Umayyad Palace at ‘Anjar”, Ars Orientalis 5 (1963) 17–25. Chellini R. (2002) Acque sorgive salutari e sacre in Etruria (Italiae regio VII): ricerche archeologiche e di topografia antica (BAR-IS 1067) (Oxford 2002). Chitty D. J. (1966) The desert a city: an introduction to the study of Egyptian and Palestinian monasticism under the Christian Empire (Oxford 1966). Choukri S. A. (2007) “Un quartier d’habitation de l’époque byzantine dans la zone dite ‘el-Nahda’ au sud-est d’Alexandrie” in Proceedings of the ninth International Congress of Egyptologists = Actes du neuvième congrès international des égyptologues: Grenoble, 6–12 septembre 2004, edd. J.-C. Goyon and C. Cardin (Orientalia Lovaniensia analecta 150) (Leuven 2007) 323–31. Christern J. (1976) Das frühchristliche Pilgerheiligtum von Tebessa. Architektur und Ornamuntik einer spätantiken Bauhütte in Nordafrika (Wiesbaden 1976). Christie N. (2006) From Constantine to Charlemagne: an archaeology of Italy, A.D. 300–800 (Aldershot 2006). Christie N. (2011) The fall of the Western Roman Empire: an archaeological and historical perspective (Historical endings) (London 2011). Cinamon G. (2013) “Ha’on”, Excavations and Surveys in Israel 125 [Online]. Available: http://www.hadashot-esi.org.il/ report_detail_eng.aspx?id=4361&mag_id=120 [Accessed on: 28/04/2015]. Ciurletti G. (2003) “Il caso Tridentum” in Abitare in città la Cisalpina tra impero e medioevo: convegno tenuto a Roma il quattro e cinque novembre 1999 = Leben in der Stadt: Oberitalien zwischen römischer Kaiserzeit und Mittelalter: Kolloquium am vierten und fünften November 1999 in Rom, edd. J. Ortalli and M. Heinzelmann (Palilia 12) (Wiesbaden 2003) 37–45. Claridge A. (1985a) “Castel Porziano, la zona del vicus Augustanus 1984”, Archeologia Laziale 7 (1985) 218–23.

472 Claridge A. (1985b) “Il vicus di epoca imperiale nella tenura presidenziale di Castelporziano: indagini archeologiche 1984” in Castelporziano I. Campagna di scavo e restauro 1984, ed. Anonymous (Rome 1985) 71–78. Clédat J. (1909) “Recherches et fouilles au mont Casios et au lac Sirbonis”, CRAI (1909) 764–74. Coarelli F. (1995a) “Forum Boarium”, LTUR 2 (1995) 295–97. Coarelli F. (1995b) “Forum Holitorium”, LTUR 2 (1995) 299. Coarelli F. (1995c) “Basilica Constantiniana, B. Nova” in LTUR 1 (1995) 170–73. Coarelli F. (1999a) “Piscina Publica”, LTUR 4 (1999) 93–94. Coarelli F. (1999b) “Saturnus, aedes”, LTUR 4 (1999) 234–36. Coarelli F. (2008) Roma (Guide archeologiche Laterza 9) (Bari 2008). Coates-Stephens R. (2001a) “Gli impianti ad acqua e la rete idrica urbana” in Atti del colloquio internazionale Il Liber Pontificlais e la storia materiale (Roma 21–22 febbraio 2001), ed. H. Geertman (Meded 60–61) (Rome 2001) 135–53. Coates-Stephens R. (2001b) “Muri dei bassi secoli in Rome: observations on the re-use of statuary in walls found on the Esquiline and Caelian after 1870”, JRA 14 (2001) 217–38. Coates-Stephens R. (2003a) “The Water-Supply of Rome from Late Antiquity to the Early Middle Ages”, Acta ad archaeologiam et atrium historiam pertinentia 17 (2003) 165–86. Coates-Stephens R. (2003b) “Gli acquedotti in epoca tardoantica nel suburbio” in Suburbium: Il suburbio di Roma dalla crisi del sistema delle ville a Gregorio Magno edd. P. Pergola, R. Santagnelli Valenzani and R. Volpe (CÉFR 311) (Rome 2003) 415–36. Coates-Stephens R. (2006) “La committenza edilizia bizantina a Roma dopo la riconquista” in Le città italiane tra la tarda antichità e l’alto Medioevo: atti del Convegno (Ravenna, 26–28 febbraio 2004), ed. A. Augenti (Florence 2006) 299–316. Cocchiaro A. (1988) “Mesange (Brinidisi), MalvindiCampofreddo”, Taras 8 (1988) 167–68. Cocchiaro A., Marinazzo A. and Travaglini A. (1990) “Monete dagli scavi di Brindisi (1984–1988)”, Annali del istituto italiano di numismatica 36 (1990) 81–133. Cohen M. and Israel Y. (1996) “‘En Hazeva—1990–1994”, Excavations and Surveys in Israel 15 (1996) 110–16. Colini A. (1944) Storia e topografia del Celio nell’antichità (AttiPontAcc (Serie III). Memorie 7) (The Vatican 1944). Colledge M. (2000) “Art and Architecture”, CAH 11 (2000) 966–83. Contiero G. (1987) “Terme e battisteri” in Terme romane e vita quotidiana, ed. M. Pasquinucci (Modena 1987) 108–109. Corbier M. (2009) “Timgad, son développement urbain, ses notables” in Urbanisme et urbanisation en Numidie militaire. Actes du colloque organisé les 7 et 8 mars 2008 par l’Université Lyon 3, ed. A. Groslambert (Collection du Centre d’Études et de Recherches sur l’Occident Romain) (Paris 2009) 181–98.

Bibliography Corvo C. (2000) “Le terme” in Volterra. Il teatro e le terme. Gli edifici, lo scavo, la topographia, edd. M. Munzi and N. Terrenato (Florence 2000) 63–106. Cosentini C. (1966) “Le antiche terme di Acireale”, Memorie e rendiconti dell’Accademia di scienze lettere e belle arti degli Zelanti e dei Dafnici 6 (1966) 1–37. Costamagna L. (2009) “Un mosaico di VI secolo sotto Palazzo Mauri a Spoleto” in Atti del XIV colloquio dell’Associazione italiana per lo studio e la conservazione del mosaico per il patrocinio del ministero per i Beni e le attività culturali (Spoleto, 7–9 febbraio 2008), ed. C. Angelelli (Tivoli 2009) 17–26. Courtois C. (1950) Timgad, antique Thamvgadi (Algiers 1950). Courtois C. (1954) “Ruines romaines du Cap Bon”, Carthage 5 (1954) 182–202. Courtois C. (1955) Les Vandales et l’Afrique (Paris 1955). Cowper B. (1865) “The Chronicle of Edessa”, Journal of Sacred Literature (N.S.) 5 (1865) 28–45. Cozza L. (1975) “I recenti scavi delle Sette Sale”, AttiPontAcc (Serie III). Rendiconti 47 (1975) 81–101. Crindle J. (1897) The Christian topography of Cosmas, an Egyptian monk (London 1897). Crislip A. T. (2005) From monastery to hospital: Christian monasticism and the transformation of health care in Late Antiquity (Ann Arbor 2005). Cross F. L. and Livingstone E. (2005) The Oxford dictionary of the Christian Church (Oxford and New York 2005). Crova B. (1956) Le terme romane nella Campania, Atti dell’ VIII Convegno Nazionale di Storia dell’ Architettura (1956) 271–88. Cuffel A. (2009) “Polemicizing women’s bathing among medieval and early modern muslims and christians” in The nature and function of water, baths, bathing, and hygiene from antiquity through the Renaissance, edd. C. Kosso and A. Scott (Technology and change in history 11) (Leiden 2009) 171–88. Cultrera G. (1954) “Siracusa. Il bagno di Dafne”, NSc (1954) 124–30. Dall’Aglio P. L., De Maria S. and Podini M. (2007) “Territory, city, and private life at Suasa”, JRA 20 (2007) 177–201. Dalton O. (1915) The Letters of Sidonius Apollinaris (Oxford 1915). Darby R. (2015) “The late Roman military baths of the Wadi Arabah: a survey of recent archaeological fieldwork”, Syria 92 (2015) 67–84. Darby R. and Darby E. (2010) “Ayn Gharandal Survey and Preservation Project”, AJA 114 (2010) 534–35. Darby R. and Darby E. (2012) “The Ayn Gharandal Archaeological Project: A Preliminary Report on the 2010 and 2011 Seasons”, ADAJ 56 (2012) 405–422. Darby R. and Darby E. (2015) “The Late Roman fort at ‘Ayn Gharandal, Jordan: interim report on the 2009–2014 field seasons”, JRA 28 (2015) 461–70.

Bibliography Daressy G. (1912) “A travers les Kôms du Delta”, ASAE 12 (1912) 169–213. Dauphin C. (1989) “Golan Survey—1988”, Excavations and Surveys in Israel 9 (1989) 6–7. David M. (1996) I pavimenti decorati di Milano antica I secolo A.C.—VI secolo D.C. Volume 1, Testo (Notizie dal chiostro del monastero maggiore. Supplemento XVI) (Milan 1996). David M. (2013) “Un nuovo sectile policromo dalle Terme del Sileno nel quartiere fuori porta Marina a Ostia” in Atti del XVIII Colloquio dell’Associazione italiana per lo studio e la conservazione del mosaico: Cremona, 14–17 marzo 2012, ed. C. Angelelli (Tivoli 2013) 705–716. David M., Carinci M., Graziano S. M. et al. (2014) “Nuovi dati e argomenti per Ostia tardoantica dal Progetto Ostia Marina”, MÉFRA [En ligne] 126. Davies G. and Magness J. (2007) “Yotvata—2006”, Excavations and Surveys in Israel 119 [Online]. Available: http://www .hadashot-esi.org.il/report_detail_eng.aspx?id=477&mag_ id=112 [Accessed on: 19/05/2015]. Davies G. and Magness J. (2008) “Yotvata—2007”, Excavations and Surveys in Israel 120 [Online]. Available: http://www .hadashot-esi.org.il/report_detail_eng.aspx?id=807&mag_ id=114 [Accessed on: 18/05/2015]. Davies G. and Magness J. (2011) “The Roman fort at Yotvata: a foundation under Valens?”, JRA 24 (2011) 469–80. Davoli P. (2017) “A New Public Bath in Trimithis (Amheida, Dakhla Oasis)” in Collective Baths in Egypt 2. New Discoveries and Perspectives, ed. B. Redon (Études urbaines 10) (Cairo 2017) 193–220. De Angelis d’Ossat M. M. (1990) “Como (Comum). L’organ­ izzazione urbanistica” in Milano capitale dell’ Impero Romano 286–402 d.C. (Milano, Palazzo Reale, 24 gennaio22 aprile 1990), ed. Anonymous (Milan 1990) 163. de Bonneville F. (1998) The book of the bath (London 1998, Eng. Translation [1997]). De Caro S. (2001) “L’attività della Soprintendenza archeologica di Napoli e Caserta nel 2000” in Problemi della chora coloniale dall’occidente al Mar Nero: atti del quarantesimo Convegno di studi sulla Magna Grecia, Taranto 29 settembre-3 ottobre 2000, edd. A. Stazio and S. Ceccoli (Taranto 2001) 865–905. De Fine Licht K. (1974) Untersuchungen an den Trajansthermen zu Rom (AnalRom. Supplementum 7) (Copenhagen 1974). De Haan N. (1997) “Nam nihil melius esse quam sine turba lavari. Privatbäder in den Vesuvstädten”, Meded. Antiquity 56 (1997) 205–226. De Haan N. (2005) “Sic calet tamquam furnus. Fornaci e ipocausti” in Omni pede stare: saggi architettonici e circumvesuviani in memoriam Jos de Waele (Naples 2005) 141–46. De Haan N. (2006) “‘… zoveel verfijning en glans’. Privébaden in Romeinse woningen”, Lampas 39 (2006) 343–61.

473 De Haan N. (2007) “Terme Romane. Tipologie tra uso e utilità” in Bains curatifs et bains hygiéniques en Italie de l’antiquité au moyen âge, edd. M. Guérin-Beauvois and J. M. Martin (CÉFR 383) (Rome 2007) 37–51. De Haan N. (2010) Römische Privatbäder: Entwicklung, Verbreitung, Struktur und sozialer Status (Frankfurt and Berlin 2010). De Marinis G., Silvestrini M. and Palermo L. (2008) “Lo Scavo e il Restauro delle Terme di Santa Lucia (Sassoferrato— AN)” in Sentinum 3. Sentinum 295 A.C. Sassoferrato 2006. 2300 Anni Dopo La Battaglia. Una Città Romana Tra Storia E Archeologia. Convegno Internazionale Sassoferrato 21–23 Settembre 2006, ed. M. Medri (Studia Archeologica 163) (Rome 2008) 205–211. De Miranda A. (2010) L’Hammam nell’Islam Occidentale fra l’VIII e il XIV secolo (Studia Archaeologica (Roma) 176) (Rome 2010). De Simone G. F., Lubrano M., Torino M., et al. (2012) “La villa con terme di Pollena Trocchia in località Masseria De Carolis: architettura, abitanti, eruzioni”, Università degli Studi Suor Orsola Benincasa. Annali. Archeologia. Studi e ricerche sul campo (2011–2012) 195–217. De Simone G. F., Perrotta A. and Scarpati C. (2011) “L’eruzione del 472 d.C. ed il suo impatto su alcuni siti alle falde del Vesuvio”, RStPomp 22 (2011) 61–71. De Spirito G. (1999a) “Thermae Olympiadis”, LTUR 5 (1999) 62–63. De Spirito G. (1999b) “Thermae Sallusti/Sallustianae”, LTUR 5 (1999) 63–64. De Spirito G. (1999c) “Thermae Diocletiani (in fonti agiografiche)”, LTUR 5 (1999) 58. De Vos Raaijmakers M. and Attaoui R. (2013) Rus Africum I. Le paysage rural antique autour de Dougga et Téboursouk: cartographie, relevés et chronologie des établissements (Bari 2013). De Vries B. and Lain A. (2006) “The Legionary Baths (Area C10)” in The Roman frontier in central Jordan: final report on the Limes Arabicus Project, 1980–1989, ed. S. T. Parker (Dumbarton Oaks Research Library and Collection) (Washington D.C. 2006) 213–26. DeForest D. (2013) “Emperors, Baths, and Public Space: The Imperial Thermae in Rome’s Late Antique Landscape” in Perspectives on Public Space in Rome, from Antiquity to the Present Day, edd. G. Smith and J. Gadeyne (Farnham and Burlington 2013) 43–64. Degbomont J.-M. (1984) Le chauffage par hypocauste dans l’habitat privé: de la place St-Lambert à Liège à l’Aula Palatina de Trèves (Études et recherches archéologiques de l’Université de Liège 17) (Liège 1984). Degrassi D. (1996) “Insula Tiberina”, LTUR 3 (1996) 99–101. DeLaine J. (1987) “The ‘Cella Solearis’ of the Baths of Caracalla: a Reappraisal”, PBSR 55 (1987) 147–56.

474 DeLaine J. (1988) “Recent research on Roman baths”, JRA 1 (1988) 11–32. DeLaine J. (1989) “Some observations on the transition from Greek to Roman baths in hellenistic Italy”, MeditArch 2 (1989) 111–25. DeLaine J. (1992) “New Models, Old Modes: Continuity and Change in the Design of Public Baths” in Die Römische Stadt im 2. Jahrhundert n. Chr. Der Funktionswandel des öffentlichen Raumes, edd. H.-J. Schalles, V. von Hesberg and P. Zancker (Xantener Berichte 2) (Cologne 1992) 257–75. DeLaine J. (1997) The Baths of Caracalla. Astudy in the design, construction, and economics of large-scale building projects in imperial Rome (JRA Supplementary Series 25) (Portsmouth). DeLaine J. (1999a) “Bathing and Society” in Roman baths and bathing: proceedings of the first International Conference on Roman Baths held at Bath, England, 30 March–4 April 1992, edd. J. DeLaine and D. E. Johnston (JRA Supplementary Series 37) (Portsmouth) 7–16. DeLaine J. (1999b) “Baths. The urban phenomenon” in Roman baths and bathing: Proceedings of the 1st International Conference on Roman Baths held in Bath, England, 30 march–4 april 1992, edd. J. DeLaine and D. E. Johnston (JRA. Supplementary series. 37) (Portsmouth) 157–64. DeLaine J. (2001) “Bricks and mortar: exploring the economics of building techniques at Rome and Ostia” in Economies beyond agriculture in the classical world, edd. D. J. Mattingly and J. Salmon (New York 2001) 230–68. DeLaine J. (2006) “Baths and Bathing in late antique Ostia” in Common ground: archaeology, art, science and humanities, edd. C. Mattusch, A. Donohue and A. Brauer (Oxford 2006) 338–43. DeLaine J. (2007) “Historiography. Origins, evolution and Convergence” in Bains curatifs et bains hygiéniques en Italie de l’antiquité au moyen âge, edd. M. Guérin-Beauvois and J. M. Martin (CÉFR 383) (Rome 2007) 21–35. Delattre A. (1882) “Découverte d’un cimetière chrétien dans les ruines de Carthage”, Missions catholiques 14 (1882) 117–20. Delattre A. (1883) “Fouilles et découvertes dans un ancien cimetière chrétien de Carthage situé près de la Malga”, Missions catholiques 15 (1883) 93–97. Delattre A. (1886) “Fouilles du cimetière chrétien et de la basilique de Damous-Karita, à Carthage”, BAC (1886) 220–40. Delattre A. (1908) “Fouilles dans le flanc sud de la colline de Bordj-Djedid, à Carthage (1908)”, CRAI 52 (1908) 592–601. Delogu P. (2000) “Solium Imperii—Urbs Ecclesiae. Roma fra la tarda antichità e l’Alto Medioevo” in Sedes regiae (ann. 400–800), edd. G. Ripoll and J. M. Gurt (Barcelona 2000) 83–108. Delorme J. (1960) Gymnasion: étude sur les monuments consacrés à l’éducation en Grèce (des origines à l’Empire Romain) (BÉFAR 196) (Paris 1960).

Bibliography Denizeau V. and Redon B. (2009) “Glossaire. Le vocabulaire du bain collectif, du balaneion au hammām” in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études Urbaines 7) (Cairo 2009) 399–407. Denoix S. (1992) Décrire le Caire. Fustât-Misr d’après Ibn Duqmâq et Maqrizi: l’histoire d’une partie de la ville du Caire d’après deux historiens égyptiens des XIV e–XV siècles (Études urbaines 3) (Cairo 1992). Denoix S. (2009) “Des thermes aux hammams: nouveaux modèles ou recompositions?” in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études Urbaines 7) (Cairo 2009) 12–32. Devoulx A. (1875) “Alger. Étude archéologique et topographie sur cette ville aux époques romaine (Icosium), arabe (Djezaïr Ben Maz’renna) et turque (El-Djezair)”, RAfr 19 (1875) 289–428. Dewing H. (1935) Procopius. The Anecdota or Secret History. Translated by H. B. Dewing (Loeb Classical Library 290) (Cambridge, Mass. 1953). Dewing H. (1940) Procopius. On Buildings. General Index. Translated by H. B. Dewing (Loeb Classical Library 343) (Cambridge, Mass. 1940). Di Capua F. (1940) “Appunti su l’origine e sviluppo delle terme romane”, RendNap N.S. 20 (1940) 81–160. Di Capua F. (1941) L’idroterapia ai tempi dell’impero romano (La scienza e la tecnica ai tempi di Roma imperiale 10–11) (Rome 1941). Di Cola V. (2013) “Le fasi edilizie” in Ostia. V, Le terme del nuotatore, cronologia di un’insula ostiense, edd. M. Medri and V. Di Cola (Studi miscellanei 36) (Rome 2013) 113–99. Di Meo S. and Zaccagnini R. (2000) “Roma. S. Croce in Gerusalemme. Nuove acquisizioni topografiche”, BullCom 99 (2000) 246–50. Di Stefano Manzella I. (1999) “Thermae Falerianae”, LTUR 5 (1999) 58–59. Dionisotti A. C. (1982) “From Ausonius’ Schooldays? A Schoolbook and Its Relatives”, JRS 72 (1982) 83–125. Donati P. (1990) “Un vicus in capo al Verbano: Muralto” in Milano capitale dell’ Impero Romano 286–402 d.C., ed. Anonymous (Milan 1990) 243. Dow M. (1996) The Islamic Baths of Palestine (British Academy Monographs in Archaeology 7) (Oxford 1996). Drabkin I. E. (1950) Caelius Aurelianus. On Acute Diseases and On Chronic Diseases (Chicago 1950). Drappier L. (1911–1913) “Les thermes de Thaenae”, Bulletin de la Société archéologique de Sousse 16 (1911–1913) 23–28. Drappier L. (1920) “Les thermes de Thuburbo Maius”, BAC (1920) 55–75. Dubiez A. (1906) “Mosaïques de Sidi Nasseur Allah”, Bulletin de la Société archéologique de Sousse 7 (1906) 20–21.

Bibliography Duffy J. (1984) “Byzantine Medicine in the Sixth and Seventh Century: Aspects of Teaching and Practice”, DOP 38 (1984) 21–27. Dumaine H. (1910) “Bains”, DACL II, c.72–117. Dumas A. (1855) Impressions de voyage, par Alexandre Dumas. Quinze jours au Sinai (Paris 1855). Dunbabin K. M. (1978) The Mosaics of Roman North Africa. Studies in Iconography and Patronage (Oxford 1978). Dunbabin K. M. (1989) “Baiarum grata voluptas. Pleasures and dangers of the baths”, PBSR 57 (1989) 6–46. Dunbabin K. M. (1999) Mosaics of the Greek and Roman World (Cambridge 1999). Durante A. and Gervasini L. (2000) Luni. Zona archeologica e Museo Nazionale (Rome 2000). Duval N. (1971a) Les églises africaines à deux absides: recherches archéologiques sur la liturgie chrétienne en Afrique du Nord. 1. Les basiliques de Sbeitla à deux sanctuaires opposés (basiliques I, II et IV ) (Paris 1971). Duval N. (1971b) “Églises et thermes en Afrique du Nord. Note sur les installations chrétiennes dans les constructions thermales à propos de Madaure et de Mactar”, BAC 7 (1971) 297–317. Duval N. (1982) “L’urbanisme de Sufetula = Sbeitla en Tunisie”, ANRW II 10 (1982) 596–632. Duval N. (1989) “Inventaire des inscriptions latines païennes de Sbeitla”, MÉFRA 101 (1989) 403–88. Duval N. (1990) “Sufetula: l’histoire d’une ville romaine de la haute-steppe à la lumière des recherches récentes” in L’afrique dans l’occident romain (Ier siècle av. J.-C.-IV e siècle ap. J.-C.): actes du colloque organisé par l’Ecole française de Rome sous le patronage de l’Institut national d’archéologie et d’art de Tunis (Rome, 3–5 décembre 1987), ed. Anonymous (CÉFR 134) (Rome 1990) 495–535. Duval N. and Baratte F. (1973) Les ruines de Sufetula: Sbeïtla (Tunis 1973). Dvorjetski E. (2007) Leisure, Pleasure and Healing. Spa Culture and Medicine in Ancient Eastern Mediterranean (Supplements to the Journal for the Study of Judaism 116) (Leiden and Boston 2007). Écochard M. and Le Cœur C. (1943), Les bains de Damas (Damascus 1943). Eger A. (2007) “Age and Male Sexuality: ‘Queer Space’ in the Roman Bath-house?” in Age and Ageing in the Roman Empire, edd. M. Harlow and R. Laurence (JRA Supplementary Series 65) (Portsmouth 2007) 131–52. Eisenberg M. and Jastrzebska E. (2010) “The southern bathhouse (SWL-I)” in Hippos-Sussita 2010. Eleventh season of excavations, edd. A. Segal, M. Schuler and M. Eisenberg (Haifa 2010) 44–48. Ellis S. P. (1989) “Fouilles. Archéologie de sauvetage. Fouilles Britaniques 1988”, CEDAC Carthage bulletin 10 (1989) 13. Elter R. and Hassoune A. (2004) “Le monastère de Saint-Hilarion à Umm-el-’Amr (bande de Gaza) (note d’information)”, CRAI 148 (2004) 359–82.

475 Elter R. and Hassoune A. (2008) “Le complexe du bain du monastère de Saint-Hilarion à Umm el-ʿAmr, première synthèse architecturale”, Syria 85 (2008) 129–44. Empereur J.-Y., Hesse A. and Picard O. (1994) “Alexandrie (Égypte) 1992–1993”, B.C.H 118 (1994) 503–19. Ennabli A. (1976a) “Bulla Regia Tunisia”, PECS (1976) 171–72. Ennabli A. (1976b) “Mactaris Tunisia”, PECS (1976) 540. Ennabli A. (1976c) “Thuburbo Maius Tunisia”, PECS (1976) 916–17. Ennabli A. (1976d) “Thugga Tunisia”, PECS (1976) 917–19. Ennabli A. (1986) “Les thermes du thiase marin de Sidi Ghrib (Tunisie)”, MonPiot 68 (1986) 1–59. Ennabli A. and Neuru L. (1994) “Excavation of the Roman villa at Sidi Ghrib, Tunisia, 1985–1992”, EchCl 38 (N.S. 13) (1994) 207–20. Ennabli L. (1997) Carthage. Une métropole chrétienne du IVe à la fin du VIIe siècle (Études d’Antiquités africaines) (Paris 1997). Ennaïfer M. (1976) La cité d’Althiburos et l’édifice des Asclepieia (Bibliothèque Archéologique 1) (Tunis 1976). Episcopo S. (1993) “S. Balbina, titulus”, LTUR 1 (1993) 155. Episcopo S. (1996a) “Hierusalem, Basilica, Ecclesia”, LTUR 3 (1996) 28. Episcopo S. (1996b) “S. Marcelo”, LTUR 3 (1996) 211. Episcopo S. (1999) “S. Sabina, basilica, titulus”, LTUR 4 (1999) 221–23. Erickson-Gini T. (2011) “Be’er Shema”, Excavations and Surveys in Israel 123 [Online]. Available: http://www.hadashot -esi.org.il/report_detail_eng.aspx?id=1641&mag_id=118 [Accessed on: 05/05/2015]. Espinar Moreno M. (2014) Baños Árabes de Granada y su Provincia. Materiales para la Arqueología y Cultura Material (Annales Academiae Scientiarum Fennicae. Humaniora 367) (Helsinki 2014). Es-Sadra L. (2012) “Transformation du paysage urbain volubilitain à l’époque préislamique”, AfrRom 19 (2012) 637–54. Étienne R. (1960) Le quartier Nord-Est de Volubilis (Paris 1960). Ettlinghausen R. (1972) From Byzantium to Sasanian Iran and the islamic world. Three modes of artistic influence (Leiden 1972). Euzzenat M. (1976) “Banasa (Sidi Ali bou Djenoun) Morocco”, PECS (1976) 139–40. Evelyn-White D. (1919) Ausonius. Volume I (Loeb classical library 96) (London 1919). Fabbricotti E. (1976) “I bagni nelle prime ville romane”, CronPomp 2 (1976) 29–111. Fabian P. (1995) “The late-Roman military camp at Beer Sheba: a new discovery” in The Roman and Byzantine Near East: Some Recent Archaeological Research, ed. J. H. Humphrey (JRA Supplementary Series 14) (Portsmouth) 235–41. Fagan G. G. (1993) “Pliny, Naturalis historia 36, 121 and the number of balnea in early Augustan Rome”, CP 88 (1993) 333–35.

476 Fagan G. G. (1996a) “Sergius Orata. Inventor of the hypocaust?”, Phoenix 50 (1996) 56–66. Fagan G. G. (1996b) “The Reliability of Roman Rebuilding Inscriptions”, PBSR 64 (1996) 81–93. Fagan G. G. (1999a) Bathing in public in the Roman world (Ann Arbor 1999). Fagan G. G. (1999b) “Interpreting the evidence: Did slaves bathe at the baths?” in Roman baths and bathing: proceedings of the first International Conference on Roman Baths held at Bath, England, 30 March–4 April 1992, edd. J. DeLaine and D. E. Johnston (JRA Supplementary Series 37) (Portsmouth) 25–34. Fagan G. G. (2000) “Hygienic conditions in Roman public baths” in Cura aquarum in Sicilia: proceedings of the tenth International Congress on the History of Water Management and Hydraulic Engineering in the Mediterranean Region: Syracuse, May 16–22, 1998, ed. G. C. M. Jansen (BABesch Supplement 6) (Leuven 2000) 281–87. Fagan G. G. (2001) “The genesis of the Roman public bath. Recent approaches and future directions”, AJA 105 (2001) 403–26. Fagan G. G. (2006) “Bathing for Health with Celsus and Pliny the Elder”, The Classical Quarterly 56 (2006) 190–207. Fakharani F. (1983) “Recent Excavations at Marea in Egypt” in Das Römisch-Byzantinische Agypten: Akten des internationalen Symposions 26–30 September 1978 in Trier, ed. Anonymous (Aegyptiaca Treverensia 2) (Mainz 1983) 175–86. Fales F., Maselli Scotti F., Rubinich M., et al. (2003) “Università di Udine. Aquileia: scavi dell’edificio pubblico detto “delle grande terme”. Campagne 2002–2003”, AquilNost 74 (2003) 182–287. Farrington A. (1995) The Roman baths of Lycia: an architectural study (British Institute of Archaeology at Ankara 20) (London 1995). Fasolo M. (2014) Tyndaris e il suo territorio (Rome 2014). Faucher T. and Redon B. (2014) “Le prix de l’entrée au bain en Égypte hellénistique et romaine d’après les données textuelles et numismatiques” in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.3) (Cairo 2014) 835–55. Fendri M. (1964) “Les thermes des mois à Thina”, Cahiers de Tunisie 12 (1964) 47–57. Fendri M. (1985) “Cités antiques et villas romaines de la région sfaxienne, Africa”, Fouilles, monuments et collections archéologiques en Tunisie 9 (1985) 151–208. Fendri N. (1964) “Les Thermes des mois à Thina. Inventaire préliminaire du trésor et présentation des differents types de monnaies”, Cahiers de Tunisie 12 (1964) 59–67. Fentress E. (1989) “Sétif, les thermes du Ve siècle”, AfrRom 6 (1989) 321–37.

Bibliography Fentress E. and Limane H. (2010) “Excavations in Medieval Settlements at Volubilis. 2000–2004” in Miscelánea de historia y cultura material de al-Andalus. Homenaje a Maryelle Bertrand, edd. P. Cressier, I. Montilla Torres, J. R. Sánchez Viciana and A. Vallejo Triano (Cuadernos de Madinat al-Zahra 7) (Cordoba 2010) 105–122. Février P.-A. (1964) “Notes sur le développement urbain en Afrique du Nord, les exemples comparés de Djemila et de Sétif”, Cahiers Archéologiques 14 (1964) 1–47. Février P.-A. (1976) “Cuicul (Djemila) Algeria”, PECS (1976) 249–50. Fiema Z. (1993) “Tuwaneh and the Via Nova Trajana in Southern Jordan. A short note on the 1992 season”, ADAJ 37 (1993) 549–50. Fiorelli G. (1877) “Notizie degli scavi 1877”, NSc (1877) 1–336. Fiorelli G. (1878) “Notizie degli scavi. Gennaio”, NSc (1878) 3–43. Fischer M. (1986) “Khirbet Zikhrin”, Revue Biblique 93 (1986) 269–76. Fischer M. (1994) “Horbat Zikhrin—1987/1989”, Excavations and Surveys in Israel 12 (1994) 40–45. Flavia P., Giuliani R. and Leone D. (2000) “L’area delle terme (Saggio III. 1997–1998)” in Ordona X. Ricerche archeologiche a Herdonia (1993–1998), ed. G. Volpe (Bari 2000) 127–201. Flemming R. (2013) “Baths and Bathing in Greek Medicine” in Greek Baths and Bathing Culture. New discoveries and approaches, edd. S. K. Lucore and M. Trümper (BABesch Supplement 23) (Leuven 2013) 23–32. Floriani Squarciapino M. (1961) “La Synagoga di Ostia”, BdA 46 (1961) 326–37. Floriani Squarciapino M. (1974) “Terme di Porta Marina o della Marciana”, FA 24–25 (1974) 565 (nr. 8343). Floriani Squarciapino M. (1987) “Nuovi mosaici ostiensi”, AttiPontAcc (Serie III). Rendiconti 58 (1987) 87–114. Fontana M. V. (2011) “Il hammām e l’esercizio fisico in epoca ommayade” in Hammam: le terme nell’Islam: convegno internazionale di studi, Santa Cesarea Terme, 15–16 maggio 2008, edd. R. D’Amora and S. Pagani (Aquae 9) (Florence 2011) 101–105. Fontanille M.-T. (1985) “Les bains dans la médecine Gréco-Romaine” in La Médecine en Gaule: villes d’eaux, sanctuaires des eaux, ed. A. Pelletier (Paris 1985) 15–24. Foucher L. (1958a) “Note sur des signatures de mosaïstes”, Carthage 9 (1958) 129–36. Foucher L. (1958b) Thermes romains des environs d’Hadrumete (Institut National d’Archéologie et Arts. Notes et Docu­ ments (Nouvelle Série) 1) (Tunis 1958). Foulché A. L. (2011) “Les Balnea de Rome à travers la Forma Vrbis Severiana et les Catalogues Régionnaires”, MÉFRA [En ligne] 123.2. Available: http://mefra.revues.org/452 [Accessed on: 20/05/2015]. Foulché A. L. (2013) “À propos des thermae et balnea romains: des transferts de propriété ?” in Spolier et confisquer dans

Bibliography les mondes grec et romain, edd. M.-C. Ferriès and F. Delrieux (Chambéry 2013) 441–59. Fournet T. (2012a) “The ancient baths of southern Syria in their Near Eastern context. Introduction to the Balnéorient project” in Spa, sanitas per aquam, Proceedings of the International Frontinus-Symposium on the technical and cultural history of ancient baths, Aachen, March 18–22, 2009, edd. R. Kreiner and W. Letzner (BABesch Supplement 21) (Leuven 2012) 327–36. Fournet T. (2012b) “Thermes impériaux et monumentaux de Syrie du Sud et du Proche-Orient” in Actes du colloque L’eau en Méditerranée de l’Antiquité au Moyen-Âge, edd. J. Jouanna, P. Toubert and M. Zink (Cahier de la villa Kérylos 23) (Paris 2012) 185–246. Fournet T. (2013) “Map: Location of Greek Public Baths” in Greek Baths and Bathing Culture. New discoveries and approaches, edd. S. K. Lucore and M. Trümper (BABesch Supplement 23) (Leuven 2013) 264. Fournet T. and Lepetz S. (2014) “Faire feu de tout bois. L’utilisation des os et des déjections animales comme combustible dans les thermes byzantins de Bosra (Syrie du Sud)” in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient, organisé par l’Institu français de Proche-Orient et la Direction Générale des Antiquités et des Musées de Syrie (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 611–28. Fournet T., Lucore S. K., Redon B., et al. (2013) “Catalog” in Greek Baths and Bathing Culture. New discoveries and approaches, edd. S. K. Lucore and M. Trümper (BABesch Supplement 23) (Leuven 2013) 269–333. Fournet T. and Reddon B. (2009) “Les bains souterrains de Taposiris Magna et le bain de tradition hellénique en Égypte” in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études urbaines 7) (Cairo 2009) 113–38. Fournet T. and Reddon B. (2013) “Heating Systems of Greek Baths. New Evidence from Egypt” in Greek Baths and Bathing Culture. New discoveries and approaches, edd. S. K. Lucore and M. Trümper (BABesch Supplement 23) (Leuven 2013) 239–63. Fournet T. and Redon B. (2017a) “Romano-Byzantine Baths of Egypt: The Birth and Spread of a Little-Known Regional Model” in Collective Baths in Egypt 2. New Discoveries and Perspectives, ed. B. Redon (Études urbaines 10) (Cairo 2017) 279–321. Fournet T. and Redon B. (2017b) “Catalogue of the Small Greek Baths of Egypt” in Collective Baths in Egypt 2. New Discoveries and Perspectives, ed. B. Redon (Études urbaines 10) (Cairo 2017) 437–50. Fournet T., Redon B. and Vanpeene M. (2017) “Catalogue of the Roman and Byzantine Baths of Egypt” in Collective Baths

477 in Egypt 2. New Discoveries and Perspectives, ed. B. Redon (Études urbaines 10) (Cairo 2017) 451–523. Fournet T. and Tholbecq L. (2015) “Les bains de Sabra: un nouvel édifice thermal aux portes de Petra”, Syria 92 (2015) 33–43. Fournier C. (2016) Les Bains d’al-Andalus VIII e–XV e siècle (Histoire) (Rennes 2016). Fowden E. K. (1999) The Barbarian Plain: Saint Sergius between Rome and Iran (Berkeley and Los Angeles 1999). Fowden G. (2004) Qusayr ‘Amra: art and the Umayyad Elite in late antique Syria (The transformation of the classical heritage 36) (Berkeley 2004). Fraisse A. (2002) Cassius Felix. De la Médicine (Collection des Universités de France) (Paris 2002). Frank K. S. (2006) “Cassianus, Iohannes”, New Pauly Online. Available: http://referenceworks.brillonline.com/entries/ brill-s-new-pauly/cassianus-iohannes-e227940 [Accessed on: 09/12/2013]. Frontori I. (2014) “Nora (Pula, Cagliari). Indagini archeologiche nel quartiere delle Terme Centrali, Area E (Campagna di scavo 2013)”, The Journal of Fasti Online [Online]. Available: http://www.fastionline.org/docs/FOLDER-it-2014-303.pdf [Accessed on: 28/05/2015]. Gallo P. (2009) “Un bain à la grecque dans l’île de Nelson” in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études urbaines 7) (Cairo 2009) 65–72. Gambini V. and Catani E. (1976) “Nuove terme bizantine nei dintorni di Cirene”, QAL 8 (1976) 449–63. Garbrecht G. and Manderscheid H. (1994) Die Wasser­ bewirtschaftung römischer Thermen: archäologische und hydrotechnische Untersuchungen. (Mitteilungen / Leichtweiss-Institut für Wasserbau der Technischen Universität Braunschweig 118A) (Braunschweig 1994). Gardner E. (1911) The Dialogues of Saint Gregory, surnamed the Great: Pope of Rome & the first of that name. Divided into Four Books, wherein he entreateth of the Lives and Miracles of the Saints in Italy and of the Eternity of Men’s Souls (London 1911). Garrison M., Foss P. and Wells C. (1993) “A newly-discovered cryptoporticus and bath at Carthage”, JRA 6 (1993) 251–60. Garzya A. (1994) “L’eau dans la littérature médicale de l’antiquité tardive” in L’eau, la santé et la maladie dans le monde grec: actes du colloque organisé à Paris (CNRS et Fondation Singer-Polignac) du 25 au 27 novembre 1992, edd. R. Ginouvès, A.-M. Guimier-Sorbets, J. Jouanna, et al. (B.C.H supplement 28) (Athens 1994) 109–119. Gasperini L. (1988) “Gli Etruschi e le sorgenti termali” in Etruria Meridionale. Conoscenza, conservazione, fruizione. Atti del Convegno, Vitterbo, 29/30 Novembre–1 Dicembre 1985, edd. G. Colonna, C. Bettini and R. A. Staccioli (Rome 1988) 27–35.

478 Gatier P.-L. (2009) “Bains, monastères et pèlerinages au Proche Orient et en Egypte à l’époque protobyzantine (IV e–VII e s. apr. J.-C.)” in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études Urbaines 7) (Cairo 2009) 275–86. Gatti E. and De Spagnolis M. (1980) “Un intervento nel centro storico di Roma. Impianto termale all’estremità della Regio VII”, Archeologia Laziale 4 (1980) 132–41. Gauckler P. (1896) “Le domaine des Laberii à Uthina”, MonPiot 3 (1896) 177–229. Gauckler P. (1902) “Séance de la commission de l’Afrique du Nord”, BAC (1902) CCIII–CCXVII. Gauckler P. (1907) “Les thermes de Gebamund à Tunis”, CRAI 51 (1907) 790–95. Gavault P. (1883a) “Note sur les ruines antiques de Toukria”, RAfr 27 (1883) 231–40. Gavault P. (1883b) “Tipasa”, RAfr 27 (1883) 400–404. Gawlikowski M. (1997) “The Oriental City and the Advent of Islam” in Die Orientalische Stadt: Kontinuität, Wandel, Bruch. 1. Internationales Colloquium der Deutschen Orient-Gesellschaft 9.-10. Mai 1996 in Halle/Saale, ed. G. Wilhelm (Colloquien der Deutschen Orient-Gesellschaft 1) (Berlin 1997) 339–50. Gayraud R.-P. (1998) “Fostat: évolution d’une capitale arabe du VII e au XII e siècle d’après les fouilles d’Istabl `Antar” in Colloque international d’Archéologie islamique, 3–7 février 1993, ed. R.-P. Gayraud (Cairo 1998) 435–60. Genequand D. (2002) “Une mosquée à Qusair Amra”, ADAJ 46 (2002) 583–89. Genequand D. (2009) “Économie de production, affirmation du pouvoir et dolce vita: aspects de la politique de l’eau sous les Omeyyades au Bilad al-Sham”, Stratégies d’acquisition de l’eau et société au Moyen-Orient depuis l’Antiquité [Online]. Available: http://books.openedition .org/ifpo/1321 [Accessed on: 11/08/2015]. Gentili G. (1956) La Villa imperiale di Piazza Armerina (Itinerari dei musei e monumenti d’Italia 87) (Rome 1956). Gering A. (2010) “Genußkultur und Ghettobildung: Ist Ostia repräsentativ für spätantike Metropolen?” in Die antike Stadt im Umbruch: Kolloquium in Darmstadt, 19. Bis 20. Mai 2006, edd. N. Burkhardt and R. Stichel (Wiesbaden 2010) 92–108. Gering A. (2014) “Le ultime fasi della monumentalizzazione del centro di Ostia tardoantica”, MÉFRA [En ligne] 126. Germain S. (1969) Les mosaïques de Timgad: Étude descriptive et analytique (Centre de recherches sur l’Afrique méditerranéenne. Série Archéologie) (Paris 1969). Ghini G. (1985) “Terme Neroniano—Alessandrine” in Roma Archeologia nel Centro II: la ‘città murata’, ed. A. M. Bietti Sestieri (Rome 1985) 395–99. Ghini G. (1999) “Thermae Neronianae”, LTUR 5 (1999) 60–65. Ghiotto A. (2003) “Gli impianti termali” in Amplissimae atque ornatissimae domus: l’edilizia residenziale nelle città della

Bibliography Tunisia romana, edd. S. Bullo and F. Ghedini (Rome 2003) 221–32. Ghirardini G. (1917) “Scavi nell’ area del Palazzo di Teodorico a Ravenna”, Atti e Memorie della Deputazione di Storia Patria per le Province di Romagna 4 (1917) 234–98. Ghirardini G. (1918) “Gli scavi del Palazzo di Teodorico a Ravenna”, Monumenti Antichi pubblicati per cura dell’ Accademia Nazionale dei Lincei 24 (1918) 737–838. Giannotta M. (1980) Metaponto ellenistico romana (Galatina 1980). Gibson R. and Morello R. (2012) Reading the Letters of Pliny the Younger: An Introduction (Cambridge 2012). Gichon M. (1979) “The Roman Bath at Emmaus: Excavations in 1977”, IEJ 29 (1979) 101–110. Gichon M. (1993) “Emmaus. The Southern Baths” in The New Encyclopedia of Archaeological Excavations in the Holy Land II, edd. E. Stern, A. Levinzon-Golbo’a and J. Aviram (New York 1993) 387–89. Gieryn T. (2002) “What buildings do”, Theory and Science 31 (2002) 35–74. Ginouvès R. (1955) “Sur un aspect de l’évolution des bains en Grèce vers le IV e siècle de notre ère”, B.C.H 79 (1955) 135–52. Ginouvès R. (1962) Balaneutike: recherches sur le bain dans l’antiquite grecque (BÉFAR 200) (Paris 1962). Ginouvès R., Coarelli F. and Martin R. (1998) edd. Dictionnaire méthodique de l’architecture grecque et romaine. Tome III, Espaces architecturaux, bâtiments et ensembles (Rome and Athens 1998). Giorgi C. (2013) “’Terme di Elagabalo’. Il balneum tardoantico: studio archeologica e rilievo 3D” in Valle del Colosseo e pendici nord-orientali del Palatino, edd. C. Panella and L. Saguì (Materiali e contesti 1) (Rome 2013) 55–86. Giorgi E. (2012) “Nuovi dat dagli scavi di Suasa sulla genesi e lo sviluppo dell’abitato” in I processi formativi ed evolutivi della città in area adriatica, edd. G. de Marinis, G. M. Fabrini, G. Paci, et al. (BAR-IS 2419) (Oxford 2012) 345–62. Giuliani C. F. and Verduchi P. (1995) “Forum Romanum (età tarda)”, LTUR 2 (1995) 342–43. Goldmann Z. (1993) “Tell Acco” in The New Encyclopedia of Archaeological Excavations in the Holy Land I, edd. E. Stern, A. Levinzon-Golbo’a and J. Aviram (New York 1993) 17–27. Goodchild R. G. (1964) “Teuchira—Arsinoe (Tocra)”, LibAnt 1 (1964) 144–45. Goodchild R. G. (1965) “The unfinished “Imperial” baths of Leptis Magna”, LibAnt 2 (1965) 15–28. Goodchild R. G. (1971) Kyrene und Apollonia (Ruinenstädte Nordafrikas 4) (Zürich 1971). Goodchild R. G. (1976) Apollonia, the port of Cyrene: excavations by the University of Michigan, 1965–1967 (Supplements to LibAnt 4) (Tripoli 1976). Goodchild R. G. and Kraeling C. (1962) “The “odeon” and related structures” in Ptolemais. City of the Libyan Pentapolis, ed. C. Kraeling (Chicago 1962) 89–96.

Bibliography Gorm Andersen F. (1993) “Die Funde” in Gadara—Umm Qēs III. Die byzantinischen Thermen, edd. I. Nielsen, F. Gorm Andersen and S. Holm-Nielsen (Abhandlungen des deutschen Palästinavereins 17) (Wiesbaden 1993) 153–201. Gorzalczany A. (2009) “Ramla (South)”, Excavations and Surveys in Israel 121 [Online]. Available: http://www .hadashot-esi.org.il/report_detail_eng.aspx?id=1254&mag_ id=115 [Accessed on: 19/05/2015]. Gourevitch D. (2011) Pour une archéologie de la médecine romaine (Collection Pathographie 8) (Paris 2011). Grabar O. (1978) City in the Desert. Qasr-al-Hayr East (Cambridge, Mass. 1978). Grabar O. (1993) “Minnim, Horbat” in The New encyclopedia of archaeological excavations in the Holy Land, ed. E. Stern, A. Levinzon-Golbo’a and J. Aviram (Jerusalem 1993) 1050–51. Grassmann H.-C. (1994) “Wirkungsweise und Energieverbrauch der Römischen Thermen in Weissenburg/Bay”, Balnearia 2 (1994) 4–7. Grassmann H.-C. (2011) Die Funktion von Hypokausten und Tubuli in antiken römischen Bauten, insbesondere in Thermen. Erklärungen und Berechnungen (BAR-IS 2309) (Oxford 2011). Gräzer A. (2009) “Hygiène et sécurité dans l’habitat égyptien d’époque pharaonique” in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études urbaines 7) (Cairo 2009) 33–63. Green R. (1996) “Ausonius, Decimus Magnus”, OCD (1996) 223. Grelle F. (1992) “La città tardoantica” in Principi imperatori vescovi: duemila anni di storia a Canosa, ed. R. Cassano (Venice 1992) 821–23. Gros P. (1996) L’architecture romaine: du début du III e siècle av. J.-C. à la fin du Haut-Empire (Les manuels d’art et d’archéologie antiques) (Paris 1996). Grossmann P. (1973) “Abu Mena. Grabungen von 1961 bis 1969”, ASAE 61 (1973) 37–48. Grossmann P. (2009) “Badeeinrichtungen in ägyptischen frühchristlichen Klöstern” in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études urbaines 7) (Cairo 2009) 287–95. Grossmann P., Brooks Hedstrom D. and Mohamad Osman S. M. (2009) ‘Second Report on the Excavation in the Monastery of Apa Shenute (Dayr Anba Shinuda) at Suhag”, DOP 63 (2009) 167–219. Grotzfeld H. (1970) Das Bad im arabisch-islamischen Mittel­ alter: eine kulturgeschichtliche Studie (Wiesbaden 1970). Gsell S. (1894) “Tipasa, ville de la Maurétanie césarienne”, MÉFR 14 (1894) 291–450. Gsell S. (1898) “Chronique archéologique africaine”, MÉFR 18 (1898) 64–140.

479 Gsell S. (1899) “Chronique archéologique africaine”, MÉFR 19 (1899) 35–83. Gsell S. (1901) Les monuments antiques de l’Algérie I (Paris 1901). Gsell S. (1905) “Mosaïque romaine de Sila”, Receuil des notices et mémoires de la société archéologique, historique et géographique du département de Constantine 39 (1905) 1–7. Gsell S. and Graillot H. (1893) “Explorations archéologiques dans le département de Constantine (Algérie). Ruines romaines au nord de l’aurès”, MÉFR 13 (1893) 461–541. Gsell S. and Joly C.-A. (1922) Khamissa, Mdaourouch, Announa II (Paris 1922). Guérin-Beauvois M. (2007) “Les aquae: sujet médico-religieux ou thème littéraire” in Bains curatifs et bains hygiéniques en Italie de l’antiquité au moyen âge, edd. M. Guérin-Beauvois and J. M. Martin (CÉFR 383) (Rome 2007) 93–114. Guérin-Beauvois M. (2015) Le thermalisme romain en Italie: aspects sociaux et culturels aux deux premiers siècles de l’Empire (BÉFAR 364) (Rome 2015). Guérin-Beauvois M. and Martin J. M. (2007) “Introduction Méthodologique” in Bains curatifs et bains hygiéniques en Italie de l’antiquité au moyen âge, edd. M. Guérin-Beauvois and J. M. Martin (CÉFR 383) (Rome 2007) 1–19. Gui I., Duval N. and Caillet J.-P. (1992) Basiliques chrétiennes d’Afrique du Nord: Inventaire des monuments de l’Algérie (Collection des études augustiniennes. Antiquité 129–130) (Paris 1992). Guidobaldi F. (1993) “S. Clemens”, LTUR 1 (1993) 278–79. Guspini L. (2007) “Roma. Nuovi ritrovamenti in Via Ariosto sull’Esquilino”, The Journal of Fasti Online [Online]. Available: http://www.fastionline.org/docs/FOLDER-it -2007-85.pdf [Accessed on: 16/11/2015]. Guzzo P. G. (1976) “Attività dell’uficio scavi sibari nel 1975” in La Magna Grecia nell’età romana: atti del quindicessimo convegno di studi sulla Magna Grecia, Taranto 5–10 ottobre 1975, ed. G. Carratelli (Naples 1976) 615–31. Guzzo P. G. (1986) “Il territorio dei Bruttii dopo il secolo II D.C.” in Società romana e impero tardoantico III, ed. A. Giardina (Bari 1986) 531–41. Håland E. J. (2009) “’Take, Skamandros, my virginity’: ideas of water in connection with rites of passage in Greece, modern and ancient” in The nature and function of water, baths, bathing, and hygiene from antiquity through the Renaissance, edd. C. Kosso and A. Scott (Technology and change in history 11) (Leiden 2009) 109–148. Hamilton R. (1959) Khirbat al Mafjar: an Arabian mansion in the Jordan Valley (Oxford 1959). Hanoune R., Olivier A. and Thébert Y. (1983) “Les thermes au nord-ouest du théâtre” in Recherches archéologiques Franco-Tunisiennes à Bulla Regia I. Miscellanea 1, edd. A. Beschaouch, R. Hanoune and M. Khanoussi (CÉFR 28.I) (Rome 1983) 63–92.

480 Hansen C. G. (2002) Carthage. Results of the Swedish excavations 1979–1983. Vol. I. A Roman bath in Carthage (ActaRom series 4 LIV: I) (Stockholm 2002). Hansen M. F. (2015) The Spolia Churches of Rome. Recycling Antiquity in the Middle Ages (Aarhus 2015). Harrison R. (1962) “An Orpheus Mosaic at Ptolemais in Cyrenaica”, JRS 52 (1962) 13–18. Hartranft C. (1886) The Ecclesiastical History of Sozomen (A select library of the Nicene and Post-Nicene Fathers of the Christian Church, Second Series 2) (Ediburgh 1886). Häuber C. (1990) “Zür Topographie der Horti Maecenatis und der Horti Lamiani auf dem Esquilin in Rom”, Kölner Jahrbuch für Vor- und Frühgeschichte 23 (1990) 11–107. Häuber C. (2014) The Eastern Part of the Mons Oppius in Rome. The Sanctuary of Isis et Serapis in Regio III, the Temples of Minerva Medica, Fortuna Virgo and Dea Syria, and the Horti of Maecenas (BullCom Supplementi 22) (Rome 2014). Haug A. (2003) Die Stadt als Lebensraum. Eine kulturhistorische Analyse zum spätantiken Stadtleben in Norditalien (Internationale Archäologie Band 85) (Rahden 2003). Haynes D. E. L. (1965) An archaeological and historical guide to the pre-Islamic antiquities of Tripolitania (Tripoli 1965). Heather P. (1996) The Goths (The Peoples of Europe) (Oxford 1996). Heinz W. (1983) Römische Thermen: Badewesen und Badeluxus im Römischen Reich (Edition AntW) (Munich 1983). Heinz W. (1996) “Antike Balneologie in späthellenistischer und römischer Zeit: Zur medizinischen Wirkung römischer Bäder”, ANRW II.37 (1996) 2411–32. Heinz W. (2012) “Von der Antike zum Mittelalter: Balneologie im Wandel” in Spa, sanitas per aquam. Proceedings of the International Frontinus-Symposium on the Technical and Cultural History of Ancient Baths, Aachen, March 18–22, 2009, edd. R. Kreiner and W. Letzner (BABesch Supplement 21) (Leuven 2012) 313–18. Heinzelmann M. (2001) “Ostia, Regio III. Untersuchungen in den unausgegrabenen Bereichen des Stadtgebietes. Vorbericht zur dritten Grabungskampagne 2000”, RM 108 (2001) 313–28. Heres T. L. (1979) “La storia edilizia delle Terme di Nettuno (II, 4, 2) ad Ostia II: Lo sviluppo nei secoli IV–V d. C.”, Meded. Antiquity 41 (1979) 35–42. Heres T. L. (1980) “Alcuni appunti sulla basilica cristiana (III, 1, 4) di Ostia antica”, Meded. Antiquity 42 (1980) 87–99. Heres T. L. (1982) Paries. A proposal for a dating system of Late-Antique masonry structures in Rome and Ostia, A.D. 235–600 (Amsterdam 1982). Heres T. L. (1992) “La storia edilizia delle Terme dei Sette Sapienti (III, X 2) ad Ostia antica. Uno studio preliminare”, Meded 51 (1992) 76–113. Hestrin R. (1993) “Beth-Yerah” in The New encyclopedia of archaeological excavations in the Holy Land 1, edd. E. Stern, A. Levinzon-Golbo’a and J. Aviram (Jerusalem 1993) 255–59.

Bibliography Heymans E. (2013) “The Bronze Boxer from the Quirinal revisited: a construction-related deposition of sculpture”, BABesch 88 (2013) 229–44. Hickey T. M. (2008) “Papyrology” in The Oxford Handbook of Byzantine Studies, ed. E. Jeffreys (Oxford 2008) 115–28. Hillenbrand C. (2005) “Muhammed and the rise of Islam” in The New Cambridge Medieval History I, c. 500–c. 700, ed. P. Fouracre (Cambridge 2005) 317–45. Hillenbrand R. (1999) “Anjar and Early Islamic Urbanism” in The Idea and Ideal of the Town between Late Antiquity and the Early Middle Ages, edd. G. P. Brogiolo and B. Ward-Perkins (The Transformation of the Roman World 4) (Leiden 1999) 59–98. Hirschfeld Y. (1993) “Tiberias” in The New Encyclopedia of Archaeological Excavations in the Holy Land IV, edd. E. Stern, A. Levinzon-Golbo’a and J. Aviram (New York 1993) 1464–70. Hirschfeld Y. (1997) “Tiberias”, Excavations and Surveys in Israel 16 (1997) 35–42. Hirschfeld Y. and Amitai-Preiss N. (1997) The Roman baths of Hammat Gader: final report (Jerusalem 1997). Hodge A. T. (1991) Roman Aqueducts & Water Supply (London 1991). Hodgkin T. (1886) The Letters Of Cassiodorus Being A Con­ densed Translation Of The Variae Epistolae Of Magnus Aurelius Cassiodorus Senator (London 1886). Hoffmann M. (1999) Griechische Bäder (Quellen und Forschungen zur Antiken Welt Bd 32) (Munich 1999). Holder-Egger O. (1878) “Agnelli qui et Andreas Liber Pontificalis Ecclesiae Ravennatis” in Scriptores Rerum Langobardorum et Italicarum I. Saec. VI–IX, ed. Anonymous (Monumenta Germaniae Historica) (Hannover 1878). Holm-Nielsen S. (1993) “Verlauf der Ausgrabungen” in Gadara—Umm Qēs III. Die byzantinischen Thermen, edd. I. Nielsen, F. Gorm Andersen and S. Holm-Nielsen (Abhandlungen des deutschen Palästinavereins 17) (Wiesbaden 1993) 1–10. Holm-Nielsen S., Nielsen I. and Andersen F. G. (1986) “The excavation of Byzantine Baths in Umm Qeis”, ADAJ 30 (1986) 219–32. Holum K., Hohlfelder R., Bull R., et al. (1988) King Herod’s Dream. Caesarea on the Sea (New York 1988). Honoré T. (1996a) “Theodosian Code”, OCD (1996) 1501. Honoré T. (1996b) “Justinian’s codification”, OCD (1996) 803–804. Horton F. (1996) “A Sixth-Century Bath in Caesarea’s Suburbs and the Transformation of Bathing Culture in Late Antiquity” in Caesarea Maritima: A Retrospective After Two Millennia, edd. A. Raban and K. Holum (Leiden 1996) 177–89. Hoss S. (2005) Baths and bathing: the culture of bathing and the baths and thermae in Palestine from the Hasmoneans to the Moslem conquest (BAR-IS 1346) (Oxford 2005).

Bibliography Hoss S. (2010) “Die spätantike Blüte römischer Thermen in Palästina” in Die antike Stadt im Umbruch: Kolloquium in Darmstadt, 19. Bis 20. Mai 2006, edd. N. Burkhardt and R. Stichel (Wiesbaden 2010) 165–77. Hoss S. (2012) “From rejection to incorporation. The Roman bathing culture in Palestine” in Spa, sanitas per aquam, Proceedings of the International Frontinus-Symposium on the technical and cultural history of ancient baths, Aachen, March 18–22, 2009, edd. R. Kreiner and W. Letzner (BABesch Supplement 21) (Leuven 2012) 259–64. Hoss S. (2018) “Small Finds in Roman Thermae: an Introduction” in Thermae in context: Roman bathhouses in the town and in daily life, edd. H. Pösche, A. Binsfeld and S. Hoss (Archaeologia Mosellana 10) (Luxembourg City 2018) 47–56. Hubert E. (2007) “Les bains à Rome et dans le Latium au Moyen Age. Textes et archéologie” in Bains curatifs et bains hygiéniques en Italie de l’antiquité au moyen âge, edd. M. Guérin-Beauvois and J. M. Martin (CÉFR 383) (Rome 2007) 127–42. Hudson P. J. (1990) “Pavia (Ticinum), L’organizzazione urbanistic e la topographia christiana” in Milano capitale dell’Impero Romano 286–402 d.C. (Milano, Palazzo Reale, 24 gennaio-22 aprile 1990), ed. Anonymous (Milan 1990) 174–76. Hülsen C. (1910) Die Thermen des Agrippa. Ein Beitrag zur Topographie des Marsfeldes in Rom (Rome 1910). Hüser H. (1979) “Wärmetechnische Messungen an einer Hypokaustenheizung in der Saalburg”, SaalbJb 36 (1979) 12–30. Huß W. (2006a) “Lambaesis”, New Pauly Online. Available: http://referenceworks.brillonline.com/entries/brill-s-new -pauly/lambaesis-e629340 [Accessed on: 09/12/2015]. Huß W. (2006b) “Thugga”, New Pauly Online. Available: http:// referenceworks.brillonline.com/entries/brill-s-new-pauly/ thugga-e1212600 [Accessed on: 09/12/2015]. Husson G. (1983) OIKIA. Le vocabulaire de la maison privée en Égypte d’après les papyrus grecs (Papyrologie 2) (Paris 1983). Israel Y. (1993) “Ashqelon”, Excavations and Surveys in Israel 13 (1993) 100–105. Jackson B. (1892) The Ecclesiastical History, Dialogues, and Letters Of Theodoret (A select library of the Nicene and Post-Nicene Fathers of the Christian Church, Second Series 3) (Edinburgh 1892). Jackson B. (1895) The Treatise De Spiritu Sancto, The Nine Homilies Of The Hexæmeron And The Letters Of Saint Basil The Great Archbishop Of Cæsaria (A select library of the Nicene and Post-Nicene Fathers of the Christian Church, Second Series 8) (Edinburgh 1895). Jackson R. (1999) “Spas, waters and hydrotherapy in the Roman World” in Roman baths and bathing: proceedings of the first International Conference on Roman Baths held at Bath, England, 30 March–4 April 1992, edd. J. DeLaine and

481 D. E. Johnston (JRA Supplementary Series 37) (Portsmouth 1999) 107–116. Jansen-Winkeln K. (2006b) “Pelusium”, New Pauly Online. Available: http://referenceworks.brillonline.com/entries/ brill-s-new-pauly/pelusium-e912920 [Accessed on: 13/05/ 2015]. Jaritz F. (1993) Die arabischen Quellen zum heiligen Menas (Abhandlungen des Deutschen Archäologischen Instituts Kairo. Islamische Reihe 7) (Heidelberg 1993). Jeddi N. (1994) “Une mosaïque inédite d’Ouled Hafouz (Tunisie)” in La mosaïque gréco-romaine IV (Trèves 8–14 août 1984), edd. J.-P. Darmon and A. Rebourg (Paris 1994) 273–79. Jeffreys E., Jeffreys M. and Scott R. (1986) The Chronicle of John Malalas (Byzantina Australiensa 4) (Melbourne 1986). Johne K.-P. (2006) “Historia Augusta”, New Pauly Online. Available: http://referenceworks.brillonline.com/entries/ brill-s-new-pauly/historia-augusta-e515750 [Accessed on: 05/02/2014]. Johnson P. (2012) Economic evidence and the changing nature of urban space in late antique Rome (Instrumenta 42) (Barcelona 2012). Joly E. (2000) “Sabratha”, Rivista di Topografia Antica 8 (2000) 129–50. Jones G. D. B. (1983) “Excavations at Tocra and Euhesperides, Cyrenaica 1968–1969”, LibSt 14 (1983) 109–121. Jones G. D. B. (1984) “The Byzantine Bath-house at Tocra: a Summary Report”, LibSt 15 (1984) 107–111. Jones M. W. (2000) Principles of Roman architecture (New Haven 2000). Jouffroy H. (1992) “Les aquae africaines” in Les eaux thermales et les cultes des eaux en Gaule et dans les provinces voisines: actes du Colloque, 28–30 septembre 1990, Aix-les-Bains, ed. R. Chevalier (Caesarodunum. Bulletin de l’Institut d’études latines et du Centre de recherches A. Piganiol 26) (Tours 1992) 87–96. Joukowski M. (2006) “Exciting Developments: The Brown University 2006 Petra Great Temple Excavations”, 2006 ADAJ Report [Online]. Available: https://www.brown . e d u / D e p a r t m e n t s / Jo u kow s ky _ I n s t i t u te / Pe t ra / excavations/2006-adaj.html [Accessed on: 10/12/2016]. Kahwagi-Janho H. (2014) “Étude comparative des deux bains de l’hippodrome de Tyr”, in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 399–412. Kenawi M. (2014) Alexandria’s hinterland: archaeology of the Western Nile Delta, Egypt (Oxford 2014). Kennedy H. (1985) “From Polis to Madina: Urban Change in late antique and Early Islamic Syria”, PastPres 106 (1985) 3–27.

482 Kennedy H. (2007) The Great Arab Conquest. How the Spread of Islam Changed the World We Live In (New York 2007). Kenrick P. M. (2009) Tripolitania (Libya Archaeological Guides) (London 2009). Kenrick P. M. (2013) Cyrenaica (Libya Archaeological Guides) (London 2013). Khanoussi M. (2001) “Les mosaïques des thermes romains de Baten Zammour (région de Gafsa, Tunisie; Capsa, Byzacène)”, BAntFr (2001) 88. Khachab A. (1949) Ptolemaic and Roman baths of Kōm el Ahmar (Cairo 1949). Khachab A. (1957) “Les hammams du Kôm Trougah”, ASAE 54 (1957) 117–39. Khachab A. (1978) [Ta Sarapeia] à Sakha et au Fayum, ou, Les bains thérapeutiques (Supplément aux ASAE; cahier no. 25) (Cairo 1978). Khayari A. A. (1994) “Les thermes extra muros à Volubilis”, AfrRom 10 (1994) 301–312. Kisilevitz S. (2014) “Kefar Uriyya”, Excavations and Surveys in Israel 126 [Online]. Available: http://www.hadashot-esi.org .il/report_detail_eng.aspx?id=13705&mag_id=121 [Accessed on: 30/04/2015]. Kloner A. (2000) “The Cave of the Birds—A Painted Tomb on the Mount of Olives” in Ancient Jerusalem Revealed, ed. H. Geva (Jerusalem 2000) 306–310. Kloner A. and Cohen M. (1999) “Bet Guvrin, Area 750”, Excavations and Surveys in Israel 19 (1999) 78*–79*. Knudstad J. E. (1962) “The square of the cisterns”, in Ptolemais. City of the Libyan Pentapolis, ed. C. Kraeling (Chicago 1962) 62–67. Koetschet P. (2014) “Quelques aspects du bain dans la médecine arabe médiévale”, in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 7.4) (Cairo 2014) 1011–1024. Köhler J. (2006) “Forschungen zu römischen Thermalbädern”, in Cura aquarum in Ephesus: proceedings of the twelfth International Congress on the History of Water Management and Hydraulic Engineering in the Mediterranean Region: Ephesus/Selçuk, Turkey, October 2–10, 2004, ed. G. Wiplinger (BABesch Supplement 12/Sonderschriften Österreichisches Archäologisches Institut 42) (Leuven 2006) 437–42. Koɫątaj W. (1972) “Les fouilles polonaises à Kôm el-Dikka (Alexandrie)”, ÉtTrav 6 (1972) 148–67. Koɫątaj W. (1976) “La dernière période d’utilisation et la déstruction des thermes romains tardifs de Kom el-Dikka”, ÉtTrav 9 (1976) 217–29. Koɫątaj W. (1992) Alexandrie VI. Imperial baths at Kom el-Dikka (Warsaw 1992). Kolb B. and Dumreicher H. (2008) “The hammām—A living cultural heritage”, International Journal of Architectural Research 2 (2008) 17–28.

Bibliography Korenfeld I. and Bar-Nathan R. (2014) “Yehud”, Excavations and Surveys in Israel 126 [Online]. Available: http://www .hadashot-esi.org.il/report_detail_eng.aspx?id=10581& mag_id=121 [Accessed on: 04/05/2015]. Kraeling C. (1962) Ptolemais: city of the Libyan pentapolis (University of Chicago Oriental Institute publications 90) (Chicago 1962). Krautheimer R. and Corbett S. (1967) “S. Pudenziana”, in Corpus basilicarum Christianarum Romae III, edd. R. Krautheimer, S. Corbett, V. Frankl et al. (The Vatican 1967) 277–302. Kreiner R. and Letzner W. (2012) edd. Spa, sanitas per aquam. Tagungsband des Internationalen Frontinus-Symposiums zur Technik und Kulturgeschichte der antike. Thermen. Aachen, 18.–22. März 2009 (BABesch Supplement 21) (Leuven 2012). Krell O. (1901) Altrömische Heizungen (Munich/Berlin 1901). Krencker D., Krüger E. and Lehman H. (1929) Die Trierer Kaiserthermen: mit einer Übersicht über die wichtigsten Thermenanlagen des römischen Reiches (Trierer Grabungen und Forschungen 1: 1–2) (Augsburg 1929). Kretzschmer F. (1953) “Hypokausten”, SaalbJb 12 (1953) 7–41. Kretzschmer F. (1955) “Die Heizung der Aula Palatina in Trier”, Germania 33 (1955) 200–210. Kucharczyk R. (2009) “Windowpanes from the Bath in Marea. A look at late antique Glazing Techniques”, in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études urbaines 7) (Cairo 2009) 255–62. Kuhoff W. (2010) “Sufetula: Der Wandel eines städtischen Zentrums im spätrömischer Afrika”, in Krise und Kult. Vorderer Orient und Nordafrika von Aurelian bis Justinian, edd. D. Kreikenbom, K.-U. Mahler, P. Schollmeyer et al. (Millenium-Studien zu Kultur und Geschichte des ersten Jahrtausends n. Chr. 28) (Berlin 2010) 279–315. La Follette L. (1985) “Le terme Decianee sulll’Aventino”, Archeologia Laziale 7 (1985) 139–44. La Follette L. (1994) “The Baths of Trajan Decius on the Aventine”, Rome Papers (1994) 1–88. La Pera S. (1993) “Aqua Virgo”, LTUR 1 (1993) 72–73. Lamei S. (2012) “Egyptian baths and cultural tradition in the medieval Cairo”, in Spa, sanitas per aquam, Proceedings of the International Frontinus-Symposium on the technical and cultural history of ancient baths, Aachen, March 18–22, 2009, edd. R. Kreiner and W. Letzner (BABesch Supplement 21) (Leuven 2012) 307–312. Lancaster L. (2012) “A new vaulting technique for early baths in Sussex: the anatomy of a Romano-British invention”, JRA 25 (2012) 419–40. Lancel S. (1992) Carthage (Africana) (Tunis 1992). Lanciani R. (1875) “Di un gruppo di lapidi e latercoli militari scoperti sull’esquilino”, BullCom (1875) 77–82. Lanciani R. (1879) “Regio XII”, NSc (1879) 14–15. Lanciani R. (1881) “Regio XII”, NSc (1881) 89–90.

Bibliography Lantier R. (1921) “Découvertes archéologiques sur la colline dite ‘de Junon’ à Carthage”, BAC (1921) 87–94. Lantier R. (1931) “Les grands champs de fouilles de l’Afrique du Nord”, AA (1931) 462–575. Laronde A. (1997) “Mission archéologique française en Libye. Rapport sur la campagne de fouilles d’août 1995 à Leptis Magna”, LibAnt n.s. 2 (1997) 195–98. Laronde A. (1998) “Mission archéologique française en Libye. Rapport sur la campagne de fouilles d’août 1996 à Leptis Magna”, LibAnt n.s. 3 (1998) 251–57. Laronde A. and Michel V. (2009) “La cité d’Érythron en Cyrenaïque”, RA 47 (2009) 176–85. Lassus J. (1930) “Autour des basiliques chrétiennes de Tipasa”, MÉFR 47 (1930) 222–43. Lassus J. (1959) “L’archéologie algérienne. Henchir Safia”, Libyca 7 (1959) 325–43. Lassus J. (1962) “Fouilles à Sétif, 1958–1962”, BAC (1961–1962) 95–107. Lassus J. (1976) “Thamugadi (Timgad) Algeria”, PECS (1976) 899–902. Lassus J. (1981) La forteresse byzantine de Thamugadi. Fouilles à Timgad 1938–1956 (Paris 1981). Lauffray J. (1991) Halabiyya-Zenobia 2. Place forte du limes oriental et la Haute-Mésopotamie au 6e siècle (Bibliothèque Archéologique et Historique (Beyrouth) 138) (Paris 1991). Lavan L. (2012) “Public Space in late antique Ostia: Excavation and Survey in 2008–2011”, AJA 116 (2012) 649–91. Lavan L. (Forthcoming) Public Space in the late antique City (Late Antique Archaeology. Supplementary Series 5) (Leiden). Le Bomin J. (2015) “Les thermes byzantins de Taposiris Magna: campagne 2015 [Online]”. Available: http://balneorient.hypotheses.org/2946 [Accessed on: 29/10/2015]. Le Bomin J., Marchand J. and Vanpeene M. (2016) “The Byzantine Baths of Taposiris Magna—2016 Campaign [Online]”. Available: http://balneorient.hypotheses.org/ 3039 [Accessed on: 15/10/2016]. Leclerq H. (1910) “Baptistère”, DACL II.1 (1910) c.381–470. Le Glay M. (1985) “Les premiers temps de la Carthage romaine: pour une révision des dates”, in Histoire et archéologie de l’Afrique du Nord. IIe colloque international (Grenoble 5–9 avril 1983), edd. S. Lancel, A. M. Bisi, A. Ciasca et al. (Bulletin archéologique du CTHS-N.S.-19B) (Paris 1985) 235–47. Lehar H. (2012) Die römische Hypokaustheizung. Berechnungen und Überlegungen zu Leistung, Aufbau und Funktion (Aachen 2012). Lenoir E. (1985) “Volubilis du Bas-Empire à l’époque islamique”, in Histoire et archéologie de l’Afrique du Nord. IIe colloque international (Grenoble 5–9 avril 1983), edd. S. Lancel, A. M. Bisi, A. Ciasca et al. (Bulletin archéologique du CTHS-N.S.-19B)(Paris 1985) 425–28. Lenoir E. (1991) “Thermes romains de Maurétanie Tingitane” in Les thermes romains: actes de la table ronde (Rome, 11–12

483 novembre 1988), ed. Anonymous (CÉFR 142) (Rome 1991) 151–60. Lenoir E. (1995) “Thermes et palestres à l’époque romaine”, BAssBudé (1995) 62–76. Leone A. (2002) “Le sepolture nello spazio urbano a Cartagine tra V e VII secolo dC”, AnTard 10 (2002) 233–48. Leone A. (2007) Changing townscapes in North Africa from Late Antiquity to the Arab conquest (Studi storici sulla tarda antichità 28) (Bari 2007). Leone A. (2013) The end of the pagan city: religion, economy, and urbanism in late antique North Africa (Oxford 2013). Lepaon T. (2015) “Les bains de Placcus à Gerasa de la Décapole. Synthèse des recherches et premières remarques”, Syria 92 (2015) 105–22. Lepelley C. (1979) Les cités de l’Afrique romaine au Bas-Empire: 1. La permanence d’une civilisation municipale (Paris 1979). Lepelley C. (1981) Les cités de l’afrique romaine au bas-empire. Vol. 2: Notices d’histoire municipale (Paris 1981). Lepelley C. (1994) “Le musée des statues divines. La volonté de sauvegarder le patrimoine artistique païen à l’époque théodosienne”, Cahiers de Tunisie 42 (1994) 5–15. Leschi L. (1940) “Découvertes effectuées à Lambèse”, BAC (1938–1940) 265–76. Leschi L. (1953) Djemila, antique Cuicul (Algiers 1953). Leveau P. (1984) Caesarea de Maurétanie, une ville romaine et ses campagnes (CÉFR 70) (Rome 1984). Levi D. (1947) Antioch Mosaic Pavements (Princeton 1947). Lézine A. (1968) Carthage. Uthique. Études d’architecture et d’urbanisme (Paris 1968). Lézine A. (1969) Les thermes d’Antonin à Carthage (Tunis 1969). Lézine A. and Elouard P. (1961) Architecture romaine d’Afrique: recherches et mises au point (Publications de l’Université de Tunis, Faculté des lettres et sciences humaines de Tunis Première série, Archéologie-histoire 9) (Paris 1961). Lézine M., Picard C. and Picard G.-C. (1956) “Observations sur la ruine des thermes d’Antonin à Carthage”, CRAI (1956) 425–30. Lichoka B. (2008) “The early Byzantine coins” in Marea Vol. 1. Byzantine Marea. Excavations in 2000–2003 and 2006, edd. H. Szymanska and K. Babraj (Biblioteka Muzeum Archeologicznego w Krakowie 4) (Kraków 2008) 145–52. Liebeschuetz J. H. W. G. (2003) “Gens into Regnum: the Vandals” in Regna and Gentes: The Relationship between late antique and Early Medieval Peoples and Kingdoms in the Transformation of the Roman World, edd. H.-W. Goetz, J. Jarnut and W. Pohl (Leiden 2003) 55–83. Lipinska J. (1966) “Polish excavations at the Kom al-Dikka in Alexandria”, ÉtTrav 1 (1966) 181–99. Lippolis E. and Baldini Lippolis I. (1997) “La formazione e lo sviluppo del centro urbano di Brundisium: aspetti e problemi della ricerca”, Taras 17 (1997) 305–41. Lipps J. (2013) “Alarichs Goten auf dem Forum Romanum? Überlegungen zu Gestalt, Chronologie und Verständnis

484 der spätantiken Platzanlage”, in The Sack of Rome in A.D. 410: the event, its context and its impact: proceedings of the conference held at the German Archaeological Institute at Rome, 04–06 November 2010, edd. J. Lipps, C. Machado and P. Rummel (Palilia 28) (Wiesbaden 2013) 103–22. Lissi Caronna E. (1993) “Castra Praetoria”, LTUR 1 (1993) 251–54. Lohmann H. (1979) “Beobachtungen zum Stadtplan von Timgad”, in Wohnungsbau im Altertum, ed. W. Hoepfner (Diskussionen zur Archäologischen Bauforschung 3) (Berlin 1979) 167–87. Lomas K. (1995) “Urban elites and cultural defintion: Romanisation in south Italy”, in Urban society in Roman Italy, edd. T. J. Cornell and K. Lomas (London 1995) 107–20. Lopreato P. (1994) “Le grandi terme di Aquileia. I mosaici del frigidarium”, in La mosaïque gréco-romaine IV (Trèves 8–14 août 1984), edd. J.-P. Darmon and A. Rebourg (Paris 1994) 87–99. Lowther Clarke W. (1918) The Lausiac History of Palladius (Translations of Christian Literature. Series I) (New York 1918). Lucore S. K. (2013) “Bathing in Hieronian Sicily”, in Greek Baths and Bathing Culture. New discoveries and approaches, edd. S. K. Lucore and M. Trümper (BABesch Supplement 23) (Leuven 2013) 151–79. Luni M. (1975) “Le terme bizantine”, LibAnt 11–12 (1975) 266–76. Lusuardi Siena S. (1986) “Milano: la città nei suoi edifici. Alcuni problemi”, in Atti del X Congresso internazionale di studi sull’Alto Medioevo Milano 26–30 settembre, ed. Anonymous (Spoleto 1986) 209–40. Lux U. (1966) “Der Mosaikfußboden eines spätantiken Bades in umm-qēs”, ZDPV 82 (1966) 64–70. MacDonald W. L. (1986) The architecture of the Roman Empire. 2, An urban appraisal (Yale publications in the history of art 35) (New Haven 1986). MacMullen R. (1982) “The epigraphic habit in the Roman Empire”, AJP 103.3 (1982) 233–46. Magen I. (1993) “Martyrius Monastery”, in The New Encyclopedia of Archaeological Excavations in the Holy Land III, edd. E. Stern, A. Levinzon-Golbo’a and J. Aviram (New York 1993) 968–72. Magie D. (1921) Historia Augusta, Volume I: Hadrian. Aelius. Antoninus Pius. Marcus Aurelius. L. Verus. Avidius Cassius. Commodus. Pertinax. Didius Julianus. Septimius Severus. Pescennius Niger. Clodius Albinus. Translated by David Magie. (Loeb classical library 139) (Cambridge, Mass. 1921). Magie D. (1924) Historia Augusta, Volume II: Caracalla. Geta. Opellius Macrinus. Diadumenianus. Elagabalus. Severus Alexander. The Two Maximini. The Three Gordians. Maximus and Balbinus. Translated by David Magie (Loeb Classical Library 140) (Cambridge, Mass. 1924). Magie D. (1932) Historia Augusta, Volume III: The Two Valerians. The Two Gallieni. The Thirty Pretenders. The Deified Claudius. The Deified Aurelian. Tacitus. Probus.

Bibliography Firmus, Saturninus, Proculus and Bonosus. Carus, Carinus and Numerian. Translated by David Magie (Loeb Classical Library 263) (Cambridge, Mass. 1932). Mahdavi A. and Orehounig K. (2008) “An inquiry into the thermal, acoustical, and visual aspects of indoor environment in traditional hammams”, International Journal of Architectural Research 2 (2008) 136–49. Mahjoubi A. (1978) Recherches d’histoire et d’archéologie à Henchir El-Faouar (Tunisie) (Publications de l’Université de Tunis, Faculté des lettres et sciences humaines, 1ère série 12) (Tunis 1978). Maioli M. G. (1987) “L’edilizia privata tardoantica in Romagna: appunti sulla pavimentazione musiva”, in XXXIV corso di cultura sull’arte ravennate e bizantina: seminario internazionale di studi su “Archeologia e arte nella Spagna tardoromana, visigota e mozarabica”: Ravenna, 4–11 aprile 1987, ed. Anonymous (Ravenna 1987) 209–51. Maioli M. G. (1988) “Ravenna”, in Flumen aquaeductus: nuove scoperte archeologiche dagli scavi per l’acquedotto della Romagna, ed. L. Prati (Bologna 1988) 71–93. Maioli M. G. (1998) “Nuovi rinvenimenti di mosaici a Cesena (Forli)”, in Atti del V Colloquio dell’Associazione italiana per lo studio e la conservazione del mosaico Paribeni, edd. A. Guidobaldi and F. Guidobaldi (Ravenna 1998) 75–82. Maioli M. G. (2000) “Cesena”, in Aemilia. La cultura romana in Emilia Romagna dal III secolo a.C. all’età costantiniana. Catalogo della Mostra (Bologna, Pinacoteca Nazionale 2000), ed. M. Calvani (Venice 2000) 495–500. Maioli M. G. (2003a) “La topografia di Ravenna e Classe in età romana e bizantina”, in Viaggio nei siti archeologici della provincia di Ravenna, ed. G. Montevecchi (Ravenna 2003) 9–15. Maioli M. G. (2003b) “La stratificazione urbana di via d’Azeglio”, in Viaggio nei siti archeologici della provincia di Ravenna, ed. G. Montevecchi (Ravenna 2003) 16–23. Maioli M. G. and G. Montevecchi (2003) “Palazzolo: scavi del Palazzo di Teodorico”, in Viaggio nei siti archeologici della provincia di Ravenna, ed. G. Montevecchi (Ravenna 2003) 84–85. Maischberger M. (1999a) “Testaceus Mons”, LTUR 5 (1999) 28–30. Maischberger M. (1999b) “Transtiberim”, LTUR 5 (1999) 77–83. Maisler B., Stekelis M. and Avi-Yonah M. (1952) “The excavations at Beth Yerah (Khirbet el-Kerak) 1944–1946”, IEJ 2 (1952) 218–19. Maiuri A. (1950) “Significato e natura del solium nelle terme romane”, PP 5 (1950) 223–27. Majcherek G. (1990) “Excavations at Kom el-Dikka in Alexandria in the 1989 season”, Polish Archaeology in the Mediterranean 1 (1990) 75–83. Majcherek G. (1992) “Excavations in Alexandria in 1990–91”, Polish Archaeology in the Mediterranean 3 (1992) 5–14.

Bibliography Majcherek G. (2008) “The pottery assemblage from the baths and sāqiyah”, in Marea Vol. 1. Byzantine Marea. Excavations in 2000–2003 and 2006, edd. H. Szymanska and K. Babraj (Biblioteka Muzeum Archeologicznego w Krakowie 4) (Kraków 2008) 105–27. Malacrino C. G. (2014) “I nuclei termali delle ville romane calabresi fra il II e il IV secolo d.C.: Roggiano Gravina, Malvito e Casignana”, in La villa restaurata e i nuovi studi sull’edilizia residenziale tardoantica. Atti del Convegno internazionale del Centro interuniversitario di studi sull’edilizia abitativa tardoantica nel Mediterraneo (CISEM), Piazza Armerina 7–10 novembre 2012, edd. P. Pensabene and C. Sfameni (Bari 2014) 289–302. Malarczyk D. (2008) “Ummayad coins”, in Marea Vol. 1. Byzantine Marea. Excavations in 2000–2003 and 2006, edd. H. Szymanska and K. Babraj (Biblioteka Muzeum Archeologicznego w Krakowie 4) (Kraków 2008) 153–59. Manacorda D. (1993) “Crypta Balbi”, LTUR 1 (1993) 326–29. Manderscheid H. (1981) Die Skulpturenausstattung der kaiserzeitlichen Thermenanlagen (Monumenta artis Romanae 15) (Berlin 1981). Manderscheid H. (1988a) “Römische Thermen: Aspekte von Architektur, Technik und Ausstattung”, in Die Wasserversorgung antiker Städte: Mensch und Wasser, Mitteleuropa, Thermen, Bau-materialien, Hygiene, ed. Anonymous (Geschichte der Wasserversorgung 3) (Mainz 1988) 99–125. Manderscheid H. (1988b) Bibliographie zum römischen Badewesen: unter besonderer Berücksichtigung der öffentlichen Thermen (Munich and Berlin 1988). Manderscheid H. (1994) “Aspekte der Mosaikausstattung in öffentlichen und privaten Thermenanlagen”, in La mosaïque gréco-romaine IV (Trèves 8–14 août 1984), edd. J.-P. Darmon and A. Rebourg (Paris 1994) 61–66. Manderscheid H. (2000a) “The Water Management of Greek and Roman Baths”, in Handbook of Ancient Water Technology, ed. Ö. Wikander (Leiden 2000) 467–538. Manderscheid H. (2000b) “Ohne Wasser kein Bad”, in Cura aquarum in Sicilia: proceedings of the tenth International Congress on the History of Water Management and Hydraulic Engineering in the Mediterranean Region (Syracuse, May 16–22, 1998), ed. G. C. M. Jansen (BABesch Supplement 6) (Leiden 2000) 173–80. Manderscheid H. (2004) Ancient baths and bathing: a bibliography for the years 1988–2001 (JRA Supplementary Series 55) (Portsmouth 2004). Manderscheid H. (2012) “Ein spätes Kaiserbad in Rom oder von der Genese eines Faktoids”, in Spa: sanitas per aquam: Proceedings of the International Frontinus-Symposium on the technical and cultural history of ancient baths, Aachen, March 18–22, 2009, edd. R. Kreiner and W. Letzner (BABesch Supplement 21) (Leuven 2012) 265–73.

485 Manzelli V. (2000) Ravenna (Atlante tematico di topografia antica. Supplemento 8) (Rome 2000). Manzelli V. (2003) “Lo scavo del Palazzo di Teodorico”, in Domus del Triclinio. Alla scoperta di Ravenna romana. Mosaici e altri tesori mai visti, ed. Anonymous (Ravenna 2003) 69–77. Mar R. (1990) “Las termas de tipo medio de Ostia y su inserción en el espacio urbano. Estudio preliminar”, Italica 18 (1990) 31–77. Mar R. (2005) El Palatí: la formació dels palaus imperials a Roma (Sèrie Documenta 3) (Taragona 2005). Marcillet-Jaubert J. (1976) “Lambaesis (Algeria), Tazoult”, PECS (1976) 478–79. Marcuse J. (1903) Bäder und Badewesen in Vergangenheit und Gegenwart: eine kulturhistorische Studie (Stuttgart 1903). Mare W. H. (1994) “The 1992 season of excavation at Abila of the Decapolis”, ADAJ 38 (1994) 359–78. Maréchal S. (2012) “Research on Roman bathing: old models and new ideas”, RBPhil 90 (2012) 143–64. Maréchal S. (2013) “Roman Public Baths in Modern Libya”, BABesch 88 (2013) 205–28. Maréchal S. (2015) “Lavacrum: just another word for baths? How the terminology of baths may have reflected changes in bathing habits”, RBPhil 93 (2015) 139–77. Maréchal S. (2016) “Public and Private Bathing in late antique North Africa. Changing Habits in a Changing Society?” in TRAC 2015: Proceedings of the Twenty-Fifth Theoretical Roman Archaeology Conference, Leicester 2015, edd. M. J. Mandich, T. J. Derrick, S. G. Sanchez, G. Savani and E. Zampieri (Oxford 2016) 215–40. Maréchal S. (2017) “A note on the drainage of pools in Roman baths”, BABesch 92 (2017) 179–86. Maréchal S. (2018) “Romeinse badhuizen: dissectie van een succesverhaal”, TMA 58 (2018), 30–36. Maréchal S. (Forthcoming) “The tubuli from Ostia”, in Roman Brick and Tile. Past, present and future of the study of Roman ceramic building materials, ed. T. Clerbaut (Oxford). Mari Z. (1993) “Aqua Claudi”, LTUR 1 (1993) 63–64. Marinucci A. and Cicerchia P. (1992) Scavi di Ostia XI. Le terme del Foro o di Gavvio Massimo (Rome 1992). Markschies C. (2006). “Evagrius Scholasticus, ancient Greek church historian”, New Pauly Online. Available: http://referenceworks.brillonline.com/entries/brill-s-new-pauly/ evagrius-e403570 [Accessed on: 07/05/2014]. Martin A., Heinzelmann M., De Sena E., et al. (2002) “The Urbanistic project on the previously unexcavated areas of Ostia (DAI-AAR 1996–2001)”, MAAR 47 (2002) 259–304. Marzano A. (2007) Roman villas in central Italy: a social and economic history (Columbia studies in the classical tradition 30) (Leiden and Boston 2007). Masekh S., Fournet T., Piraud-Fournet P. and Boraik M. (2017) “The Roman Baths at Karnak, Between River and Temples.

486 Architectural and Urban Context”, in Collective Baths in Egypt 2. New Discoveries and Perspectives, ed. B. Redon (Études urbaines 10) (Cairo 2017) 221–66. Matthews J. F. (1996a) “Ammianus Marcellinus”, OCD (1996) 73–74. Matthews J. F. (1996b) “Historia Augusta”, OCD (1996) 713–14. Mazar A. (1993) “Qasile. Excavations in area C”, in The New encyclopedia of archaeological excavations in the Holy Land IV, edd. E. Stern, A. Levinzon-Golbo’a and J. Aviram (New York 1993) 1204–12. Mazor G. and Bar-Nathan R. (1994) “Scythopolis—Capital of Palestina Secunda”, Qadmoniot 107–108 (1994) 117–37. Mazor G. and Bar-Nathan R. (1998) “The Bet She’an Excavation Project—1992–1994. Antiquities Authority Expedition”, Excavations and Surveys in Israel 17 (1998) 6–36. Mc Donough S. J. (2009) “‘We and those waters of the sea are one’: baptism, bathing, and the construction of identity in late ancient Babylonia”, in The nature and function of water, baths, bathing, and hygiene from antiquity through the Renaissance, edd. C. Kosso and A. Scott (Technology and change in history 11) (Leiden 2009) 263–73. Mc Parland L. C., Hazell Z., Campbell G., et al. (2009) “How the Romans got themselves into hot water: temperatures and fuel types used in firing a hypocaust”, Environmental Archaeology 14 (2009) 176–83. Medri M. (2013) “Le terme del nuotatore in Ostia antica”, in Ostia. V, Le terme del nuotatore, cronologia di un’insula ostiense, edd. M. Medri and V. Di Cola (Studi miscellanei 36) (Rome 2013) 27–109. Meiggs R. (1960) Roman Ostia (Oxford 1960). Melek Wasif P. (1979) “A Graeco—Roman Bath at Tell Sersena”, ASAE 63 (1979) 177–82. Menchelli S. and Vaggioli M. A. (1987) “Le terme di San Gaetano di Vada”, in Terme Romane e vita quotidiana, ed. M. Pasquinucci (Modena 1987) 121–27. Meneghini R. (1999) “Edilizia pubblica e privata nella Roma altomedievale. Due episodi di riuso”, MÉFRM 111 (1999) 171–82. Mercier G. (1870) “Note sur la mosaïque de Léda trouvée à Aumale (Auzia)”, RAfr 14 (1870) 434–40. Merlin A. (1903) “Les fouilles de Dougga en 1902”, NouvArch 11 (1903) 1–116. Merlin A. (1915) “Découvertes archéologiques récentes en Tunisie”, BAC (1915) 1–109. Merlin A. (1946–1949) “Séance de la commission de l’Afrique du Nord”, BAC (1946–1949) 229–30. Merten E. (1983) Bäder und Badegepflogenheiten in der Darstellung der Historia Augusta (Antiquitas Reihe 4, Beiträge zur Historia-Augusta-Forschung Bd 16) (Bonn 1983). Meshel Z. (1989) “A Fort at Yotvata from the Time of Diocletian”, IEJ 39 (1989) 228–38.

Bibliography Messina A. (1993) “Le terme di Dafne e di Baia a Siracusa”, Rivista di Topografia Antica 3 (1993) 201–202. Meyer B. (1989) “Problèmes du combustible dans les bains publics de l’Égypte grecque et romaine”, in Egitto e storia antica dall’Ellenismo all’età araba: bilancio di un confronto (atti del colloquio internazionale, Bologna, 31 agosto–2 settembre 1987), edd. L. Criscuolo and G. Geraci (Bologna 1989) 565–71. Meyer B. (1992) “Les femmes et les bains publics dans l’Égypte grecque, romaine et byzantine”, in Proceedings Of The XIXth International Congress of Papyrology (Cairo 2–9 September 1989), ed. A. Mosalamy (Cairo 1992) 51–60. Meyer B. (1994) “L’eau et les bains publis dans l’Égypte ptolemaïque, romaine et byzantine”, in, Les problèmes institutionnels de l’eau en Égypte ancienne et dans l’Antiquité méditerranéenne (colloque AIDEA Vogüé 1992), ed. B. Menu (Bibliothèque d’étude 110) (Cairo 1994) 273–79. Meyer B. (1997) ““Gymnase” et “Thermes” dans l’Égypte romaine et byzantine”, in Akten des 21. internationalen Papyrologenkongresses (Berlin 13.–19.8.1995), ed. B. Krämer (Stuttgart and Leipzig 1997) 691–95. Meyer B. (2001) “Magie et Bains publics”, in Atti del XXII congresso internazionale di papirologia (Firenze, 23–29 agosto 1998), edd. I. Andorlini and G. Bastianini (Florence 2001) 937–42. Michel V. (2011a) “Nouvelles recherches en Cyrenaïque: les thermes romains d’Erythron, Latrun (note d’information)”, CRAI (2011) 205–220. Michel V. (2011b) “Nouvelles recherches en Cryénaïque (Érythron, Latrun)”, Les Nouvelles de l’Archéologie 123 (2011) 28–33. Michel V. (2014) “Les thermes romains d’Érythron en Cyrénaïque, Libye”, in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 369–84. Milella A. (1996) “S. Susanna, titulus”, LTUR 4 (1996) 387–88. Millet M. (2013) “Previous Archaeological work”, in Ocriculum (Otricoli, Umbria). An archaeological survey of the Roman town, edd. S. Hay, S. Keay and M. Millet (Archaeological monographs of the British School at Rome 22) (London 2013) 13–24. Mingana A. (1920) “New documents on Philoxenus of Hierapolis and on the Philoxenian version of the bible”, The Expositor 19 (1920) 149–60. Mirabella Roberti M. (1959) “Cinque anni di lavori per le antichità in Lombardia”, in Cisalpina. Vol. I, Atti del Convegno sull’attività archeologica nell’Italia Settentrionale: (Villa Monastero di Varenna) 9–15 giugno 1958, ed. Anonymous (Milan 1959) 72–86.

Bibliography Mirabella Roberti M. (1963) “Archeologia ed arte di Brescia romana”, in Storia di Brescia I. Dalle origini alla caduta della Signoria Vescontea (1426), ed. G. Trecciani degli Alfieri (Brescia 1963) 231–320. Miselli M. (2001) “I forni nelle abitazioni di Pompei”, Orizzonti 2 (2001) 229–41. Misiewicz K., Maɫkowski W. and Muszyńska M. (2010) “The topography of Ptolemais. Results of non-destructive survey campaigns (2002–2005)”, in Cirene e la Cirenaica nell’antichità, ed. M. Luni (Monografie di archeologia libica 30) (Rome 2010) 197–203. Mitchell S. (2007) A history of the later Roman Empire, A.D. 284–641: the transformation of the ancient world (Blackwell history of the Ancient World) (Malden 2007). Mittmann S. (1966) “Die Inschriften des spätantiken Bades in umm-qēs”, ZDPV 82 (1966) 71–73. Mohamedi A., Benmansour A., Amamra A., et al. (1991) Fouilles de Sétif 1977–1984 (Supplement au BAAlg 5) (Casbah 1991). Mohamedi A. and Fentress E. (1985) “Fouilles de Sétif 1978– 1982: la construction de la ville islamique”, in Histoire et archéologie de l’Afrique du Nord. IIe colloque international (Grenoble, 5–9 avril 1983), edd. S. Lancel, A. M. Bisi, A. Ciasca et al. (Bulletin archéologique du CTHS-N.S.-19B) (Paris 1985) 469–77. Mollo Mezzena R. (1992) “Augusta Praetoria tardoantica. Viabilità e territorio”, in Felix temporis reparatio: atti del Convegno archeologico internazionale Milano capitale dell’ Impero Romano, Milano, 8–11 Marzo 1990, edd. G. Sena Chiesa and E. A. Arslan (Milan 1992) 273–320. Monceaux P. (1922) “Découverte d’un groupe d’édifices chrétiens à Djemila”, CRAI 66 (1922) 380–407. Monceaux P. (1923) Cuicul chrétien (Numidie) (AttiPontAcc Serie III. Memorie 1) (The Vatican 1923). Montevecchi G. (2004) “Introduzione”, in Archeologia urbana a Ravenna. La ‘Domus dei Tappeti di Pietra’. Il complesso archeologico di via D’Azeglio, ed. G. Montevecchi (Ravenna 2004) 1–20. Montevecchi G. and Leoni C. (2004) “Settore settentrionale. Impianto termale (III–IV d. C.)”, in Archeologia urbana a Ravenna. La ‘Domus dei Tappeti di Pietra’. Il complesso archeologico di via D’Azeglio, ed. G. Montevecchi (Ravenna 2004) 48–55. Montserrat D. (2005) “‘Carrying on the work of the earlier firm’. Doctors, medicine and Christianity in the Thaumata of Sophronius of Jerusalem”, in Health in Antiquity, ed. H. King (London 2005) 230–42. Moore W. and Wilson H. A. (1892) Select Writings and Letters of Gregory, Bishop of Nyssa (A select library of the Nicene and Post-Nicene Fathers of the Christian Church, Second Series 5) (Edinburgh 1892). Morley N. (2005) “The salubriousness of the Roman city”, in Health in Antiquity, ed. H. King (London 2005) 192–204. Moro P. (1956) Iulium Carnicum (Zuglio) (Rome 1956).

487 Morris J. B. and Simcox W. H. (s.d.) “The Homilies of St. John Chrysostom, Archbishop of Constantinople, On the Epistle of St. Paul the Apostle to the Romans. Translated by Rev. J. B. Morris and Rev. W. H. Simcox”, in Saint Chrysostom: Homilies on the Acts of the Apostles and the Epistle to the Romans, ed. P. Schaff (A Select Library Of The Nicene And Post-Nicene Fathers Of The Christian Church, Series I. 11) (Edinburgh s.d.) 329–565. Morselli C. (1995) “Forum Iulium”, LTUR 2 (1995) 299–306. Mossakowska-Gaubert M. (2009) “Les bains et les moines. Le rôle hygiénique, thérapeutique et symbolique de la toilette corporelle dans la vie des moines égyptiens durant les premiers siècles du mouvement monastique”, in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études urbaines 7) (Cairo 2009) 297–303. Müller-Wiener W. (1966) “Abu Mena. 4. Vorläufiger Bericht”, MDIK 21 (1966) 171–87. Müller-Wiener W. (1967) “Abu Mena. 5. Vorläufiger Bericht”, MDIK 22 (1967) 206–24. Müller-Wiener and Grossmann P. (1967) “Abu Mena. 5. Vorläufiger Bericht”, AA 22 (1967) 457–80. Mulryan M. (2014) Spatial ‘Christianisation’ in Context: Strategic Intramural Building in Rome from the 4th–7th c. A.D. (Oxford 2014). Munn M. (2009) “Earth and water: the foundations of sovereignty in ancient thought”, in The nature and function of water, baths, bathing, and hygiene from antiquity through the Renaissance, edd. C. Kosso and A. Scott (Technology and change in history 11) (Leiden 2009) 191–210. Munzi M. (2000). “La città in declino”, in Volterra. Il teatro e le terme. Gli edifici, lo scavo, la topographia, edd. M. Munzi and N. Terrenato (Florence 2000) 193–95. Munzi M. and Terrenato N. (2000) “Introduzione”, in Volterra. Il teatro e le terme. Gli edifici, lo scavo, la topographia, edd. M. Munzi and N. Terrenato (Florence 2000) 1–21. Musil A. (1907) Kusejr ‘Amra (Vienna 1907). Musil A. (1908) Arabia Petraea II. Edom. Topographischer Reisebericht (Vienna 1908). Musso L. and Matug J. (2012) “Hunting Baths/Terme della Caccia: topographical context, distribution of rooms, plan type”, in Lepcis Magna. Hunting Baths. Building, Restoration, Promotion, edd. B. Biancchi and L. Musso (Rome 2012) 21–40. Musurillo H. (1972) The acts of the Christian martyrs (Oxford early Christian texts) (Oxford 1972). Nagorski A. (2009) “Bet Nehemya”, Excavations and surveys in Israel 121 [Online]. Available: http://www.hadashot-esi.org .il/report_detail_eng.aspx?id=1218&mag_id=115 [Accessed on: 19/05/2015]. Nardelli M. (2012) “Il sistema di distribuzione e smaltimento delle acque”, in Le Terme di Caracalla, ed. Piranomonte (Rome 2012) 120–24.

488 Nassery S., Wagner G. and Castel G. (1976) “Un grand bain gréco-romain à Karanis”, BIFAO 76 (1976) 231–75. Negev A. (1988) The architecture of Mampsis: Final Report. Volume I: The Middle and Late Nabatean Periods (Qedem 26) (Jerusalem 1988). Negev A. (1993) “Elusa”, in The New Encyclopedia of Archaeo­ logical Excavations in the Holy Land II, edd. E. Stern, A. Levinzon-Golbo’a and J. Aviram (New York 1993) 380. Negev A. (1994) “Be’er Sheva, Bathhouse”, Excavations and Surveys in Israel 14 (1994) 147. Negev A. (1997) The architecture of Oboda. Final Report (Qedem 36) (Jerusalem 1997). Negm M. A. and Kosciùk J. (1990) “The private Roman bath found near Abu Mina (Egypt)”, in Akten des 13. Internationalen Kongresses für Klassische Archäologie, Berlin 1988, ed. Anonymous (Mainz 1990) 442–45. Negrelli C. (2008) Rimini Capitale. Strutture insediative, sociali ed economiche tra V e VII secolo (Borgo San Lorenzo 2008). Nielsen I. (1985) “Considerazioni sulle prime fasi dell’evoluzione dell’edificio termale romano”, AnalRom 14 (1985) 81–112. Nielsen I. (1993a) Thermae et Balnea. The Architecture and Cultural History of Roman Public Baths. I. Text (Aarhus 1993, 2nd ed. [1990]). Nielsen I. (1993b) Thermae et Balnea. The Architecture and Cultural History of Roman Public Baths. II. Catalogue and Plates (Aarhus, 2nd ed. [1990]). Nielsen I. (1993c) “The baths: problems of terminology”, Balnearia 1 (1993) 3–4. Nielsen I. (1993d) “Geschichte und Funktion der Anlage”, in Gadara—Umm Qēs III. Die byzantinischen Thermen, edd. I. Nielsen, F. Gorm Andersen and S. Holm-Nielsen (Abhandlungen des deutschen Palästinavereins 17) (Wiesbaden 1993) 132–46. Nielsen I. (1993e) “Beschreibung des Gebäudes”, in Gadara— Umm Qēs III. Die byzantinischen Thermen, edd. I. Nielsen, F. Gorm Andersen and S. Holm-Nielsen (Abhandlungen des deutschen Palästinavereins 17) (Wiesbaden 1993) 11–119. Nielsen I. (1993f) “Die Technische Installationen”, in Gadara— Umm Qēs III. Die byzantinischen Thermen, edd. I. Nielsen, F. Gorm Andersen and S. Holm-Nielsen (Abhandlungen des deutschen Palästinavereins 17) (Wiesbaden 1993) 120–31. Nielsen I. (1999) “Early provincial baths and their relations to early Italic baths”, in Roman baths and bathing: proceedings of the first International Conference on Roman Baths held at Bath, England, 30 March–4 April 1992, edd. J. DeLaine and D. E. Johnston (JRA Supplementary Series 37) (Portsmouth 1999) 35–43. Nielsen I., Gorm Andersen F. and Holm-Nielsen S. (1993) edd. Gadara—Umm Qēs III. Die byzantinischen Thermen (Abhandlungen des deutschen Palästinavereins 17) (Wiesbaden 1993).

Bibliography Nielsen I. and Schiøler T. (1980) “The Water System in the Baths of Mithras in Ostia”, AnalRom 9 (1980) 149–59. Nordh A. (1949) Libellus de regionibus Urbis Romae (Lund 1949). Novara P. (2003) “Scavi del cosidetto Palazzo di Teodorico”, in Viaggio nei siti archeologici della provincia di Ravenna, ed. G. Montevecchi (Ravenna 2003) 58–62. Nutton V. (1984) “From Galen to Alexander. Aspects of Medicine and Medical Practise in Late Antiquity”, DOP 38 (1984) 1–14. Nutton V. (2004) Ancient medicine (Sciences of antiquity) (London and New York 2004). Nutton V. (2006) “Caelius Aurelianus, Physician from Sicca Veneria, c. A.D. 400”, New Pauly Online. Available: http:// referenceworks.brillonline.com/entries/brill-s-new-pauly/ caelius-e223580 [Accessed on: 08/07/2014]. Oetelaar T., Hughes L., Humphrey J. W., et al. (2014) “A computational investigation of the thermal environment of the caldarium in a replica Roman bath,” JRA 27 (2014) 392–402. Oetelaar T., Johnston C., Wood D., et al. (2013) “A computational investigation of a room heated by subcutaneous convection—A case study of a replica Roman bath”, Energy and Buildings 63 (2013) 59–66. Oleson J. P. (1984) Greek and Roman mechanical water-lifting devices: the history of a technology (Phoenix. Supplementary volume 16) (Dordrecht 1984). Oleson J. P. (1990) “The Humeima Hydraulic Survey”, EchCl 34 (1990) 145–63. Olevano F. and Rosso M. (2001) “Il mosaico a grandi tessere marmoree delle terme ‘della Marciana’ a Ostia”, in Atti dell’VIII colloquio AISCOM, edd. F. Guidobaldi and A. Paribeni (Ravenna 2001) 561–72. Oliverio G. (1931) “Campagna di scavi a Cirene nell’estate del 1928”, AfrIt 3 (1931) 141–229. Omeri I. (2014) “A Bath from Late Antiquity Discovered in Khirbat al-Maqsurah”, in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 509–18. Onn A., Weksler-Bdolah S. and Bar-Nathan R. (2011) “Jerusalem, The Old City, Wilson’s Arch and the Great Causeway”, Excavations and Surveys in Israel 123 [Online]. Available: http://www.hadashot-esi.org.il/report_detail_ eng.aspx?id=1738&mag_id=118 [Accessed on: 07/05/2015]. Oren E. (1993a) “Sera, Tel”, in The New Encyclopedia of Archaeological Excavations in the Holy Land IV, edd. E. Stern, A. Levinzon-Golbo’a and J. Aviram (New York 1993) 1329–35. Oren E. (1993b) “Sinai”, in The New encyclopedia of archaeological excavations in the Holy Land IV, edd. E. Stern, A. Levinzon-Golbo’a and J. Aviram (New York 1993) 1340.

Bibliography Oriolo F. (2001) “L’edificio termale di Iulium Carnicum: nuove considerazioni”, in Iulium Carnicum: centro alpino tra Italia e Norico dalla protostoria all’età imperiale: atti del Convegno, Arta Terme-Cividale, 29–30 settembre 1995, edd. G. Bandelli and F. Fontana (Rome 2001) 275–95. Othman A. (2014) “Rapport préliminaire sur les bains de l’Antiquité tardive à Sura (Syrie, Moyen Euphrate)”, in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 495–508. Pacciarelli M. (1997) ed. Acque, grotte e dei: 3000 anni di culti preromani in Romagna, Marche e Abruzzo (Imola 1997). Packer J. (1995) “Forum Trajani”, LTUR 2 (1995) 348–56. Pagani S. and D’Amora R. (2011) “Dalle terme al hammām e dal hammām al bagno turco: percorsi di lettura”, in Hammam: le terme nell’Islam: convegno internazionale di studi, Santa Cesarea Terme, 15–16 maggio 2008, edd. R. D’Amora and S. Pagani (Aquae 9) (Florence 2011) 7–15. Page M. J. (1993) “Roman Baths—a Health Warning!”, Balnearia 1 (1993) 2–3. Paley S. and Porath Y. (1993) “Hefer, Tel”, in The New encyclopedia of archaeological excavations in the Holy Land, edd. E. Stern, A. Levinzon-Golbo’a and J. Aviram (New York 1993) 612–13. Palladino S. (1996) “Le terme Eleniane a Roma”, MÉFRA 108 (1996) 855–71. Pallarès F. (1998) “La città di Ventimiglia nel passagio tra tarda Romanità e Medioevo”, Rivista ingauna e intemelia n.s. 51 (Gennaio–Dicembre 1996) 23–32. Palombi D. (1988) “Contributo alla topografia della via Sacra dagli appunti inediti di Giacomo Boni”, Quaderni di topografia antica 10 (1988) 77–97. Palombi D. (1993a) “Aqua Drusia”, LTUR 1 (1993) 66. Palombi D. (1993b) “Aqua Mercurii”, LTUR 1 (1993) 69. Panazza G. (1959) “Appunti su Brescia Romana”, in Cisalpina I. Atti del Convegno sull’attività archeologica nell’Italia Settentrionale: (Villa Monastero di Varenna) 9–15 giugno 1958, ed. A. Calderini (Milan 1959) 116–46. Panella C. (1999) “Porticus Liviae”, LTUR 4 (1999) 127–29. Panella C. and Rizzo G. (2014) Ostia VI. Le terme del nuotatore (Studi miscellanei 38) (Rome 2014). Pani Ermini L. (1979) “Il territorio portuense dopo il IV secolo alla luce degli scavi all’isola sacra”, Archeologia Laziale 2 (1979) 243–49. Papi E. (1999a) “Il balneum nelle dimore di Roma dall’età repubblicana al I secolo D.C.”, MÉFRA 111 (1999) 695–728. Papi E. (1999b) “Palatium (età repubblicaana—64 d.C.)”, LTUR 4 (1999) 22–28. Papi E. (1999c) “Palatium (64 d.C.—V sec. d. C.)”, LTUR 4 (1999) 28–38.

489 Parmegiani N. and Pronti A. (1993) “S. Caecilia”, LTUR 1 (1999) 206–207. Pasquinucci M. (1987) ed. Terme romane e vita quotidiana (Modena 1987). Patrich J. (1999) “The Warehouse Complex and the Governor’s Palace”, in Caesarea Papers 2. Herod’s Temple, the Provincial Governor’s Praetorium and Granaries, the Later Harbour, a Gold Coin Hoard, and Other Studies, edd. K. Holum, A. Raban and J. Patrich (JRA Supplementary Series 35) (Portsmouth 1999) 70–107. Paulin E. (1890) Thermes de Dioclétien (Restauration des monuments antiques par les architectes de l’Académie de France à Rome 7) (Paris 1890). Paulin M., Dagnas G. and Bonifay M. (2011) “Les Thermes du Levant à Leptis Magna (Libye)”, Les Nouvelles de l’Archéologie 123 (2011) 23–27. Paulin M., Dagnas G. and Bonifay M. (2012) “Les Thermes du Levant à Leptis Magna: présentation architecturale et périodisation générale”, AntAfr 46–48 (2012) 89–135. Pavolini C. (1980) “Saggi di scavo lungo la via severiana ad Ostia”, Archeologia Laziale 3 (1980) 113–22. Pavolini C. (1993) “Gli scavi nell’Ospedale Militare (1987– 1992)”, RM 100 (1993) 443–56. Pavolini C. (2000) “Le domus del Celio”, in Aurea Roma. Dalla città pagana alla città cristiana, edd. S. Ensoli and E. La Rocca (Rome 2000) 147–48. Pavolini C. (2006) Ostia (Guide archeologiche Laterza 11) (Bari 2006). Pazzini A. (1941) L’organizzazione sanitaria in Roma imperiale (La scienza e la tecnica ai tempi di Roma imperiale 4) (Rome 1941). Pedley J. (1967) “Excavations at Apollonia, Cyreniaca. Second Preliminary Report”, AJA 71 (1967) 141–47. Pedley J. (1976) “The Byzantine Baths”, in Apollonia, the port of Cyrene: excavations by the University of Michigan, 1965–1967, ed. R. Goodchild (Supplements to LibAnt 4) (Tripoli 1976) 225–43. Peilstöcker M. and Burke A. (2009) “Yafo, Ganor Compound”, Excavations and Surveys in Israel 121 [Online]. Available: http://www.hadashot-esi.org.il/report_detail_eng. aspx?id=1049&mag_id=115 [Accessed on: 05/05/2015]. Peleg M. (1987) “The South Bath house”, Excavations and Surveys in Israel 6 (1987) 43–44. Peleg M. (2004) “A Late Roman-Byzantine Bathhouse at Bet She’an”, ‘Atiquot 46 (2004) 51*–81*. Pellacchia D. (2014) “I “Bagni del clero”. Edilizia termale nel quartiere episcopale ravennate (V–IX sec.)”, in La villa restaurata e i nuovi studi sull’edilizia residenziale tardoantica atti del Convegno internazionale del Centro interuniversitario di studi sull’edilizia abitativa tardoantica nel Mediterraneo (CISEM), Piazza Armerina 7–10 novembre 2012, edd. P. Pensabene and C. Sfameni (Bari 2014) 205–12.

490 Pelliccioni G. (1973) Le nuove scoperte sulle origini del battistero lateranense (AttiPontAcc. Serie III. Memorie 12) (The Vatican 1973). Pensabene P. (2007) Ostiensium marmorum decus et decor: studi architettonici, decorativi e archeometrici (Rome 2007). Pensabene P. (2009) “I mosaici della villa romana del Casale: distribuzione, programmi iconografici, maestranze”, in Mosaici Mediterranei, ed. M. C. Lentini (Caltanisetta 2009) 69–98. Pensabene P. (2010) “Villa del Casale e il territorio di Piazza Armerina tra Tardoantico e Medioevo. Le nuove ricerche del 2004–2009”, in Piazza Armerina. Villa del Casale e la Sicilia tra tardoantico en medioevo, ed. P. Pensabene (Studia Archaeologica 175) (Rome 2010) 1–32. Pensabene P. (2014) “Nuove scoperte alla Villa del Casale di Piazza Armerina: magazzini, terme e fornaci”, in La villa restaurata e i nuovi studi sull’edilizia residenziale tardoantica: atti del Convegno internazionale del Centro interuniversitario di studi sull’edilizia abitativa tardoantica nel Mediterraneo (CISEM), Piazza Armerina 7–10 novembre 2012, edd. P. Pensabene and C. Sfameni (Bari 2014) 9–18. Pentiricci M. (2010) “L’attività edilizia a Leptis Magna tra l’età tetrarchica e il V secolo: una messa a punto”, in Leptis Magna. Una città e le sue iscrizioni in epoca tardoromana, edd. I. Tantillo and F. Bigi (Cassino 2010) 97–171. Peretz I. and Nahshoni P. (2009) “El-Burj”, Excavations and Surveys in Israel 121 [Online]. Available: http://www .hadashot-esi.org.il/report_detail_eng.aspx?id=1226&mag_ id=115 [Accessed on: 19/05/2015]. Peréx M. J., Fernández Ochoa C., García-Entero V., et al. (2014) “Thermes et balaneia en Hispanie. Un état de la question”, in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.1) (Cairo 2014) 67–82. Pesando F. and Guidobaldi F. (2006) Pompei, Oplontis, Ercolano, Stabiae (Guide archeologiche Laterza 14) (Bari 2006). Pesce G. (1950) Il “Palazzo delle Colonne” di Tolemaide di Cirenaica (Monografie di archeologia libica 2) (Rome 1950). Pettenò E. (1996) “Acque termali e medici dell’ Africa romana”, AfrRom 11 (1996) 385–402. Pfretzschner E. (1909) Die Grundrissentwicklung der römischen Thermen: nebst einem Verzeichnis der erhaltenen altrömischen Bäder mit Literaturnachweis (Zur Kunstgeschichte des Auslandes 65) (Strasbourg 1909). Pharr C. (2001) The Theodosian code and novels, and the Sirmondian constitutions (Union 2001). Pia Rossignani M. (1985) “L’area a nord del Foro”, in Luni— guida archeologica, ed. A. Frova (Sarzanna 1985) 55–62.

Bibliography Picard G. (1949) “Séance de la commission de l’Afrique du Nord”, BAC (1946–1949) 215–30. Picard G. (1950) “Séance de la commission de l’Afrique du Nord. I. Carthage”, BAC (1950) 74–81. Picard G. (1956) “Rapport sur les campagnes de fouilles effectuées à Mactar en 1954 et 1955”, BAC (1955) 173–79. Picard G. (1974) “Les fouilles de Mactar (Tunisie) 1970–1973”, CRAI (1974) 9–33. Pietrangeli C. (1943) Ocriculum (Otricoli) (Rome 1943). Pilkington J. (1943) The Confessions of St. Augustine (New York 1943). Piranomonte M. (2012) ed. Le Terme di Carcalla (Rome 2012). Piraud-Fournet P. (2014) “Les bains d’apparat des résidences urbaines du Proche-Orient aux époques byzantine et omeyyade”, in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 687–710. Pisani Sartorio G. (1996) “Macellum Liviae”, LTUR 3 (1996) 203–204. Platnauer M. (1922) Claudian. Volume II. On Stilicho’s Consulship 2–3. Panegyric on the Sixth Consulship of Honorius. The Gothic War. Shorter Poems. Rape of Proserpina. Translated by M. Platnauer (Loeb classical library 136) (Cambridge, Mass. 1922). Platner S. B. and Ashby T. (1929) A Topographical Dictionary of Ancient Rome (London 1929). Poccardi G. (2001) “Thermes et bains de l’Ostie antique”, in Ostia: port et porte de la Rome antique, ed. J.-P. Descœudres (Geneva 2001) 161–71. Poccardi G. (2006) “Les bains de la ville d’Ostie à l’époque tardo-antique (fin IIIe—début VIe siècle)”, in Les cités de l’Italie tardo-antique, IVe–VIe siècle: institutions, économie, société, culture et religion, edd. M. Ghilardi, C. J. Goddard and P. Porena (CÉFR 369) (Rome 2006) 167–86. Poinssot L. and Lantier R. (1922) “Séance de la commission de l’Afrique du Nord”, BAC (1922) LX–LXIV. Poinssot L. and Lantier R. (1925) “Séance de la commission de l’Afrique du Nord”, BAC (1925) XXVIII–XL. Porath Y., Raban A. and Patrich J. (1998) “The Caesarea Excavation Project—March 1992–June 1994”, Excavations and Surveys in Israel 17 (1998) 38–49. Poulle A. (1878) “Le bains de Pompeianus”, Receuil des notices et mémoires de la société archéologique, historique et géographique du département de Constantine 19 (1878) 431–54. Pringle D. (1981) The Defence of Byzantine Africa from Justinian to the Arab Conquest. Vol. I (BAR-IS 99) (Oxford 1981). Pronti A. (1996) “S. Maria Trans Tiberim, titulus”, LTUR 3 (1996) 219–20.

Bibliography Purcell N. (1995a) “Forum Romanum (the Republican Period)”, LTUR 2 (1995) 325–36. Purcell N. (1995b) “Forum Romanum (the Imperial Period)”, LTUR 2 (1995) 336–42. Puschmann T. (1878) Alexander von Tralles: Original-Text und Übersetzung nebst einer einleitenden Abhandlung: ein Beitrag zur Geschichte der Medizin, Band 1 (Vienna 1878). Puschmann T. (1879) Alexander von Tralles: Original-Text und Übersetzung nebst einer einleitenden Abhandlung: ein Beitrag zur Geschichte der Medizin, Band 2 (Vienna 1879). Rapoport A. (1974) “Symbolism and Environmental Design”, Journal of Architectural Education 27 (1974) 58–63. Rapoport A. (1976) “Sociocultural Aspects of Man-Environment Studies” in The Mutual Interaction of People and Their Built Environment. A Cross-Cultural Perspective, ed. A. Rapoport (The Hague 1976) 7–35. Rea R. (1993) “Amphitheatrum”, LTUR 1 (1993) 30–35. Rebuffat R. (1970a) “Bu Njem, 1968”, LibAnt 6–7 (1970) 9–107. Rebuffat R. (1970b) “Bu Njem, 1970”, LibAnt 6–7 (1970) 107–68. Rebuffat R. (1975) “Bu Njem, 1971”, LibAnt 11–12 (1975) 189–242. Rebuffat R. (1977) “Bu Njem, 1972”, LibAnt 13–14 (1977) 33–78. Rebuffat R. (1991) “Vocabulaire thermal: Documents sur le bain romain”, in Les thermes romains: actes de la table ronde (Rome, 11–12 novembre 1988), ed. Anonymous (CÉFR 142) (Rome 1991) 1–34. Rebuffat R., Joulia J. C. and Monthel G. (1987) “Note préliminaire sur les grands thermes d’Apollonia”, LibAnt 15–16 (1987) 263–78. Reddé M. (2009) “Trois petits balnéaires du désert Oriental d´Égypte”, in Le bain collectif en Égypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études Urbaines 7) (Cairo 2009) 213–20. Redon B. (2009) “L’armée et les bains en Égypte hellénistique et romaine”, BIFAO 109 (2009) 407–50. Redon B. (2012a) “L’insertion spatiale et économique des établissements balnéaires en Égypte aux époques hellénistique et romaine”, in ‘Quartiers’ artisanaux en Grèce ancienne: une perspective méditerranéenne, edd. A. Esposito and G. Sanidas (Villeneuve d’Ascq 2012) 57–80. Redon B. (2012b) “Établissements balnéaires et présences grecque et romaine en Égypte”, in Grecs et Romains en Égypte: territoires, espaces de la vie et de la mort, objets de prestige et du quotidien, ed. P. Ballet (Bibliothèque d’étude 157) (Cairo 2012) 155–69. Redon B. (2017a) ed. Collective Baths in Egypt 2 (Études urbaines 10) (Cairo 2017). Redon B. (2017b) “The Missing Baths of the First and Second Centuries in Egypt: A Tentative Explanation”, in Collective Baths in Egypt 2. Balaneia, thermae, hammam, ed. B. Redon (Études urbaines 10) (Cairo 2017) 267–78. Reekmans L. (1989) “L’implantation monumentale chrétienne dans le paysage urbain de Rome de 300 à 850”, in Actes du XIe congrès international d’archéologie chrétienne: Lyon,

491 Vienne, Grenoble, Genève et Aoste, 21–28 septembre 1986, ed. N. Duval (CÉFR 123/Studi di Antichità cristiana 41) (Rome 1989) 861–915. Reich R. and Shukron E. (1994) “Jerusalem, Mamilla”, Excavations and Surveys in Israel 14 (1994) 92–96. Reis M. P. (2004) Las termas y balnea romanos de Lusitania (Studia Lusitania 1) (Madrid 2004). Reynolds J. (2003) Christian monuments of Cyrenaica by J. B. Ward-Perkins and R. G. Goodchild (Society for Libyan Studies. Monographs 4) (London 2003). Reynolds J. and Lloyd J. A. (1996) “Cyrene”, CAH 10 (1996) 619–40. Riad H. (1975) “Anciens bains d’Alexandrie”, BSRAA 43 (1975) 113–22. Ricciardi M. A. and Scrinari S. M. V. (1996a) La civiltà dell’acqua in Ostia Antica. Volume II (Rome 1996). Ricciardi M. A. and Scrinari S. M. V. (1996b) La civiltà dell’acqua in Ostia Antica. Volume I (Rome 1996). Richardson L. (1992) A new topographical dictionary of ancient Rome (Baltimore 1992). Ring J. W. (1996) “Windows, baths, and solar energy in the Roman empire”, AJA 100 (1996) 717–24. Risse M. (2001) Volubilis. Eine römische Stadt in Marokko von der Frühzeit bis in die islamische Periode (Mainz 2001). Robinson O. (1992) Ancient Rome: city planning and administration (London and New York 1992). Rodríguez Almeida E. (1981) Forma Urbis Marmorea. Aggiornamento Generale 1980 (Rome 1981). Rodríguez Almeida E. (1984) Il Monte Testaccio: ambiente, storia, materiali (Rome 1984). Rodziewicz E. and Rodziewicz M. (1983) “Alexandrie 1976”, ÉtTrav 12 (1983) 241–61. Rodziewicz M. D. (1979) “Thermes romains près de la Gare Centrale d’Alexandrie”, ÉtTrav 11 (1979) 108–38. Rodziewicz M. D. (1984) Les habitations romaines tardives d’Alexandrie: à la lumière des fouilles polonaises à Kôm elDikka (Warsaw 1984). Rodziewicz M. D. (2009) “Ancient Baths in Alexandria”, in Le bain collectif en Égypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études Urbaines 7) (Cairo 2009) 191–201. Rolfe J. C. (1950) Ammianus Marcellinus. History, Volume I. Books 14–19 (Loeb Classical Library 300) (Cambridge, Mass. 1950). Rolfe J. C. (1952) Ammianus Marcellinus. History. Volume III. Books 27–31 (Loeb classical library 331) (Cambridge, Mass. 1952). Rook T. (1978) “The Development and Operation of Roman Hypocausted Bath”, JAS 5 (1978) 269–82. Rook T. (1979) “The effect of the evolution of flues upon the development of architecture”, in Roman brick and tile. Studies in manufacture, distribution and use in the Western empire, ed. A. McWhirr (BAR-IS 68) (Oxford 1979) 303–308.

492 Rook T. (1994) “Thoughts on hypocausts”, Balnearia 2 (1994) 10–11. Rose V. (1894) Theodori Prisciani Euporiston libri III cum Physicorum fragmento et additamentis pseudo-Theodoreis (Leipzig 1894). Rossi F. (1990) “Brescia”, in Milano capitale dell’ Impero Romano 286–402 d.C. (Milano, Palazzo Reale, 24 gennaio-22 aprile 1990), ed. Anonymous (Milan 1990) 153–54. Rossiter J. J. (1998) “A Roman bath-house at Bir el Jebbana: preliminary report on the excavations (1994–1997)”, in Carthage papers: the early colony’s economy, water supply, a public bath, and the mobilization of state olive oil, edd. J. T. Peña, J. J. Rossiter, A. Wilson, et al. (JRA Supplementary Series 28) (Portsmouth 1998) 103–115. Rossiter J. J. (2009) “The ‘neighbourhood baths’ of Roman Carthage”, in Contrôle et distribution de l’eau dans le Maghreb antique et médiéval, ed. V. Bridoux (CÉFR 426) (Rome 2009) 177–97. Rostovtzeff M. (1903–1905) Tesserarum urbis Romae et Suburbi plumbearum sylloge (St-Petersburg 1903–1905). Rousseau P. (1996a) “Basil of Caesarea”, OCD (1996) 234. Rousseau P. (1996b) “Gregory of Nyssa”, OCD (1996) 657. Rousselle A. (1988) Porneia: on desire and the body in antiquity (Family, sexuality, and social relations in past times) (Cambridge, Mass. 1988; Eng. translation [1983]). Rousset M.-O. and Rochette M. (2014) “Des bains à Qinnasrin/ al-’Is (Syrie)”, in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.2) (Cairo 2014) 519–34. Rubinich M. (2006) “Aquileia (UD). Le indagini dell’università di Udine alle ‘Grandi Terme’, località Braida Murada (scavi 2005–2006)”, Notiziario della Soprintendenza per i beni archeologici del Friuli Venezia Giulia 1 (2006) 151–58. Rubinich M. (2007) Aquileia (UD). Le indagini dell’università di Udine alle ‘Grandi Terme’ in località Braida Murada, Notiziario della Soprintendenza per i beni archeologici del Friuli Venezia Giulia 2 (2007) 133–40. Rubinich M. (2008) “Aquileia (UD). Le indagini dell’Università di Udine alle ‘Grande Terme’ in località Braida Murada (2008)”, Notiziario della Soprintendenza per i beni archeologici del Friuli Venezia Giulia 3 (2008) 161–66. Rubinich M. (2013) “Le Thermae Felices Constantinianae”, in Costantino e Teodoro. Aquileia nel IV secolo, edd. C. Tiussi, L. Villa and M. Novello (Milan 2013) 85–90. Russell B. (2013) The economics of the Roman stone trade (Oxford studies on the Roman economy) (Oxford 2013). Sadek M. (1992) “The Baths at the Ancient Harbour of Marea”, in Atti del sesto congresso internazionale di Egittologia, ed. J. Leclant (Turin 1992) 549–54.

Bibliography Saguì L. (1990a) “Lo scavo”, in Archeologia urbana a Roma: il progetto della Crypta Balbi. 5, L’esedra della Crypta Balbi nel medioevo (XI–XV secolo), edd. L. Saguì and L. Paroli (Florence 1990) 15–94. Saguì L. (1990b) “‘Balnea’ medievali: trasformazione e continuità della tradizione classica”, in Archeologia urbana a Roma: il progetto della Crypta Balbi. 5, L’esedra della Crypta Balbi nel medioevo (XI–XV secolo), edd. L Saguì and L. Paroli (Florence 1990) 98–116. Saguì L. (2009) “Pendici nord-orientali del Palatino: le ‘terme di Elagabalo’. Indagini archeologiche e prime riflessioni”, ArchCl 60 (2009) 235–74. Saguì L. (2013) “Area delle ‘Terme di Elagabalo’: tre milleni di storia alle pendici del Palatino”, in Scavare nel Centro di Roma, ed. C. Panella (Rome 2013) 133–52. Saguì L. and Cante M. (2015) “Archeologia e architettura nell’area delle ‘Terme di Elagabalo’, alle pendici nordorientali del Palatino. Dagli isolati giulio-claudii alla chiesa paleocristiana”, Thiasos 4 (2015) 37–75. Saliou C. (2014) “Bains et histoire urbaine. L’exemple d’Antioche sur l’Oronte dans l’Antiquité”, in 25 siècles de bain collectif en Orient. Proche-Orient, Égypte et péninsule Arabique. Balaneia, Thermae, Hammam. Actes du 3e colloque international Balnéorient (Damas/Syrie 2–6 nov. 2009), edd. M.-F. Boussac, S. Denoix, T. Fournet and B. Redon (Études urbaines 9.3) (Cairo 2014) 657–86. Salvatore M. (1984) “Area c.d. delle terme”, in Venosa: un parco archeologico ed un museo. Come e perché, ed. M. Salvatore (Mediterraneo tardoantico e medievae: Scavi e ricerche 2) (Taranto 1984) 82–85. Salvatore M. (2004) “Verso il Medioevo”, in Custodire l’antichità. La Soprintendenza per i Beni Archeologici dell’Umbria 1964– 2004, ed. M. Saioni (Perugia 2004) 55–61. Salza Prina Ricotti E. (1971) “Le ville maritime di Silin (Leptis Magna)”, RendPontAcc 43 (1971) 135–63. Sánchez J. A. J. and Carbonell J. S. (2004) “Termas e iglesias durante la Antigüedad Tardía: ¿reutilización arquitectónica o conflicto regigioso? Algunos ejemplos hispanos”, in Sacralidad y Arqueología, edd. J. M. Blázquez Martínez and A. G. Blanco (Antigüedad y Cristianismo. Monografías Históricas sobre la Antigüedad Tardía 21) (Murcia 2004) 185–201. Santangelli Valenzani R. and Meneghini R. (1996) “Fasi tarde di occupazione dell’isolato con domus e balnea”, in Antiche Stanze. Un quartiere di Roma imperiale nella zona di Termini, edd. M. Barbera and R. Paris (Milan 1996) 172–77. Saradi-Mendelovici H. (1990) “Christian Attitudes towards Pagan Monuments in Late Antiquity and Their Legacy in Later Byzantine Centuries”, DOP 44 (1990) 47–61. Sarnowski T. (1978) Les représentations de villas sur les mosaïques africaines tardives (Archiwum filologiczne 37) (Wroclaw 1978).

Bibliography Sartori N. (2015) “Le bain tardif de Dharih (Jordanie méridionale)”, Syria 92 (2015) 123–31. Sartre M. (2006) Histoires grecques (Paris 2006). Saumagne C. (1925) “Fouilles récemment exécutées à Carthage par la Direction des Antiquités”, BAC (1925) XCVI–XCVIII. Saumagne C. (1928) “Notes de topographie carthaginoise”, BAC (1928) 629–64. Saunders J. J. (1985) A History of Medieval Islam (London 1985). Scarborough J. (1969) Roman medicine (Aspects of Greek and Roman life) (London 1969). Schaff P. (s.d.) Saint Chrysostom: Homilies on Galatians, Ephesians, Philippians, Colossians, Thessalonians, Timothy, Titus and Philemon (A select library of the Nicene and Post-Nicene Fathers of the Christian Church 13) (Edinburgh s.d.). Schaff P. and Wace H. (s.d.) The Seven Ecumenical councils (A Select Library of the Nicene and Post-Nicene Fathers of the Christian Church, Second Series) (Edinburgh s.d.). Schick R. (2007) “Byzantine Jerusalem”, in Jerusalem before Islam, edd. Z. Kafafi and R. Schick (BAR-IS 1699) (Oxford 2007) 169–88. Schiebold H. (2010) Heizung und Wassererwärmung in römischen Thermen: Historische Entwicklung, Nachfolgesysteme, Neuzeitliche Betrachtungen und Untersuchungen (Schriften der Deutschen Wasserhistorischen Gesellschaft Sonderbd 3) (Siegburg 2010, 2nd edition). Schmidt K. (2012) Das umayyadische “Wüstenschloss” und die Siedlung am Ǧabal Says. Band I, Architektur (Damaszener Forschungen 13) (Darmstadt and Mainz 2012). Schmölder-Veit A. (2010) “Nymphäumsräume: Neue Treffpunkte in der spätantiken Stadt”, in Die antike Stadt im Umbruch: Kolloquium in Darmstadt, 19. Bis 20. Mai 2006, edd. N. Burkhardt and R. Stichel (Wiesbaden 2010) 109–119. Schöllgen G. (1995) “Balnea mixta. Entwicklungen der spätantiken Bademoral im Spiegel der Textüberlieferung der ‘Syrischen Didaskalie’”, in Panchaia. Festschrift für Klaus Thraede (Münster 1995) 182–94. Scholz P. (2004) “Elementarunterricht und intellektuelle Bildung im hellenistische Gymnasion”, in Das hellenistische Gymnasium, edd. P. Scholz and D. Kah (Wissenskultur und gesellschaftlicher Wandel 8) (Berlin 2004) 103–128. Scobie A. (1986) “Slums, Sanitation and Mortality in the Roman World”, Klio 68 (1986) 399–433. Scourfield J. H. D. (1996) “Jerome (Eusebius Hieronymus)”, OCD (1996) 794. Sears G. (2007) Late Roman African Urbanism. Continuity and transformation in the city (BAR-IS 1693) (Oxford 2007). Sedlak D. (2014) Water 4.0: the past, present, and future of the world’s most vital resource (New Haven 2014). Segal A. (2013) “Urban plan and city landscape”, in Hippos-Sussita of the Decapolis. The first twelve seasons of excavations 2000–2011. Volume 1, edd. A. Segal, M. Eisenberg, J. Młynarczyk, et al. (Haifa 2013) 64–85.

493 Seriy G. (2011) “Khirbat Lasan”, Excavations and Surveys in Israel 123 [Online]. Available: http://www.hadashot-esi.org .il/report_detail_eng.aspx?id=1725&mag_id=118 [Accessed on: 05/05/2015]. Serr P and Bacco G. (1998) “Forum Traiani: il contesto termale e l’indagine archeologica di scavo”, AfrRom 12 (1998) 1213–55. Sgobbo I. (1929) “Terme Flegree ed origine delle terme romane”, in Atti del I Congresso Nazionale di Studi Romani, ed. Anonymous (Rome 1929) 186–94. Shaw B. D. (1984) “Water and society in the ancient Maghrib: technology, property and development”, AntAfr 20 (1984) 121–73. Shaw J. (2012) “Bathing at the Mycenaean Palace of Tiryns”, AJA 116 (2012) 555–71. Sheehan P. (2010) Babylon of Egypt: the archaeology of old Cairo and the origins of the city (Cairo 2010). Shenhau A. (1986) “Horvat Hanot (Khirbet el-Khan)”, Excava­ tions and Surveys in Israel 5 (1986) 46–47. Shenouda S. (1973) “Alexandria University excavations on the cricket playgrounds in Alexandria”, OpRom 9 (1973) 193–205. Shiyyab A. (2014) “Recent Discoveries in City Center Of Petra, Jordan: A Preliminary Excavation Report”, Mediterranean Archaeology and Archaeometry 14.1 (2014) 103–118. Sidebotham S. (1993) “University of Delaware Fieldwork in the Eastern Desert of Egypt”, DOP 48 (1993) 263–75. Sion O. (2009) “Ashqelon, el-Qabu”, Excavations and Surveys in Israel 121 [Online]. Available: http://www.hadashot-esi.org .il/report_detail_eng.aspx?id=1105&mag_id=115 [Accessed on: 19/05/2015]. Sion O. (2010) “Ramla”, Excavations and Surveys in Israel 122 [Online]. Available: http://www.hadashot-esi.org.il/ report_detail_eng.aspx?id=1539&mag_id=117 [Accessed on: 06/05/2015]. Sijpesteijn P. M. (2007) “The Arab conquest of Egypt and the beginning of the Muslim rule”, in Egypt in the Byzantine world, 300–700, ed. R. Bagnall (Cambridge 2007) 437–59. Sklar-Parnes A. (2006) “Jerusalem, Shu’fat”, Excavations and Surveys in Israel 118 [Online]. Available: http://www .hadashot-esi.org.il/report_detail_eng.aspx?id=347&mag_ id=111 [Accessed on: 19/05/2015]. Smith A. M. (2005) “Bir Madhkur Project: A Preliminary Report on Recent Fieldwork”, BASOR 340 (2005) 57–75. Smith A. M. (2010) “The Bir Madhkur Project: A Preliminary Report on Recent Fieldwork”, ADAJ 58 (2010) 143–52. Smith R. W. and Bugni V. (2006) “Symbolic Interaction Theory and Architecture”, Symbolic Interaction 29 (2006) 123–55. Solimany G. (2012) “Neve Shalom (East)”, Excavations and Surveys in Israel 124 [Online]. Available: http://www .hadashot-esi.org.il/report_detail_eng.aspx?id=2062& mag_id=119 [Accessed on: 29/04/2015].

494 Souville G. (1976) “Madauros (Mdaourouch) Algeria”, PECS (1976) 541. Spruit J. E. (1997) Corpus iuris civilis: tekst en vertaling. 4: Digesten 25–34 (Zutphen 1997). Spurza J. (1999) “The Building History of the Palazzo Imperiale at Ostia: Evolution of an Insula on the Banks of the Tiber River”, Meded 58 (1999) 129–42. Squatriti P. (1998) Water and society in early medieval Italy, A.D. 400–1000 (Cambridge 1998). Staccioli R. A. (1958) “Sugli edifici termali minori”, ArchCl 10 (1958) 273–78. Staccioli R. A. (1961) “Terme minori e balnea nella documentazione della Forma Urbis”, ArchCl 13 (1961) 92–102. Staccioli R. A. (1968) “Una probabile sede di corporazione nella Pianta Marmorea di Roma antica”, Quaderni di topografia antica 5 (1968) 23–26. Stager L. and Esse D. (1986) “Ashkelon”, Excavations and Surveys in Israel 5 (1986) 2–6. Stasolla F. R. (2002a) Pro labandis curis: il balneum tra tarda antichità e medioevo (Tardoantico e medioevo: studi e strumenti di archeologia 4) (Rome 2002). Stasolla F. R. (2002b) “Balnea ed edifici di culto: relazioni e trasformazioni tra tarda antichità e alto medioevo”, in Ecclesiae urbis: atti del Congresso internazionale di studi sulle chiese di Roma (IV–X secolo), Roma, 4–10 settembre 2000, edd. F. Guidobaldi and A. Guidobaldi (Studi di antichità cristiana 59) (The Vatican 2002) 143–51. Stasolla F. R. (2008) “Tra igiene e piacere: thermae e balnea nell’alto medioevo”, in L’acqua nei secoli altomedievali: Spoleto, 12–17 aprile 2007, ed. Anonymous (Settimane di studio della Fondazione Centro italiano di studi sull’alto medioevo 55) (Spoleto 2008) 873–925. Stieglitz R. (2006) Tell Tanninim. Excavations at Krokodeilon Polis 1996–1999 (Boston 2006). Stöger H. (2011) Rethinking Ostia: A Spatial Enquiry into the Urban Society of Rome’s Imperial Port-Town (Archaeological Studies Leiden University 24) (Leiden 2011). Stommel E. (1959) “Christliche Taufriten und antike Badesitten”, JAC 2 (1959) 5–14. Storz S. (1994) Tonröhren im antiken Gewölbebau: mit einer Rekonstruktion des Schalungstragwerkes für die Trompengewölbe der Kobbat Bent el Rey in Carthage (Sonderschriften Deutsches Archäologisches Institut. Römische Abteilung 10) (Mainz 1994). Strong D. (1994) Roman museums: selected papers on Roman art and architecture (London 1994). Stucchi S. (1975) Architettura cirenaica (Monografie di archeologia libica 9) (Rome 1975). Subias Pascual E. (1991) La Domus dels Dioscurs d’Ostia Antica (Tarragona 1991). Suleiman A., Albini P. and Migliavacca P. (2004) “A short introduction to historical earthquakes in Libya”, Annals of Geophysics 47 (2004) 545–54.

Bibliography Synek E. (1998) “Christliche Badekultur”, in Alltägliches Altertum, ed. E. Specht (Frankfurt 1998) 227–50. Szymanska H. and Babraj K. (2008) Marea Vol. 1. Byzantine Marea. Excavations in 2000–2003 and 2006 (Biblioteka Muzeum Archeologicznego w Krakowie 4) (Kraków 2008). Szymanska H. and Babraj K. (2009) “Les bains de Marea”, in Le bain collectif en Égypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Études urbaines 7) (Cairo 2009) 247–53. Tabaie A. and Carrez-Maratray J.-Y. (2009) “Les bains romains de Farama. État des lieux et projets”, in Le bain collectif en Égypte. Balaneia, thermae, hammam, ed. M.-F. Boussac, T. Fournet and B. Redon (Études urbaines 7) (Cairo 2009) 221–28. Tabar M. A. and Gendelman P. (2013) “Horbat Mesar: A Roman And Byzantine-Period Bathhouse”, Excavations and Surveys in Israel 125 [Online]. Available: http://www.hadashot-esi .org.il/images//Meiser.pdf [Accessed on: 15/06/2015]. Tagliamonte G. (1999) “Palatium. Palatinus Mons (fina alla prima età repubblicana)”, LTUR 4 (1999) 14–22. Tamburino A. (2010) “Water Technology in Ancient Mesopotamia”, in Ancient Water Technologies, ed. L. W. Mays (New York 2010) 29–51. Tantillo I. (2010) “I costumi epigrafici. Scritture, monumenti, pratiche”, in Leptis Magna. Una città e le sue iscrizioni in epoca tardoromana, edd. I. Tantillo and F. Bigi (Cassino 2010) 173–203. Tantillo I and Bigi F. (2010) edd. Leptis Magna. Una città e le sue iscrizioni in epoca tardoromana (Cassino 2010). Thatcher E. D. (1956) “The Open Rooms of the Terme del Foro at Ostia”, MAAR 24 (1956) 167–264. Thébert Y. (1983) “L’évolution urbaine dans les provinces orientales de l’Afrique romaine tardive”, Opus 2 (1983) 99–130. Thébert Y. (1991) “Les sodalités dan les thermes d’Afrique du Nord”, in Les thermes romains: actes de la table ronde (Rome, 11–12 novembre 1988), ed. Anonymous (CÉFR 142) (Rome 1991) 193–204. Thébert Y. (2003) Thermes romains d’Afrique du Nord et leur contexte méditerranéen: études d’histoire et d’archéologie (BÉFAR 315) (Rome 2003). Thirion J. (1957) “Un ensemble thermal avec mosaïques à Thina (Tunisie)”, MÉFR 69 (1957) 207–245. Tholbecq L., Delcros S. and Paridaens N. (2015) “Les bains du Jabal Khubthah (Pétra, Jordanie)”, Syria 92 (2015) 23–32. Thomas E. and C. Witschel (1992) “Constructing reconstruction: claim and reality of Roman rebuilding inscriptions from the Latin West”, PBSR 60 (1992) 135–77. Thorndike L. (1928) “Sanitation, baths and street-cleaning in the Middle Ages and Renaissance”, Speculum 3 (1928) 192–203. Thouvenot R. (1941) “La maison d’Orphée à Volubilis”, Publications du Service des antiquités du Maroc 6 (1941) 42–66.

Bibliography Thouvenot R. (1945) “Les thermes du Nord”, Publications du Service des antiquités du Maroc 7 (1945) 156–65. Thouvenot R. (1948a) “La maison aux travaux d’Hercule”, Publications du Service des antiquités du Maroc 8 (1948) 69–108. Thouvenot R. (1948b) “Le quartier Nord-Est. La rive droite du decumanus maximus”, Publications du Service des antiquités du Maroc 8 (1948) 109–143. Thouvenot R. (1958) Maisons de Volubilis: le palais dit de Gordien et la maison à la mosaïque de Venus (Publications du Service des antiquités du Maroc) (Rabat 1958). Thouvenot R. and Luquet A. (1951) “Les thermes de Banasa”, Publications du Service des antiquités du Maroc 9 (1951) 9–62. Tohme L. (2011) “Between balneum and hamam: the role of Ummayad baths in Syria”, in Bathing culture of Anatolian civilizations: architecture, history, and imagination, ed. N. Ergin (Ancient Near Eastern Studies. Supplement 37) (Leuven 2011) 65–75. Tomasello F. (1984) “Un edificio termale a Misterbianco (Catania)”, Cronache di Archeologia 18 (1979) 187–204. Tomasello F. (1997) “L’edificio termale di Misterbianco. Problemi di metodologia progettuale antica”, Cronache di Archeologia 31 (1997) 117–22. Tortorici E. (1994) “Terme Severiane, terme ‘Severiane’ e terme Septimianae”, BullCom 92 (1994) 161–72. Tourrenc S. (1968) “La dédicace du temple du Génie de la colonie à Timgad”, AntAfr 2 (1968) 197–220. Trümper M. (2009) “Complex Public Bath Buildings of the Hellenistic Period. A Case Study in Regional Differences”, in Le bain collectif en Egypte. Balaneia, thermae, hammam, edd. M.-F. Boussac, T. Fournet and B. Redon (Etudes Urbaines 7) (Cairo 2009) 139–79. Trümper M. (2013) “Urban Context of Greek Public Baths”, in Greek Baths and Bathing Culture. New discoveries and approaches, edd. S. K. Lucore and M. Trümper (BABesch Supplement 23) (Leuven 2013) 33–72. Trümper M. (2014) “‘Privat’ versus ‘öffentlich’ in hellenistischer Bädern”, in Stadtkultur im Hellenismus, edd. A. Matthaei and M. Zimmermann (Die hellenistische Polis als Lebensform 4) (Heidelberg 2014) 206–249. Tsafrir Y. (1988) “An Introduction to Rehovot-in-the-Negev”, in Excavations at Rehovot-in-the-Negev. Volume I: The Northern Church, edd. Y. Tsafrir, J. Patrich, R. Rosenthal-Heginbottom et al. (Qedem 25) (Jerusalem 1988) 1–16. Tsafrir Y. and Förster G. (1994) “From Scythopolis to Baysan— Changing Concepts of Urbanism”, in The Byzantine and early Islamic Near East. II, Land use and settlement patterns: papers of the second workshop on Late Antiquity and early Islam, edd. G. R. D. King and A. Cameron (Princeton 1994) 95–115.

495 Tsafrir Y. and Förster G. (1997) “Urbanism at Scythopolis— Beth Shean in the Fourth to Seventh Centuries”, DOP 51 (1997) 85–146. Tsiolis V. (2001) “Las termas de Fregellae. Arquitectura, tecnología y cultura balnear en el Lacio durante los siglos III y II a.C”, Cuadernos de prehistoria y arqueología 27 (2001) 85–114. Tsiolis V. (2006) “Fregellae: il complesso termale e le origini degli edifici balneari urbani nel mondo romano”, in Sicilia ellenistica, consuetudo italica: alle origini dell’architettura ellenistica d’occidente ; [atti del convegno] Spoleto, Complesso monumentale di S. Nicolò, 5–7 Novembre 2004, edd. M. Osanna and M. Torelli (Biblioteca di “Sicilia antiqua” 1) (Pisa 2006) 243–55. Tsiolis V. (2013) “The Baths at Fregellae and the Transition from Balaneion to Balneum”, in Greek Baths and Bathing Culture. New discoveries and approaches, edd. S. K. Lucore and M. Trümper (BABesch Supplement 23) (Leuven 2013) 89–111. Turci M. (2014) Un complesso termale tardo-antico. Il balneum di Musiciolus (IV, XV, 2) nel quartiere fuori Porta Marina, MÉFRA [En ligne] 126. Tzaferis V. (1980) “A Roman bath at Rama”, ‘Atiqot 14 (1980) 66–75. Tzaferis V. and Israeli S. (1999) “Banias—1996”, Hadashot Arkheologiyot 109 (1999) 1–2. Ugurlu E. and Böke H. (2009) “The use of brick-lime plasters and their relevance to climatic conditions of historic bath buildings”, Construction and Building Materials 23 (2009) 2442–50. Van Vaerenberg J. (2011) “Flush Water for Toilets in and near the Baths”, in Roman toilets: their archaeology and cultural history, edd. G. C. M. Jansen, A. Koloski.-Ostrow. and E. Moormann (BABesch Supplement 19) (Leuven 2011) 78–86. Vaultrin J. (1932) “Les mosaïques chrétiennes de Carthage”, RAfr 73 (1932) 182–318. Veloccia Rinaldi M. L. (1974) “Scavi alle Terme Maritime”, FA 24–25 (1974) 563–65 (nr. 8342). Veloccia Rinaldi M. L. (1975) “Il pons matidiae”, in Ricerche archeologiche nell’Isola sacra, edd. M. L. Veloccia Rinaldi and P. Testini (Instituto nazionale d’archeologia e storia dell’arte. Monografie 2) (Rome 1975) 13–39. Vergara Caffarelli E. (1957) “Leptis Magna”, FA 12 (1957) 351–52. Versluys M. J. (2014) “Understanding objects in motion. An archaeological dialogue on Romanization”, Archaeological Dialogues 21 (2014) 1–20. Vigarello G. (1985) Le propre et le sale: l’hygiène du corps depuis le Moyen Age (Paris 1985). Villard L. (1994) “Le bain dans la médicine hippocratique”, in L’eau, la santé et la maladie dans le monde grec: actes du

496 colloque organisé à Paris (CNRS et Fondation Singer-Polignac) du 25 au 27 novembre 1992, edd. R. Ginouvès, A.-M. Guimier-Sorbets and J. Jouanna (B.C.H supplément 28) (Athens 1994) 41–60. Villicich R. (2002) “Nouve ricerche archeologiche nell’are del ‘Palazzo di Teoderico’ a Galeata (campagne di scavo 1998– 2001)”, Ocnus 9–10 (2002) 251–57. Villicich R. (2012) “Scavi nell’area della villa di Teoderico a Galeata (FC): i nuovi dati”, The Journal of Fasti Online [Online]. Available: http://www.fastionline.org/docs/ FOLDER-it-2012-261.pdf [Accessed on: 26/01/2015]. Villicich R. (2014) “La villa teodericiana di Galeata: risultati e prospettive dopo le recenti campagne di scavo”, in La villa restaurata e i nuovi studi sull’edilizia residenziale tardoantica atti del Convegno internazionale del Centro interuniversitario di studi sull’edilizia abitativa tardoantica nel Mediterraneo (CISEM), Piazza Armerina 7–10 novembre 2012, edd. P. Pensabene and C. Sfameni (Bari 2014) 241–50. Vilucchi S. (1985) “Terme di Costantino”, in Roma Archeologia nel Centro II: la ‘città murata’, ed. A. M. Bietti Sestieri (Rome 1985) 357–59. Vilucchi S. (1989) “Le terme di Costantino sul Quirinale e gli edifici privati di età precedente”, BullCom 91 (1989) 350–55. Vitelli G. (1981) Islamic Carthage. The archaeological, historical and ceramic evidence (CEDAC Dossier) (Tunis 1981). Vitto F. (1987) “The Beth Shean Project: E. Excavations West of the South Bathhouse”, Excavations and Surveys in Israel 6 (1987) 45. Vladu A.-M. (2017). “The Theft of Water from the Aqueducts of the City of Rome in Frontinus’ Time”, Rivista Cicsa serie noua 3 (2017) 4–18. Volpe G. (2000a) “La domus delle Sette Sale”, in Aurea Roma. Dalla città pagana alla città cristiana, edd. S. Ensoli and E. La Rocca (Rome 2000) 161–62. Volpe G. (2000b) “Herdonia romana, tardoantica e medievale alla luce di scavi recenti”, in Ordona X. Ricerche archeologiche a Herdonia (1993–1998), ed. G. Volpe (Bari 2000) 491–530. Volpe G., Annese G. and Favia P. (2007) “Terme e complessi religiosi paleocristiani. Il caso di San Giusto”, in Bains curatifs et bains hygiéniques en Italie de l’antiquité au moyen âge, edd. M. Guérin-Beauvois and J. M. Martin (CÉFR 383) (Rome 2007) 217–61. Volpe G., Romano A. and Turchiano M. (2013) “San Giusto, L’ecclesia e il saltus carminianensis: vescovi rurali, insediamenti, produzioni argricole e artigianali. Un approccio globale allo studio della cristianizzazione delle campagne”, in Acta XV Congressus Internationalis Archaeologiae Christianae, ed. Anonymous (The Vatican 2013) 559–80. Volpe R. (2007a) “Le Terme di Traiano e la la ξυστιχή σύνοδος”, in Res Bene Gestae: ricerche di storia urbana su Roma antica

Bibliography in onore di Eva Margareta Steinby, edd A. Leone, D. Palombi and S. Walker (Rome 2007) 427–37. Volpe R. (2007b) “La villa cd. ad duas Lauros. Trasformazioni e sviluppo”, in Centocelle 2. Roma S.D.O. Le indagini archeologiche, ed. R. Volpe (Studi e materiali dei Musei e Monumenti comunali di Roma) (Rome 2007) 15–47. Waille V. (1886) “Note relative à des fouilles exécutées à Cherchel au mois de mai 1886”, CRAI 30 (1886) 301–305. Waille V. (1887) “Note sur les fouilles archéologiques exécutées à Cherchel (Algérie), par M. Victor Waille, professeur à l’école supérieure des lettres d’Alger”, CRAI 31 (1887) 53–58. Wallis Budge E. (1894) The Discourses of Philoxenus, Bishop of Mabbogh, A.D. 485–519 (London 1894). Ward R. B. (1992) “Women in Roman baths”, HTR 85 (1992) 125–47. Ward-Perkins B. (1977) “Ricerche su Luni Medievale. Lo scavo nella zona nord del Foro”, in Scavi di Luni 2. Relazione delle campagne di scavo 1972–1973–1974, ed. A. Frova (Rome 1977) 633–38. Ward-Perkins B. (1978) “L’abbandono degli edifici pubblici a Luni”, Quaderni. Centro studi lunensi 3 (1978) 33–46. Ward-Perkins B. (1984) From classical antiquity to the Middle Ages: urban public building in northern and central Italy, A.D. 300–850 (Oxford historical monographs) (Oxford and New York 1984). Ward-Perkins J. B. (1981) Roman imperial architecture (Pelican history of art) (Harmondsworth 1981). Ward-Perkins J. B., Little J. and Mattingly D. (1986) “Town Houses at Ptolemais, Cyrenaica: A Summary Report of Survey and Excavation Work in 1971, 1978–1979”, LibSt 17 (1986) 109–153. Ward-Perkins J. B. and Toynbee J. M. C. (1949) “The hunting baths at Lepcis Magna”, Archaeolgia 93 (1949) 166–95. Wareth U. and Zignani P. (1992) “Nag al-Hagar. A Fortress with a Palace of the Late Roman Empire. Second Preliminary Report”, BIFAO 92 (1992) 185–210. Watson P. (1996) “Pella Hinterland Survey 1994: Preliminary Report”, Levant 28 (1996) 63–76. Weber M. (1996) Antike Badekultur (Beck’s archäologische Bibliothek) (Munich 1996). Weeber K.-W. (2007) Baden, Spielen, lachen. Wie die Römer ihre Freizeit verbrachten (Geschichte erzählt) (Darmstadt 2007). Weiss Z. and Neyzer E. (1994) “Zippori—1992/1993”, Excava­ tions and Surveys in Israel 14 (1994) 40–46. Whitehead J. (2010) “The Baths at Carsulae. Excavations and Survey 2004–2008”, The Journal of Fasti Online [Online]. Available: http://www.fastionline.org/docs/FOLDER-it -2010-187.pdf [Accessed on: 27/05/2015]. Whitmore A. (2018) “Artefact Assemblages from Roman Baths: Expected, Typical, and Rare Finds” in Thermae in context: Roman bathhouses in the town and in daily life, edd.

Bibliography H. Pösche, A. Binsfeld and S. Hoss (Archaeologia Mosellana 10) (Luxembourg City 2018) 57–77. Wickham C. (2005) Framing the Early Middle Ages. Europe and the Mediterranean 400–800 (Oxford 2005). Wiedler S. (1999) Aspekte der Mosaikausstattung in Bädern und Thermen des Maghreb (Antiquitates. Archäologische Forschungsergebnisse 18) (Hamburg 1999). Wikander Ö. (1996) “Senators and equites, 6. Caius Sergius Orata and the invention of the hypocaust”, OpRom 20 (1996) 177–82. Wilkes J. J. (1999) “Approaching Roman Baths”, in Roman baths and bathing: proceedings of the first International Conference on Roman Baths held at Bath, England, 30 March–4 April 1992, edd. J. DeLaine and D. E. Johnston (JRA Supplementary Series 37) (Portsmouth 1999) 17–23. Williams F. (2009) The Panarion of Epiphanius of Salamis. Book I (Sects 1–46) (Nag Hamadi and Manichean Studies 63) (Leiden 2009). Wilson A. (1998) “Water supply in ancient Carthage”, in Carthage papers: the early colony’s economy, water supply, a public bath, and the mobilization of state olive oil, edd. J. T. Peña, J. J. Rossiter, A. Wilson et al. (JRA Supplementary Series 28) (Portsmouth 1998) 65–102. Wilson A. (2001) “Urban Water Storage, Distribution, and Usage in Roman North Africa”, in Water use and hydraulics in the Roman city, Colloquia and conference papers 3, ed. A. Koloski-Ostrow (Dubuque 2001) 83–96. Wilson A. (2003) “Classical water technology in the early Islamic world”, in Technology, ideology, water: from Frontinus to the Renaissance and beyond. Papers from a conference at the Institutum Romanum Finlandiae, May 19–20, 2000, edd. C. Bruun and A. Saastamoinen (Rome 2003) 116–41. Wilson R. J. A. (1983) Piazza Armerina (London 1983). Wilson R. J. A. (1992) “Terracotta vaulting tubes (tubi fittili). On their origin and distribution”, JRA 5 (1992) 97–129. Wilson R. J. A. (1996) “Lambaesis”, OCD (1996) 812. Wipszycka E. (2007) “The institutional church” in Egypt in the Byzantine world, 300–700, ed. R. Bagnall (Cambridge 2007) 331–49. Wiseman T. P. (1993) “Campus Martius”, LTUR 1 (1993) 220–24. Wissemann M. (1984) “Das Personal des antiken römischen Bades”, Glotta 62 (1984) 80–89. Woolf G. (2014) “Romanization 2.0 and its alternatives”, Archaeological Dialogues 21 (2014) 45–50. Wright F. A. (1933) Jerome. Select Letters. Translated by F. A. Wright (Loeb Classical Library 262), (Cambridge, Mass. 1933). Wright G. (1957) “Cyrene: A Survey of Certain Rock-Cut Features to the South of the Sanctuary of Apollo”, JHS 77 (1957) 301–310. Wright W. (1921) Philostratus, Eunapius. Lives of the Sophists. Eunapius: Lives of the Philosophers and Sophists. Trans­ lated by Wilmer C. Wright. (Loeb Classical Library 134) (Cambridge, Mass. 1921).

497 Wright W. (1923) Julian. Letters. Epigrams. Against the Galilaeans. Fragments. Translated by Wilmer C. Wright (Loeb Classical Library 157) (Cambridge, Mass. 1923). Yannai E. (2014à) “Mishmar David”, Excavations and Surveys in Israel 126 [Online]. Available: http://www.hadashot-esi.org .il/report_detail_eng.aspx?id=11621&mag_id=121 [Accessed on: 04/05/2015]. Yegül F. (1979) “The small city bath in classical antiquity and a reconstruction study of Lucian’s ‘Baths of Hippias’”, ArchCl 31 (1979) 108–131. Yegül F. (1981) “[Rez.zu]: Die Stabianer Thermen in Pompeji”, AJA 85 (1981) 507–508. Yegül F. (1992) Baths and bathing in classical antiquity (New York and Cambridge, Mass. 1992). Yegül F. (2003) “Cilicia at the crossroads: transformations of baths and bathing culture in the Roman East”, Olba 8 (2003) 55–72. Yegül F. (2010) Bathing in the Roman world (New York and Cambridge 2010). Yegül F. (2013) “Thermal Matters: Intersected Legacies of Greek and Roman Baths and Bathing Culture”, in Greek Baths and Bathing Culture. New discoveries and approaches, edd. S. K. Lucore and M. Trümper (BABesch Supplement 23) (Leuven 2013) 73–88. Yegül F. and Couch T. (2003) “Building a Roman bath for the cameras”, JRA 16 (2003) 153–77. Zajac N. (1999) “The thermae. A policy of public health or personal legitimation?”, in Roman baths and bathing: proceedings of the first International Conference on Roman Baths held at Bath, England, 30 March–4 April 1992, edd. J. DeLaine and D. E. Johnston (JRA Supplementary Series 37) (Portsmouth 1999) 99–105. Zanker P. (1990) The power of images in the age of Augustus (Ann Arbor 1990). Zanotti M. G. (1995) “Ss. Gervasius et Prostasius, titulus”, LTUR 2 (1995) 371. Zanotti M. G. (1996) “S. Maria Antiqua”, LTUR 3 (1996) 214–16. Zanotti M. G. (1999) “S. Prisca, titulus”, LTUR 4 (1999) 162–63. Zehnacker H. and Hallier G. (1965) “Les premiers thermes à Volubilis et la maison à la citerne”, MÉFR 77 (1965) 89–152. Zellinger J. (1928) Bad und Bäder in der altchristlichen Kirche. Eine Studie über Christentum und Antike (Munich 1928). Ziemssen H. (2010) “Roma Auctrix Augusti. Die Veränderungen des römischen Stadtbilds unter Kaiser Maxentius (306–312 n. Chr.)”, in Die antike Stadt im Umbruch: Kolloquium in Darmstadt, 19. Bis 20. Mai 2006, edd. N. Burkhardt and R. Stichel (Wiesbaden 2010) 16–27. Zingboym O. (2011) “Khirbat Ta’ena”, Excavations and Surveys in Israel 123 [Online]. Available: http://www.hadashotesi.org.il/report_detail_eng.aspx?id=1683&mag_id=118 [Accessed on: 05/05/2015]. Zuffa M. (1962) “Nuove scoperte di archeologia e storia riminese”, Studi Romagnoli 13 (1962) 85–132.

Indices Note: The indices only pertain to the discussion, the Gazetteer is not included. Ancient Sources Acta Iohannis 60n326 Alexander of Tralles 63–64, 68 Ambrose 42, 50, 197, 198n88 Ammianus Marcellinus 34–36, 68–69, 183, 192, 210 Ananias of Shirak 67n438, 208 Anonymus Valesianus 61, 218 Antidotarium Bruxellense 61 Anthologia Latina 42, 54, 65n399, 69, 217 Anthologia Palatina 65–66, 69, 198n87 Apophthegmata Patrum 57 Apuleius 36, 66 Asterius of Emessa 49 Athanasius 55n264 Augustine 44n125, 49–50, 54, 56, 68–70, 75n521, 80, 101n79, 155, 170, 186–88, 208 Aulus Gellius 36 Aurelius Victor 41, 83n592 Ausonius 40, 48–51, 69, 183, 195n67, 196–97, 199, 203–204, 206–207 Basil of Caesarea 45–46, 66n419, 68–69, 196 Caelius Aurelianus 52–53, 68, 189, 199, 218 Cassiodorus 58–60, 69, 97, 106, 200–201, 204, 219 Cassius Dio 13 Cassius Felix 53, 61, 65n401, 68, 189, 218 Celsus 29, 68 Chronicle of Edessa 67 Cicero 12, 50n193, 92 Claudian 53 Clemens of Alexandria 75n518 Codex Iustinianus 73, 76, 201n110, 208n148 Codex Theodosianus 71–73, 76, 88n666, 201n113, 203 Colloquia Monacensia 19 Commodianus 42, 47n167, 54, 69 Constitutiones Apostolorum 75 Cosmas Indicopleustes 66, 196 Council in Trullo 75–76 Cyprian 75n518, 150 Digesta 73–75, 80, 89n677, 102, 199n97, 207, 209n144 Epiphanius 43–44 Epitome de Caesaribus 41–42, 61n329, 66n415 Eunapius 48–49, 69–70, 187 Eutropius 41 Eusebius of Caesarea 42–43, 50, 54–56, 69, 210 Evagrius Scholasticus 64, 66n415, 70, 197–98 Faventinus 22n94 Galen 29, 38n47, 40, 44, 68, 79, 195 Gregory of Nazianzus 46, 69–70, 187 Gregory of Nyssa 46–47, 69 Gregory of Tours 60–61, 66n415, 68–69, 195, 204, 218 Gregory the Great 60, 69–70, 187 Herodotos 10, 52 Hieronymus, see Jerome Institutiones 73, 75 Isidorus of Sevilla 65n399, 67 Itinerarium Burdigalense 41, 69, 208 Jerome 38–40, 42, 56, 69, 80, 97, 183, 191, 208 John Cassian 57

John Chrysostom 47–48, 54, 56, 69 John Malalas 61–63, 65n398, 67–70, 170, 184, 186, 188, 197–98, 200, 204 John Moschos 66 John of Ephesus 66–67, 69–70 John of Nikiu 67–70, 200n101, 208 Jordanes 58, 61, 68–70, 218 Joshua Stylites 58, 68, 70 Julian 43–44, 69 Libanius 43, 69, 192 Liber Pontificalis (Rome) 101–102, 107n140, 108n153–54, 121, 138, 205n127, 219 Livy 72, 110 Lucian 19, 30, 32n222 Marcellinus Comes 61 Marcus Aurelius 45 Martial 102, 105n113 Nilus of Ancyra 57 Novellae 73, 75–76 Olympiodorus 57, 97, 183, 197 Optatus of Milevis 49, 69 Oribasius 44–45, 50n200, 53, 68, 88n666, 189, 216 Palladius of Helenopolis 56–57, 69–70, 187, 197, 201 Palladius, R. T. Aemilianus 22n94 Philostorgius 42n95, 61n329, 66n415 Philoxenos of Hierapolis 57–58, 69 Photius 42n95, 57 Pliny the Elder 12, 17, 29, 75n519, 192n46 Pliny the Younger 19, 50, 79, 195 Plutarch 38, 50n193, 68 Polemius Silvius 53, 97, 103, 109 Procopius 64–65, 68–70, 84, 98, 106, 109, 152, 156, 165, 183, 200, 217–18 Pseudo-Theodorus 40–41 Quintilian 76n519 Scriptores Historia Augusta 15n51, 36–38, 42, 49n188, 63, 68–69, 104, 110, 183 Seneca 26, 30, 37n40, 38n47, 51, 69, 209 Sidonius Apollinaris 42n95, 50–52, 61n329, 66n415, 69, 79, 89n676, 97, 189n30, 195n67, 197–98, 201, 206 Socrates Scholasticus 54–55, 68–70, 170, 186–88 Sophronius of Jerusalem 68–69 Sozomen 43n117, 55, 69–70, 187 Statius 30, 40, 105n113, 206 Strabo 52 Suetonius 36–38, 68 Synod of Laodicea 75 Tacitus 29, 34, 70 Tertullian 36, 48n179 Theodoret 55–56, 68–69, 192, 203, 208 Theodorus Priscianus 40–41, 53, 61, 68, 189 Vegetius 42, 63 Venantius Fortunatus 61, 196, 206 Victor of Vita 155 Vitruvius 21–22, 24, 26–27, 207n141 Zacharias of Mytilene 57, 69–70, 200, 204, 208 Zosimus 66

499

Indices Themes Adultery 54, 58, 70, 73, 75–76 Aqueducts in general 26–27, 38, 59, 62, 64–65, 89, 95, 109–110, 112, 130, 135, 143, 145, 148, 150, 155, 165, 200–202 Aqua Alexandrina 97, 109 Aqua Alsietina 109 Aqua Anena 109 Aqua Antoniniana 18, 27, 110 Aqua Appia 92n12 Aqua Claudia 59, 92, 109 Aqua Damnata 109 Aqua Drusia 109 Aqua Iulia 92, 94, 109 Aqua Marcia 18, 27, 92n12, 109 Aqua Mercurii 110 Aqua Virgo 109 Zaghouan aqueduct 150, 155–56 Tributum 73 Vectigal 73, 201 Bathing, other activities Ball game 19, 52, 188, 189n30 Depilation 47, 70, 89, 145, 223 Massage 11, 20, 53, 145 Bathing Personnel Balneator 37–38, 74, 76, 208 Capsarius 60, 74, 76 Gymnasiarch 88 Logistès 208 Servant, see Slave Bathing Utensils Bathing costume 38 Nitrum 19, 45 Oil 11, 19, 44–45, 52–53, 55, 63, 68, 90, 140 Salt 49n188, 72, 88 Situla 74 Soap 19, 45, 63, 88, 90 Strigilis 19 Towel 19, 47, 52, 74, 87–88 Buildings Amphitheatre 35, 92, 107–108, 150, 154–55, 158, 160, 162, 171, 202 Basilica 41, 62, 105–106, 116, 139, 142–43, 148, 155, 170–71 Circus 38, 40, 92, 95, 103, 106–107, 150, 155, 225 Latrine 27, 123, 136, 142, 145, 147, 160, 165, 167, 201 Library 145 Nymphaeum 16, 18, 28, 62, 73, 94, 104, 129, 131–32, 135–36, 138, 160, 170–71, 204, 219 Theatre 35, 38, 40, 46, 64, 94–95, 110–11, 129–30, 132, 143, 148, 150, 155, 158, 160, 162, 164–65, 170–71, 202, 225 Christianity in general 3, 5, 8, 32, 34, 38–40, 42–43, 46–50, 54–57, 60–61, 63, 66–70, 75–77, 101, 106, 108, 119–20, 148, 150, 158, 170, 187, 189, 191–92, 196–98, 204–206, 208, 218 Alousia 39, 44, 56 Ascetism 39, 43, 45–46, 50, 53–57, 69, 75, 191, 208–209 Baptism 32, 36, 44, 47, 49–50, 63, 69, 101, 208–209 Baptistery 41, 57, 92, 101, 108, 109n168, 140, 143, 148, 162, 192, 203–204, 209 Church, building 44, 47–48, 58, 60, 63–64, 70, 72, 86, 97, 98n61, 99, 101–102, 106–109, 120, 130, 132, 140, 143, 148, 155–57, 168, 171, 192, 202–206, 209, 219 Donatism 49, 80, 143, 145, 147–48 Ebionites 44

Gnosticism 43 Hemerobaptism 44 Martyr 42, 45, 47–48, 63, 69, 108, 129, 132, 150, 155, 208 Monastery 38, 57–59, 61, 69, 88, 156, 171, 189, 192, 201, 204 Monophysitism 57 Nestorianism 66 New Testament 39, 57 Novatianism 54 Pilgrimage 3, 41, 70, 102, 132, 157, 185, 192n43, 203–204, 208–209, 217, 219–20 Presbytery 120 Saint 44, 60n326, 68–69, 80, 155, 203, 205, 208 Titulus 102, 107–109, 203 Virgin 39–40, 49, 56, 66, 69, 80, 208 Xenodochium 98, 109 Civic offices Comes 71, 143 Consul 40, 55, 58, 72, 78–79, 86, 92 Corrector 79 Curator 73–74, 78–80, 82–83, 89, 103, 109, 158 Magister militum 71, 79, 85 Legatus 82, 140 Praefectus annonae 78, 124, 131 Praefectus urbi 41, 79, 97–98, 112 Proconsul 82, 84, 154–55 Rector 78 Vir clarrisimus 77, 78n540, 79 Vir perfectissimus 77 Collegium 102, 111–12, 128, 203, 207 Divorce 73, 76 Earthquake 55, 58, 62, 64, 67, 70, 77, 79, 81, 86, 167, 170–71, 203–204, 219 Eunuch 38–39 Flamen perpetuus 82 Forma Urbis Roma 13, 92, 94, 102–104 Fortress 85, 145, 148, 168, 180, 184, 189–90, 197, 200–202, 204, 213, 220 Forum 16, 35, 59, 92, 94, 101–102, 105–107, 110–111, 120, 124–25, 129–132, 139–40, 142–43, 148, 150, 155–58, 165, 170–71, 173, 221 Fuel Bone 200 Chaff 87–89, 200 Charcoal 25, 87 Dung 49, 200 Flax 87, 89, 200 Olive pits 200 Wood 25, 36, 71, 73–74, 76, 87–89, 191n41, 200 Furnace in general 10, 21–27, 32–33, 38, 48–49, 55–56, 58, 66, 73, 76–77, 88, 113, 175–80, 182n553, 184–85, 190, 195–96, 199–201, 209, 223 Boiler 10, 22, 27, 63, 67, 83, 165, 176, 181–82, 190, 199–202, 209, 223 Fire 10, 13, 22, 30, 36, 52, 54–55, 62, 64, 68–69, 73–74, 77, 81, 85–86, 88, 105–106, 120, 150 Chimney 21, 23–25, 178, 199n98 Garden 16, 20, 28, 43, 72–73, 95, 98, 111, 115, 167, 205 Hammam in general 6, 12n33, 25n122, 31–33, 38n50, 50n193, 60n326, 70–71, 140, 175–82, 186, 221, 223–25 Djinn 60n326, 70 Gurn 181, 223 Gusl 70–71 Wudû 70 Judaism in general 8, 29, 42, 44, 47, 54n255, 71n457, 75, 170, 203 Miqveh 42

500 Judaism (cont.) Talmud Yerushalmi 28n161, 54n255, 198, 207n143 Torah 71n457 Justinianic Plague 65, 170, 219 Light in general 16–18, 26, 30, 51, 54, 72, 85, 196 Claustra 198 Lamps 28, 72 Window 19, 26, 30, 31, 51, 58, 77, 87 Magic in general 35, 41, 48, 55, 60, 69, 88 Bulla 88 Defixio 155 Manicheism 44, 49–50, 54, 192 Matronikon 88, 195 Medicine in general 19, 29, 38n47, 40, 44–46, 52–53, 57, 63–64, 66, 68–70, 192 Abortion 41 Asthma 52 Diarrhoea 56 Dropsy 44, 52 Elephantiasis 44 Epilepsy 41, 52 Eyes 44, 46 Fever 53, 63 Gout 41, 52 Headache 52–53 Infection 28n162, 41 Insanity 52 Leprosy 48, 85, 171 Pregnancy 41, 45, 48 Scabies 41, 53 Skin problems 39, 41, 44, 45, 51, 53 Stomach ache 35, 44 Sunburn 53 Vomiting 44 Murder 38, 42, 61, 63, 66, 68, 72 Neoplatonism 48–49 Nika Riot 62 Peoples Amazigh 8, 158, 206 Arabs 12n33, 70–71, 84–85, 157, 162, 165, 168, 170–71, 175–82, 219–24 Austoriani 81 Avars 67, 70 Goths 50, 57, 61, 65, 67, 70, 79, 97, 98, 108–110, 216, 218, 221 Persians 57–58, 60, 64–65, 67, 70 Scythes 10 Vandals 3, 65, 70, 79, 83–86, 106, 140, 143, 154–55, 158, 216–18, 221 Prisoner 72–73, 76 Prostitutes 38, 74n508, 208 Road in general 53–54, 56, 78–79, 82, 92, 123, 129–31, 135–36, 143, 145, 147–48, 150, 154, 158, 162, 202, 204, 217, 219 Via Appia 78–79, 92 Via Clodia 79 Via Domitiana 79 Via Latina 78–79, 92 Via Salaria 78 Via Traiana 79, 202 Sack of Rome AD 410 49, 66, 79, 106, 108, 216 Sack of Rome AD 455 79, 216 Slave 10, 30, 34, 19, 27, 35, 37–38, 47–48, 52, 55–56, 60–61, 69, 73, 75, 192, 207n143

Indices Stable 66, 89 Sweat hut 10, 52 Tetrarchy 3, 112, 140, 148, 158, 160, 183 Thermal Baths in general 2, 11–12, 34, 40, 43, 48, 53, 54n255, 55, 59–61, 65–69, 82, 85, 175, 192, 196, 208, 217 Baiae, see Places Thief 19, 74–76 Venationes 28, 107 Villa 2, 4, 13, 16, 31, 33, 40, 50–52, 61, 84, 94–95, 101n86, 154, 160, 164–65, 186, 188, 202, 204, 206–207 Water Boiler, see Furnace Castellum aquae 27, 138, 142 Cistern 10–11, 18, 27, 56–57, 63, 83, 89, 109–110, 145, 150, 154, 165, 167–68, 175, 200–201 Noria 27, 201 Rainwater 26–27, 95, 113, 145, 165, 168, 170, 191, 201 Reservoir 11, 27, 41, 92, 97, 101, 103–104, 107, 109–110, 113, 122, 150, 155–56, 160, 165, 168, 176, 178–80, 201 River 10, 28, 40, 43, 47, 51–52, 66, 78–79 Saqiyah 89, 179–80, 201 Sea water 49, 208 Well 27, 52, 88–89, 95, 105, 112, 140, 142–43, 150, 165, 167, 180, 200–201 See also Aqueducts Xystus 88n652 Zoroastrianism 58 People Abba Macarius 57 Adrian I, pope 109 Agathias Scholasticus 65 Agrippa 13, 15 Alaric 66, 108, 131 Alexander Severus, emperor 15, 36, 37, 42 Al-Magusi 71 Al-Walid, caliph 180 Amalasuntha 58, 61 Ananias of Shirak 67n438, 208 Anastasius, emperor 57, 61, 63, 168 Antoninus Pius, emperor 36, 67, 101, 111, 113, 115, 119, 124, 170 Arcadius, emperor 61, 71–72, 109, 154, 168 Asclepiades of Bithynia 29 Athalaric, Gothic King 58 Aurelian, emperor 36 Avicenna 71 Avitus, emperor 50 Bathsheba 58 Belisarius 65, 84, 109, 156 Caligula, emperor 37 Caracalla, emperor 36, 38, 94, 126, 145 Caradoc of Llanczarvan 34n1 Carinus, emperor 37 Celsus, tyrant 37 Celsus, physician 29, 68 Christodorus of Coptus 65–66, 198 Clemens of Alexandria 75n518 Commodus, emperor 37–38, 42, 53, 61–63, 111 Constantine, emperor 2–4, 18, 31, 42–43, 55, 61–63, 66, 73, 77, 95, 97, 101, 103, 105–109, 112, 119, 138, 185, 225 Constantius, emperor 15, 35 Constantius II, emperor 54, 77, 103, 106–107

501

Indices Cyprian of Carthage 75n518, 150 Cyril of Alexandria 55 Cyrus 68 Damacharis Scholasticus 65 David, Biblical Jewish king 58 Decius, Emperor 41, 54n258 Diocletian, emperor 3–4, 6, 31, 52, 61, 77, 95, 97, 103–104, 106, 119, 158, 165, 167, 199, 207n138, 210 Domitian, emperor 61, 92, 94, 110, 113 Elagabalus, emperor 36–38, 92 Elisha, prophet 41 Fausta, empress 42, 66 Frontinus 72n480, 109 Gaius, jurist 73 Gelimer 84 Gordian III, emperor 36 Gordian Junior 37 Gratian, emperor 72, 77 Hadrian, emperor 13, 16, 36–37, 61, 64, 73, 94, 99, 101, 110–111, 115, 117, 119–20, 122 Helena, empress 97, 108 Hippocrates 53 Honorius, emperor 71–72, 79, 98, 109, 168, 205 Honorius I, pope 106 Ibn Abd-al Hakam 175 Ibn Asakir 71 Ibn Duqmaq 175 Ibn Rustah 10n4 Innocent I, pope 102 Irenaeus 42 Johannes, St 68 Johannes Grammaticus 66, 198 John, apostle 42, 44 John VII, pope 107 Julius Caesar 61, 66, 92, 94, 150 Justin, emperor 57, 61, 67 Justinian, emperor 8, 61, 64–67, 69, 71, 73, 84, 156, 171, 183, 208 Kawad, Persian king 57–58 Khusro I, Persian king 64, 67 Leontius Scholasticus 65 Licinius, emperor 45 Lucius Verus, emperor 43 Macedonius the Consul 65 Marcus Aurelius, emperor 37, 43, 45, 62, 124, 126, 143, 170 Marianus Scholasticus 65 Maurice, Emperor 60 Maximinus, Emperor 43 Moses, Biblical prophet 47 Nero, emperor 92, 94 Olympiodorus 57, 97, 183 Optatus of Milevis 49, 69 Orata, C. Sergius 12, 29 Origen 42, 46 Palladio 15, 97–98 Paulinus of Nola 53 Paulus Silentiarius 65 Pertinax, emperor 38 Pius I, pope 101 Polemius Silvius 53, 97, 103, 109 Polycarp 42 Pompey 66 Radegunda 61 Severa, empress 54 Silvester, pope 102, 107 Sixtus III, pope 102

Solomon, Jewish king 41, 67 Soranus of Ephesus 52 Tacitus, emperor 36–37 Theodahad 61 Theodora, empress 65–66 Theodosius, emperor 47, 72, 77, 154, 208 Theodosius II, emperor 61–62, 64, 71–73 Theophanes, civil servant in Egypt 87, 208 Thrasamund, Vandal King 84–85 Tiberius, emperor 61, 94, 110 Tiberius II, emperor 66 Trajan, emperor 15, 42, 61, 63, 95, 101, 110–111, 115–17, 143n325 Valens, emperor 35–36, 54, 56, 61, 66, 72, 77, 80, 85, 97, 148, 160, 208 Valentinian I, emperor 35, 65n398, 71n463, 72, 77, 80, 85, 97, 148, 160, 168 Valentinian II, emperor 42, 54, 77, 154 Valentinian III, emperor 72 Victor of Vita 155 Vindicianus 40, 53 Zeno, emperor 61, 73 Places Abbir Maius, Africa 83 Abu Mena, Egypt 185, 203, 208, 217, 220 Abu Sha’ar, Egypt 189 Aelia Capitolina, see Jerusalem Ain En-Ngila, Africa 202 Ain Gharandal, Palestine 189, 197 Aïn Temouchent, Africa 83 al-Bara, Syria 178, 180 Alexandria, Egypt 56, 62, 64, 66–69, 175, 184–86, 188–90, 192, 197–199, 202, 204–205, 213, 216 Allifae, Italy 78 Al-Quds, see Jerusalem Althiburos, Africa 207 Amida, Mesopotamia 57–58 Ammaedara, Africa 82 Amman, Arabia 33, 180–81 Anchiali, Thrace 61, 65–66, 69 Anjar, Phoenica 177 Antinopolis, Egypt 88, 184 Antioch, Syria 35–36, 43, 47, 56, 61–64, 67–68, 72, 75, 184–85, 198, 202n120, 219 Antium, Italy 78 Aphrodisium, Asia 75 Aphroditopolis, Egypt 88 Apollonia, Libya Superior (Cyrenaica) 165, 175, 182, 218 Aponum, Italy 59 Aquileia, Italy 28, 78–79, 183–84, 186, 203, 205–206, 210 Arsinoe, Egypt 86, 89 Aspona, Galatia 56 Athens, Achaea 46, 49 Auzia, Africa 82 Avitacum, Lugdunensis 51 Baiae, Italy 40, 51, 54, 59n311, 69, 82–83, 85, 196, 217 Be’er Shema, Palestine 204 Be’er Sheva, Palestine 204 Belalis Maior, Africa 191 Beneventum, Italy 79 Bir Madhkur, Palestine 189 Bordeaux, Aquitania 41 Bormio, Italy 59n311 Brad, Syria 187–88

502 Brescia, see Brixia Brixia, Italy 216 Bu Njem, see Gholaia Bulla Regia, Africa 81, 145, 202–203 Burdigala, see Bordeaux Busir, Egypt 67 Caesarea Maritima, Palestine 41, 62, 183, 185, 200, 202, 216 Caesena, see Cesena Calama, Africa 82 Canusium, Italy 191 Capsa, Africa 82 Capua, Italy 79 Carthage, Africa 4, 50, 54n255, 72, 82, 84, 150–57, 175, 183, 185, 196, 200, 205–206, 219 Casignana, Italy 206n135 Castellum Tingitanum, Africa 85 Catania, Sicily 203 Cesena, Italy 52 Chalcedon, Pontus 35, 54, 64 Cirta, Africa 82–83 Clysma, Egypt 198 Constantinople, Thrace 35, 43, 46, 54–55, 57, 61–62, 64–66, 68, 72–73, 75, 184–85, 198, 203 Cuicul, Africa 82, 138–143, 192, 202–203, 207–210 Cumae, Italy 79 Cures Sabini, Italy 78 Curinga, Italy 184, 204 Cyrene, Libya Superior (Cyrenaica) 82n578, 183, 185, 191, 196, 202, 218 Damascus, Phoenica 32, 71 Dara, Arabia 57, 62–64 Dharih, Palestine 190n34 Edessa, Augusta Euphratensis 58, 67 El-Burj, Palestine 201 Elusa, Palestine 201 Emmaus, Palestine 55 En-Qobi/Ein el-Qabu, Palestine Erythron, Libya Superior (Cyrenaica) 205 Fanum, Italy 78 Flegrean Fields, Italy 12 Fregellae, Italy 13, 22n94 Fustat, Egypt 175, 220 Gabal Says, Syria 180 Gadara, Palestine 43, 48, 185 Galeata, Italy 186–87, 197, 204–206 Gela, Sicily 10n13 Gerasa, Arabia 188 Gheriat al-Gharbia, Africa 200 Gholaia, Africa 81 Gigthis, Africa 81 Halabiyya-Zenobia, Augusta Euphratensis 178, 188 Hammam as-Sarah, Arabia 33, 180 Ha’on, Palestine 176 Helenopolis, Bithynia 55 Herculaneum, Italy 13, 193, 208n147 Herdonia, Italy 140, 202 Hermopolis Magna, Egypt 86–88, 184, 213 Hippo Diarrhytus, Africa 82–83 Horbat Tarbenet, Palestine 201 Interamna Lirenas, Italy 79 Iol Caesarea, Africa 82–83, 198 Jericho, Palestine 41 Jerusalem, Palestine 41, 176, 184, 186, 189, 192, 196–97, 199, 201, 203–204, 209

Indices Kairouan, Africa 175 Karanis, Egypt 86–87, 182, 197, 199–200, 223 Karm Kandara, Egypt 193 Kerkouane, Africa 202 Khirbet al-Mafjar, Palestine 176 Khirbet al-Maqsurah, Phoenica 180 Khirbet al-Minya, Palestine 176 Khoms, Africa 81 Kysis, Egypt 86–87, 89 Lambaesis, Africa 82, 192, 194 Legio, Arabia 187, 197, 199, 204, 205n128 Leptis Magna, Africa 81, 183, 186, 191, 201–202, 205–206 Limisa, Africa 82–83 Liternum, Italy 79 Loadicea, Asia 62, 75 Ma’ale Adumin, Palestine 189, 192, 201, 203–204, 208, 217 Mactaris, Africa 203–205 Madauros, Africa 83–84, 189, 191, 202–203, 204–205 Mampsis, Palestine 197, 202 Marea, Egypt 185, 198–99, 203, 217, 220 Mechili, Libya Superior (Cyrenaica) 189, 197, 200, 202 Mediolanum, Italy 40, 203–204 Megara Hyblaea, Sicily 10n13 Membressa, Africa 84 Menouthis, Egypt 68 Mesange, Italy 189 Milan, see Mediolanum Misterbianco, Sicily 184 Morgantina, Sicily 10n13, 22n96 Mount Sion, Palestine 41 Mount Syna, Palestine 41 Musti, Africa 82 Mycenae, Achaea 10 Nag el-Hagar, Egypt 197, 204, 213 Nahf, Palestine 176, 201, 204 Nicomedia, Bithynia 55, 62 Nicoterra, Italy 206n135 Oboda, Palestine 197 Ocriculum, Italy 78–79 Oued Athmenia, Africa 185, 204, 206–207 Ouled Hafouz, Africa 196, 209 Oxyrhynchus, Egypt 86–89, 184, 197, 213 Padua, Italy 60 Palazzolo, Italy 204 Panopolis, Egypt 184 Pavia, see Ticinum Pelusium, Egypt 204, 213, 220 Piazza Armerina, Sicily 33, 185–86, 188, 194, 196–97, 204, 206, 207n136 Pompeii, Italy 13, 23, 27, 49n188, 193, 208 Portus, Italy 131, 199 Ptolemais, Libya Superior (Cyrenaica) 85, 163–68, 183, 187, 202, 204–205, 218 Pylos, Achaea 10 Qalandia, Palestine 202 Qasr al-Hayr, Arabia 178, 188 Qinnasrin, Syria 180 Qusair Amra, Arabia 179–80 Ravenna, Italy 59, 77, 80, 91, 143, 183, 195, 202–205, 209, 216, 218–19 Regium Iulium, Italy 78 Rehovot, Palestine 197 Rome, Italy 2, 13, 15–18, 27–31, 33–36, 38, 40–41, 43, 52–53, 57, 59–60, 65, 68, 72, 77–80, 86, 91–110, 143, 183–86, 188, 191, 194–98, 200, 202–205, 207, 209–210, 216, 218–19, 225

503

Indices Rusellae, Italy 78 Sabratha, Africa 81, 194, 202, 217–18, 226 San Giusto, Italy 197, 202, 219 Sbeitla, see Sufetula Scythopolis, Palestine 85, 168–74, 183–86, 188, 192, 204, 206, 216, 221 Serdjilla, Syria 187 Setif, see Sitifis Sicily 10, 13, 186, 195, 197, 202, 206 Sidi Ghrib, Africa 83, 145, 186, 189, 194, 196, 201, 204–207 Sitifis, Africa 175, 190, 219 Smyrna, Asia 65 Spoletium, Italy 78 Spoleto, see Spoletium Sufetula, Africa 82, 157–63, 185–86, 189, 191, 202–203, 209 Sura, Syria 178 Sussex, Britain 26 Syracuse, Sicily 10, 27n150 Tamuda, Africa 200 Taposiris Magna, Egypt 200, 202 Tarentum, Italy 78 Tarquinii, Italy 80 Tarracina, Italy 78 Taucheira, Libya Superior (Cyrenaica) 85, 165, 175, 180, 182, 184, 186, 189–90, 197, 198n90, 199–201, 204, 218–19, 223 Télanissos, Syria 187 Tell Kanaïs, Egypt 202 Tellesia, Italy 78 Thamugadi, Africa 81–82, 143–50, 185–86, 189–90, 192, 197, 202–204, 209–210, 217–18

Thebes, Egypt 184 Thenae, Africa 186 Thuburbo Maius, Africa 82, 186, 189, 191 Thugga, Africa 175, 194, 218 Thysdrus, Africa 82 Tiberias, Palestine 44, 175 Tibur, Italy 16, 73, 207 Ticinum, Italy 61, 79 Timgad, see Thamugadi Tipasa, Africa 101, 186, 192, 196, 202–203, 208–209 Tivoli, see Tibur Trier, Germany 30, 185 Trimithis, Egypt 196, 198 Tripolis, Phoenica 62, 198 Tubactis Municipium, Africa 197 Tunis, Africa 175 Turris Libisonis, Sardinia 203 Umm el-Amr, Palestine 203–204, 208 Venusia, Italy 191, 205 Verona, Italy 61, 77 Vicus Augustanus, Italy 189 Volaterrae, Italy 185–86, 202 Volsinii, Italy 79–80 Volterra, see Volaterrae Walila, Africa 175, 182 Xanxaris, Cappadocia 46 Xois, Egypt 217 Yotvata, Palestine 184, 189, 197, 200, 204

LATE ANTIQUE ARCHAEOLOGY Series Editor LUKE LAVAN Late Antique Archaeology is published annually by Brill, based on papers given at the conference series of the same title, which meets annually in London. Its Supplementary Series aims to publish thematic monographs which address life within the Roman Empire or its successor states in the period AD 283–650, as informed by material evidence, supported by other sources. All publication proposals are subject to satisfying the comments of two anonymous referees, managed at the discretion of the editor. We take the unusual route of accepting PhD theses plus examiner’s reports for review. We then produce a potential publication plan for candidates to satisfy, with an idea of the support we can provide. The editorial committee includes Albrecht Berger, Will Bowden, Kimberly Bowes, Averil Cameron, Beatrice Caseau, James Crow, Jitse Dijkstra, Sauro Gelichi, Jean-Pierre Sodini, Bryan Ward-Perkins, Emanuele Vaccaro and Enrico Zanini. Journal abbreviations follow those used by the American Journal of Archaeology, whilst literary sources are abbreviated according to the Oxford Classical Dictionary (3rd ed. Oxford 1999) xxix–liv and when not given here, following A. H. M. Jones The Later Roman Empire (Oxford 1964) vol. 2, 1462–76, then G. W. H. Lampe A Patristic Greek Lexicon (Oxford 1961). For notes for contributors, with contact details, visit: www.lateantiquearchaeology.wordpress.com For submissions and ordering information visit: www.brill.com/publications/late-antique-archaeology www.brill.com/products/series/late-antique-archaeology-supplementary-series